PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DAIEDUE

DAIEDUE

DAIEDUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:ICIRC/DaIeDue.indd-pt1

 

 

STUDIES ON 15-15’-B-CAROTENE DIOXYGENASE AND REENGINEERING
CELLULAR RETINOIC ACID BINDING PROTEIN H INTO A RETINAL
BINDING PROTEIN AND ITS INTERACTION WITH RETINAL MIMICS

VOLUME I

By

Montserrat Rabago-Smith

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Chemistry

2006

 

ABSTRACT
STUDIES ON 15-15’-B-CAROTENE DIOXYGENASE AND REENGINEERING
CELLULAR RETINOIC ACID BINDING PROTEIN II INTO A RETINAL
BINDING PROTEIN AND ITS INTERACTION WITH RETINAL MIMICS
By
Montserrat Rabago-Smith

Vitamin A and its derivatives are C20 isoprenoids and its metabolic pathway is
tied to the activity of B-Carotene Dioxygenase (BCDOX). Although it has been
known since 1930 that vitamin A is derived in viva from B-carotene, the enzymatic
origin of B-carotene cleavage is not fully understood. The current view is that central
cleavage is clearly the predominant pathway but there is also evidence that excentric
cleavage of carotenoids occurs in plants and microorganisms. Even though in the last
decade a large amount of information has been gathered for BCDOX, many questions
remain unanswered in regard to its mechanism of action. Further proof of whether
the. cleavage of B-carotene occurs via a monooxygenase or dioxygenase mechanism is
needed. Because of the importance of B-carotene as the main source of retinoids in
vivo, and because the oxidation of B—carotene is such a unique example for
regiospecific oxidation, we were interested in studying the mode of control for such
oxidation, as well as how such oxidation occurs for an inactivated olefin. Therefore
the BCDOX protein was expressed in E. coli. When a weak promoter was used the
protein was expressed in low yield as soluble protein. This protein had an activity of
194 pmol retinal/mg BCDOXomin and a KM =3 uM. Due to the low expression yield
the protein was cloned into a plasmid that contained a stronger promoter, but the

protein was only obtained as inclusion bodies. Our interest in the mechanistic aspects

 

of BCDOX led us to synthesize the tritiated photoaffinity label [10’-3H]-8’-ap0—B-
carotenoic acid. In an attempt to develop a partition assay for BCDOX the following
tritiated compounds were prepared trans-[15,15’-3H]-B-carotene, all-trans-[15-3H]-
retinal and all-trans-[15-3H]-retinol were prepared. The mechanism by which we see
colors has been long studied, but even today, it is not fully understood. So far it is
known that there are four proteins in the eye which allow us to see: rod rhodopsin and
three cone rhodopsins, red, blue, and green. Each of the four proteins binds the same
compound, ll-cis-retinal, as a protonated Schiff base (PSB) with an active site lysine
residue. The key to color vision is due to different interactions between ll-cis-retinal
and each one of the four proteins in the eye. Understanding these protein substrate
interactions is the topic of study for many researchers. Various theories have been
developed in an attempt to explain this phenomenon. But due to the difficulties in
expressing, handling, and studying rhodopsin proteins, these studies have been
hampered. To solve this problem a rhodopsin surrogate that is easier to handle,
manipulate and crystallize was engineered. By using rational protein design and
through site directed mutagenesis CRABPII has been converted into a retinal binding
protein. The. ll-cis-retinal was synthesized and binds to the engineered CRABPII
protein as a protonated Schiff base through an engineered lysine residue. Two
different all-trans-retinal analogs were synthesized, and the mode of binding has been
found to be very similar to what is observed in the all-trans-retinal system. The
synthesis of [6-13C]-L-Lysine was performed. This labeled amino acid would be used

to calculate the pKa of the Lysine residue that forms the PSB with retinal.

 

Dedicated to my parents

for their love and unconditional support.

ACKNOWLEDGMENTS

My deepest thanks to all the people that helped me accomplish the completion
of my degree, in all aspects research physically, emotionally, or spiritually.

First I would like to thank Babak, for his mentoring and patience. For the time
that we spent together, I am very grateful for the times when he taught me how to
trust myself. He always pushed me to the end of my limits. He taught me how to
approach the problems that I encountered in my research on my own, and allowed me
to make my own decisions and mistakes, in order to learn from them. Also I
appreciate all his ability to motivate me when times were difficult. Babak has he been
an excellent mentor, and a great friend. He has helped to make my time at MSU a
wonderful and fulfilling experience.

I must also thank Dr. Chrysoula Vasileiou and Dr. Rachael Crist for always
being there, both as friends and excellent co-workers. My progress and level of
understanding for this research have benefited from the intellectual foresight of
Chrysoula and Rachel. I the years that I have been here, they have been a constant
source of wisdom and insight, as well as friendship. Special thanks are due for certain
people for their efforts and assistance in various aspects of my research. I thank Dr.
Qifei Yang, Dr. Radha Narayan, Marina Tasanova and Dr. Meenakshi Sivakumar for
their useful discussions we had covering all aspects of life and science. I also would
like to thank all the members of the lab for their help entertainment and friendship.
Dan, Jennifer, Jun Shang, Sonmath, Stewart, Tao, Xiaofei, Xiaoyong, Ali, Golala. I

thank Dr. Pulgam Veera Reddy for the collaboration in the synthesis of 10-3H-apo

carotenoic acid. I cannot forget the undergraduate students Laura, Sarah and Rocky
for their efforts and enthusiasm both in learning all aspects of the project helping out
wherever possible. Without the collaborative efforts of all these individuals, the
progress of this research would have been severely hampered.

My thanks also goes out to our collaborators on the protein crystallography
efforts, Professor James Geiger (Department of Chemistry, Michigan State
. University) and students Erika Mathes and Soheila Vaezeslami. I also wish to
thankProfessor Honggao Yan (Department of Biochemisty, Michigan State
University) who provided both the instrumentation and helpful staff for aquiring all
mass spectral data. The genomics facility (Michigan State University) provided all
DNA sequencing analysis and the Macromolecular Structure Facility (Department of
Biochemistry, Michigan State University) provided all the primers.

A special mention is due for some close friend who, throughout the years
provided the emotional support necessary for maintaining an acceptable level of
sanity. These people are Chrysoula, Glenn, Mapitso, Rachael and Connie.

I must also thank all of the friends and family who have provided a constant
stream of support. My Mom and Dad have always been there for me their love
encouragement of my dreams. Specifically, I wish to thank my Mom for her
unwavering confidence in me, as well as for her great sense of humor and strength. I
thank my Dad for his example and continuous encouragement to pursue my interests.
My brother also supported me by helping me to keep sight on what is really important
in life. I thank my Grandma and aunt Amelia Smith for all their love and support and

for being proud of me.

vi

Finally, I would like to thank my husband Steve for all the moments we share
during this time We have laughed together, we have cried together and trained
together His patience and understanding have been key overcome all of my
difficulties. His love and interior peace had help me to see the light at the end of the

tunnel. Thank you for all your love and support.

vii

 

TABLE OF CONTENTS

List of Tables ..................................................................................... xi
List of Figures .................................................................................. xiii
List of Schemes .................................................................................. xx
Key to Symbols and Abbreviations ......................................................... xxiii
Part I.

Chapter 1

I. Introduction

1.1 Carotenoids as netabolic source of retinoids ................................. 1

1.2 B—Carotene Dioxygenase ....................................................... 5

1.3 15,15’ B-Carotene Dioxygenase ............................................... 8

1.4 Mechanism of action .......................................................... 13

1.5 References ....................................................................... 23
Chapter 2

II. Expression of BCDOX

2.1 Expression of using a T5 promoter .......................................... 30
2.2 B-Carotene Dioxygenase activity ........................................... 33
2.3 Expression of BCDOX with rare codons ................................... 36
2.4 Expression of BCDOX in vivo ................................................ 43
2.5 Expression of BCDOX using a T7 promoter .............................. 46
2.6 Cloning of BCDOX into a T7 promoter vector ............................ 46
2.7 Expression of pET-29b(+)-BCDOX ........................................ 52
2.8 Optimization of pBCDOX expression ...................................... 52

2.9 Attempts to express the recombinant BCDOX in a soluble form... ....53
2.9.1 Decrease the rate of expression, use of different temperatures, time and

IPT G ............................................................................. 53
2.9.2 Different E. coli strain ......................................................... 56
2.9.3 Over expression of chaperones .............................................. 58
2.10 Activity in vivo ................................................................ 61

viii

 

2.11 Expression of the native BCDOX under the control of a T7

promoter ........................................................................ 63
2.11.1 Mutation of BCDOX to add a stop codon to avoid translation of the

6xHis tag and thrombine cut site ............................................ 63
2.11.2 Cloning BCDOX .............................................................. 64
2.11.3 Expression of pBCDOX-3; native BCDOX ............................... 66
2.12 Isolation of inclusion bodies ................................................. 67
2.13 Materials and Methods ....................................................... 71
2.14 References ....................................................................... 85

Chapter 3

III. Studies on BCDOX

3.1 Introduction .................................................................... 92
3.2 Synthesis of all-trans-[10’-3H]-8’apo-B-carotenoic acid ................ 95
3.3 Enzymatic studies using of all-trans-8’apo-B-carotenoic acid and
BCDOX ........................................................................ 98
3.4 Failed photolabeling attempts using all-trans-S’apo-B-carotenoic
acid and BCDOX ............................................................ 100
3.5 Attempts to develop a rapid and efficient assay for BCDOX ......... 102
3.6 Materials and Methods ...................................................... 108
3.7 References ..................................................................... 125
Part II
Chapter 4

IV Introduction

4.1 Vision .......................................................................... 129
4.1.1 Process of vision ............................................................. 130
4.1.2 Color vision .................................................................. 136
4.2 Designing of a rhodopsin surrogate ........................................ 145
4.3 Wild Type CRABPII binding properties ................................. 149
4.4 References ..................................................................... 162
Chapter 5
V Attempts to label Lys132 to determine the pKa of CRABPII mutants
5.1 Introduction .................................................................. 178
5.2 Non selective labeling of CRABPII ....................................... 183
5.2.1 Use of 15N-NMR to determine the pK. of Lys ........................... 183

 

5.2.2
5.2.3
5.3

5.3.1
5.3.2
5.3.3
5.3.4
5.3.5
5.3.6
5.4

5.4.1
5.4.2
5.5

5.6

Chapter 6

Use of ”C-NMR ............................................................. 184

Synthesis of 6-13C-lysine of vision ........................................ 185
Specific labeling of Lysine ................................................. 195
Chemical ligation ............................................................ 195
Native chemical ligation .................................................... 196
Expression of CRABPH fused to a mini intein .......................... 201
Attempts to solubilize fusion protein ..................................... 202
Refolding of Mxe GyrA intein system .................................... 203
Use different intein .......................................................... 208
Labeling of Lys with e-lsN-Lys or 6-‘3c-Lys ........................... 213
Conversion of Cys to Thio Lys ............................................ 214
Synthesis of the alkylating agent to convert Cys to Thio-Lys. . . . . ....224
Materials and Methods ...................................................... 230
References .................................................................... 252

VI Interaction of ll-cis retina] and other chromophores with the protein
rhodopsin mimics

6.1
6.2
6.3
6.3.1
6.3.2
6.4
6.4.1
6.4.2
6.5
6.6

Chapter 7

Introduction .................................................................. 263
Preparation of mutants to study wavelength regulation ................ 273
Studies using ll—cis-retinal ................................................. 276
Synthesis using ll-cis-retinal ............................................. 277
Protein substrate interactions using ll-cis-retinal ...................... 283
Retinal analogs ............................................................... 288
Synthesis of different chromophores ...................................... 292
Binding studies wit merocyanine 3 and azulene 4 ...................... 300
Materials and Methods ...................................................... 313
References .................................................................... 338

VIIAttempts to develop an assay to identify rhodopsin surrogates

7.1
7.2
7.3

7.4
7.5

 

Introduction .................................................................. 347

In vitro test by using pORANGE, pBCDOX-l and pCRABPII ...... 348

In viva test for protonated PBS formation using pORANGE pRET-

CRABPII ...................................................................... 353

Materials and Methods ...................................................... 362

References .................................................................... 368
X

LIST OF TABLES

Table 1-1.

Table 2-1.

Table 2-2.

Table 2-3.

Table 2-4.

Table 2-5.

Table 2-6.

Table 2-7.

Table 2-8.

Table 2-9.

Table 5-1.

Table 5-2.
Table 5-3.

Table 5-4.

Table 6-1.

Different substrates tested for BCDOX
activity ........................................................................... 20

Comparison of protein activity and expression after addition of
glucose .......................................................................... 35

Protein expression after addition of rare codons using XLl-Blue as
host .............................................................................. 40

Expression of recombinant pBCDOX-2 in Rasetta(DE3)pLysS,
induced at different concentrations of IPTG and different
temperatures .................................................................... 55

Expression of recombinant pBCDOX-Z in Rosetta(DE3)pLysS,
induced by addition of different portions IPTG of OD600~ 1.0 for
overnight at 26 °C .............................................................. 55

Expression of recombinant pBCDOX-2 using different host induced by
addition of IPTG at different portions ....................................... 57

Expression of recombinant BCDOX in the presence of its native
substrate B-carotene .......................................................... 58

Overexpression of chaperones to improve the solubility of the
recombinant BCDOX ......................................................... 60

Attempts to test the recombinant mice B-carotene-lS-IS’ dioxygenase
pBCDOX-2 in viva ............................................................ 62

Expression of recombinant pBCDOX—3 in Rosetta(DE3)pLysS,
induced at different concentrations of IPTG and temperatures... .......66

Attempts to selectively reduce N-CBZ-Glutamic acid OL-methyl

ester ............................................................................ 187
Attempts to hydrogenate nitrile 19 ......................................... 188
Reduction of nitrile using Raney-Ni ...................................... 193
Use of different E. cali. hosts to test solubility of the fusion intein
protein ......................................................................... 202
Spectroscopic characteristics of rhodopsin proteins of different
species ........................................................................ 272
xi

 

Table 6-2.

Table 6-3.

Table 6-4.

Table 6-5.

Table 6-6.

Table 6-7.

Table 7-1.

Table 7-2.

Reduction of alkyne 14 to obtain ll-cis-retinal (1) ..................... 280

Titration of different CRABPII mutants with ll—cis-retinal and all-
trans-retinal ................................................................... 288

HWE between ylide 30 and aldehyde 27 to obtained chloro ester
31 ............................................................................... 295

Spectroscopic data of merocyanine 3 and the bacteriorhodopsin
(th) .......................................................................... 302

Titration of different CRABPII mutants with merocyanine 3 and all-
trans-retinal ................................................................... 305

Titration of different CRABPII mutants with azulene 4 and all-trans-
retinal ........................................................................... 311

Expression of carotenoid genes from E. uredavara in different E. calf.
strain ........................................................................... 353

Results from the attempt to develop an assay CRABPII-mutants that
form PSB with all-trans-retinal ............................................ 358

xii

LIST OF FIGURES

Note:

Figure 1-1.
Figure 1-2.

Figure 1-3.

Figure 1-4.

Figure 1-5.

Figure 2-1.
Figure 2-2.

Figure 2-3.
Figure 2-4.
Figure 2-5.
Figure 2-6.
Figure 2-7.

Figure 2-8.

Figure 2-9.

Figure 2-10.

 

Select figures in this manuscript are presented in color.
Various retinoids ................................................................ 1
Representation of Schiff base formed in rhodopsin ........................ 2

Activation of the GPCR rhodopsin, the G-protein transducin, and the
enzyme PDE as the visual process is initiated .............................. 3

Sequence comparison of human BCDOX (Ii-BCDOX), mouse
BCDOX (M-BCDOX), chicken BCDOX (C-BCDOX), and

Drasaphila BCDOX (D-BCDOX), and human
RPE65 .......................................................................... 10
Different substrates used to evaluate activity in chicken
BCDOX ......................................................................... 19

Depiction of pQE30 plasmid with mice BCDOX gene subcloned in.30
Gel depiction of the purification of His-tagged BCDOX protein ...... 31

HPLC chromatographs of enzymatic reaction of BCDOX and B-
carotene ........................................................................ 34

Gel depiction of the purification of His-tagged BCDOX protein with
impurities ....................................................................... 37

Depiction of the sequence of mice BCDOX gene showing its 29 rare
codons ........................................................................... 38

Comparison of relative activity and expression of recominant mice

BCDOX using different E. coli hosts ....................................... 42
Depiction of pORANGE plasmid containing a gene cluster to express
B-carotene ...................................................................... 43
In vivo activity of recombinant BCDOX ................................... 45

Comparison of the amino acid sequence of the reported mice BCDOX
and the obtained BCDOX ................................................... 48

Depiction of pET-29b(+) plasmid with mice BCDOX gene subcloned
in ................................................................................. 51

xiii

Figure 2-11.

Figure 2.12.

Figure 3-1.

Figure 3-2.

Figure 3-3.

Figure 4-1.

Figure 4-2.

Figure 4-3.

Figure 4-4.

Figure 4-5.

Figure 4-6.

Figure 4-7.

Figure 4-8.

Figure 4-9.

Figure 4-10.

Depiction of pET-29b(+) plasmid with mice BCDOX gene subcloned
in, with out a His tag .......................................................... 65

Gel depiction of the purification of BCDOX protein as inclusion

bodies ............................................................................ 68
Inhibition studies of BCDOX using all-trans-[10’-3H]-8’ap0-B-
carotenoic acid ................................................................ 99
Depiction of a partition assay to test BCDOX activity ................. 103

Comparison of extraction of retinal, retinol and B-carotene using a
partition assay ................................................................ 105

Schematic representation of the human eye .............................. 129

Schematic representation of the rod and cone cells in

retina ........................................................................... 130
Schematic representation of rhodopsin in rod cells ..................... 131
Representation of Schiff base formed in rhodopsin ..................... 132

Schematic representation of the intermediates observed in the
isomerization of 1 l—cis—retinal to all-trans-retinal in
rhodopsin ...................................................................... 133

Schematic representation of the change observed in rhodopsin after the
isomerization of 11-cis-retinal to all-trans-retinal ....................... 134

Activation of the GPCR rhodopsin, the G-protein transducin, and the
enzyme PDE as the visual process is initiated ........................... 135

Schematic .. representation of the visual transduction
pathway ....................................................................... 136

Representation of the visible region of the electromagnetic spectrum
and the rhodopsin absorbance patterns that result in color

vision .......................................................................... 137

Comparison of absorbances of 11-cis-retinal as free aldehyde, as a
Schiff base (SB) and as a protonated Schiff base (PSB) ............... 138

xiv

 

Figure 4-11.

Figure 4.12.

Figure 4-13.

Figure 4-14.

Figure 4.15.

Figure 4-16.

Figure 4-17.
Figure 4-18.

Figure 4.19.

Figure 4-20.

Figure 4.21.

Figure 4-22.

Figure 4-23.

Figure 4-24.

 

Distance of the counter anion and placement of other charges or
dipoles along the backbone of the polyene may be a possible mode of
wavelength regulation ...................................................... 139

Absorption maxima and apparent pKa values of Schiff base (SB) and
protonated SB (PSB) of retinal with various amino acids. The relative
position of the carboxylic acid dictates the 7km and the pK.
values .......................................................................... 141

Twisting of the planes about the double bonds of 11-cis-retinal may be
a possible mode of wavelength regulation ............................... 142

Representation of 11-cis-retinal, this molecule is not planar due to
steric interactions ............................................................. 143

Representation of the crystal structure of CRABPH bound to all-trans-
retinoic acid ................................................................... 148

Representation of interactions between wild type CRABPII and all-

trans-retinoic acid ........................................................... 149
Fluorescence and UV-vis data for CRABPII-WT ....................... 150
Fluorescence and UV-vis data for CRABPII-R132K ................... 152
Fluorescence, UV-vis and MALDI-TOF data for CRABPII-
R132K::Y134F ............................................................... 154
Representation of the crystal structure of CRABPII-R132KzzY134F

bound to all-trans-retinoic acid ............................................. 156

Representation of the conformation of all-trans-retinoic acid bound to

CRABPII-R132KzzY134F ................................................... 157
Fluorescence, UV-vis and MALDI-TOF data for CRABPII-
R132K::R111L::L121E ...................................................... 158
Representation of the crystal structure of CRABPII-
R132K::R1 l lL::L121E bound to all-trans-retinal ....................... 159

Representation of the proposed mechanism for the Schiff base
formation between CRABPII- R132K::R111L::L121E and all-trans-
retinal ........................................................................... 160

XV

Figure 5-1.

Figure 5-2.

Figure 5-3.

Figure 5-4.

Figure 5-5.

Figure 5-6.

Figure 5-7.

Figure 5-8.

Figure 5-9.

Figure 5-10.

Figure 5-11.
Figure 5-12.

Figure 5-13.

Figure 5-14.

Figure 5-15.

Figure 6-1.

Comparison of UV-vis spectrum of CRABPII-WT, CRABPII-
R132K::R111L::L121E and CRABPII-R132K::Y134F::R111L::
L1121E ........................................................................ 178

Acid/base titration of the triple mutant CRABPII-
R132K::R111L::L121E ..................................................... 180

Schematic representation of the intein-mediated native chemical
ligation process ............................................................... 197

Schematic representation of the formation of 15N-Lys132 CRABPII via
intein-mediated native chemical ligation ................................. 199

Gel depiction of the over expression of the small intein fusion

CRABPII protein ............................................................... 201
Gel depiction of the small intein fusion CRABPII protein after
dialysis .......................................................................... 205
Gel depiction of the over expression of the large intein fusion
CRABPII protein .............................................................. 210
Gel depiction of the binding of over expression of the large intein
fusion CRABPII protein to chitin beads ................................... 211
Gel depiction of the cleavage with MESNA and DTT of large intein
fusion CRABPII protein ...................................................... 212
Gel depiction of the cleavage with MESNA and DTT of large intein
fusion CRABPII protein ...................................................... 213
Comparison of lysine and thiolysine ........................................ 214

Representation of the three Cys in CRABPII216

UV-vis data for the titration of CRABPII-R132C::Rl11L::L121E and
all—trans-retinal ................................................................. 218

UV-vis data for the titration of CRABPII-R132C::R111L::L121E and
all-trans-retinal ................................................................. 221

Fluorescence, UV-vis and MALDI-TOF data for CRABPII-
R132C::R111L::L121E after alkylation ................................... 223

Representation of ll-cis-retinal and all-trans-retinal ................... 263

xvi

Figure 6-2.
Figure 6-3.
Figure 6-4.

Figure 6-5.

Figure 6-6.
Figure 6-7.

Figure 6-8.

Figure 6-9.

Figure 6-10.

Figure 6-11.

Figure 6-12.

Figure 6-13.

Figure 6-14.

Figure 6-15.

Figure 6-16.

Figure 6-17.

 

Binding pocket of the rod rhodopsin from its crystal structure showing
the PSB between the ll-cis-retinal and Lys296, the important
interactions between amino acids have been labeled .................... 266

Binding site of ion pump bacteriorhodopsin .............................. 270

Representation of the important amino acids in the crystal structures of
retinal binding site of bacteriorhodopsin ................................... 271

Distance of the counter anion and placement of other charges or
dipoles along the backbone of the polyene may be a possible mode of
wavelength regulation ...................................................... 273

Representation of the amino acid residues within 7 of the polyene in

the triple mutant CRABPII-R132KzzR11lL::L121E .................... 274
Representation of 11-cis-retinal, this molecule is not planar due to
steric interactions ............................................................. 277
Separation of alkyne precursors of 11-cis-retinal ....................... 279
UV-vis spectra of 11-cis-retinal ............................................ 282

Representation of two possible conformation of ll-cis-retinal, in the
triple mutant CRABPII-R132K::R111L::L121E ........................ 284

Comparison of absorbances of ll-cis-retinal as free aldehyde, as a
Schiff base (SB) and as a protonated Schiff base (PSB). . . . . . ..........285

Model structure of the binding cavity of the triple mutant CRABPII-
R132K: :R1 1 lL::L121E and the mutated amino acids 1 l-cis-
retinal .......................................................................... 286

UV-vis absorbances of the ligand ll-cr’s-retinal with different
rhodopsin surrogates ........................................................ 287

Acid/base titration of the triple mutant CRABPII-
R132K::L121E ............................................................... 287

Representation of merocyanines that form a very stable PSB due to the
fact that the positive charge is stabilized all around the polyene. . . ...29l

Representation of azulene and its remarkable polarizability. . . . ....291

Representation of the stabilization of positive charges in azulenes..292

xvii

Figure 6-18.

Figure 6-19.

Figure 6-20.

Figure 6-21.

Figure 6-22.

Figure 6-23.

Figure 6-24.

Figure 6-25.

Figure 6-26.

Figure 7-1.

Figure 7-2.

Figure 7-3.

Figure 7-4.
Figure 7-5.

Figure 7-6.

 

Different aldehydes commercially available ............................. 297

Fill in model of binding of merocyanine 3 to the triple mutant

CRABPII-R132K::R111L:L121E ........................................ 301

UV-vis data for the titration of merocyanine 3 with the triple mutant
CRABPII-R132K::R111L::L121E ......................................... 303
Comparison of UV-vis data for the titration of merocyanine 3 with
CRABPII-R132K::Rl 1 lL::L121E, CRABPII-
R132K::R111L::L121E::R59W, CRABPII-WT ......................... 304

Comparison of UV-vis data for the titration of merocyanine 3 with
CRABPII-R132K::R11lL::L121E, CRABPII-R132K::Rl l lL::L121Q,

CRABPII-R132KzzR1 1 1L ................................................... 306
UV-vis data for the titration of merocyanine 3 with CRABPH-
R132K::Rl11L::L121E::R59W ............................................ 307
Fill in model of binding of azulene 4 to the triple mutant CRABPII—
R132K::R111L::L121E ...................................................... 308
Comparison of UV-vis data for the titration of azulene 4 with
CRABPII-R132K::R111L::L121E, CRABPII-R132Kzle 1 1L
CRABPII-WT .................................................................. 309

Acid / base titration of the triple mutant CRABPH-
R132K::R111L::L121E bound to azulene 4 .............................. 310

Representation of different position in CRABPH to possibly insert the
nucleophilic lysine ............................................................ 347

Absorption differences between retinal and the PSB of
retinal .......................................................................... 347

Depiction of pQE30 plasmid with mice BCDOX gene subcloned inand
a pORANGE plasmid that contains the genes to express B-carotene in

viva ............................................................................ 348
Production of retinal in vivo in )CLl-Blue E. coli ....................... 349
Production of retinal in viva in different E. coli .......................... 350

Representation of an assay to identify mutants that form a PSB
between retina] and CRABPII mutants .................................... 351

xviii

Figure 7-7.

Figure 7-8.

mania

nymrm.

Depiction of pET17b plasmid with CRABPII gene subcloned in. ...352

Fluorescence, UV-vis data for CRABPII-
R132K::Y134F::R11lL::L121E and CRABPII-WT ................... 355

Depiction of pQE30 plasmid with CRABPII gene and mice BCDOX
subcloned in ................................................................... 357

Representation of an E. cali system to assay CRABPII mutants that
form PSB with retinal ....................................................... 357

xix

LIST OF SCHEMES

Scheme 1-1.

Scheme 1-2.

Scheme 1-3.

Scheme 14.
Scheme 1-5.
Scheme 1-6.
Scheme 1-7.

Scheme 2-1.

Scheme 2-2.

Scheme 3-1.

Scheme 3-2.

Scheme 3-3.

Scheme 34.
Scheme 3-5.

Scheme 3-6.

Scheme 3-7.

Scheme 5-1.

Biosynthesis of retinoids, B-carotene oxidation provides retinoids ...... 4

B-Carotene oxidation by BCDOX. Central cleavage (15-15’ oxidation)
vs excentric cleavage ........................................................... 6
9-cis-Epoxycarotenoid . dioxygenase VP14, excentric oxidative
cleavage .......................................................................... 7
General mechanism for activation of 02 in dioxygenases ............... 14
Carotene dioxygenase mechanism .......................................... 15
Carotene monooxygenase mechanism ...................................... 17
Alternate carotene dioxygenase mechanism ............................... 18
Kinetics of the recombinant B-carotene 15-15’ dioxygenase.
Monitoring the conversion of B—carotene to retinal ....................... 35
Biosynthesis of B-carotene in E. cali. starts with the condensation
between DMAPP and IPP .................................................... 44

The enzymatic mechanism of BCDOX is intriguing in that utilizes a
molecule of oxygen to cleave an electronically undistinguished double
bond with a high degree of selectivity ...................................... 92

Postulated mechanisms. A) Dioxygenase, suggest the involvement of
a dioxetane intermediate. B) Monoxygenase, suggest the involvement

of an epoxide intermediate ................................................... 93
NaB3H4 was chosen to be the source of tritium and thus required a
reductive step to incorporate the label .................... . ................... 95

Synthesis of all-trans-[lO’-3I-I]-8’apa-B-carotenoic acid. Part 1.. . .....96
Synthesis of all-trans-[10’-3H]-8’apa-B-carotenoic acid. Part II. ......97

The proposed mechanism for cross linking of orb-unsaturated acids
involves a diradical intermediate ........................................... 101

Synthesis of 3H labeled B-carotene and retinal .......................... 104

In the formation of a protonated Schiff base various equilibrium are
involved ........................................................................ 177

XX

Scheme 5-2.
Scheme 5-3.
Scheme 54.
Scheme 5-5.
Scheme 5-6.
Scheme 5-7.
Scheme 5-8.

Scheme 5-9.

Scheme 5-10.

Scheme 5-11.

Scheme 5-12.

Scheme 6-1.

Scheme 62.

Scheme 6-3.
Scheme 64.
Scheme 6-5.
Scheme 6-6.
Scheme 6-7.
Scheme 6-8.

Scheme 6-9.

Retrosynthetic analysis for the preparation of [6-13C]-L-Lysine ...... 185
Synthesis of N-CBZ-Glutamic acid or-methyl ester 12 ................. 186

Selective reduction of N-CBZ-Glutamic acid or-methyl ester 12.. . ..189

Preparation of nitrile 18 .................................................... 190
Proposed mechanism for the hydrogenation of nitriles ................. 191
Reduction of nitrile using Raney-Ni, Pd/C ............................... 194
Conversion of Cys to Thiolysine via alkylation of cysteine 25 ....... 215
Formation of a PSB between thiolysine and all-trans-retinal. . . . . .....220

The label ethanolamine could be used to prepare iodoethyl-
trifluoroacetamide. and bromo ethyl amine ............................... 225

Synthesis of alkylating agents using ethanol amine as the starting
material ........................................................................ 227

Possible mechanisms involved in the bromination of the label
ethanolamine ................................................................... 228

Synthesis of acetyline (9) intermediate in the synthesis of ll-cis-

retinal ........................................................................... 277
Synthesis of acetyline (14) precursor of the Zn mediated reduction
critical step in the synthesis of ll-cis-retinal ............................. 277
Retrosynthesis of ylide 22 ................................................... 293
Synthesis of aldehyde 27 .................................................... 293

Synthesis of ylide 30 starting from 3,3, dimethyl acrylic acid (24)....294

Synthesis of ylide 33 .......................................................... 296

Synthesis of various aldehydes .............................................. 297

HWE using ylide 33 and various aldehydes ............................... 298

Synthesis of merocyanine 3 ................................................... 299
xxi

Scheme 6-10. Synthesis of azulene 4 .......................................................... 300

xxii

KEY TO SYMBOLS AND ABBREVIATIONS

Abs, A

ALBP

Amino Acids:

Ala, A
Arg, R
Asn, N
Asp, D
Cys, C
Gln, Q
Glu, E
His, H
Leu, L
Lys, K
Phe, F

Pro,P

Angstrom

Extinction coefficient
Micro grams

Micro liters

Micro molar
Maximal wavelength
Absorbance

Adipocyte Lipid Binding Protein

Alanine
Arginine
Asparagine
Aspartic acid
Cysteine
Glutarnine
Glutamic acid
Histidine
Leucine
Lysine
Phenylalanine

Proline

xxiii

Thr, T
Trp, W
Tyr, Y

Val, V

BCDOX
th

BSA

bp

br

CAP

CD
CDNA
C-BCDOX
cGMP
CIP

Clrn
CRABPI
CRABPII

CRBP

Threonine

Tryptophan

Tyrosine

Valine

Ampicillin

Adenine Monophosphate
[fl-Carotene dioxygenase
Bacteriorhodopsin

Bovine Serum Albumin

Base pairs

Broad

Catabolite activation

Circular Dichroism
Deoxyribonucleic acid

Chicken B—Carotene dioxygenase
Cyclic-Guanidine Monophosphate
Calf Intestine Phosphatase
Chloramphenicol

Human Cellular Retinoic Acid Binding Protein I
Human Cellular Retinoic Acid Binding Protein II
Cellular Retinol-Binding Proteins
Doublet

Daltons

xxiv

D-BCDOX
DCM
DIBAL
DMAPP
DMF
DMSO
DTT

DNA

E. cali
EDTA
E120
EtOAc
GDP
GFP
GMP
GTP

GPCR

H—BCDOX
Hex

HMQC
I-IPLC

Drasaphila B-Carotene dioxygenase
Dichloromethane

Diisobutyl aluminum hydride
Dimethyl allyl pyrophosphate
Dimethyl formamide

Dimethyl sulfoxide
dithiothreitol

Deoxyribonucleic acid

trans

Escherichia cali
Ethylenediaminetetra acetic acid
Eher

Ethyl acetate

Guanidine Diphosphate

Green Fluorescent Protein
Guanidine Monophosphate
Guanidine Triphosphate
G-Protein Coupled Receptor
Hour

Human B-Carotene dioxygenase
Hexanes

Heteronuclear Multiple Quantum Coherence

High pressure liquid chromatography

HSPS
HWE
Hz

iLBPs

heat shock proteins
Homer-Wadsworth-Emmons
Hertz

Intracellular lipid-binding proteins

1MPAC'rTM-CN System Intein Mediated Purification with an Affinity Chitin-

IPP

IUPAC

J

KDa

Kb

Kd

KM

LB

LPA

m
MALDI-TOF
M-BCDOX
MESNA
MeOH

mL

MHz

binding Tag, New England Biolabs)
Isopropylthiogalactoside
Isopentyl pyrophosphate
International Union of Pure and Applied Chemists
J coupling constant
Kilo daltons
Kilo-Base pairs
Dissociation canstant
Michaelis Menten constant
Luria bertani
Lipid-protein aggregate
Multiplet
Matrix assisted laser desorption ionization — Time of flight
Mice B-Carotene dioxygenase
2-mercaptoethanesulfonic acid
Methanol
Mililiters

Mega hertz

xxvi

Min
MRNA
MSU
mCi
N-CBZ

nrn

ON
Op
PBS
PCC
PCR
PDE
pmol
PSB
RA

RARS

RPEG5

Rt

Minute

Messenger ribonucleic acid
Michigan State University
Mili curies
N-benzyloxycarbonloxy)
Nanometer

Nanomolar
N-methylmercaptoacetamidc
Nuclear Magnetic Resonance
Over night

Opsin

Phosphate buffered saline
Pyridinium chlorochromate
Polymerase chain reaction
Phosphodiesterase

pico moles

Protonated Schiff base
Retinoic acid

Retinoic acid receptors
Serum Retinol-Binding Proteins
Rhodopsin

Retinal Pigment Epithelium

Retinal

xxvii

 

RT

SB
SDS-PAGE
t

THF

TLC

Tris

tRN A
VMAX

WT

Z

Room temperature

Retinoid X receptors

Singlet

Schiff base

Sodium dodecyl sulfate - Polyacrylamide Gel Electrophoresis
Triplet

Tetrahydrofuran

Thin layer chromatography
2-amino-hydroxymethyl-l ,3-propanediol
Transfer ribonucleic acid

Maximum velocity

Wild type

Cis

Mutant proteins are designated by the one letter abbreviation for the wild type amino

acid, followed by the amino acid position number, followed by the one letter

abbreviation for the new mutant amino acid.

Note: Select Figures in this manuscript are presented in color

xxviii

 

Chapter 1

Introduction

1.1 Carotenoids as a metabolic source of retinoids

Carotenoids are C40 isoprenoids with a wide variety of structures and
biological activities. Carotenoid biosynthesis is mostly found in plants, fungi and
several bacterial species?“4 Carotenoids contain a characteristic polyene structure,
which is responsible for light absorption as well as for singlet oxygen quenching

or inactivation of aggressive radicals. More importantly, carotenoids and their

War W0

1 all trans-retinol 2 all trans-retinal

H

3 all trans-retinoic acid

\ \
\ \ \ \
\ \

5 11-ds-retlnal
4 O-cle-retlnoic acid

Figure 1-1. Different retinoids. Retinoids are isoprene derivatives.

 

oxidation products (retinoids) play an essential role in various pathways such as
cellular physiology, visual transduction,5 gene regulatory control, cancer° and
other disease prevention.7'9

Retinoids are a group of molecules derived from four isoprene units joined
head to tail (Figure l.l).‘° Retinol is known as vitamin A. ll-Cis-retinal Sis
essential for vision in the entire animal
kingdom. The visual pigments (rhodopsin) \ \ \

. , \
of animals are composed of a retinal

\NH
chromophore (vitamin A aldehyde) bound to 0
a lysine in the protein (opsin) via a Schiff
base as shown in Figure l-2.5"‘ Light Ham 1.2. A Schiff base Is
induces isomerization of the chromophore, formed between 11.0];{etina'
and 'thus leading to a conformational change and a lysine residue in
activating the visual pigments (rhodopsin). Rhodops in.
Activation of rhodopsin triggers a G protein-coupled receptor cascade that
leads to changes in the permeability of the photoreceptor cell membranes, as
shown in Figure 1-3. It is also known that different physiological concentrations
of retinoic acid induce cell differentiation at the embryonic stage. The retinoic
acid receptors (RARS) and retinoid X receptors (RXRS) are members of the
steroid receptor super family of proteins, which function as ligand dependent

transcription factors.12 All trans-retinoic acid 3 and 9-cis-retinoic acid 4 can

dietate signal transduction pathways in viva, which is key in controlling

 

 

GTP

GDP
Figure 1-3. The activation of G-proteln transducin

   

upon binding to the activated rhodopsin (R') Active PDE

catalyzes a series of events that triggers a series of
changes (in ion concentrations in the cell resulting
in electrical signals being sent to the brain)
resulting activation of phosphodiesterase that

catalyzes the hydrolysis of cyclic GMP to GMP.

transcriptional activity of RARs and RXRs. Therefore, the concentration of
retinoic acid in cells most be tightly controlled.

It has been suggested that the concentration of retinoids in the cell is
regulated by a number of specific carrier or binding proteins that belong to the
super-family of intracellular lipid-binding proteins (iLBPs). These proteins
include: Serum Retinol-Binding Proteins (RBP), transporters of retinol from
storage in the liver; Cellular Retinol-Binding Proteins (CRBP), specific carriers of

the alcoholic form of vitamin A 1; Cellular Retinoic Acid-Binding Protein

(CRABP), carriers of retinoic acid; Cellular Retinal-Binding Protein and
Interstitial Retinol Binding Proteins.”l6

Even though retinoids are essential for protein function, animals cannot
biosynthesize them. However, it has been estimated that up to 80% of vitamin A
in mammals is accumulated through the oxidative metabolism of carotenoids. Of
all the known carotenoids 50-60 display pro-vitamin A activity, in particular, B-
carotene (1).”“8 B-Carotene is oxidized to form retinal which can be reoxidized

19-21

to produce retinoic acid (3, 4), or reduced to form retinol 1, (Scheme 1-1).

Therefore, B—carotene is considered to be the inlet for many physiologically
important retinoid compounds, which affect a multitude of biological systems”;

“ B-Carotene seems to be absorbed intact by passive diffusion at the intestinal

\\\\\\\\\
6

i

We
2

\\
\\\
4 \ / \

\ 5 \

O H
H
'WQH \\\\\

1 3

Scheme 1-1. Biosynthesis of retlnolds. B—carotene oxidation provides retlnoids.

 

brush border and it shows no toxicity at high doses.” This suggests that B-
carotene could be used as a safe precursor for the in viva production of retinoids in
mammals. Except vision, retinoic acid (trans-retinoic acid 3 and 9-cis-retinoic
acid 4) exerts most of the physiological functions. It has been suggested that
oxidation of retinal is the most important metabolic pathway for the production of

these retinoic acids.”21

1.2 B-Carotene Dioxygenase

Vitamin A and its derivatives are C20 isoprenoids and its metabolic pathway is
tied to the activity of B-Carotene Dioxygenase (BCDOX). Although it has been
known since 193028 that vitamin A is derived in viva from B-carotene, the
enzymatic origin of B-carotene cleavage was only shown in 1965 when Goodman
and Olson were able to demonstrate that cell free enzymatic preparations isolated
from rat intestinal mucus convert B-carotene to retinal.”30 Up to the mid 1960s,
the central cleavage of B-carotene to two molecules of vitamin A (aldehyde) was
the most popular pathway for the oxidation of B-carotene to afford two molecules
of retinal (Scheme 1-2).29'3o In vitra studies have demonstrated a stoichiometric
formation of two moles of retinal per mole of B—carotene cleaved, suggesting a 15-
15 ’C oxidative cleavage.3 1'32

However, it was not until the nineties that Wang et al. provided evidence for
in viva and in vitro production of apo-carotenals from B-carotene. These results

supported a mechanism previously proposed by Glover,33 involving the excentric

cleavage of B-carotene to ape-carotenoids. Thus, a renewed interest in the

\\\\\7§'/‘W]/XJQ
\—w-—’

Exoentrlc
\ cleavage

Central cleavage

15—15‘ oxidation
\ \ \ \ \ \ \ \
+
ill 0\

\\\\\o \\\\‘0

Scheme 1-2. B-Carotene oxidation by BCDOX. Central cleavage (15-

15' oxidation) vs excentric cleavage.

stepwise cleavage of B-carotene to one molecule of vitamin A has recently
resurged, (Scheme l-2).3"32'3"38 The current view is that central cleavage is
clearly the predominant pathway in healthy animals and human subjects}1
although stepwise cleavage is enhanced in intestinal tissue by oxidative
str'ess,35'39"° and there is also evidence that excentric cleavage of carotenoids
occurs in plants and r'l'ricroorganisms.“’42

In plants, a protein called VP14 catalyzes the oxidative cleavage of 9-cis-
violaxanthin, a molecule similar to B-carotene as part of the biosynthetic pathway
of the plant hormone abscisic acid as revealed in Scheme 13.4”3 VP14 was the
first carotenoid dioxygenase that was molecularly identified; and is an example of

an excentric carotenoid cleavage. There are several other examples for excentric

Cleavage reactions, such as the formation of saffron“ in crocus or apa-carotenals

in citrus fruits;
however, whether or
not this process
occurs in mammals
remains to be seen.
Dimitrovskii has
reported the presence
of an apa-carotene
dioxygenase that
cleaves apo-
carotenoids to B-apa-
14’-carotenal. In

2001, von Lintig

  

HO

  

9-cis-violaxanthin

VP14 0“
o\ \
\
W0
+
H0
ng-epoxy-apo-aldehyde H
11 xanthoxin
. \ COOH
OH
O
Abelslc acid (ABA)

Scheme 1-3. 9-cis-Epoxycarotenold dioxygenase VP14.

excentric oxidative cleavage.

reported the discovery of a B-carotene dioxygenase that cleaves at the 9’-10’

double bond of B-carotene, resulting in the formation of B-apo—lO’carotenal and

B-ionone.“

The fact that different oxidation products are obtained with different enzyme

preparations suggest that the presence of the observed apa-carotenoids could be

due to the presence of an B—carotene dioxygenase isozyme that cleaves B-carotene

at a different double bond.46 Therefore, the enzyme responsible of the central

cleavage of B-carotene to retinal was called 15-15’-B-carotene dioxygenase (15-

l 5 ’-BCDOX).

1.3 15,15’-BoCarotene Dioxygenase

The past few years have witnessed a resurgence of activity in the study of
BCDOX, mainly as a result of the identification and isolation of the gene
responsible for BCDOX in a number of organisms.“"“5'47"9 Von Lintig and co-
workers expressed the BCDOX from Drasophila melanogaster into the genetic
background of an E. cali that accumulates [3-carotene,so which resulted in the
formation of retinoids that could be monitored spectrosc0pically.” The CDNA
obtained encodes a protein of 620 amino acids with an estimated molecular mass
of 70 kDa and an apparent KM of 5 uM. Blaner and colleagues used the cDNA
reported by Von Linting to obtain a cDNA encoding a mouse BCDOX and found
a protein of 566 amino acids with an estimated molecular mass of 64 kDa and an
apparent KM of 0.95 M. The human BCODX was characterized and purified by
Andersson and Lindqvistis. They purified the protein from baculovirus-infected
Spadoptera frugiperda 9 insect cells. The human BCDOX had an estimated
molecular mass of 62 KDa and an apparent KM of 7 M." This protein expresses
highly in the Retinal Pigment Epithelium (RPE65), kidney, intestine, liver, brain,
stomach and testis. The gene is ~20 Kbp and is composed of 11 exons and 10
introns that maps to chromosome 16, q2l-q23. The open reading frame of human
BCDOX contains 547 amino acids. It is interesting to notice that the position of
the BCDOX gene is close to the 8852 locus for Bardet Bield syndrome (an
autosomal recessive disease associated with pigrnentary retinopathy, mental
retardation, polydactylity, obesity and hypogenitalims). Andersson reported that

CXpression of human BCODX in bacteria was problematic, thus they used insect

cells to express BCODX. All the previous reports suggest BCDOX to be a
monomer, however their observations suggest that the active BCDOX was a
tetramer. Although this is the first evidence that BCDOX might not be a
monomer, the expression of BCDOX as a tetramer could be an artifact of
expression in insects, or it could be just simply due to a different form of the
protein in different organisms.

Many other BCDOX’s have been isolated and cloned, and so far all the
reported proteins verify that BCDOX’s are cytosolic proteins that have an open
reading frame (ORF) of about 1.7 to 3.1 Kbp, 544~620 amino acids and a
molecular weight of 62-70 I<Da.‘""‘5""“‘9 So far all the reported proteins are
monomers with the exception of human BCDOX, which as mentioned previously
was suggested to be a tetramer. The enzymes are dependant on molecular 02 and
iron cation; the pH for optimum activity is slightly alkaline (7.8-8.2). It has been
reported that this enzyme is inhibited by sulfhydryl alkylating reagents, and it is
either protected or activated by the presence of thiols such as cysteine or
glutathione.29 The Michaelis-Menten constant for the native BCDOX enzymes of

a variety of animal species relative to B-carotene has been estimated to be in the

range of 0.95-10 M. It has also been suggested that BCDOX might have an
amphiphilic character; some reports suggest that this enzyme is associated to a
high-molecular weight lipid-protein aggregate (LPA), which might be necessary
for optimal activity.’2

In chickens, mice and humans, it has been established that the mRNA for

BCDOX was localized primarily in duodenal viles as well as in tubular structures

H-BCDOX MDI-IFG---
M-BCDOX .E.-...-—-
C-BCDOX .ET-..N---
D-BCDOX .AAGV.KSFM
H-RPESS S.-QVE--—
H-BCDOX ----EPVRAK
M-BCDOX -—--...Q..
C-BCDOX ----..IK.E
D-BCDOX REIVD.IEGH
H-RPESS ----S.LT.H
H-BCDOX FTIRDGEVYY
H-BCDOX .S......F.
C-BCDOX .TFKN .....
D-BCDOX .A..N.R.T.
H-RPE65 .DFKE.H.T.
H-BCDOX FSYLSHTIPD
H-BCDOX . ........
C-BCDOX ......... E
D-BCDOX .AAIFRPDSG
H-RPEGS ...F-RGV-E
H-BCDOX YRKYVAVNLA
M-BCDOX ..........
C-BCDOX .S ........
D—BCDOX 'I'I‘DF .G . VNI-l
H-RPE65 LCN..S G

H-BCDOX KQGKSPWKHT
M-BCDOX .K....V..A
C-BCDOX .K...CF..L
D-BCDOX --.---FEDA
H-RPEGS --.-D.ISKS
H-BCDOX MATAY-IRRM
M-BCDOX .-M.GV
C-BCDOX L....-..GV
D-BCDOX YIK.Q-LGGQ
H—RPEGS FLSSWSLWGA
Fleur. 1-4A.

Sequence comparison

.DFFAVKYDE
HPAGGYK---

VTGKIPAWLQ

0000000000

eeeeeeeeee

 

H. VATL. C. W
.IVVQF.CSD

SWASCLAFHR
..... MS.D.
N ..... S.HK
NLSA..KWFE
NYMD.FESNE

QRNDPQA.R.
-.LFETV.E.

 
 

#11:}! EM
:Ve: '51.?“- e .913?-
S, . .'...r.,,S
$914. until-t

NTNIEANRIV

PF.DQY.TFT
PV...Y..CT

NVLNMGTSIV

eeeeeee Q‘,+

~Q. ---4.'-‘a6.
x. H. 6. Mad“

arx...vgg.

EEKTYIHIID

DGNGRLYPNC
s .........

HTVGESRYNH
...... K. .
I. DTK.
WK. .DMTFG.
FE. .SEPFY.

VSEFGTMAYP

..PFMHR...
O .FOT .....

EKGKTKYVIF

:oaevogefiaoo

0‘: Ker. VI. '
-ra F3??? IV

aiasggxa V

QRTRQPVQTK

..K..P..
RK.KKE.S..
RVSGKL...-
KKRKKYLNN.

  

.......

..........

KIPATVPEGK
....... DS.
...SS...KE
SF.--HG.QM
...PLQADKE

EQPFRLDILK
IOOKUOIOO

..... L. mi,
vase. .FFYLfi
YR. 59m.”

13

13

50
23

109
109
109
149
119

159
159
159
198
167

209
209
209
245
217

259
259
258
289
263

308
308
307
337
313

Sequence comparison of human BCDOX (H-BCDOX), mouse

BCDOX (M-BCDOX), chicken BCDOX (C-BCDOX), and Drosophlla BCDOX (D-

BCDOX), and human RPE65. The most conserved regions are shaded and

labeled A-B. Part I.

using GeneWorks 2.5.1 (Oxford Scientific).2

10

These homology was generated by Zack and colleagues

 

 

350
350
349
387
359

400
400
398
437
408

440
440
438
487
452

485
485
483
537
498

532
533
526
587
533

547
566

H-BCDOX _\ ‘fiEDGC IVFDVIAYED -NSLYQLFY— -LANL—— —NQD an---NSRL
u-acnox aQFFéF.-. VL ........ -s ....... - - . -- .x. .a.-—-x...
c-acuox as; ea.. H v. .Iv. .R. -N. .DM. .- - .KK. --DK. .EV---.NK.
D—BCDOX theeraa. a v. v. Ics. RN PEMINCMYLE AI. MQT PN YATLFRGRP.
H-RPEGS eggggabn F LIv. LCCWKG FEFV. NYL. - -. .--REN we.vxx.x.x
n-ecoox TSVPTLRRPA VPLHVDKNAE VCTNLIKVAS TTATALKEED GQVYCQPEFL
M-BCDOX . .................. 0.. s. .v. .s. ...... K. . ...... v.
c-ecnox . I..CK. v .QY. .D. .s. .v. L- P .s...v.. x. .81 ..... I.
D-BCDOX RF LP.GTIP PASIAKRGLV xsrs. AGLSA PQVSRTMKHS vso. ADITYM
n-apass APQ.EV. vv L. NI. - .D r. x. .VTLPN ..... ILcs. ETIWLE. .v.
a-ecoox -YEG—-——L— —ELPRVNY-~ A-HNGKQYRY vrarcvowsp IPTKIIKYDI
M-BCDOX - -——- - -....r..-— .-v...p... I..AE ..... v....L....
c-ecoox -C..--——I- -....v .-- D- Y. .K. x. .v. rs ..... V....A.LNV
o-acnox PTN.KQATAG e s. xabea GRYEEENLVN Lvrnecsoxe AFQGTNGIQL
H-RPE65 —FS.PRQAF. — .F. 01. —- QKYC. .P. r. AYGL. LNHF- V.DRLC.LNV
n—accox LTKSSLKWRE , “ ‘_,,. ILs AIVSTDP QK
M'BCDOX e ...... S. ES... aw,l§_l_:.;5’I‘.-—--.g, ‘lnlde eee ..... e -
c-ecoox Q EV.I-I c .I-I...S. engines? mane. arr. 1' cv. vsa. -N.
D-BCDOX RPEMLCD.GC FTPRIYEEHE? NGKNYRYFYev ESSLHFAVNP GTLIKVDVWN
H-RPESS x .ETWV.Q P. St. Srmxu 8H. -—--D§T‘i planed vv .PGAG. .
a-eccox LPFLLILDAK SFTELARASV DVDMHMDLHG LFITDMDWDT KKQ—-AASE-
M-BCDOX .................... .A...L.... ...P.A..NA V..--TPA.T
c-acoox A ......... r.x..c..r. N.E..L.... M..PQN.LGA are -------
D—BCDOX KSC.TWCEEN VYPSEPIFVP SP.PKSEDD. VILAS.VLGG LNDRYVGLIV
a-RFaes PAY....N.. DLS.V...E. EINIPVTF.. . -------- .s -------
H-BCDOX ---EQRDRAS D --------------- CHGA PLT

n-ecoox QEV.NS.HPT .PTAPELSHS ENDFTAG..G SSL

Figure 1-48. Sequence comparison of human BCDOX (H-BCDOX), mouse

BCDOX (M-BCDOX), chicken BCDOX (C-BCDOX), and Dresophlla BCDOX (D-

BCDOX), and human RPE65. The most conserved regions are shaded and

labeled A-D. Part II.

using GeneWorks 2.5.1 (Oxford Scientific)?

11

These homology was generated by Zack and colleagues

of lungs and kidneys. But its activity has also been found in intestines, brains,
stomachs and testes. Comparison of the deduced amino acid sequences revealed
that humans, chickens, mice and the Drosaphila enzymes share overall sequence
homology with a distinct pattern of highly identical conserved stretches (Figure 1-
4a and l-4b).

Most of the highly conserved regions are located in the terminal section of

‘7'49‘535‘ A sequence alignment performed by Yan and 9011938095

the protein.
between BCDOX from different organisms and human RPE65 reveals four areas
of increased sequence conservation. It is possible that those regions could belong
to the binding site of B—carotene.2 It is also possible that these conserved regions
are associated to the binding site of the iron cation. In particular, region A
contains a very conserved arginine, a couple of histidines and a carboxylate that
could form the binding site for the iron cation. In contrast to the vertebrate
enzyme, the insect enzyme possesses a long extension close to the C-terminus.
The role of this extension is not clear. Also, it is interesting to notice that
alignment Shows that around 8% of the highly hydrophobic amino acids are
conserved in the enzyme, suggesting that these amino acids participate in the
formation of the substrate-binding pocket.51 Besides the homology amongst each
other, the various BCDOX proteins share significant sequence homology to
RPE65. Whether the homology between BCDOX and RPE65 is due to a

conserved carotenoid/retinoid-binding motif or because of a related biochemical

function remains to be investigated.

12

Interestingly, a homology analysis of human, mouse, chicken and fly
BCDOX and the RPE proteins, performed by Cunningham and co workers, shows
that ten residues are absolutely conserved, 4 histidines and 6 acidic residues. It is
also possible that those 4 histidines are involved in the binding of the iron cation.
Also it has been suggested that a conserved region (consensus sequence
EDDGVVLSXVV S) at residues 469-480 of mouse BCDOX, might be considered
a family signature sequence.55 A detail examination of the different domains in
the protein does not provide any obvious evidence for a transmembrane domain,
or any potential site for N-linked glycosilation. This supports the observed results
that BCDOX is a cytosolic protein. However, it does not eliminate the possibility

that it might be a tetramer or a lipid associated protein.

1.4 Mechanism of action

15-15’ B-Carotene Dioxygenase belongs to the enzyme family of
oxygenases. Oxygenases are enzymes that catalyze the incorporation of oxygen
from molecular oxygen (02) into an organic substrate. There are two classes of
oxygenases: dioxygenases in which both oxygen atoms from molecular oxygen
(02) are incorporated into the substrate(s), and monooxygenases in which one
oxygen atom from molecular oxygen (02) is incorporated into the substrate while
the other is reduced to water. As previously mentioned, it has been observed that
BCDOX is an iron dependant enzyme. Oxygenases commonly use iron (Fe) as a
cofactor to activate the molecular oxygen. The iron in oxygenases could be part

of a heme group or a non-heme form binding to some residues in the protein. 15-

I3

15’ B-Carotene Dioxygenase is a nonheme iron—containing enzyme. In most of
the non-heme Feu dependant oxygenases the iron has a five or six coordination
sphere that binds to the molecular oxygen.

Oxidation by dioxygenases first requires the activation of molecular
oxygen. However, to date, the precise mode for the" activation of molecular
oxygen has eluded researchers. Up to now, most researchers agree that first; Fell
exists in a six coordinate state when it is relatively unreactive toward 02. When
the substrate binds, one of the ligands dissociates from Fe", generating a 5
coordinate complex. 02 binds to this complex generating a highly reactive iron-

oxygen intermediate, which could oxidize the substrate; (Scheme l-4).5‘5'57

Substrate
[2+ [2+ 02 3+
II. .r‘ h. .e\ —. I" "‘ ___.. lo I 2‘3-
, Fie‘ ____.L.. ’ FB‘ (3‘ ___-_.———-> 'Fle" +
O, .
raie‘lv‘fi‘e straw”

Scheme 14. General mechanism for activation 0102 in dioxygenases.

The ligands present in non-heme enzymes so far identified are tyrosine,
histidine and glutamate.”58 It is known that BCDOX is a non-heme Fe" oxidase.
However, identification of the amino acids that bind to Fen has not been
accomplished. Therefore, the mechanism of action remains unknown.

Oxidative cleavage of B-carotene is an unprecedented process for a
dioxygenase enzyme. Typically, dioxygenases require a hydrophilic handle such

as a hydroxyl group to proceed.’9 Enzyme/substrate interactions that lead to

Stereospecific catalytic action through hydrogen bonding and dipolar

l4

 

 

communication are common. However, the enzymatic mechanism of BCDOX is
intriguing in that it uses a molecule of oxygen to cleave an electronically
undistinguished double bond with a high degree of selectivity.

It was not until after 40 years of the discovery of carotene dioxygenase in
1965, that Olson and Hayaishi demonstrated that molecular oxygen is
incorporated into vitamin A. When 1802 gas was used, a retinal adduct that
contained an M42 mass peak was observed, but not when H2180 was used. Since
then the B-carotene oxygenase was believed to be a dioxygenase (Scheme l-SA).
Later, Goodman postulated a mechanism that suggested the interrnediacy of a
dioxetane intermediate. This mechanism supposes a 2+2 cycloaddition of

molecular oxygen to the central double bond of B-carotene, followed by a retro

 

al a ‘

I’M W

BCDOX l 02

 

fife/Ker MW

Scheme 1-5. Carotene dioxygenase mechanism. A) Both atoms from molecular

 

 

 

 

 

oxygen ("’02) are added to retinal. 8) Proposed mechanism of oxidation, dioxetane

lnterrnedlate.

15

 

2+2 cleavage of the carbon-carbon bond (Scheme 1-5B). Although this could be a
viable mechanism, cleavage of inactivated olefins to aldehydes by dioxygenases is
unprecedented. Also the postulated dioxetane is a high energy intermediate and
its decomposition is highly exothermic (~70 kcal/mol),60 followed by
chemiluminescence. This is the result of producing a triplet state oxygen, which
upon relaxation emits light. This is the proposed mechanism for bioluminescence
for the firefly luciferase reaction.60

For another 40 years the dioxygenase mechanism was accepted as the true
mode of action for B-carotene oxygenase. It was not until 2001 that Woggon and
co-workers challenged the oxidation mechanism. Woggon provided evidence for
the incorporation of one 180 atom of molecular oxygen and the incorporation of
one ”0 atom from labeled water (Scheme 1-6A). Accordingly, and in contrast to
earlier findings, the reaction mechanism of enzymatic central B-carotene cleavage
does not agree with a dioxygenase catalyzed mechanism. The proposed
mechanism by Goodman would require the incorporation of one complete oxygen
molecule into the product aldehydes and no oxygen originated from labeled water
should be present (Scheme 1-6B). The discrepancy could be due to the different
conditions used in the experiments or the different origins of the BCDOX used.
Olson and Hayaishi used crude enzyme preparations from rat liver and rat
intestine.29 Woggon and colleagues used baby hamster kidney cells.“ Woggon
and colleagues proposed an oxidation mechanism that first involves the
epoxidation of the 15-15’ double bond; they suggest that the oxygen from this

epoxidation comes from molecular oxygen. As a second step, the epoxide formed

16

 

xwy

BCDOX
1802: 17OH2

J\*\/CH19170

fifiw

 

 

 

 

 

Scheme 1-6. Monooxygenase mechanism. A) One atom from

molecular oxygen is added to one retinal ("’02) and one molecule

of water is added to retinal H2‘7O. 8) Proposed mechanism of

oxidation. epoxidation by molecular oxygen follow by addition of a

molecule of water.

afford two molecules of retinal.

17

is attacked by a molecule of water to afford a diol; this diol could fragment to

Alternate mechanisms for oxidation of B-carotene can be envisioned in
Which two molecules of oxygen are used to cleave the olefin. The presence of
Fe2+ can lead to the generation of hydroperoxy radicals that can attack the central
01efinic bond, and via capture of another oxygen molecule, yield a vicinal bis-

h)’<Tlroperoxy intermediate. This intermediate could undergo fragmentation to

 

produce retinal. Similarly, attack of a hydroxyl radical and capture of an oxygen
molecule can result in a hydroxy-hydroperoxy intermediate as shown in Scheme

1-7.

(°” H
3‘wa _. VKXQA ___. 2\ v0 +H20

0940H

Scheme 1-7. Alternate mechanisms. A hydroperoxy radical could be
formed, and attack the central oleiinic bond. Capture of another oxygen
molecule can yield a vicinal bis-hydroperoxy intermediate. This
intermediate could undergo fragmentation to produce retinal. Similarly,
attack of a hydroxyl radical and capture of an oxygen molecule can result in

a hydroxy-hydroperoxy intermediate.

Woggon and co-workers synthesized a few substrate analogues of B-
carotene.I When a single bond was replaced at the lS-lS’C, no activity was
observed, but a slow oxidation of (b) to B—carotene (a) was observed, followed by
oxidative cleavage to retinal (15%). A change in the configuration of the 15-

15 ’C=C bond in B-carotene (trans to cis) or substitution to a single C-C or a CEC
bond yields no product at all. Thus it was concluded that any deviation from the
‘rod like’ B-carotene structure is not tolerated (Figure 1-5 d, e and 0. When the
Positions of the methyl groups were modified (Figure 1-5 a, b and c) a significant
change in the activity was observed. The position and presence of methyl groups

attached to the polyene chain is significant. Also it was evident that at least one

18

-—~—H _ufi A m .1._-—

v—H

 

 

 

 

 

 

Substrate] Yb” 9"

 

 

oi retinal
a 100
c 2
d 15
e O
l O

 

 

 

 

 

 

Figure 1-5. Different substrates used to evaluate activity in chicken BCDOX.
a-c were used to evaluate the importance oi methyl groups. Variation of the

geometry in 15-15'0 d-i was used to evaluate the effect in function of BCDOX.‘

‘natural’, unsubstituted B-ionone ring was required for the oxidative cleavage to
take place. Woggon suggested a hydrophobic barrel-like substrate-binding site, in
which the protein’s amino acid residues interact with the methyl groups to direct
the central double bond (15-15’) to the correct place for optimal distance to the
active site metal-0x0 center, thus supporting the unique observed regiospecificity
in the cleavage of B-carotene to retinal.

Cama and Sing performed a number of substrate specificity experiments

With BCDOX using different apo-carotenals and carotenoids.62 They found that

19

the BCDOX from pig and rabbit showed remarkable similarity with respect to
cleavage of carotenoids. With respect to apo—carotenoids, it was found that 10’-
apo—carotenoid alcohol and aldehyde forms are more readily cleaved, as compared
with other apo-carotenals tested (Table 1-1). However, all of these experiments
were performed using purified protein from intestines that was not completely
pure.

Even though in the last decade a large amount of information has been
gathered for BCDOX, many questions remain unanswered in regard to its
mechanism of action. More specifically, it is not understood how the enzyme

directs the oxidation of an undistinguished double bond In addition, further proof

Table 1-1. Different substrates tested for BCDOX activity.
Guinea pig Rabbit

 

 

 

Substrate relative relative

activity activity

A B-B—carotene 1.00 1.00
B B—s—carotene 0.72 0.56
C 5.8-epoxy-5.6—dihydro- B-B—carotene 0.32 0.25
D 5.8-epoxy-5,8-dlhydro- [HS—carotene 0.19 0.11
E 5.8.5'.8’-diepoxy-5.8,5’8’-tetradihydro- [HS—carotene 0.02 0.02
F 5.6-epoxy-5.6-dihydro~ p—a—carotene 0.25 0.22
G 5.8-epoxy-5.8.dihydro- B—s-carotene 0.09 0.09
H B—B-carotene-a’pl 0.04 0.03
l 3',4’-didehydro-B-B—carotene-S-ol 0.03 0.02
J 8’-apo-B-carotenal 0.09 0.03
K 10'-apo-B—carotenal 0.57 0.55
I. 12’-apo-B—carotenal 0.12 0.07
M 8'-apo-B—caroten-8'-yl acetate 0.16 0.10
N 8'-apo~B—carotenol 0.07 0.02
O 10'-apo-B-carotenol 0.50 0.43
P 12'-rpo-fi—carotenol 0.08 0.07

20

 

R, a, =CHO (J)
\ \ \ \ \ \ \ \ =CH20H(N)
a 0(0)ocn3 (M)

\ \ \ \ \ \ \ R2
R2=CHO (K)
=CH20H(O)
\ \ \ \ \ \ R3
R3=CH0 (L)

= CHZOH (P)

of whether the cleavage of [ii-carotene occurs via a monooxygenase or
dioxygenase mechanism is needed. Because of the importance of B-carotene as
the main source of retinoids in vivo, and because the oxidation of B-carotene is
such a unique example for regiospecific oxidation, we were interested in
understanding the mode of control for such oxidation, as well as how such
oxidation occurs for an inactivated olefin. Finally we were interested in
discovering whether the oxidation mechanism occurred via a dioxygenase or

monooxygenase mechanism.

22

fifl 4f.—

 

 

 

 

21

1.5

References

G. M. Wirtz; C. Bomemann; A. Giger; R. K. Muller“, H. Scheinder; G.
Schlotterbeck; G. Schiefer; W. Woggon, "The substrate specificity of beta
carotene 15-15'monooxygenase." Helvetica Chimica Acta 2001, 84, 2301-
2315.

W. Yan; G.-F. Jang; F. Haeseleer; N. Esumi; J. Chang; K. Palcewski; D.
Zack, "Cloning and characterization of a human beta-carotene." Genomics
MI, 72, 193-202.

G. A. Armstrong, "Genetics of eubacterial carotenoid biosynthesis: A
colorful tale." Annual Review of Microbiology 1997, 51, 629-659.

G. E. Bartley; P. A. Scolnik, "Plant carotenoids: Pigments for
photoprotection, visual attraction, and human health." Plant Cell 1995, 7,
1027-1038.

R. A. Morton, Photochemistry of Vision. ed.; Springer-Verlag: New York,
1972; 'Vol.' p 33.

P. Palozza, "Regulation of cell cycle progression and apoptosis by beta-
carotene in undiferentiated and differentiated HL-60 leukemia cells:
Possible involvement of a redox mechanism." International Journal of
Cancer 2002, 97, 593-600.

L. J. Gudas, In The retinoids: Biology chemistry and medicine. ed.; New
York, 1994; 'Vol.' p 443-520.

A. C. Ross, "Cellular-metabolism and activation of retinoids-roles of
cellular retinoid-binding proteins." FASEB Journal 1993, 7, (2), 317-327.

C. Merlet-Benichou; J. Vilar; M. Lelievre-Pegorier; T. Gilbert, "Role of
retinoids in renal development: pathophysiological implication." Current
Opinion in Nephrology and Hypertension 1999, 8, (1), 39-43.

23

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

T. E. Gndersen; R. Blomhoff, "Qualitative and quantitative liquid
chromatographic determination of natural retinoids in biological samples."
Journal of Chromatography A 2001, 935, 13-43.

L. Stryer, "Vision: From photon to perception." Proceedings of the
National Academy of Sciences of the United States of America 1996, 93,
(2), 557-559.

D. J. Mangelsdor; R. M.Evans, "The RXR heterodimers and orphan
receptors." Cell 1995, 83, (6), 841-850.

D. E. Ong, "A novel retinol-binding protein form rat." The Journal of
Biological Chemistry 1984, 259, (10), 1476-1482.

J. E. Smith; D. S. Goodman, "Retinol-binding protein and the regulation of
vitamin A transport." Federal Proceedings 1979, 38, 2504-2509.

S. Futterman, and Saari, J. C., "Occurrence of 11-cis-retinal-binding
protein restricted to the retina." Investigative Ophthalmology and Visual
Science 1977, 16, 768-771.

Y. L. W. Lai, B., Liu, Y.P., and Chader, G. J., "Interphotoreceptor retinol-
binding proteins: possible transport vehicles between compartments of the
retina." Nature 1982, (298), 848-859.

J. A. Olson; N. I. Krinsky, "Introduction: the colorful, fascinating world of
the carotenoids: important physiologic modulators." FASEB Journal 1995,
9, 1547-1550.

R. S. Parker, "Absorption, metabolism, and transport of carotenoids."
FASEB Journal 1996, 10, 542-551.

A. A. Dmitrovskii, Metabolism of natural retinoids and their functions. In
: New trends in Biological Chemistry. ed.; Japan Science Society: Tokyo,
Japan, 1991; 'Vol.' p 297-308.

J. A. Olson, "Some aspects of vitamin A metabolism." Vitamines and
Hormones 1968, 26, 1-63.

24

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

G. A. J. Pitt, Carotenoids. ed.; Isler, O: Birkhauser Verlag, Basel, 1971;
'Vol.' p 717-742.

T. R. Evans; S. B. Kaye, "Retinoids: present role and future potential."
Cancer 1999, (80), 1-8.

S. A. Kliewer“, J. M. Lehmann; T. M. Willson, "Orphan nuclear receptors:
Shifting endocrinology into reverse." Science 1999, 284, 757-760.

G. M. W. Morris-Kay, S. J., "Retinoids and mammalian development."
International Reviews in Cytolology 1999, 188, 73-131.

M. D. M. Collins, G. E, "Teratology of retinoids." Annual Reviews in
Pharmacology and Toxicology 1999, 39, 399-430.

A. B. Goodman, "Three independent lines of evidence suggest retinoids as
causal to schizophrenia. " Proceedings of the National Academy of
Sciences of the United States of America 1998, 95, 7240-7244.

D. Hollander, Ruble, P.E. Jr, "beta-Carotene intestinal absorption: bile,
fatty acid, pH, and flow rate effects on transport." American Journal of
Physiology 1978, 235, E686-E69l.

T. Moore, "Vitamin A and carotene." Biochemical Journal 1930, 24, 692-
702.

J. A. Olson; O. Hayaishi, "The enzymatic cleavage of beta-carotene into
vitamin A by soluble ezymes of rat liver and intestine." Proceedings of the
National Academy of Sciences of the United States of America 1965, 54,
1364-1370.

D. S. Goodman; H. S. Huang, "Biosynthesis of vitamin A with rat
intestinal enzymes." Science 1965, 149, 879-880.

A. Nagao; A. During; C. Hoshino; J. Terao; J. A. Olson, "Stoichiometric
conversion of all trans-beta-carotene to retinal by pig intestinal extract."
Archives of Biochemistry and Biophysics 1996, 328, (1), 57-63.

25

32.

33.

34.

35.

36.

37.

38.

39.

41.

J. Devery; B. V. Milborrow, "beta-Carotene-15,l5'-dioxygenase (Ec-
1.13.11.21) isolation reaction-mechanism and an improved assay
procedure." British Journal of Nutrition 1994, 72, (3), 397-414.

J. Glover, "The conversion of beta-carotene into vitamin A." Vitamines
and Hormones 1960, 18, 371-386.

M. R. Lakshman; I. Mychkovsky; M. Attlesey, "Enzymatic conversion of
all-trans-beta-carotene to retinal by a cytosolic enzyme from rabbit and rat
intestinal mucosa." Proceedings of the National Academy of Sciences of
the United States of America 1989, 86, 9124-9128.

K. Wang, "The bioconversion of beta-carotene into retinoids." Subcellar
Biochemistry 1998, 30, 159-180.

T. V. Vliet; R. V. Schaik; H. V. D. Berg; W. H. P. Schreurs, "Effect of
vitamin A and beta-carotene intake on dioxygenase activity in rat
intestine." Annals of the New York Academy of Sciences 1993, 691, 220-
222.

A. Yang; R. K. Tume, "Comparison of beta-carotene-splitting activity
isolated from intestinal mucosa of pasture-grazed sheep, goats and cattle."
Biochemistry. Molecular Biology International 1993, 30, 209-217.

G. Scita; G. W. Aponte; G. Wolf, "Uptake and cleavage of beta-carotene
by cultures of rat small intestinal cells and human lung fibroblasts."
Journal of Nutritional Biochemistry 1992, 3, 118-123.

J. Olson, In Modern Nutrition in Health and Disease. 9th ed.; Williams
and Wilkins: Baltimore, MD, 1999; 'Vol.' p 193-221.

J. A. Olson, "The effects of iron and copper status and of dietary
carbohydrates on the activity of rat intestinal beta-carotene 15,15'
dioxygenase." British Journal of Nutrition 2000, 84, 3-4.

8. H. Schwartz; B. C. Tan; D. A. Gage; J. A. D. Zeevaart; D. R. McCarty,
"Specific oxidative cleavage of carotenoids by VP14 of maize." Science
1997, 276, (5320), 1872-1874.

26

42.

43.

45.

47.

48.

49.

50.

S. H. Schwartz; X. Qin; J. A. D. Zeevaart, "Characterization of a novel
carotenoid cleavage dioxygenase from plants." The Journal of Biological
Chemistry 2001, 276, (27), 25208-25211.

S. Kamoda; Y. Saburi, "Strucrural and enzymatical comparison of
lignostilbene-alpha,beta-dioxygenase isozymes, I, II, and III, from
Pseudomonas paucimobilis TMY1009." Bioscience Biotechnology and
Biochemistry 1993, 57, 931-934.

P. Winterhalter; M. Straubinger, "Saffron-renewed interest in an ancient
spice." Food Review. International 2000, 16, 39-59.

C. Kiefer; S. Hesse]; J. M. Lampert; K. Vogt; M. O. Lederer; D. E.
Breithaupt; J. von Lintig, "Identification and characterization of a
mammalian enzyme catalyzing the asymmetric oxidative cleavage of
provitamin A." Journal of Biological Chemistry 2001, 276, (17), 14110-
14116.

A. A. Dmitrovskii; G. N. N; S. B. Gomboeva; V. Ershov; V. Bykhovsky,
"Enzymatic oxidation of beta apo-8'-carotenol to beta apo-l4'-carotenal by
an enzyme different from beta carotene 15-15' dioxygenase." Biochemistry
1997, 62, (7), 917-923.

J. von Lintig; A. Wyss, "Molecular analysis of vitamin a formation:
Cloning and characterization of beta-carotene 15,15 '-dioxygenases."
Archives of Biochemistry and Biophysics 2001, 385, (1), 47-52.

N A. Wyss; G. Wirtz; W. D. Woggon; R. Brugger; M. Wyss; A. Friedlein; H.

Bachmann; W. Hunziker, "Cloning and expression of beta,beta-carotene
15,15 '-dioxygenase." Biochemical and Biophysical Research
Communications 2000, 271, (2), 334-336.

A. Wyss; G. M. Wirtz; W. D. Woggon; R. Brugger; M. Wyss; A.
Friedlein; G. Riss; H. Bachmann; W. Hunziker, "Expression pattern and
localization of beta,beta-carotene 15,15 '-dioxygenase in different tissues."
Biochemical Journal 2001, 354, 521-529.

J. von Lintig; K. Vogt, "Filling the gap in vitamin A research - Molecular
identification of an enzyme cleaving beta-carotene to retinal." Journal of
Biological Chemistry 2000, 275, (16), 11915-11920.

27

51.

52.

53.

54.

55.

56.

57.

58.

59.

A. Lindqvist; S. Anderson, "Biochemical properties of purified
recombinant human beta carotene 15-15'-monooxygenase." Journal of
Biological Chemistry 2002, 277, (26), 23942-23948.

D. Sklan, "Carotene-cleavage activity in chick intestinal mucosa cytosol:
association with a high-molecular-weight lipid-protein aggregate fraction
and partial characterization of the activity." British Journal of Nutrition
1983, 50, 417-425.

J. Paik; A. During; E. H. Jamison; C. L. Mendelsohn; K. Lai; W. Blaner,
"Expression and characterization of a murine enzyme able to cleave beta
carotene." The Joumal of biological chemistry 2001, 276, 32160-32168.

von Lintig; A. Dreher; C. Kiefer; M. F. Wernet; K. Vogt, "Analysis of the
blind drosophila mutant ninaB identifies the gene encoding the key
enzyme for vitamin A formation in vivo." Proceedings of the National
Academy of Sciences of the United States of America 2001, 98, 1130-1135.

M. T. Redmond; S. Gentleman; T. Duncan; S. Yu; B. Wiggert; E. Gantt; S.
Cuningham, "Identification, expression and substrate specificity of a
mammalian beta-carotene 15-15'-dioxygenase." Journal of Biological
Chemistry 2001, 276, (9), 6560-6565.

F. Takuzo, Oxygenases and model systems. ed.; Klwer Academic
Publishers: Norwell, Ma, 1997 ; 'Vol.' p.

E. 1. Solomon; T. C. Brunold; M. 1. Davis; J. N. Kemsley; S. K. Lee; N.
Lehnert; F. Neese; A. J. Skulan; Y. S. Yang; J. Zhou, "Geometric and
electronic structure/function correlations in non-heme iron enzymes."
Chemical Reviews 2000, 100, (1), 235-349.

L. Que; R. Y. N. Ho, "Dioxygen activation by enzymes with mononuclear
non-heme iron active sites." Journal of Chemical Reviews 1996, 96, (7),
2607-2624.

B. 1. Solomon; T. C. Brunold; M. I. Davis; J. N. Kemsley; L. Sang-Kyu;
N. Lehnert; F. Neese; A. J. Skulan; Y. Yang; J. Zhou, "Geometric and
electronic structure/function correlations in non-heme iron enzymes."
Journal Chemical Reviews 2000, 100, 235-349.

28

60. T. Wilson, "Comments on the mechanism of chemi- and
bioluminescence." Photochemistry and Photobiology 1995, 62, 601-606.

61. M. G. Leuenberger; C. Engeloch-Jarnet; W.-D. Woggon, "The reaction
mechanism of the enzyme catalyzed central cleavage of beta-carotene to
retinal." Angewandte Chemie International Edition 2001, 40, 2614-2617.

62. H. Singh; H. R. Cama, "Enzymatic cleavage of carotenoids." Biochimica
et Biophysica Acta 1974, 370, 49-61.

29

Chapter 2
Expression of BCDOX

2.1 Expression of BCDOX using a T5 promoter

The usefulness of retinoids in the visual process has been well understood
for many decades."3 Also, their vital role in cellular and genetic control has been
well documented recently.“'7 However, what is not well understood is the
biogenesis of retinoids in mammals. Retinoids are biosynthesized by oxidative
cleavage of B-carotene by B-carotene-lS-IS’ dioxygenase (BCDOX) (EC
1.13.11.21) to yield retinal."13 And as discussed previously, we are interested in
understanding the mechanism and function of BCDOX. The first step towards the
understanding of BCDOX function, regulation and mechanism would be the
expression of the protein in large quantities.

Professor Lintig (University of
Freiburg) graciously provided us with vector
(pQE30) that contained the gene for BCDOX DB CD 0x. 1
from mice (AJ278064).” The expression of
this BCDOX gene is under the control of a

Amp

phage T5 promoter transcription-translation

Figure 2—1. Gene of mice B-

System, which is recognized by the E. coli
carotene-15-15’ dioxygenase

l erase. This s stem contains two lac
p0 ym y cloned in a pQE30 vector

Operator sequences, which increase lac (pgopoxq),

30

repressor binding and ensure efficient repression. Also, in the N terminal
sequence of the protein a His tag (6 His in a row) is included to facilitate BCDOX
purification (Figure 2-1). Expression of mice BCDOX has been previously

‘3‘“ chicken,” and

described” in various systems and species such as drosophila,
human”18 but no report has been made using a pQE30 vector. Qiagenl9 suggests
the use of either JM109 or XLlBlue E. coli host to overexpress proteins cloned
into pEQ30 plasmids.” Thus pQE30-mice-BCDOX (pBCDOX-l) was
transformed into both XLlBlue and JM109. Initial attempts to express pBCDOX-
1 using different IPTG concentrations and different temperatures proved
unsuccessful. To accelerate the screening of optimal conditions to . express
BCDOX, the supernatant was analyzed using Anti-His HRP-Conjugate. Anti-His
HRP-Conjugate allows for the identification of recombinant proteins that

contained a 6x Histidine tag. This Anti-His HRP Conjugate21 consists of a mouse

monoclonal IgGi Anti-His ,g.
180 000 In:

    

antibody coupled to { ~
116 000* ‘
up

horseradish peroxidase. In 34 000' "

the presence of peroxide and

 

58 000
4-chloro-1-naphthol, the
horseradish peroxidase 36 000‘ i ..fi
Figure 2-2. Purification of recombinant B-carotene-
produces a colored

15-15'-dioxygenase. Purification was performed by
precipitate.”24 Initially four
the use of an NTA resin. 1. Marker. 2. BSA (60KDa).
d‘fferemE‘ callsmnsmrl‘ 3. Supernatant. 4. Run throw. 5. Wash. 6,7. Eluted

Blue, I M109, BCDOX.

31

 

Tuner(DE3)pLacI, and BL21(DE3)pLysS), and 1 mM IPTG were used to induce
protein expression at 28 °C for three or six hours. But again no recombinant
protein was found either in the soluble fraction or the pellet. When different
amounts of IPTG were used to express the recombinant protein, no improvement
in the expression level of the protein was observed. It is known that low or no
expressions of proteins occurs when the expressed protein is toxic to the bacteria.

To address this problem, the most common approach is the use of low
concentrations of IPTG for the expression of the protein. But when this fails, the
addition of glucose to the media is recommended to increase plasmid stability.”
When glucose is available as an energy source, it is used preferentially over other
sugars. In more detail, glucose causes catabolite repression by reducing the level
of cyclic AMP. The protein responsible for the catabolite activation (CAP) is
active only in the presence of cyclic AMP. The presence of glucose in the media
decreases the basal noise, increasing the plasmid stability. When 1% of glucose
was used, the recombinant protein was obtained in low yield, about 0.7 mg per
liter of culture.

Purification of the enzyme was achieved by the use of the 6x histidine tag
present in the protein. Proteins that contain a histidine tag bind strongly and
selectively to a NTA resin (Contains Nickel, resin form Qiagen) and can be eluted

by the addition of imidazole or a change in the pH.

32

2.2 B-Carotene Dioxygenase activity

The activity of BCDOX was analyzed using an assay previously
developed by Olson.26 This assay consists of monitoring the oxidation of
B—carotene to retinal.

In particular, about 20 ug of BCDOX (purified recombinant protein) was
incubated with B-carotene at 30 °C (For a more detailed description refer to
Materials and Methods). The reaction was quenched at different times. Next, the
carotenoids and retinoids were extracted with hexanes, and were analyzed by
High Pressure Liquid Chromatography (HPLC), (Figure 2-3). A normal phase
column (C18) was used as stationary phase, and hexane: ethyl acetate (95:5) as a
mobile phase at a flow rate of l mIJmin.

The enzyme was found to have an activity of 25.5 pmol retinal/mg
BCDOX-min (Scheme 2-1). It has been reported that addition of cofactors to the
growth media promotes correct folding of proteins, and therefore, increases the
enzymatic activity.27 Thus addition of FeSOa to the LB media was tried, and as
reported, a 7x increase in the BCDOX activity (194 pmol retinal/mg
BCDOX-min) and a KM =3 M was observed. However, the yield of the protein
obtained did not change much (Table 2-1). These values slightly differ from the
reported value (36 pmol retinal/mg BCDOX-min and a KM =6 uM)."

The activity of BCDOX observed seems to be relatively low compared to
other enzymes. So far it is not clear whether the enzyme activity is low due to

lack of correct folding. However, since the activity reported from other groups is

33

 

 

 

 

 

50.
4m-
sol
20.
g 10 Retinal
l i
g 0 [”4 é . r— .

 

 

minutes

Figure 2-3. HPLC chromatographs of enzymatic reaction
using BCDOX and B-carotene. The formation of retinal was
monitored In an HPLC. A) Without enzyme, B) Assay uslng
BCDOX isolated from a culture with out FeSO. 25.5 pmol
retinal /mg of BCDOX-min. C) Assay using BCDOX isolated
from a culture with FeSO, 194 pmol retinal /mg of BCDOX-

min.
also low, this led us to believe that BCDOX rate of conversion of fi-carotene to
retinal is slow. It has been shown that BCDOX is an important source of retinoic
acid in vitro in testis as well as in small intestine, liver, kidney and lung.28 Also,
the concentration of retinoic acid in the cytoplasm is very tightly controlled, thus
it is possible that BCDOX is a check point in the biosynthesis of retinoids. And it

is possible that retinoids work as allosteric inhibitors to control their own

biosynthesis. Also, even though there is an overwhelming agreement that

34

prnoles of retinal

 

l l 1

0 50 1 00 1 50 200 250 300 350

minutes

Scheme 2-1. Kinetics of the recombinant B-carotene 15-15’ dioxygenase. Monitoring the

conversion of B-carotene to retinal.

BCDOX is a cytosolic protein, it is important to keep in mind the lipophilicity of
B—carotene. Carotenoids and their derivatives are non-polar molecules with the
tendency to participate in membranes and are generally insoluble in aqueous

solutions. Thus, it is possible that for optimal activity of BCDOX, various lipid

Table 2-1. Comparison of protein activity shows that glucose addition is essential for

BCDOX expression and addition of iron results in a higher protein activity.

 

Protein expression

 

 

5- 00” ug of protein BCDOX activity
per 1L pmol retinal/ mg
BCDOXomin
XL1-Blue-pBCDOX-1 0 , 0
XL1-Blue-pBCDOX-1+ glucose 736 25.5
XL1-Blue-pBCDOX-1+giucose+ FeSO, 865 194

 

35

aggregates should be present, and recombinant protein expression does not
provide the Optimal conditions for oxidation of B—carotene.

Unfortunately it was found that the protein is not very stable. After
protein purification, if the protein is frozen it loses activity after 48 hours. When
10% of glycerol is added to the storing buffer, it loses 63% of its activity after one
week. When the recombinant protein is exposed to room temperature for more
than three hours the activity decreases significantly. Instability of recombinant
BCDOX has been observed by other groups. For example, Wyss and co workers
observed that during a 50-fold increase in the concentration of the protein 75% of

activity is lost, which they believed is due to aggregation.10

2.3 Expression of BCDOX with rare codons

The fact that all the previous efforts to improve the expression of BCDOX
did not have any significant impact in the protein led us to believe that there was
something else interfering with the expression. One reason for the low yield
could be an insufficient tRNA pool for the codon usage in the BCDOX gene.”3o
Low tRNA can lead to translational stalling, premature translation termination,
translation frame shifting and amino acid rrrisincorporation.31'32 Evidence that this
could be the reason was supported by the fact that every time BCDOX was
expressed and purified, four bands co-eluted with the elution buffer. These bands
could correspond to premature termination of the BCDOX due to codon bias
usage. The isolated fragments could be premature termination of the protein, and

since the recombinant BCDOX contained an N-terrninus histidine tag, the

36

fragments would bind to 1 2 3 4 5 6 7 8 9

the column as well, and co- 180 000 i
1 1

elute with the BCDOX, 6 000 I:

84 000 «

WCDOX
} hatpuritiee

Figure 2-4. Agarose gel shows the purified

Figure 24. 58 ooo;"-‘ 5

It is known that in f
36000“ ..

E. coli some tRNA codons

 

it!“

are underrepresented. In
recombinant mice of B-carotene-15-15' dioxygenase.

particular, Arg codons _
The gel shows more than one protein, the shown

(AGA’ AGO CGA)’ the Ile proteins co-elute with the recombinant BCDOX. The

codon (AUA) and the Leu bands with low molecular weight correspond to
codon (CU A) all represent premature termination of proteins. 1. Marker. 2. BSA

60 KDa. 3. Su rnatant. 4. Run throu h. 5. 6.
less than 8% of their ( ) pa 9

Wash. 7-9. Eluted BCDOX.

corresponding population
of codons while the Gly codon (GGA), Arg codon (CGG) and Pro codon (CCC)
fall to less than 2% of their respective population. An analysis of the BCDOX
sequence showed: Arg (AGA)=2 codons, Arg (AGG)=10 codons, Ile (AUA)=5
codons and Pro (CCC)=12 codons.29‘30 Figure 2-5 shows the sequence of
BCDOX and rare codons (codons that are expressed in low amounts in E. coli) are
represented in blue.

When E. coli is used to express proteins, the expression of affected

recombinant genes could be salvaged either by replacing them with synonymous

codons that are more frequently used using site-directed mutagenesis or by

37

ATG
MUS

GAG
Glu

AAA
Lys

GTG
val

CCC
RIO

GGG
Gly

GCC
Ala

CTT
Leu

TAC CTG
2!; Lou

TCT GAG
Ser Glu

ATA
Ile

ACA
Thr

ATG
net

CTC
leu

CAG
Gln

TTC
Phe

—_‘

GCT TTC
Ala Phe

AAC ATC
Asn Ile

AGO AAA
It! Lys

TAT GTG
Tyr val

GTC CTT
val Leu

TTT AAG
Phe Lys

AAG CAC
Aye His

TAC TAC
EXT TYr

TTT AAG TTG GAT

TCC
Ser

ATG
M68

ATC
Ile

GCG
Ala

AAC
Asn

ATC
Ile

GCG
Ala

CAC
His

AIL
Ile

TTT
Phe

GGC

GLXVSer

CAC
His

ACA
Th:

CAC
His

AGT
Ser

AGT
Ser

GAC
Asp

ACC
Thr

GGA
Gly

TAC
IYI

TTG
Leu

TGT
Cys

AAA
Lys

GAC CCC
AspvPro

AAC
Asn

GTA
val

ATG

Met Glu

CCT
Pro

GCC
Ala

GAA
Glu

GTT
val

AGC
Ser

TTT
Phe

ATC

Phe Lzs Len Asp Ile

GCT TCC TGT ATG TCA
Ala Ser Cys nee Ser

CAG AGO ACC AGG AAG
Gln Arggmhr Agggbys

GGC CAG AAT

Gly

ATT
Ile

GTG
val

TTT
Phe

ACC
Thr

ATG
Mét

TCT
Ser

GGA
Gly

CAG
Gln

CTG
Leu

ACA
Thr

ACA
Thr

TTC
Phe

GGT
Gly

CTC
Leu

TTC
Phe

CCT
Pro

Gly

CCA
Pro

GGA
Gly

TCC
Ser

TAC
Tyr

GCC
Ala

CAC

His

Glu

ACC
Thr

GCT
Ala

TCC
Ser

GTG
val

TGC
Cys

GTC
val

AAG
Lys

GAC
Asp

GTG
val

Asp

GCA
Ala

GAG
Glu

ATC
Ile

ATC
Ile

TAC
Tyr

ACC
Thr

GAC
Asp

CTA
Leu

ACC
Thr

GTC
val

CCA
Pro

TCC
Ser

ACG
Thr

ATG
Met

AGO

AAG
Lys

TIP

AGC
Ser

GCC
Ala

CCG
Pro

ATC
Ile

TTC
Phe

GAG
Glu

TCG
Ser

GTG
V31

GAC
Asp

ATT
Ile

GAG
Glu

GCC
Ala

GAG

Arg Glu

CCT

ACC

Lys

CTG
Leu

AAG
gys

GAT
ASE

AAC
Asn

GAC
Asp

ccc
Pro

TAT

ACC
Thr

CAC
His

GAC
ASP

AGC
Ser

TCC
Se:

AAC
Asn

ACC
Tyr

GAC

Glu

CAG
Glu

TAC
Ty:

GGG
Gly

ATT
Ile

CCC
Pro

GAC
Asp

Ala

TTG
Leu

CCT
Pro

Lys

AAG
Lys

TCC
Ser

TAT
.ZYT

GCA
Ala

AAG

Asp Lys

AAG

TTC

Pro Th: Lys Phe

CAG CTG
Gln Leu

GGG ACC
GLXVThr

AAC CAT
Asn His

GAG GTC
Glu val

GAG GCC
Glu Ala

TGC AAA
Cys Lys

ACA
Thr

TTC
Phe

ACC ACG
Th: Th:

AAG
Lys

GAG
Glu

TAT
.2Zr

CAT
His

GGG AGO

Glu

AAG
Lys

AAA
Lys

CGC TCG
Arg Ser

GTG
val

GTG
val

TAC ATG
Tyr.Met

TAC
my:

ACA
Th:

TAC ACA

ASH

GAG
Glu

CCA
Pro

CTG
Leu

CTC

TTT
Twp Phe
TAC

Phe mgr

an
L3!

ATC
Ile

AAC

AAC
ASR

GAC
Asp

AAC
Asn

ACC
Thr

GAG
Glu

GAT
A80

GTT
val

GAC GAG
Asp Glu

ACA AAA

Arg Th: Lys

AAG
Lys

GGG
Gly

CTG
Lou

CTC
Leu

TTT
Phe

CTG
Leu

AGO
lg:

GGA
Gly

CAT
His

ATT
Ile

GAT GCC

Tyr Thr Asp Pro

GTT
val

CGA

Lew 1gp Asn

GAT
Asp

Agy Ser

ATC
Ile

TTT
Phe

TGT
Cys

AAC

Asn

TAC

CAG
Gln

GCC
Ala
AAC GGG
Gly

GGC CTG
Gig Lou

AGC AAA

Lys

GTG
val

GTG
val

TCC AAA
Se: Lys

CTG ATC
Leu Ile

TAC ATC
,gyr Ile

CGG AAG

ryr Ari gs

GCT
Ala

TAC
TY?

AGT
Se:

TC?
59:

GAC
Glu

GTG
val

ATC
Ile

ATG
MO:

GGG AAT
Gly'Asn

GTG
val

ATA
Ile

CGC
P10

GTG
val

ccc
Pro

AGC
Ser

CAG
Gln

CCT
Pro

AGC
Ser

TGG
TIP

ATC
Ile

GAC
Asp

GTG GTC
9;; val

Figure 2-5A. The sequence of mice of B-carotene-15-15’ dioxygenase contains a

total of 29 rare codons in its sequence. Arg(AGA)= 2 codons. Arg(AGG)=10 codons,

IIe(AUA)=5 codons and Pro(CCC)=12 codons. sequence continues in Figure 2.58.

38

Pho

GCC
Ala

GAC
Asp

GTG
val

TCA
502‘

Pro

Gly
ACC
Thr

Glu

Glu

CCT
Pro

GTT
val

Ala

CAT
His

CAT
His

GAG
Glu

TAT
Tyr

TTC
Pho

GAG
Glu

CGC
Pro

CTC
Lou

TCT ACA

Sor

GTC

Glu val

AAG
Ed’s

AAG
EYE

CCA
Pro

ATA
Ile

TGC
Cys

AGC
Sor

GAT
Asp

GAT
Asp

TTA

Pho Lou

GAT
Asp

GCG
Ala

AAT
Asn

ccc
Pro

GGT
Gly

GCA
Ala

ACA

Thr

cA'r
His

Figure 2-53.

GTC
val

GAC
Asp

GAG
Glu

CAT
His

ACT

CTC
Lou

TAT

CTG
Lou

TGG
TIP

GGA
Gly

CTC
Leu

GAC
Asp

GTG
val

GAT
Asp

GGT
Gly

AAT
AS}!

GCC
Ala

AGC
Sor

AGC
Sor

AAG
Lys

TCC
Sor

GTG
val

GAC
Asp

GCA
Ala

ACA

TAC GAA

Glu

CGC
Arg

TAC
Tyr

TAT

Lys Tyr

CCA
Pro

GCA
Ala

GTC
val

ATT
Ilo

ATT
Ilo

CTG
Lou

ATG
Mot

CAC
His

AAG
Lys

CCG
Pro

GGC

CAG
Gln

ACA
Thr

TCA

GlygSor

TAC

CTC
Lou

AGC
A£F

AAG
Lys

GCC
Ala

GGC
GAY

ATC
Ile

GAC
Asp

GAG
Glu

TTA
Lou

GAT
Asp

CTG
Lou

ACT
Thr

GCA
Ala

AGT
Ser

GAG
Glu

TAT
TX!

CTG
Lou

GAT
Asp

CTG
Lou

CTA
Leu

TTT
Pho

ATT
Ile

CCT
Pro

TCA
Ser

GCC
Ala

GAC
A59

CCA
Pro

CCA
Pro

CTT
Lys

GAG
Glu

CAG
Gln

ACC
Tyr

GCA
Ala

AAG
Lys

GAG
Glu

GCA
Ala

CTC
Lou

CTG
Lou

GCC
Ala

Lys

CTT
Lou

GCT
Ala

GAA
Glu

TAA

GAC
Asp

CTC
Lou

TCA
Sor

GAA
Glu

GAG
Glu

CTC
Lou

GCT
Ala

ACA
Thr

Phe

ATC
Ile

AGT
Ser

CAT
His
GAA
Glu

CTG
Lou

GGC
Glu

Pho

GTG
val

GTG
val

Lys

CCT
Pro

GAA
Glu

AAG
Lys

GTT
val

GTC
val

TTT
Pho

GGT
Gly

ACG
Thr

AGC
Sor

TGT
Cys

GTG
val

CTG
Lou

TAC
Tyr

CCT
Pro

ACC
Tyr

GGC TCA
Glx75or

GAC GGC

ASP

ATA
Ile

CGG
Arg

CAG
Gln

GTA
val

TCC
Sor

TCC
Sor

CCC
Pro

ACG
Tyr

TCT
Sor

ACG
Thr

ACG
Thr

GAA
Glu

TTA
Lou

TTT
Phe

CAA
Gln

GAG
Glu

CAC
His

AGT
Sor

CTG
Lou

GCC
Ala

CTC
Lou

AAT
Asn

CAT

Glz_Hisgy§l

AAT
Asn

roe
Tap

TTA
Lou

CCA
Pro

GAT
Asp

Lou

ATC
Ile

GTT
val

GAA
Glu

TTT GAT
Pho Asp

AAC CTG
Asn Lou

GTG ATC
val Ile

AAC AAG
Ala Lys

TTT GCT

Ax! Arg Pho Ala

TTA
Lou

GTC
val

TAT
Tyr

GTC

TAT
Ty:

GCT
Ala

AGT
Sor

CCA
Pro

AAG

Lys sz

GGT GCG
Gly Ala

ccc
Pro

CAA
Gln

CGC

Ala Arg

CCA
PI'O

GAT

AAG GTG
Lys val

TGC CAG
Cys Gln

TAC AAC
Ty: Asn

CCA
Pro

GTC
val

TCT
Sor

GAG
Glu

AAG
Lys

GAT
Asp

AAG
Lys

CTG
Lou

GCC
Ala

TCT
Sor

GCA GAC

AspiAla Asp

GAA
Glu

AAT
Asn

AAC TCA
Asn Sor

GAC TTC
Asp Phe

GAT
lip

ACA
Thr

The sequence of mice of B-carotene-15-15’ dioxygenase contains a

total of 29 rare codons in its sequence. Arg(AGA)= 2 codons, Arg(AGG)=10 codons,

IIe(AUA)=5 codons and Pro(CCC)=12 codons.

39

supplementing the host with extra copies of the equivalent tRNA genes.
Stratagene33 has generated a strain of E. coli BL21-Codon Plus-RIL (BL21-RP)
cells that contains a plasmid with extra copies of the tRNA genes for the codons
Arg (AGA, AGG), Ile (AUA) and Len (CUA), and BL21-Codon Plus-RP cells
(BLZI-RIL) that contains a plasmid with extra copies of the tRNA genes for the
codons of the Arg (AGA, AGG), and Pro (CCC). Site directed mutagenesis of 29
different rare codons is not a simple task, and it has been reported that addition of
the plasmid provides as good or better results than mutagenesis. The BL21-RIL
and BL21-RP cells were generously donated by Professor Frost (MSU).
Competent cells of BL21-RP and BLZl-RIL were prepared, followed by
transformation with pBCDOX-l. After transformation, cells from one colony
were used for further experiments after the presence of both plasmids was
confirmed.

Expression of BCDOX in BL21-RlL-i-pBCDOX-l and BL21-
RP+pBCDOX-1 was performed. The results (summarized in Table 2-2) were

unexpected. First, the amount of protein obtained did not change significantly.

Table 2-2. Protein expression after addition of rare codons using XL1-Blue as host.

 

Protein expression

 

 

E. call no of protein per 1L BCDOX activity
pmol retinal] mg
BCDOXOmin
XL1-Blue-pBCDOX 612 20
XL1-Blue pACYC-FiiL-pBCDOX-1 1312 6
XL1-Blue pACYC-RP-pBCDOX-1 1050 0

 

40

Second, the activity of the protein obtained was lower. As discussed previously
the expression of the recombinant BCDOX seems to be very tightly controlled,
and the use of a different host (BL21) might have significant impact in the control
for expression of BCDOX. In more detail, the pQE30 vector, in which the
BCDOX is cloned, contains a double lac operator for over expression of the
recombinant protein. However, the plasmid does not contain the gene to express
Lac]. The BL21 cells do not have any gene that produces LacI protein, and thus,
the expression of BCDOX is not regulated as tightly. Consequently to address
this problem, the use of an E. coli that produces Lac] protein would be necessary.
XL1-Blue E. coli contains a gene that produces the Lao] protein in its genomic
DNA. For that reason XLl-Blue cells were used as host instead of BL21 cells.
XL1-Blue pBCDOX competent cells were prepared and transformed with either
pACYC-RIL or pACYC-RP plasmid. As expected, the expression of the protein
was increased but only by a factor of two while the activity decreased showing no
overall significant improvement.

Novagen has also designed a system to express proteins containing rare
codons. This system contains a plasmid called pRARE, which encodes for all the
rarely used codons (Arg, Ile, Gly, Leu and Pro) and it is commercially available.
This plasmid is cloned in a different E. coli host called Rosetta Blue. This strain
was particularly interesting because its genome is similar to XL1-Blue and over
expresses the Loci". Therefore this was the ideal system to test if the Lac)q was
crucial for the expression of pBCDOX-l or not. The results obtained are

summarized in Figure 2-6; all results are normalized for a better comparison. It

41

 

 

Figure 2-6. Relative activity (I) and expression of recombinant (I) BCDOX. Results
using different E. coli. hosts 1. XL1-Blue-pBCDOX-1. 2. XL1-Blue-pBCDOX-1-
pRP(Ary, Pro). 3. XL1-Blue—pBCDOX-1-pRlL(Arg, Ile, Leu). 4. Rosseta-pRARE (Arg,
Ile, Gly, Lou, Pro)-pBCDOX-1.

was observed that the yield of the recombinant BCDOX increased but not
significantly, and the activity did not improve. The light bars show the yield of
protein obtained after purification. The highest protein yield was obtained when
Rosetta Blue cells were used. On the other hand, the highest activity was obtained
using the original system pBCDOX-l/XLl-Blue and the lowest using pACYC-
RP/pBCDOX-l. However, the differences are not significant, and could be due to
the inherent variability of the system and not to a real improvement of the
expression and activity of BCDOX. At this point, it was concluded that addition
of tRNA’s did not significantly improve either expression of the recombinant

BCDOX or its activity.

42

2.4 Expression of BCDOX in vivo

The expression of BCDOX in E. coli was monitored in vivo by the
colorimetric change of the oxidation of B-carotene (orange) to retinal (yellow).
Lintig and co-workers, had expressed the BCDOX into a genetic background of
an E. coli that accumulates B-carotene, which resulted in the formation of

13 Professor Lintig graciously

retinoids that could be monitored by a color shift.
provided a vector that contained the genes necessary for expression of B-carotene
in E. coli. This plasmid pORANGE‘3 (Figure 2-7) contain a gene cluster with
crtE(a), crtB(b), crtI(c) and crtY (d).3“‘35 B—Carotene is derived from the
condensation between the C5 skeleton of isopentyl pyrophosphate (IPP) and
dimethyl allyl pyrophosphate (DMAPP). This
condensation is catalyzed by crtE, to form
geranylgeranyl pyrophosphate (GGPP). Next,
the synthesis of phytoene is the first

committed step in Clo-carotenoid biosynthesis

 

and is catalyzed by crtB; it proceeds as a head-

to-head condensation of two molecules of Figure 24. pORANGE. plasmid

GGPP. Subsequently, four dehydrogenation contains a gene cluster with crtE,

steps are carried out by err! in the conversion 0113 Gill and crtY. Those genes

of phytoene to lycopene. Finally, 1y00pene is produce B-carotene from dimethyl

36 37 allyl pyrophosphate and isopentyl

cyclized by crtY to form B-carotene
pyrophosphate.

(Scheme 2-2).

43

\ \ \\\ \ \

Phytoene

\ \\\\\\\ \

Lyeopone

lenY

‘\\\\\\\\\.
B-Cemuno

Scheme 2-2. Biosynthesis of Bearotene in E. coli. starts

with the condensation between DMAPP and IPP.

XLlBlue cells were transformed with pORANGE. The cells obtained
after transformation showed an orange coloration due to the accumulation of B-

carotene (Figure 2-8). The cells were grown under red safe lights until the picture

 

was taken to decrease the isomerization of the produced B-carotene. The
competent cells were prepared and transformed with pBCDOX-l. The
transformed cells were grown and BCDOX expression was induced by addition of
isopropyl-B-D-thiogalactopyranoside (IPTG). After the cells were grown for
about six hours in the dark, the obtained cells were colored off white. The shift of
the coloration was caused by oxidation of B-carotene (orange) to retinal (yellow)

by BCDOX, as a result of BCDOX catalysis.

 

Figure 2-8. Activity of recombinant mice B-carotene-15-
15’ dioxygenase in vivo. 1. XL1-Blue transformed with
pOFiANGE and pBCDOX-i with out induction (addition of
IPTG). Orange color due to formation of B-carotene 2.
XL1-Blue transformed with pORANGE and pBCDOX-1 after
addition of IPT G. The yellow coloration is due to oxidation

of fl-carotene to form retinal.

45

2.5 Expression of BCDOX using a T7 promoter

Given the fact that the main objective for this project was the
understanding of BCDOX’s enzymatic and mechanistic properties it was
imperative to express large quantities of the enzyme. The plasmid pQE30 uses a
T5 promoter (Figure 2.1), which is recognized by the E. coli polymerase.
Nonetheless this promoter has not been widely used since it is considered a weak
promoter. As a result, the low expression of the recombinant BCDOX could be
due to the fact that the expression of the protein is under the control of a weak
promoter. Different systems to express proteins have been developed. For
example a T7-promoter (bacteriophage T7 promoter) driven system, originally
developed by Studier and colleagues,3“° is known to be a much stronger
promoter. Also since E. coli RNA polymerase does not recognize the T7
promoter, there is very low transcription of the target gene in the absence of T7
polymerase. Therefore the amount of basal transcription is tightly controlled,
increasing the stability of plasmids with toxic genes. Thus the BCDOX
expression controlled by a T7 promoter would offer a good alternate for the
expression of BCDOX. Therefore, BCDOX was to be cloned into a different

plasmid under the control of a T7 promoter.

2.6 Cloning of BCDOX into a T7 promoter vector
Initially a pET-30a was chosen as the target plasmid. The MCS of pET-
30a is under the control of a T7 promoter with a C-terminal histidine-tag with an

intervening enterokinase cut site that should allow the removal of the his-tag in

46

order to study the native protein. Primers were designed using the reported
sequence for mice BCDOX, in order to introduce two cut sites, NdeI and Kim].
The first step would be the extraction of the gene using PCR followed by
digestion and ligation into pET-30a. Many attempts to PCR the BCDOX gene
were performed. Different conditions were tried (changing the extension time,
alignment temperature, concentration of different components etc.) but no PCR
product was obtained. These results were puzzling for a while, and it was
concluded that the only reasonable explanation was that one of the primers was
not binding to the parent DNA. Therefore, a complete sequence of the gene was
performed. The results showed that the reported sequence differed from the
sequence in the vector in the last 51 base pairs, thus explaining why the attempted
PCR provided no product. Figure 2-9 shows the differences between the reported
gene and the sequenced gene.

In the course of this investigation, we decided to use a thrombin cut site to
remove the His tag instead of enterokinase. The thrombine cut site is known to be
more efficient for cleavage than the enterokinase 1:1000, 1:10
enterokinasezprotein ratio respectively. If Xhol is used to clone the desired gene
in the pET30a the thrombine cut site would not be present. pET-29b(+) contains a
T7 promoter and a thrombin site which is known to be more efficient when it
comes to enzymatic cleavage (1:1000), and there is a His-tag that would facilitate
the purification using an NTA column. Thus it was chosen to be the host for mice
BCDOX. Two cut sites were chosen to clone the BCDOX gene into pET-29b(+),

(Ndel(295) and KpnI(238)). A set of primers were designed, the first of which

47

MEIIFGQNKKEQLEPVQAKVTGSIPAWLQGTLLRNGPGMHTVGESKYNHWFDGL
MEIIFGQNKKEQLEPVQAKVTGSIPAWLQGTLLRNGPGMHTVGESKYNHWFDGL

ALLHSFSIRDGEVFYRSKYLQSDTYIANIEANRIVVSEFGTMAYPDPCKNIFSK
ALLHSFSIRDGEVFYRSKYLQSDTYIANIEANRIVVSEFGTMAYPDPCKNIFSK

AFSYLSHTIPDFTDNCLINIMKCGEDFYATTETNYIRKIDPQTLETLEKVDYRK
AFSYLSHTIPDFTDNCLINIMKCGEDFYATTETNYIRKIDPQTLETLEKVDYRK

YVAVNLATSHPHYDEAGNVLNMGTSVVDKGRTKYVIFKIPATVPDSKKKGKSPV
YVAVNLATSHPHYDEAGNVLNMGTSVVDKGRTKYVIFKIPATVPDSKKKGKSPV

KHAEVFCSISSRSLLSPSYYHSFGVTENYVVFLEQPFKLDILKMATAYMRGVSW
KHAEVFCSISSRSLLSPSYYHSFGVTENYVVFLEQPFKLDILKMATAYMRGVSW

ASCMSFDREDKTYIHIIDQRTRKPVPTKFYTDPMVVFHHVNAYEEDGCVLFDVI
ASCMSFDREDKTYIHIIDQRTRKPVPTKFYTDPMVVFHHVNAYEEDGCVLFDVI

AYEDSSLYQLFYLANLNKDFEEKSRLTSVPTLRRFAVPLHVDKDAEVGSNLVKV
AYEDSSLYQLFYLANLNKDFEEKSRLTSVPTLRRFAVPLHVDKDAEVGSNLVKV

SSTTATALKEKDGHVYCQPEVLYEGLELPRINYAYNGXPYRYIFAAEVQWSPVP
SSTTATALKEKDGHVYCQPEVLYEGLELPRINYAYNGKPYRYIFAAEVQWSPVP

TKILKYDILTKSSLKWSEESCWPAEPLFVPTPGAKDEDDGVILSAIVSTDPQKL
TKILKYDILTKSSLKWSEESCWPAEPLFVPTPGAKDEDDGVILSAIVSTDPQKL

PFLLILDAKSFTELARASVDADMHLDLHGLFIPDADWNAVKQTPAETQEVENSD
PFLLILDAKSFTELARASVDADMHLDLHGLFIPDADWNAVKQTPAETQEVENSD

HPTDPTAPBLsasmnxerroessr. Expected
HPTDPTAPEVPRVDLQPSLIS ----- Actual

Figure 2-9. Comparison of the amino acid sequence of the reported
mice BCDOX (top) and the obtained BCDOX (bottom). The last

seventeen amino acids differ from the reported sequence.

contained an NdeI cut site (5 '-CGGACATATGGAGATAATATITGGCCA
GAATAAGAAAGAACAGCTGG-3’) and the other a Kpnl cut site (5 '-GACCCGGGTA
CCCTCIUGTGCTGTCGGA TC-3’). The strategy to clone BCDOX into pET-29b(+)
consisted of amplifying the gene, digesting it and ligating into the pET—29b(+)

vector. The PCR was uneventful and provided the BCDOX gene with the cut

48

sites NdeI and KpnI (See Material and Methods for details). The next step would
be the digestion of the PCR product and the vector pET-29b(+). The double
digestion of the plasmid and the PCR product proved to be problematic.
According to the manufacturer, for a double digestion using KpnI and Ndel, the
use of NEBuffer 1 is recommended Even though the activity of NdeI in
NEBuffer l is 100%, it is not the buffer that shows the optimal activity for Ndel.“l
A digestion test using pET-29b(+) as the standard showed that at 2 and 4 hours
the digestion was not complete for NdeI. Only overnight (~14 hrs) the digestion
showed a single band. Thus, the digestion of pET-29b(+) and BCDOX gene
(from PCR) using NEBuffer l, were performed overnight at 37 °C. After
digestion, and purification of the digested vector and gene, the isolated DNA were
quantified. Ligations were performed using different mole ratio between the gene
and the plasmid, at different temperatures. After transformations of JM109 cells
with the ligation reaction products, some colonies were obtained. Analysis of
those colonies showed that no gene was cloned into pET-29b(+)-BCDOX. It has
been reported that at times, residual polymerase might fill the sticky ends to
generate blunt ends,“2 thus, interfering with the proper ligation of the gene into
the vector. To avoid polymerase interference, the PCR product was purified The
gene and plasmid were digested overnight at 37 °C. purified, and quantified.
Ligation attempts using the conditions previously mentioned were set up,
followed by transformation into JM109 which provided a few colonies. Analysis

of these colonies resulted in vectors without the desired gene.

49

 

 

 

Even though the manufacturers“ suggest the use of NEBuffer 1 for the
double digestion, it was possible that double digestion was not taking place;
therefore cloning of the gene would not be possible. To address this problem,
single digestions were performed consecutively. Also to address the re-ligation
problem, two strategies were followed. The first one was based on the principle
that linear DNA transforms with much lower efficiency than a circular DNA.
Sequence analysis of the pET-29b(+) showed a 33!” cut site between the NdeI
and KpnI, that is not present in the BCDOX gene. For that reason, digestion of
the ligation product with BglII would digest only the re-ligation products and not
the cloned products. After transformation, the majority of the colonies obtained
should contain the cloned product and not the re-ligation product, due to the low
transformation efficiency of linear DNA. Ligation reactions using the conditions
previously mentioned were set up, followed by digestion with Hg!” for two hours.
The digested ligations were transformed into JM109, and after transformation of
the ligated products, no colonies were observed.

The second strategy consisted of the removal of the terminal 5’-phosphate
groups in the digested plasmid DNA. During in vitro ligation, DNA ligase
catalyzes the formation of a phosphodiester bond between adjacent nucleotides
only if one nucleotide has a 5’-phosphate residue and the other carries a 3’-
hydroxyl. Removal of the 5’-phosphate residues can be performed with alkaline

phosphatase Calf Intestine Phosphatase (CIP).‘3'“

As previously, PCR,
purification, double digestion of plasmid and BCDOX, and purification were

performed. The digested vector was then treated with CIP for 2 hours at 37 °C.

50

 

This was followed by ligation and transformation into JM109. After overnight
incubation, 6 colonies were obtained. Only one of them seemed to contain the
BCDOX gene, Figure 2.10 (pBCDOX-Z). The cloned product (Vector-BCDOX)
was obtained when a 1:10 mole ratio between vector/BCDOX and incubation at
16 °C overnight were performed.

The obtained plasmid was sent for sequencing using the following
primers: 5 '-GA TCTCGA TCCCGCGAAATTAA TA CGA C -3’ 3330044; 5 '-
CCAGACCCTAGAGACCTTG GAGAAGG-3' BBB0045; 5'-
CGAGGAGAAGTCCAGGCTGACC-3' BBB0046; 5 ’-GATCGATCT C GATCCCGCG-3’
8830055; 5 '-GCAGACTGGAATGCAGTGAAGC-3' BBBOOSZ. The sequence showed
4-point mutations: two silent mutations and two real mutations, Cys (T GC) 102
mutated to Tyr (TAC), and Ala (GCC) 337 mutated to Thr (ACC). These plasmid

obtained would be refer as pBCDOX-Z.

  

o,»

  

Figure 2-10. Gene of mice [3-
carotene-15-15' dioxygenase cloned in
a pET-29b(+)=pBCDOX-2.

Thr=thrombine

51

 

 

2.7 Expression of pET-29b(+)-BCDOX

E. coli Rosetta(DB3)pLysS cells were chosen as the host to overexpress
the pE'IQ9b(+)-BCDOX recombinant protein. This strain was chosen because it
supplies the rare tRNAs, and it possesses a copy of the’I7 RNA polymerase.38
This gene is under the control of the lacUV5 promoter, which allows a tight
control in the expression of the recombinant protein. After the transformation of
the host with pBCDOX-Z, several colonies were obtained, and analysis of five of
them showed the pBCDOX and the codon plus plasmid. One colony was chosen

for further studies.

2.8 Optimization of pBCDOX expression

The recombinant BCDOX protein was expressed and indeed this time over
expression of the protein was obtained. Unfortunately though, all overexpressed
protein was in the form of inclusion bodies. The formation of inclusion bodies
was unfortunate but not surprising due to BCDOX’s nature and because it has
been observed that over expression of recombinant proteins in E. coli can result in
the formation of inclusion bodies in the cell cytoplasm or periplasm.‘5"° It has
been found that the tendency to form aggregates depends on the rate of expression
and the rate or refolding. If the rate of expression exceeds the rate of refolding,
co-expressed proteins come into contact with one another and aggregate prior to
reaching their native state." It is known that disulphide bonds and hydrophobic
patches may affect the solubility of a protein and result in the formation of

inclusion bodies. To improve solubility, researchers have applied a variety of

52

 

 

 

 

methods such as adjusting the amount of IPTG used,48 lower the temperature,49
co-expression of the protein of interest with chaperones and foldases,” the use of
solubilizing fusion partners?“52 different E. coli strains,” among others. So far,
all literature reports that describe the expression and isolation of BCDOX”"”3'55
agree with the fact that BCDOX is a cytosolic protein. However, there is
evidence suggesting an amphiphilic character of BCDOX and that increased ionic
strength and high protein concentrations are likely to promote aggregation. As
previously discussed it has been observed by Wyss and co workers that at a 50-
fold concentration of the protein leads to 75% loss of its activity which is
attributed to aggregation.10

Due to the nature of BCODX activity, oxidation of B-carotene, it is very
possible that this protein is associated with lipid aggregates or other proteins. The
fact that when BCDOX is expressed mostly as inclusion bodies supports the latter

findings. Not much information is available about the overexpression of BCDOX

in bacterial systems.

2.9 Attempts to express the recombinant BCDOX in a soluble form

2.9.1 Decrease the rate of expression, use of different temperatures, time
and IPTG

The rate of protein folding is a first order process. By decreasing the rate
of expression of the recombinant protein, one can favor correct protein folding.

There are several reported methods to decrease the rate of overexpression of

53

recombinant proteins, a number of which were tested to express BCDOX as
soluble protein. The results are discussed below. All the following experiments
were analyzed by 10% polyacrilamide SDS gel electrophoresis of both the
supernatant (soluble fraction) and pellet (possible inclusion bodies) of the crude
expression mixture. To assure that BCDOX was not present in small amounts in
the supernatant, the samples were treated with the Anti-His HRP conjugate.2|
Treatment with Anti-His HRP conjugate would allow for the identification of
histidine tagged proteins.

The most commonly used approach to improve protein solubility and
folding involves adjusting the amount of IPTG used48 and the temperature at
which the protein is expressed.49 The net effect when the temperature or the IPTG
concentration is decreased is to reduce the rate at which recombinant proteins are
expressed. By slowing down the rate for the protein expression proper folding is
promoted. Therefore, the expression of BCDOX was performed at different
temperatures (30, 28, 16 and 4 °C) and different concentrations of IPTG (0, 0.4, l
and 10 mM). In addition all other possible permutations were screened. The cells
were lysed and the supernatant and the pellets were analyzed. Unfortunately, the
results showed again that in all cases, the protein was in the form of inclusion
bodies and no trace of soluble protein was found using Anti-His HRP conjugate.
As expected, low temperatures and low IPTG concentration decreased the amount
of inclusion bodies formed (Table 2-3). Because T7 is a strong promoter and it is
under the control of the lac operon, the expression of BCDOX is activated by

addition of MG.

54

 

Table 2-3. Expression of recombinant pBCDOX-2 in Rosetta(DEa)pLys$,
induced at different concentrations of IPTG and different temperatures. The
results are shown as S=Supernatant, I.B.= Inclusion bodies. I- BCDOX

present, *2 No BCDOX present.

 

 

 

 

Temperature
:7: so °c as re 26 °c 16 °C
8 LB 8 LB. 8 LB. 8 LB
0,0 x x x x x x x x
0,4 x I x I x I x I
0.1 x I x I x I x I
10 x I x I x I x I

 

 

 

 

 

 

 

 

 

 

 

V A variation to the addition of IPTG was performed. Four different
portions were added over a period of eight hours and each sample was analyzed.
After SDS page analysis of the supernatant and the pellet, no change in the
amount of soluble recombinant protein BCDOX was observed. (Table 2-4).

Again all supematants were treated with histidine tag antibody and no protein was

Table 2-4. Expression of recombinant
pBCDOX—2 in Rosetta(DE3)pLysS, induced
by addition of different portions IPT G of
ODwo~ 1.0 for overnight at 26 “C. The
results are shown as s-Supematant, l.B.=
Inclusion bodies. I: BCDOX present, x:
No BCDOX present.

 

IPTG
0 portions
1 portion
4 portions
8 portions

xxxxa)
\‘x'xx'm

55

 

 

 

 

observed. After these trials it was concluded that the amount of IPTG used to
induce protein expression did not have an impact in the folding of the protein.
The expression of the proteins at 26 °C was monitored every 1, 3, 5 and 8 hours.
When the cells were lysed analysis of the soluble fraction and pellet showed that
BCDOX was expressed only as inclusion bodies. This experiment was repeated at
16 °C and 28 °C and the same results were obtained. Thus it could be concluded
that different temperatures and different induction times did not have any positive

influence in the formation of soluble proteins.

2.9.2 Different E. coli strain

It has been shown that in a bacterial host, the solubility of proteins changes
with different E. coli strains. Because of this reason, different E. coli strains were
tested for overexpression of the recombinant BCDOX. The strains chosen for
overexpression were: A) Tuner(DE3)pLacI. The difference between these cells
and Rosseta is that Tuner cells do not have the LacY mutation, which allows the
uniform absorption of IPTG. B) ER2566 these cells are derived from BL21 and
have a copy of the T7 RNA gene into lac and BL21(DE3) pLysS. Competent
cells for Tuner(DE3)pLacI, BL21(DE3)pLysS., Rosseta and ER2566 were
prepared. Each strain was transformed with pBCDOX-Z and the expression of the
recombinant protein was tested. Because the concentration of IPTG proved to be
irrelevant in the Rosseta strain the next experiments were performed at constant
IPTG concentration of 0.4 mM, added at OD600~ 1.0, and 26 °C. Aliquots of 6

mL were taken at l, 3, 5 and 8 hours after induction. The results showed (Table

56

2-5) that the protein was again over expressed in the form of inclusion bodies,
with the concentration of inclusion bodies increasing with time.

Treatment of all supematants with Histidine-tag antibody resulted in
negative results. Since no recombinant protein was identified in the soluble form
the experiment was repeated at expression temperatures of 16 °C and 9.5 °C.
Aliquots were taken at 4, 8, 19, 30 and 42 hours after induction. The expression
of the recombinant BCDOX was pretty slow (as expected) and when the
overexpression of BCDOX was achieved, the protein was formed as inclusion

bodies, as in the previous cases. No protein was found in the supernatant when

Table 2-5. Expression of recombinant pBCDOX-2 using different host
Rosetta(DE3)pLysS, Tuner(DE3)pLacl, BL21(DE3)pLysS ER2566
induced by addition of IPTG at different portions ODooo~ 1.0 for
overnight at 26 'C. The results are shown as S-Supematant. I.B.=
Inclusion bodies. /= BCDOX present, x: No BCDOX present.

 

 

 

 

 

E. coli host IPT G (mu) 8. LB.
0.0 X x
Rosetta(DEa)pLysS- 0.01 x I
pBCDOX-2 0.10 x I
1.00 x I
0.0 X 3‘
Tuner(DE3)pLacl- 0.01 x I
pBCDOX~2 0.1 0 x I
1.00 x I
0.0 x x
BL21(DE3)pLysS- 0.01 x I
pBCDOX-2 0.10 x I
1.00 x I
0.0 X x
0.01 x I
ER2566-pBCDOX-2 0.1 0 x /
1.00 x I

 

57

each supernatant was treated with the Anti-His I-IRP conjugate. At this point it
was concluded that the use of different E. coli strains did not have a significant
influence in the solubility of BCDOX. Also it has been reported that addition of
the substrate in the culture can improve the solubility of a protein. The native
substrate for BCDOX is B-carotene, and as previously discussed, the pORANGE
vector can produce B-carotene in vivo. Thus an ER2566 E. coli strain was
transformed with pORANGE. After expression of the recombinant protein no

improvement in the solubility was observed (Table 2-6).

Table 2-6. Expression of recombinant BCDOX in the presence of its native
substrate B-carotene. 532566 was chosen as the host and it was
transformed with pORANGE and pBCDOX-z. pOFiANGE is responsible for
the formation of p-carotene. The protein was express under red safe light.
The results are shown as S=Supernatant, LB.= Inclusion bodies. “a BCDOX

present, x: No BCDOX present.

 

 

E. coll host IPTG (mu) 3. LB.
0.0 x x ,
ER2566-pacoox-2 0.01 x I
pORANGE 0.10 x ’
1.00 x I

 

2.9.3 Over expression of chaperones

In E. coli cells there is a complex protein machinery to help proteins fold
after their biosynthesis 'or a heat shock.” Those complex proteins are called
chaperone proteins. Melecular chaperones possess no catalytic activity. Their
function is to prevent polypeptide chains from forming incorrect associations with

itself or another protein." For up regulation of chaperones in bacterial systems

58

two techniques are used, heat shock and addition of organic solvents, especially
ethanol. When E. coli is subjected to a variety of stresses, including temperature
up shift, exposure to organic solvents, and the accumulation of misfolded proteins,
the synthesis of heat shock proteins (HSPS) is upregulated in order to repair
cellular damage. It has been reported that solubility of proteins enhance using a
heat shock prior to protein induction and that the over expression of the
recombinant protein is not affected. It is believed that increase it the solubility is
due to the fact that some heat-shock proteins or chaperones such as GroEL and
DnaK are up regulated when the E. coli is heated at 42 °C. Two strains used in
these experiments were: Rosetta(DE3)pLysS and Tuner(DE3)pLacI. 100 mL of
cells were at 37 °C until OD600~ 0.1 and then warmed to a 42 °C until ODaoo~ 1.0.
At this point expression was induced with 0.4 mM IPTG at 16 °C. Aliquots of 6
mL were taken at 3, 6, 9 and 16 hours. The cells were collected and analyzed as
previously described. The results (Table 2-7) showed that there was not
enhancement in the solubility of BCDOX obtained.

Since increasing the temperature did not provide the expected results,
instead, to promote the expression of chaperones addition of organic solvents was
performed. Ethanol is one of the most powerful elicitors of the heat-shock
response in E. coli, and its addition to the media has been shown to increase the
solubility of a variety of proteins. The amount of ethanol used to enhance
solubility was 3%, and was added right before inoculation. The cells with ethanol
grew much slower than the ones without. The expression was induced and the

cells were collected and analyzed as previously. The results showed (Table 2-7

59

 

hat there was almost no expression of BCDOX in the pellet or in the supernatant.
In addition, the His-tag detection assay did not detect any His-tagged protein in
the supernatant solution. All attempts to express BCDOX as a soluble protein
have failed. And as previously discussed the experiments were performed in a
small scale (100 mL of culture) and the supematants were treated with the Anti-
I-Iis HRP conjugate to test for any traces of BCDOX in the supernatant. No
recombinant protein was identified from any of the previous studies. To finally

confirm that no protein was expressed in soluble form, a large-scale expression

Table 2-7. OverexpreSSion of chaperones to improve the solubility of
the recombinant BCDOX. A. Heat shock. B. Addition of ethanol.
The results are shown as S=Supematant, LB.= Inclusion bodies. I:

BCDOX present, at: No BCDOX present.

 

 

 

A. B
E. coil host Time 8 LB S LB
3 x x x x
ROSBitB(DE3) pLysS- 6 x I x x
pBCDOX-2 9 x I x I
1 6 x I x I
3 x x x x
Tuner(DES)pLacl- 6 x I x x
pBCDOX-2 9 x I x I
1 6 X I x I

 

was performed. Three liters of culture were grown and induced (for details see
Materials and Methods). The supernatant was purified using an NTA column,
under the same conditions as the recombinant protein cloned in pQE30. The
purified supernatant was not as pure as expected. The purified supernatant
showed a band corresponding to the size of BCDOX, but more than one protein
were present in the eluted fraction. Whether the presence of those proteins was
due to unspecific binding or associated proteins to BCDOX is not well
understood. The purified supematants showed no BCDOX activity (eluted

fractions).

2.10 Activity in viva

Finally to test whether small amounts of the recombinant BCDOX were
expressed in an active form in E. coli., BCDOX activity was monitored in viva by
following the colorimetric change of the oxidation of B-carotene (orange) to
retinal (yellow). The source of B-carotene in viva would be the pORANGE vector
as previously described and the host chosen was ER2566 because it contains a
copy of the T7 DNA polymerase in its genome. After transformation of ERZS66
with pORANGE the competent cells were prepared and transformed with
pBCDOX-Z. Unfortunately this test did not provide the expected results, because
when the ER2566-pORANGE was expressed the phenotype obtained was
yellowish instead of orange (Table 2-8). This experiment was repeated several
times with a great variability of different shades of yellow-orange colors obtained.

There is no obvious reason why the ER2566 E. coli strain does not work as well

61

as the XL1-Blue strain. The precursor of Bocarotene is isoprene, and the genotype
of ER2566 does not suggest any deficiency in the production of isoprene (it
should be at least as good as XL1-Blue). The experiments were repeated several
times and the results were not consistent. Variability in the expression of
carotenoids in E. coli has been previously observed and no obvious correlation
between the genotype and carotenoid expression has been found. Sandman has
suggested that this difference is due to the different accumulation of carotenoids

in the membrane.“5

Table 2-8. Attempts to test the recombinant mice B-carotene-15-15’ dioxygenase
pBCDOX-2 in viva. E82655 was transformed with pORANGE and pBCDOX-Z. The
yellow coloration should be due to oxidation of B-carotene to form retinal. However in
this case it was impossible to establish a difference between the coloration of the E.

call that produced B-carotene and the one that should produce retinal. Y=yellow,

 

 

O=orange.
Temperature 16 °C 26 °C 28 'C
lPTG Color IPTG Color IPTG Color
ER2566/pBCDOX-2 x Y x Y x Y
ER2566/pBCDOX-2 \’ Y r/ Y r/ Y
ER2566 /pOFiANGE x Y x Y x Y
ER2566 IpORANGE “ Y “ Y r’ Y
ER2566/pBCDOX-2/pOFiANGE x Y x O X Y
ER2566/pBCDOX-2/pORANGE / Y I Y 1 Y

 

 

 

 

62

2.11 Expression of the native BCDOX under the control of a T7 promoter

2.11.1 Mutation of BCDOX to add a stop eodon to avoid translation of the
(5ka tag and thrombine cut site

The BCDOX was cloned into a T7 promoter-containing pET-29b(+) to
increase the expression of the BCDOX protein. As expected the T7 promoter
increased the expression of BCDOX. However, inclusion bodies were obtained
despite all the attempts to express BCDOX in a soluble form. Improper folding of
the recombinant proteins causes the formation of inclusion bodies. The lack of
folding could be caused by the extra amino acids that formed the thrombine cut
site and the histidine tag at the C- terminus in the recombinant protein. To
eliminate the possibility that the extra amino acids were interfering with the
correct folding of the recombinant protein, a stop codon was added to the BCDOX
sequence (before the 6x histidine and thrombine). The presence of the st0p codon
would stop the translation of the extra amino acids, therefore, express only the
native protein. The first approach to add the stop codon was by using site directed
mutagenesis. Site directed mutagenesis is a technique used to introduce specific
mutations into a DNA sequence. Site directed mutagenesis uses chemically
synthesized oligonucleotides that contain the desired mutation, which serves as a
primer for DNA synthesis using a DNA polymerase.

A set of two primers that contained the stop codon were designed; Primer
1 3330078 5 '-GCACCAGAGGGTACCTAGGTG-3 ’ and primer 2 3330079, 5 ’-

CACCTAGGTA CCCI‘CTGGTGC-3'. After the PCR was performed no product was

63

observed. To address this problem, the extension time was increased, (the vector
is 7 Kbp long)” but no improvement was obtained. A gradient PCR to scan for
the optimal temperature (from 65.5 to 45 °C) did not change the results. When
site-directed mutagenesis is used, the amplification of the vector is linear and not
exponential like in PCR. Therefore, increments in the concentration of parent
DNA were increased up to 100 ng and 300 ng, but again no amplified DNA were
observed. It is possible that the lack of product is due to the fact that the
pBCDOX size was too large. Even though the amplification of long DNA (545
Kbp) has been reported!“50 it was accomplished by changing parameters such as
pH, glycerol, and DMSO concentration, decreasing denaturing times, salt
concentrations, etc. Screening for all those parameters would be costly and time
consuming. There are other commercially available options from Stratagene and
Invitrogene, but they are costly and not 100% efficient. Thus, a different

experimental design was approached.

2.11.2 Cloning of BCDOX

The next approach was to use a conventional cloning strategy. First,
mutation of the BCDOX gene to add a stop codon using normal PCR, digestion
and then ligation into a pET-29b(+) vector which contains the T7 promoter was
attempted. The same cut sites as before were chosen (NdeI and Kpnl) since they
are present in the BCDOX gene. The primers used were 5'-
CACCAGGGTACCCTATGGTGCAGT-3' and the other that binds to the beginning of the

BCODX gene 5'-GGA GATA‘I‘ACATATGGAGATAATA-3’. The PCR was simple, and

provided the BCDOX gene with the NdeI and KpnI cut sites and stop codon. As

mentioned previously double digestion using KpnI and Mia], proved problematic.

Thus, this time a single digestion followed by another single digestion using the

optimal conditions for each enzyme was performed.41 After the PCR reaction the

DNA was purified, followed by a single digestion of the vector and PCR product

using KpnI. The DNA was purified followed by another single digestion using

Ndelr DNA purification and quantification was performed before ligations. The

ligation reactions were performed at 16 °C at a
ratio of 1:5, 1:10 and 1:15 moles of
vector/BCDOX. The ligation reactions were
transformed into M 109. The gene obtained
from one of the transformed colonies was sent
for sequencing using the following primers 5'-
CAGACCCTAGAGACCTTGGAG-3' 8380080; 5'-
CGAGGAGAAGTCCAGG CTGACC—3’ BBBOO46;
5'-GATCGATCI'CGATCCCGCG-3’ 3330055; 5’-
GCAGACI‘G GAATGCAGTGAAGC-3’ 3330052.
The sequence confirmed the presence of the
stop codon, and no other mutation was found

in pBCDOX-3 (Figure 2-13).

65

 

Figure 2-13. Gene of mice 8-
carotene-15-15' dioxygenase
cloned in a pET-29b(+). A stop
codon was added before the
thrombine (thr) cut site and the

histidine tag (pBCDOX-a).

2.11.3 Expression of pBCDOX-3; native BCDOX

For over expression of the recombinant BCDOX the plasmid was
transformed into an E. coli Rosetta(DBB)pLysS. The first attempt to express the
mutated BCDOX was performed at 16 °C and 28 °C,6"62 and different
concentrations of IPTG as well as different induction times were screened. The
cells were collected and lysed and both the supernatant and the pellet were
analyzed using SDS page gel. At 0.1 mM IPT G no expression of recombinant
BCDOX was observed. When protein expression was monitored at different

times, the overexpressed protein was only found as inclusion bodies (Table 2-9).

Table 2-9. Expression of recombinant pBCDOX-3 in
Rosetta(DE3)pLysS, induced at different concentrations of
IPTG and temperatures. The results are shown as

SsSupematant, LB.= Inclusion bodies. I: BCDOX present, at:

 

 

 

 

 

IPT G Time 16 °C 28 °C

mM hours 3 LB 8 LB
3 x x x x

6 x x x x

0.00 9 x x x x
15 x x x x

3 x x x x

x x x x

0.01 g x I x I
16 x I x I

3 x x x x

6 x x x x

0'10 9 x I x I
16 x I x I

3 x x x x

x x x x

1'00 g x I x I
16 x I x I

 

66

 

Unfortunately the results obtained were the same as the recombinant protein with
the thrombine cut site and the histidine tag. Thus it was concluded that the extra
amino acids in the recombinant BCDOX were not responsible for the formation of
inclusion bodies. It is possible that the amphiphilic character of BCDOX, as well
as, the fact that the active enzyme could be a tetramer might decrease the rate of

folding, therefore enhance formation of the inclusion bodies.

2.12 Isolation of inclusion bodies

The formation of inclusion bodies could offer several advantages for the
production of recombinant proteins if they can be refolded. The proteins
produced in inclusion bodies might be unstable in the cytoplasm of E. coli due to
proteolysis and may be toxic for the host cell in the native conformation. It has
been seen that under appropriate conditions, the recombinant proteins deposited in
inclusion bodies amount to about 50% or more of the total cellular protein. Since
inclusion bodies have a relatively high density they can be isolated from the
cellular proteins, with a purity up to 90%.‘53 A number of methods have been
reported for solubilizing and refolding proteins form inclusion bodies.
Denaturants such as urea and guanidine hydrochloride, and the use of highly
alkaline pHs, strong thiol-containing reductants (dithiothreitol, dithioerythritol
mercaptoethanol), chelating reagents (EDTA), have been reported to solubilize a
significant percentage of insoluble proteins. Renaturation is usually accomplished

by the removal of excess denaturants by either dilution or a buffer-exchange step

67

(dialysis, diafiltration, gel-filtration chromatography or immobilization onto a

solid support).
As mentioned previously, attempts to
BCDOX have been

produce a soluble

unsuccessful. Therefore, optimization of the
expression and isolation of inclusion bodies was
pursued. Isolation of the inclusion bodies was
accomplished by over expression of BCDOX in
Rasetta(DE3)pLysS. The soluble proteins were
separated and the pellet was washed with a buffer
containing 0.5% Triton X-100, followed by a
wash with buffer containing 2 M urea and 0.5%
Triton X—100. The pellet obtained was
resuspended into a buffer containing 8 M urea and
DTT, or 6 M guanidine hydrochloride. It was

observed that smaller concentrations of urea (<6

M or <4 M) or guanidine did not solubilize the

12345678
" ‘.-"

  

Figure 2-14. 10% agarose
gel. Isolation of Inclusion
bodies produced by expression
of recombinant mice 8-
carotene-15-15' dioxygenase

cloned in a pBCDOX-1 in

Rosetta(DE3)pLysS. 1. Cell
pellet 2. Supernatant.
3,4.lnciusion bodies. 5, 6, 7.

Inclusion bodies washed with
Triton X-100. 8. BCDOX

marker.

inclusion bodies. It was also noticed that addition of DTT increased the solubility.

The inclusion bodies were purified almost with 90% purity and around 8 mg per

liter (Figure 2-14).

After the BCDOX was purified (from inclusion bodies) and solubilized,

refolding of the protein was the next step. Solubilization of inclusion bodies

usually occurs using strong chaotropes, such as urea, and the protein is refolded to

68

its correct structure as the denaturant concentration is reduced. In general,
refolding is initiated by diluting the denatured protein into a refolding buffer
solution (no chaotropic agent present). Once the denaturant concentration is
reduced the protein refolds at rate equivalent to first order kinetics. The unfolded
protein quickly folds to an intermediate form, and there is a slow transition from
the intermediate to the native form (limiting step) and may take minutes to days to
complete. At this point nonproductive aggregation reactions can occur, resulting
in the formation of stable aggregates.“’7 Therefore it is really important to avoid
this pathway, which can be achieved by having very low concentrations of
protein. In the case of disulfide-bonded proteins, renaturation buffers must
promote disulfide bond formation. The options available for refolding include
dilution, dialysis, diafiltration, gel filtration, and detergents among others.61 With
those facts in mind, design of a system to refolded BCDOX was performed. As
mentioned previously, there are no reports concerning BCDOX (or a similar
protein) refolding, thus, preliminary refolding experiments were performed. The
refolding method chosen was a 100-fold dilution, basically because of its
simplicity and the fact that the amounts of protein used are low. Glutathione was
used to aid in disulfide formation. Aliquots of the solution were withdrawn and
BCDOX activity was assayed, (~0.16 mg of BCDOX were used per assay). None
of the samples showed activity. Also dialysis of BCDOX protein was performed
but no active protein was obtained.

As mentioned previously, this is not the only time in which bacterial

expression is problematic. In additon it has been observed that the BCDOX

69

protein is not stable. Andersson, Zack and Cunninham reported that expression of
human BCDOX in bacteria was problematic, and even though the expression in E.
coli has been reported, there was no reference to the efficiency of the BCDOX
expressed.15"7'18 Andersson, reported that to overcome the instability of the
recombinant BCDOX protein they used insect cells to overexpress BCDOX.18
Also baby hamster HIK cells have been used to express BCDOX but whether they
chose this system because they had problems with the bacterial system and then
they had to switch to mammalian cells is unknown.

Progress in this project was hampered by the lack of expressing large
amounts of soluble BCDOX protein. Unfortunately the use of mammalian cells to
express BCDOX in our group was not possible. Thus at this point we decided not

to continue to attempt to express BCDOX.

70

2.13 Material and methods

A. Different E. coli used

 

 

E. coli strain Genotype

JM109 el4'(McrA')recAl endAl gyrA96 thi-l nst17 (rt-mu) SUPP.“ relAi
A(lac-proAB)[ii"traD36proABlaclq Z

XL1 Blue recAl endAl gyrA96 thi-l nstl7 (tr-ma) supB44relA1[F'proABiacl" z
AMIS TnlO tet’)

BL21(DE3)pLysS E. coli B F ampT hst(r3' ma') dcm’ gal (srl-
recA)306::Tn10(TcR)(DE3)pLysS(Cm )

TunefiDE3)pLacI coli B F ampT hst(r3' m3) dam+ gal lacl’l
(DE3)pLacI(CmR)

BL21-Codon 1311134111,, E. coli B F ompT hSdSU'B. mg') dcm‘ Tet' gal endA
Hre[argU ileY leuW Cam'

BL21-Codon Plus-RP E. coli B F ampT hst(r3' mg') clam+ Tet' gal endA
Hte[argU proL Cam‘]

Rosetta Blue recAI endAI gyrA96 thi-I 11.9de 7 (rm-tram) supE44
relAI lachARE [F’praA+ 8* Incl" Z AM15::Tn10 Tet’)

 

B. Typical preparation of competent cells

The E. coli strain of interest was grown (37 °C until ODgoo of 0.4-0.6, the
media contained LB, the corresponding antibiotic). After about three h The cells
were harvested by centrifugation at 5,000 RPM for 10 min at 4 °C. The cells were
re-suspended (100 mL of 0.9% NaCl) and then centrifugated at 5,000 RPM for 10
rrrin at 4 °C. The cells were re-suspended (50 mL of 100 mM CaClz) and
incubated for 30 min at 0 °C. The cells were centrifuged at 5,000 RPM for 10
rrrin at 4 °C. and the cells were re-suspended (4 mL of 100 mM CBClz, 15%
glycerol). The cells in suspension were aliquoted (0.1 mL) and were quickly

frozen with liquid nitrogen.

71

C. Typical transformation of competent cells

Sterile conditions were maintained throughout the entire protocol. The
competent cells were incubated at 0 °C for 5 min and the plasmid DNA (10
ng/uL) was added. The mixture was incubated for 30 rrrin and the cells were heat
shocked at 42 °C for 2 min at 0 °C. LB (antibiotic if necessary) was added and
the cells were incubated at 37 °C for l h. Then the cells were transfered to a plate

containing LB/antibiotic and were grown overnight at 37 °C.

D. Detection of 6xHis-tagged proteins (dot blot) using Anti-His HRP

conjugates Qiagen° Kit

A protein solution (1-100 ng/ttL) was applied directly onto a membrane.
The membrane was washed twice for 10 min with TBS buffer (10 mM Tris-Cl,
pH 8.0, 150 mM NaCI). The membrane was then incubated for 1 h in blocking
buffer at room temperature. The membrane was washed twice for 10 min with
TBS-Tween/Triton buffer (20 mM Tris-Cl, pH 8.0, 500 mM NaCl, 0.05% (v/v)
Tween 20 and 0.2% (v/v) Triton X-lOO). The membrane was washed for 10 min
with TBS buffer. Then the membrane was incubated in Anti-His HRP conjugate
solution (111500 dilution of conjugate stock solution in blocking buffer). This
was followed by a membrane washed (twice for 10 min) with TBS-Tween/I‘riton
buffer and was washed for 10 min with TBS buffer. The membrane was stained
with a solution containing 4-chloro-1-naphthol (18 mg), Tris (30 mL of solution

(0.9% NaCl (w/v) in 0.1 M Tris-Cl pH 8.0) and methanol (6 mL). This was

72

followed by addition of hydrogen peroxide (6 mL, 30% solution) to start the
catalysis of the to horseradish peroxydase. Then the membrane was rinsed twice
with water and dried. The membranes were scanned as soon as possible because

the colors fade with time.

E. Expression of BCDOX

The target gene (pBCDOX) was transformed into XL1-Blue cells
according to standard protocols and plated on Ampicillin resistant LB plates. A
single colony from the plate was inoculated (200 mL LB/carbenicillin broth
(Carbenicillin 100 rig/mL)) and grown at 37 °C. while shaking, overnight. The
overnight culture was inoculated (l L of LB/ 1% glucose / Carbenicillin medium)
and grown at 37 °C until ODeoo --= 1.0 (~ 3 h on average). The expression was
induced by the addition of IPTG (1 mM) and FeSO4 (55 M final concentration).
The culture was incubated at 26 °C for 6 h. The cells were harvested by
centrifugation (9000 rpm, 30 min) and frozen at -20 °C overnight. The cell mass
was thawed and re-suspended in Buffer A (20 mL for r L expression). The cells
were lyzed for 30 min at 0 °C with lysozyme (200 uL of 1 mg/mL), followed by
sonication for 10 min at 4 °C. Then MgS04 (300 uL 1 mM) and benzoase (30 uL,
of a 25 mm enzyme, Novagen) were added followed by incubation on ice for 30
rrrin. The mixture was spun down (9000 rpm, 30 min) and the supernatant, that
contained the soluble BCDOX was further purified. Some of the crude was saved
for BCDOX activity assay. Buffer A (50 mM NaH2P04, 300 mM NaCl, 10 mM

imidazole, pH=8)

73

F. Purification using a Histidine Purification Tag

The lysate that contained the histidine tag BCDOX protein, was applied to
a nickel resin for purification (Novagen). The lysate was allowed to flow through
the column, after which time it was washed (2x 10 mL) with the wash buffer (50
mM NaHzPO4, 300 mM NaCl, 20 mM imidazole, pH=8). The protein was eluted
with the elution buffer (50 mM NaH2P04, 300 mM NaCl, 250 mM imidazole,
pH=8), and the fractions were then analyzed by SDS-PAGE. The fractions
containing protein were then concentrated with Arrricon Centriplus YM-lO filters,
and buffer exchanged to a T ris- KOH buffer (Tris- KOH 10 mM buffer pH 8.0, 6
mM sodium taurocholate, 0.5 mM DTT 10 mM buffer pH 8.0, 6 mM sodium
taurocholate, 0.5 mM DTT). Protein concentration was calculated via BCA

protein assay.

G. Ni2+ Resin Regeneration

Resin is regenerated and recycled after each use by first stripping the
column of all Ni2+ ions and anything bound to them. The column is washed with
three column lengths of the strip buffer (100 mM EDTA, 0.5 M NaCl, 20 mM
Tris-HCl, pH=7.9), followed by three column lengths of sterile distilled water.
The column is then recharged with five column lengths of a fresh, clean Ni2+
source utilizing the charge buffer (50 mM NiSO4). Finally, the column is

equilibrated with three column lengths of binding buffer (50 mM NaH2P04, 300

74

mM NaCl, 10 mM imidazole, pH=8) to wash away any unbound Ni“ ions. The

column can be recycled as long as it retains a blue/green color after the cleaning.

H. Extended Ni2+ Resin Regeneration

When the flow rate of the column slows dramatically, or does not turn a

strong blue/green after being washed with charge buffer, the following wash

should be done (for a 2.5 mL column bed)

1.

2.

10.

ll

12.

5 mL of 6 M guanidine HCl, 0.2 M acetic acid
5 mL sterile distilled water
2.5 mL 2% SDS

2.5 mL 25% EtOH

. 2.5 mL 50% EtOH

2.5 mL 75% EtOH

12.5 mL 100% EtOH

2.5 mL 75% EtOH

2.5 mL 50% EtOH

2.5 mL 25% EtOH

. 2.5 mL sterile distilled water

12.5 100 mM EDTA, pH=8.0

13. 7.5 mL sterile distilled water

14.

7.5 20% EtOH (for prolonged storage)

75

Note that the SDS solutions and the SDS wash steps cannot be done in the cold
room as the detergent precipitates out of solution. Also, the column should be
stored in 20% EtOH when not in use for long periods of time because bacteria can

grow in the agarose resin.

1. Preparation of B-carotene micelles

A B-carotene solution was prepared by combining a B-carotene stock
solution (875 mM in hexanes, 81 ML) taurocholate (100 pL) and Tween (3340
uL, 4% w/v acetone). Then, the solution was dried and dissolved in buffer B
(Buffer B,: 10 mL of Tris-KOH 10 mM buffer pH 8.0, 6 mM sodium
taurocholate, 0.5 mM DTT). The B-carotene used for the assay most be purified

using a pipette silica column eluted with hexanes. The eluted solution was

concentrated under nitrogen, and redissolved in hexanes.

J. Assay for BCDOX, conversion of B-carotene to all trans-retinal

The final reaction mixture for the cleavage reaction contained B-carotene
(7.1 uM), Tris-KOH buffer (8 mM pH 8.0), sodium taurocholate (4.05 mM), DTT
(0.34 mM), Tween 40 (0.3%), a-tocopherol (50 uM), Peso. (1.38 mM), and
BCDOX (roughly 20 pg of protein).

The assay was started by adding the protein to solution A (solution A: 540
HL Tris-KOH buffer pH 8.0, 6 mM sodium taurocholate, 0.5 mM DTT, and 20 pl.
of FeSOa). The solution was pre-incubated for 5 min at 30 °C. Then the B-

carotene micelles solution (160 uL) was added and the mixture was then

76

incubated for 2-3 h at 30 °C. The reaction was stopped by adding formaldehyde
((200 ill... 46% in water) and was incubated for another 10 min. The carotenoids
and products were extracted sequentially with chloroformzmethanol (1.5 mL, of a
1:2, v/v, 0.01% pyrrogallol) and hexane (1.5 mL) and with chloroformzhexane
(0.5 mL: 1.5 mL). After addition of each organic solvent, the reaction mixture was "
mixed using a vortex for 40 seconds. The combined extracts were then dried
under a steam of nitrogen gas, and dissolved in isooctane: toluene (200 pLL of (8/2,

v/v) containing 0.01% butylhydroxytoluene).

K. HPLC analysis of the products obtained by a BCDOX activity

The products were analyzed using a normal phase column (C18), hexane:
ethyl acetate (95:5) was used as a mobile phase at a flow rate of l mIJmin. B-
carotene was monitored at 450 nm, retinal at 370 and 325nm; and retinol at 325

nm. B-Carotene RT=10.5, retinal RT=3.0.

L. Calculation of Michelis Menten (KM) constant (steady state).
Calculation-of KM and VMAX for the isolated recombinant enzyme was
performed. The formation of retinal was monitored, the enzymatic reaction was
quenched at different times. Analysis of the data using a double reciprocal plot
provided the values for KM (uM) and VMAX (pmol retinal/minomg), excess of

substrate to mimic a steady state conditions was used.

77

M. Plasmid Purification:

Bacteria from a single colony was grown (500 mL, LB, containing
appropriate antibiotics) and purified with via Qiagen“D column purification. Upon
completion of the Qiagen” purification, the recovered DNA was resuspended in
(400 pL) sterile water, and analyzed by UV-visible for concentration
determination and purity. The concentration of DNA obtained is rig/“L.

Aszw x (50 ug/ 1000 1.1L) x (volume of DNA used / total volume UV
sample)

Purity of DNA sample = Abszw/ Aszgo
= 1.8, pure
>1.8, RNA contamination

< 1.8, protein contamination

N. Sample preparation for sequencing DNA

In a sterilized eppendorf tube (0.5 mL), DNA (2 pg) and primer (30 pmol)
were mixed. When sequencing is performed, the largest amount of base pairs that
are sequenced with accuracy is about 400 bp. Therefore, to sequence the

complete gene, five different primers were used.

.—i
O

5 '-GA TCTCGA TCCCGCGAAA TTAA TACGAC-3 '

2. 5'-CCAGACCCTAGAGACCTTGGAGAAGG-3'
3. 5 ’-CGAGGAGAAGTCCAGGCTGACC-3 '

4. 5 ’-GA TCGA TCTCGATCCCGCG-3’

5. 5 '-GCAGACTGGAA TGCAGTGAAGC-3 '

78

The DNA must be purified by a Qiageno column and resuspended in
sterile water prior to analysis. Failure to do this will result in poor sequence
data. In addition, the sequence sample should be diluted with sterile water,

not buffer.

0. Melting temperature calculation for primers:
Primers should ideally have melting points 2 78 °C. and end in at least
one, if not more, GC base pair
Tm = 81.5 °C + (0.41)(% GC) — 675 IN -% mismatch

where N = primer length

P. Optimal PCR conditions

A small-scale gradient PCR was performed to optimize the extension

 

 

 

 

 

temperature
PCR recipe
Template DNA 100 ng
Primer 1 1 mM
Primer2 1 mM
DNTP 200 W
Deep vent 2 U
10 x deep vent buffer 5 uL (1x)
M9304 5 12M
H20 SO'RXH 11L
PCR prom
1 X 94 °C 5 min
94 °c 1 min
30 X 48 °C 3 min
72 °C 3 min
1 X 72 °C 10 min
1 X 25 °C 10 min

 

79

Q. PCR Primers

Primers Sequence PCR Template

5'-CGGACATATGGAGATAATATTTGGCCAGA
BCDOX-2 ATMGAAAGAACAGCTGG-S’ pBCDOX-1
5’ -GACCCGGGTACCCTCTGGTGCTGTCGGATC-3’

5‘-GCACCAGAGGGTACCTAGGTG-3 pBCDOX-1

300°“ 5' CACCTAGGTACCCTCTGGTGC-a’

R. Purification of DNA from agarose gel

GENCLEAN Turbo QiagenO Protocol: The desired band was excised
from the agarose gel. The gel was cut into smaller pieces and transferred into an
eppendorf tube. Then a GENCLEAN Turbo solution (100 uL per 100 pg of
agarose) was added and the mixture was melted at 55 °C for 5 min. The melted
solution was transferred into a GENCLEAN Turbo cartridge. The filter was spun
for 5 seconds. The filter was washed twice by addition of GENCLEAN Turbo
Wash (500 um. The DNA was eluted by addition of GENCLEAN Turbo solution

(30 uL). The DNA obtained was used directly in the enzyme digestions.

S. Digestions and ligation reactions
As mentioned previously, the double digestion did not afford the desired
product. Therefore, subsequent single digestions were performed. Same

protocols were followed for the gene (PCR product) and the pET-29b(+). The

80

first digestion was performed using KpnI followed by purification using
GENECLEAN. The second digestion was performed using Ndel.
The DNA was purified following the same procedure and quantified

before the li gations were performed

 

 

 

 

__Qig§stions
BCDOX pET-29m
Kpnl 1 “L . 1 “L .
Ndel - 1 uL - 1 pL
DNA30pL 30uL 30uL 30uL 30uL
Buffer10r2(10X) 1X 1X 1X 1X
H20 6 ul— 6 9L 6 ul- 6 rd.

 

 

o Incubation at 37 °C for 2 h.

T. Calf Intestine Phosphatase treatment

 

 

 

 

 

 

CIP treatment
Digested pET- Digested BCDOX
29b(+) gene
DNA obtained from DNA purification 30 pl. 30 pL
CIP 1pI 1U 1U
Buffer (10X) 1 X 1 X
H20 4 pL 4 pL
0 Incubation at 37 °C for 2 h
U. Ligations.
Lugitlone

BCDOX 4.5x10‘“moles

pET-290(+) 4.5x10‘“moles

T4 ligase 1 id

Buffer (10X) 1 pl

H20 2 pl

 

0 Incubation at 16 °C for 24 h
JM109 E. coli were transformed with 5 uL of the ligation reactions

81

V. Expression of pET-29b(+)

For over expression of pET-29b(+)-BCDOX (pBCDOX-l or -2) the
plasmid was transformed into an E. coli Rosetta(DE3)pLysS following the
previously mentioned protocol. This strain was chosen because it supplies rare
tRNAs, it carries a copy of the T7 lyzozyme gene (for tight control), and it
possesses a copy of theT7 RNA polymerase gene under the control of the lacUV5
promoter, which allows a tight control for the over expression.

A single colony from the plate was inoculated (200 mL LB/Ampicillin-
Cloramphenicol broth) and grown at 37 °C while shaking overnight. The
overnight culture was re-inoculated (1 L of LB/l% glucose / Ampicillin-
Cloramphenicol medium), the culture was grown at 37 °C until 013500 a 1.0 (~3 h
on average). The expression was induced by addition of IPTG (0.4 mM) and
FeSOa (55 M final concentration). The culture was incubated at various
temperatures 4-32 °C for 6 h or overnight. The cells were harvested by
centrifugation (9000 rpm, 30 rrrin) and frozen at —20 °C overnight. The cell mass
was thawed and re-suspended in Buffer A (20 mL for 1 L expression). The cells
were lyzed for 30 min at 0 °C with lysozyme (200 pl. of 1 mg/mL), followed by
sonication for 10 min at 4 °C. At this point MgSOa (300 mL, 1 mM) and
benzonase (30 pL, 25 mm, Novagen) were added to reduce viscosity and the
mixture was incubated on ice for 30 min. The mixture was spun down (9000 rpm,
30 min) and the supernatant and the insoluble pellet was analyzed by SDS PAGE
gel. In all cases the over expressed BCDOX was found in the insoluble fraction

as inclusion bodies.

82

W. Purification of BCDOX as inclusion bodies

Isolation of the inclusion bodies was accomplished by over expression of
pBT—29b(+)-BCDOX in Rosetta(DE3)pLysS induction by addition of IPT G (0.4
mM), OD600~1.0 for 8 h at 26 °C. The soluble proteins were separated and the
pellet was washed with a buffer containing 0.5% Triton X-100, followed by a
buffer containing 2 M urea and 0.5% Triton X-100. The pellet obtained was
finally resuspended into a buffer containing 8 M urea and DTT, or 6 M guanidine
hydrochloride. It was observed that a smaller concentration of urea <6 M or <4 M
guanidine did not solubilize the inclusion bodies. It was also noticed that addition
of D'I'l‘ increased the solubility. The inclusion bodies were purified almost with

90% purity and at around 8 mg per liter.

X. Refolding experiments

A 100 fold dilution was performed by adding Buffer A (100 mL to 1 mL
of protein solution). The 100 dilution was performed using Buffer A (Buffer A:
Tris-KOH 10 mM buffer pH 8.0, 6mM sodium taurocholate, 0.5 mM DTT. buffer
and stirred for 24 h. 20 mL were concentrated ten times using Arnicon Centriplus

YM-30 filters, and the activity test showed an inactive protein.

Y. Expression of BCDOX in viva

The expression of BCDOX in E. coli was monitored in viva by the
colorimetric change of the oxidation of B-carotene (orange) to retinal (yellow).

The resultant E. coli cells had an orange coloration due to the accumulation of B-

83

carotene in the cells. In detail an E. coli XL1-Blue or ERZS66 was transformed
with pORANGE (plasmid that contains the genes necessary for the synthesis of
B—carotene). The transformed cells were gown and they showed an orange
phenotype. These cells were converted into competent cells and then transformed
with the pBCDOX-l or pBCDOX-2 plasmids. These transformed cells (cells that
contained pORANGE and pBCDOX-l) were gown under LII/Chloramphenicol,
carbenicillin and 1% glucose (5 mL) to an OD of 1.00. Expression of BCDOX
was then induced by addition of IPTG (1 mM final concentration). This culture
was gown for 6 more h. The cells were spun down at 10000 RPM and rinsed
with water. The shift of the coloration was caused by oxidation of B-carotene
(orange) to retinal (yellow) by BCDOX. Where the cells had a yellow coloration
the activity of BCDOX was confirmed. The same protocol was attempted using
ER2566 E. coli instead of XL1-Blue but the transformed cells did not accumulate

B-carotene.

84

2.14 References

10.

R. A. Morton, Photochemistry of Vision. ed.; Springer-Verlag: New York,
1972; 'Vol.' p 33.

L. J. Gudas, In The retinoids: Biology chemistry and medicine. ed.; New
York, 1994; 'Vol.' p 443-520.

A. C. Ross, "Symposium-retinoids-cellular-metabolism and activation-
introduction." Journal afNutritian 1993, 123, (2), 344-345.

P. Palozza, "Regulation of cell cycle progression and apoptosis by beta-
carotene in undiferentiated and differentiated HL-60 leukemia cells:
Possible involvement of a redox mechanism." International Journal of
Cancer 2002, 97, 593-600.

L. J. Gudas, "Retinoids, retinoid-responsive genes, cell-differentiation, and
cancer." Cell Growth and Dijferentiatian 1992, 3, (9), 655-662.

A. C. Ross, "Cellular-metabolism and activation of retinoids-roles of
cellular retinoid-binding proteins." FASEB Journal 1993, 7, (2), 317—327.

L. Stryer, "Vision: From photon to perception." Proceedings of the
National Academy of Sciences of the United States of America 1996, 93,
(2). 557-559.

H. 0. Olson J. A., "The enzymatic cleavage of B-carotene into vitamin A
by soluble exymes of rat liver and intestine." Proceedings of the National
Academy of Sciences of the United States of America 1965, 54, 1364-1370.

D. S. Goodman; H. S. Huang, "Biosynthesis of vitamin A with rat
intestinal enzymes." Science 1965, 149, 879-880.

A. Wyss; G. M. Wirtz; W. D. Waggon; R. Brugger; M. Wyss; A.
Friedlein; G. Riss; H. Bachmann; W. Hunziker, "Expression pattern and
localization of beta,beta-carotene 15,15 '-dioxygenase in different tissues."
Biochemical Journal 2001, 354, 521-529.

85

ll.

12.

13.

14.

15.

l6.

17.

18.

19.

20.

A. Wyss; G. Wirtz; W. D. Woggon; R. Brugger; M. Wyss; A. Friedlein; H.
Bachmann; W. Hunziker, "Cloning and expression of beta,beta-carotene
15,15 '-dioxygenase." Biochemical and Biophysical Research
Communications 2000, 271, (2), 334-336.

A. Takeda; T. Morinobu; K. Takitani; M. Kimura; H. Tamai,
"Measurement of retinoids and beta-carotene 15,15 '-dioxygenase activity

in HR-l hairless mouse skin with UV exposure." Journal of Nutritional
Science and Vitaminology 2003, 49, (1), 69-72.

J. van Lintig; K. Vogt, "Filling the gap in vitamin A research - Molecular
identification of an enzyme cleaving beta-carotene to retinal." Journal of
Biological Chemistry 2000, 275. (16). 11915-11920.

Qiagen, "Protein Expression System: QIAexpress Expression System."
The QIAexpressionist 2000, l.

M. T. Redmond; S. Gentleman; T. Duncan; 8. Yu; B. Wiggert; E. Gantt; S.
Cuningham, "Identification, expression and substrate specificity of a
mammalian beta-carotene 15-15'-dioxygenase." Journal of Biological
Chemistry 2001, 276, (9), 6560-6565.

von Lintig; A. Dreher; C. Kiefer; M. F. Wernet; K. Vogt, "Analysis of the
blind drosophila mutant ninaB identifies the gene encoding the key
enzyme for vitamin A formation in vivo." Proceedings of the National
Academy of Sciences of the United States of America 2001, 98, 1130-1135.

W. Yan; G.-F. Jang; F. Haeseleer; N. Esumi; J. Chang; K. Palcewski; D.
Zack, "Cloning and characterization of a human beta-carotene." Genomics
2001, 72, 193-202.

A. Lindqvist; S. Anderson, "Biochemical properties of purified
recombinant human beta carotene 15-15'-monooxygenase." Journal of
Biological Chemistry 2002, 277, (26), 23942-23948.

Qiagen, "Escherichia coli host strains." The QIA expressionist 2000, 15-
16.

Pal-Bhowmick; Ipsita; K. Sadagopan; Vora; K. Hardeep; Sehgal; Alfica;
Shanna; Shobhona; G. K. Jarori, "Cloning, over-expression, purification

86

21.

22.

23.

24.

25.

26.

27.

28.

29.

and characterization of Plasmodium falciparum enolase." European
Journal of Biochemistry 2004, 271, 4845-4854.

Qiagen, "Monoclonal anti-His antibodies for sensitive detection of 6xHis-
tagged proteins." QIAGEN News 1997, (3), 1.

Qiagen, "QIAexpress Detection and Assay Handbook." 2000, 11, 17-27.

E. Page; v. Strandmann; C. Zoidl; H. Nakhein; R. Holewa; P. Lorenz; L.
Hipab-Klein; G. U. Ryffel, "Highly specific and sensitive detection of
6xhist-tagged proteins using MRGS-His antibody." Qiagen News 1996,
(l). 9.

P. L. Zamorano; V. B. Mahesh; L. M. D. Sevilla; L. P. Bhat; G. K. Bhat;
D. W. Brann, "Expression and localization of the leptin receptor in
endocrine and neuroendocrine tissues of te rat." Neuroendocrinology 1997,
65. 223.

Y. Zhang; L. Taiming; J. Liu, "Low temperature and glucose enhanced T7
RNA polymerase-based plasmid stability for increasing expression of
glucagon-like peptide-2 in Escherichia coli." Protein Expression and
purification 2003, 29, (1), 132-139.

J. A. Olson, "Provitamin-a function of carotenoids-the Conversion of beta-
carotene into vitamin-A." Journal of Nutrition 1989, 119, (1), 105-108.

K. W. Walker; H. W. Gilbert, "Effect of redox environment on the in vitro
and in vivo folding of RTEM- 1 beta-lactamase and Escherichia coli
alkaline phosphatase." The Journal of Biological Chemistry 1994, 269,
(45), 28487-28493.

J. L. Napoli; K. R. Race, "Biogenesis of retinoic acid from beta-carotene.
Differences between the metabolism of beta-carotene and retinal." The
Journal of Biological Chemistry 1988, 263, 17372-17377.

T. Ikemura, "Correlation between the abundance of Escherichia coli
transfer RNAs and the occurrence of the respective codons in its protein
genes." Journal of Molecular Biology 1981, 146, (1), 1-21.

87

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

H. Dong; I. Nilsson; C. G. Kurland, "Co-variation of tRNA abundance and
codon usage in Escherichia coli at different growth rates." Journal of
Molecular Biology 1996, 260, 649-663.

C. Kurland; J. Gallant, "Errors of heterologous protein expression."
Current Opinion in biotechnology 1996, 7, 489-493.

E. Goldman; A. H. Rosenberg; G. Zubay; F. W. Stuider, "Consecutive
low-usage leucine codons block translation only when near the 5' end of a
message in Escherichia coli." Journal of Molecular Biology 1995, 245,
467-473.

Strategene.

N. Misawa; Y. Stomi; K. Kondo; A. Yokoyama; S. Kajiwara; T. Saito; T.
Ohtani; W. Miki, "Structure and functional analysis of a marine bacterial
carotenoid biosynthesis gene cluster and astaxanthin biosynthetic pathway
proposed at the gene level." Journal of Bacteriology 1995, 1995, 6575-
6584.

E. T. Wurtzel; G. Valdez; P. Matthews, "Variation in expression of
carotenoid genes in transfored Escherichia coli strains." Bioresearch
Journal 1997, 1,1-11.

G. Sandmann, "Carotenoid biosynthesis and biotechnological application."
Archives of Biochemistry and Biophysics 2001, 385 , 4-12.

M. Albrecht; N. Misawa; G. Sandmann, "Metabolic engineering of the
terpenoid biosynthetic pathway of Escherichia coli for production of the
carotenoids carotene and xeaxanthin." Biotechnology Letters 1999, 21 ,
791-795.

F. W. Studier; B. A. Moffatt, "Use of bacteriophage T7 RNA polymerase
to direct selective high-level expression of cloned genes." Journal of
Molecular Biology 1986, 189, 113-130.

A. H. Rosenberg; B. N. Lade; D. Churi; S. Lin; J. J. Dunn; F. W. Studier,
"Vectors for selective expression of cloned DNAs by T7 RNA
polymerase." Gene 1987, 56, (1), 125-135.

88

40.

41.

42.

43.

45.

46.

47.

48.

49.

50.

F. W. Strudier; A. H. Rosenberg; J. J. Dunn; J. W. Dubendorff, "Use of T7
RNA polymerase to direct expression of cloned genes." Methods in
Enzymology 1990, 185, 60-89.

New England Biolabs Catalog and technical reference. ed.; 2000-2001;
'Vol.' p 204.

Sambrook; Russel, Molecular cloning. 3rd. edition ed.; 1999; 'Vol.' p 159.

O. Seeburg; J. Shine; J. A. Martial; J. D. Baxter; H. M. Goodman,
"Nucleotide sequence and amplification in bacteria of structural gene for
rat growth hormone." Nature 1977, 270, 486-494.

A. Ulrich; J. Shine; J. Chirgwin; R. Pictet; E. Tscher; W. J. Tutter; H. M.
Goodman, "Rat insulin genes: construction of plasmids containing the
coding sequences." Science 1977, 196, 1313-1319.

F. A. O. Marston, "The purification of eukaryotic polypeptides synthesized
in Escherichia coli." Journal of Biochemistry 1986, 240, 1-12.

J. F. Kane; D. L. Hartley, "Formation of recombinant protein inclusion
bodies in Escherichia coli." Trends in Biotechnology 1988, 6, 95-101.

A. D. Guise; S. M. West; J. B. Chadhuri, Molecular Biotechnology 1996,
653-664.

Blackwell; Horgan, "A novel strategy for production of a highly expressed
recombinant protein in an active form." FEBS Letters 1990, 295, 10-15.

N. Burton; B. Cavallini; M. Kanno; V. Moncollin; J. M. Egly, "Expression
in Escherichia coli: purification and properties of the yeast general
transcription factor TFIID." Protein Expression and Purification 1991, 2,
432-441.

L. Lobe]; S. Pollak; J. Klein; J. W. Lustbader, "High-level bacterial
expression of a natively folded, soluble extracellular domain fusion protein
of the human luteinizing hormone/chorionic gonadotropin receptor in the
cytoplasm of Escherichia coli." Endocrine 2002, 14, (2), 205-212.

89

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

w. A. Prinz; F. Aslund; A. Holmgren; J. Beckwith, "The role of the
thioredoxin and glutaredoxin pathways in reducing protein disulfide bonds
in the Escherichia coli cytoplasm." Journal of Biological Chemistry 1997,
272, (25), 15661-15667.

E. L. Vallie; E. A. Diblasio; S. Kovacic; K. L. Grant; R. F. Schendel; J. M.
McCoy, "A thioredoxin gene fusion expression system that circumvents
inclusion body formation in the Escherichia coli cytoplasm."
Bioflechnology 1993, 11, 187-193.

W. D. Woggon; R. M. Wenger, "Supramolecular chemistry and molecular
recognition." Chimia 2000, 54, (1-2), 9-12.

W. D. Woggon, "The central cleavage of beta,beta-carotene-A
supramolecular mimic of enzymatic catalysis." Chimia 2000, 54, (10),
564-568.

J. Paik; A. During; E. H. Jarrison; C. L. Mendelsohn; K. Lai; W. Blaner,
"Expression and characterization of a murine enzyme able to cleave beta-
carotene." The Journal of Biological Chemistry 2001, 276, 32160-32168.

M. M. Altamirano; C. Garcia; L. D. Possani; A. R. Fersht, "Oxidative
refolding chromatography: folding of the scorpion toxin CnS." Nature
Biotechnology 1999, 17, 187-191.

Sambrook; Russel, "Molecular Cloning." 1999, (8).
PCR Metods applications. ed.; 1992; 'Vol.' p 51-59.

M. R. Ponce; J. L. Micol, "PCR amplification of long DNA fragments."
Nucleic Acid Research 1992, 20, (3), 623.

R. Peek; R. W. L. M. Niessen; J. G. G. Schoenmakers; N. H. Lubsen,
"DNA methylation as a regulatory mechanism in rat-crystallin gene
expression." Nucleic Acid Research 1991, 19, (1), 77-83.

E. D. Clark, "Protein refolding for industrial processes." Current Opinion
in Biotechnology 2001, 12, 202-207.

90

62.

63.

F. Baneyx, "Recombinant protein expression in Escherichia coli." Current
Opinion in Biotechnology 1999, 10, 411-421.

T. G. Jeffrey; F. Baneyx, "Divergent effects of chaperone overexpression
and ethanol supplementation on inclusion body formation in recombinant
Escherichia coli." Protein Expression and Purification 1997, 11, 289-296.

91

3.1 Introduction

Chapter 3

Studies on BCDOX

Oxidative cleavage of B-carotene is an unprecedented process for a

dioxygenase enzyme. Typically, dioxygenases require a hydrophilic handle such

as a hydroxyl group to proceed."2 Enzyme/substrate interactions that lead to

stereospecific catalytic

action through hydrogen bonding and dipolar

communication are common. However, the enzymatic mechanism of BCDOX is

intriguing in that it uses a molecule of oxygen to cleave an electronically

undistinguished double bond with a high degree of selectivity (Scheme 3-1). As

mentioned previously,
the postulated
mechanisms suggest
either the involvement

of a monooxygenase3

(Scheme 3-2A) or a
dioxygenase4 (Scheme
3-2B). Despite the fact
that it is known that the
15-15’ double bond is

selectively oxidized, the

‘\\\\\\ \\.
1

1 ecoox

We

Scheme 3-1. The enzymatic mechanism of BCDOX is
intriguing in that it seemingly utilizes a molecule of
oxygen to cleave an electronically undistinguished double
bond with a high degree of selectivity.

92

E]
W W

02 mWo-i-RVW : \ \ )\ \/

Retinal

 

 

 

Scheme 3-2. Postulated mechanisms. A) Dioxygenase, suggest the involvement
of a dioxetane intermediate. 8) Monoxygenase, suggest the involvement of an

epoxide intermediate.

mode of activation for inactivated double bonds is unknown. More importantly,
the oxidation of a somewhat featureless hydrocarbon with such high
regiospecificity is not well understood. Therefore, understanding the enzymatic
action on hydrocarbon substrates could offer insight into nature’s methodology for
utilizing other molecular interactions for recognition and specificity.

Even though the sequence for BCDOX is known, there is no 3D structure
available. Without any 3D structure there is no insight into the selective
mechanism for the oxidation of the double bond. Homology studies suggest that
different BCDOX proteins share overall sequence homology with a distinct
pattern of highly identical conserved stretches. As mentioned previously, the
highly conserved regions are located in the C-terrninal part of the protein.

Whether the conserved amino acids belong to the binding site of the iron or B-

93

carotene is not well understood. An alignment shows that around 8% of highly
hydrophobic amino acids are conserved in the enzyme, suggesting that these
amino acids participate in the formation of the substrate-binding pocket.s Binding
of the iron cation could be aided by 4 histidines and 6 acidic residues that are
absolutely conserved. In addition, as mentioned previously, a region contains a
very conserved arginine, a couple of histidines and a carboxylate and it is possible
that this amino acids might be involved in the binding site for iron cation.°

Our interest in the mechanistic aspects of BCDOX has led us to synthesize
the tritiated photoaffinity label [10’~3H]-8’-apo—B-carotenoic acid 3. The choice
of 3 as the photoaffinity probe stems from the fact that apo—B-carotenoic acids are
bad substrates for BCDOX, and thus will bind the enzyme, its oxidation to form
retinal is very slow.7'9 On the other hand, call-unsaturated polyene carboxylic
acids are capable of photo-activation and can be used as photoaffinity probes.
This was demonstrated by Rando and coworkerslo during the isolation of retinoic
acid binding proteins in which radiolabeled retinoic acid was used as the
photoaffinity labeling agent.ll Photoaffinity labeling uses light induced chemical
activation to form a covalent bond between a biological host (an enzyme or

protein of interest) with a photoreactive substrate.‘2"4

3H
\\\\\\\\ 02H
3

all-lrans-[1O’-3H]-8'apo-B-carotenoic acid

94

3.2 Synthesis of all-trans-[10’-3H]-8’apo-B-carotenolc acid

Synthesis of radiolabeled carotenoids are challenging for two reasons.
First, the polyolefinic nature of these compounds leads to their less desirable
stability, and thus introduction of radiolabels is generally limited to mild
reactions. Although there is not a large body of literature of the synthesis of

radiolabeled carotenoids (mostly biosynthesized by feeding of labeled precursors

15.16

to various organism), related compounds such as retinoids have been

synthesized by reduction of intermediate carbonyl group with Na/LiB3H4, or

17-20

semi-hydrogenation of intermediate acetylenes with 3H2. Secondly, their

sensitivity to light forces the
. . 1 'C the
need to work 1n darkrooms With 0 a n
\ \ \ \ \ \ \
minimal red lighting and ‘
increases the difficulty of routine JJ NaB3H4

chemical manipulations. In the

3H
synthesis of any radiolabeled W014
5

compound, it is preferable to u
incorporate the label as close to 10' Carbon 3
' ' \\\\\\\\C°2H
the final step as possrble. Thrs
3

reduces the need for handling

_ Scheme 3-3. 10' carbon was chosen as the

and purification of the
site to introduce the radioiabel hydrogen.
radrolabled intermediates, and NaBaHt was chosen to be the source of tritium

increases the radioactive yield 0f and thus required a reductive step to

the labeled compound. Although incorporate the label.

95

there is a range of positions present in 8’-apo—B-carotenoic acid 3 that could be
considered as appropriate to incorporate tritium, carbon 10’ was chosen as the site
to introduce the radiolabel. NaB3H4 was chosen to be the source of tritium and
thus required a reductive step to incorporate the label (Scheme 3-3). The
subsequent steps could be performed in high yields, without the need for extensive
manipulation and purification.

The synthesis of all-trans-[10’-3H]-8’apo-B-carotenoic acid was
accomplished as follows21 (Scheme 3-4). Construction of the Crs-Ylide22 8 starts
with a Grignard reaction between B-ionone (6) and vinyl magnesium bromide to
yield the (1,8 unsaturated alcohol 7 in quantitative yield. The alcohol 7 was
reacted with triphenylphosphine hydrobromide to afford the Cys-ylide 8 in 93%
yield. Construction of the 2,7-dimethyl-2,4,6-octatrienedia123'2‘ (12) was

accomplished by reacting trans-1.4-dichloro-2-butene (9) with triethyl phosphate

‘0 CH2=CHMgBr PPMHB' a gm
_ 7 \
98% ' 0H CH30H. CHCI, H' "a
6 7 (111), 93% 8

O=P(OE1)3 051 OF NaOi-l. K2003
C'Ml ————’ ( )2 NPO(E10)2+ Mao =
9

 

 

 

 

96% 90%
10
11
M9 AoOH.THF, o
0M9 12 97% 13

Scheme 3.4. Synthesis of all-trans-[t0'-’H]-8’apo-B-carotenoic acid. Pan I.

96

13

 

 

 

 

RAgPh?Br : n \ \ \ \ H0“, B-Carotene
s NaOMe. MeOH ,4 10%
RT. 30 min, 62%
Sealed tube 15 2. M1102. CHzclzt 0 °C. 511
90 °C. 5h. 71% 67% over two steps
1. NaBaH4 (255mCi/mmol).
NaHfrHF (1:1),
0 °C. 15 min 3
Rwy/V010 = 3W0 HO
2. 001102, CHzclz. 1
15 0 °C. 5h, 7

68% over two steps

PhaP=C(CHa)CC>25t ”H co 1. KOH, i-BuOH, reflux,2h

; Et Ar
THF, reflux, 5h. 30% “W 2 2. H230, (20%). 30%

3H
W: R W02
3
21 mCi/mmol

19% overall
radiochemical eld

 

 

 

 

 

 

 

Scheme 3-5. Synthesis of all-lrans-H O’-3Hl-8’aDo-B-carotenolc acid. Part ii.

to yield 2-butenyl-1,4-bisphosphonic acid tetraethyl ester (10) in quantitative
yield. Reaction of 9 with pyruvic aldehyde dimethyl acetal (10) afforded the
protected aldehyde 11 in 90% yield. Deprotection of the acetal groups gave 2,7-
dimethyl-2,4,6-octatrienedial (12) in 97% yield, Scheme 3-4.

Wittig olefination of 13 with ylide 8 was initiated with sodium methoxide

in methanol, which delivered the C25 aldehyde 14 in good yields with 97% E

97

stereoselectivity of the newly formed double bond. Treatment of aldehyde 14
with (carbethoxymethylene)triphenylphosphorane led to 15 (97% E) in 71% yield.
The ethyl ester 15 was reduced with DIBAL, and the resulting allylic alcohol was
oxidized with Mn0225 to provide Cn-aldehyde 16. The tritium label was
incorporated by the reduction of aldehyde 16 with NaB3I-14 at 0 °C, which was
immediately reoxidized with MnOz to yield the radiolabeled Cn-aldehyde 17.
Wittig olefination of 17 with (carbethoxyethylidene)triphenylphosphorane
proceeded well to deliver ester 18 as the sole product (Z isomer was not detected
by NMR spectroscopy). Hydrolysis of the ethyl ester 18 provided the
radiolabeled acid 3 (21 mCi/mmol) with a 19% overall radiochemical yield

(Scheme 3.5).

3.3 Enzymatic studies using of all-trans-B’apo-B-carotenoic acid and
BCDOX

The obtained 8’apo-B-carotenoic acid was used to perform enzymatic
studies with BCDOX. The initial studies were performed with non-labeled 8’-
apo-B-carotenoic acid. A typical assay contained an initial concentration of 200
“M of 8’-apo-carotenoic acid, and 20 ug of BCDOX. The enzymatic reaction
was stopped at different times and HPLC was used to monitor the formation of

retinal. A normal phase column (C5), hexane: ethyl acetate (95:5) was used as a

mobile phase at a flow rate of l mllmin. The results showed that 3 was a substrate

98

pmol retinal

 

 

 

llM all-transna'apo-B-carotenoic acid

Figure 3-1. Inhibition studies using ali-trans-[1O'-3H]-8'apo-B-
carotenoic acid. This graph shows that in presence of B-
carotene, the more ali-trans-B'apo-B-carotenoic acid is present,

the less amount of retinal is formed.

with an activity of 2.1 pmol retinal/mg of BCDOXomin. When only B-carotene is
used as a substrate, the obtained VMAX was 194 pmol retinal/mg of BCDOXomin.
These results agree with previously reported data in which apo-carotenals and
ape-carotenoic acids are substrates for BCDOX, but its oxidative cleavage at the
15-15’ bond is not as efficient as B-carotene.7'9 Therefore, 3 can be considered as
a slow substrate for BCDOX. Inhibition studies with 3 were performed to
investigate if the substrate was binding the same active site as B-carotene.
Different concentrations of apo-carotenoic acid were tested. First, the BCDOX
was incubated with the apo-carotenoic acid for 15 minutes, and then B-carotene
was added. The outcome showed as expected that 8’-apo-carotenoic acid inhibits

the oxidation of B-carotene to retinal (Figure 3-1). The fact that the 8’- apo-

99

carotenoic acid binds the site of oxidation in BCDOX and is a slow substrate for
the oxidation of BCDOX satisfies one of the requirements for the development of

a photoaffinity probe for BCDOX.

3.4 Failed photolabeling attempts using all-trans-S’apo-B-carotenoic acid
and BCDOX

Preliminary crosslinking studies using [lO’-’H]-8’-apo—B-carotenoic acid
were performed to explore the viability of this method. All the experiments were
performed under red safe-light illumination. A mixture of the protein and [10’-
3H]-8’-apo-B-carotenoic acid was incubated for 15 minutes and then irradiated
with a long-wave UV light (360 nm) at 0 °C for 30 minutes. For controls, three
more samples were prepared. A sample containing the same concentration of
[10’-’I-I]-8’-apo—B-carotenoic acid and B-carotene, a sample with only B-carotene,
and a third sample with [lO’-3H]-8’-apo-B-carotenoic acid that was not subjected
to irradiation. Following irradiation, each sample was subjected to electrophoresis
to remove [10’-3H]-8’-apo-B-carotenoic acid that did not cross-linked to
BCDOX. The band that corresponded to BCDOX was separated and the
radioactivity present in each sample was measured.

The same radioactivity was found in all the samples, suggesting that no
cross-linking between the BCDOX and the acid had occurred. In the first attempt.
the buffer used contained a—tocopherol but its presence could interfere with the
mechanism of generation of the photolabel active species of ogB unsaturated

acids. It has been postulated that 01.8 unsaturated acids cross-link via a di-radical

100

intermediate, which first abstract a

OH
dro n ' 'v ' H
I: ge radical from the am e srteand [RN/(:0 ___"L_. fi"
en the two newly created radicals
combine rapidly (Scheme 3-6). It is I
known that cl-tocopherol26 is a radical H RAiH
quencher, therefore, the experiment was \ 0H .- \ i”

repeated with out a-tocopherol, but the
Scheme 3-6. The proposed mechanism

results obtained were the same. It is
for cross linking of cap-unsaturated acids

immm‘, ‘° “mice that the final involves a diradical lnterrnedlate
concentration of the enzyme was only 1.2
HM and attempts to concentrate the sample further resulted in loss of activity.
Whether the loss of activity was due to mechanical manipulation or protein
aggregation was not clear. Consequently, it is possible that low concentration of
protein resulted in no crosslinking of the protein with the [10’-3H]-8’-apo—8-
carotenoic acid. As mentioned previously 0,8 unsaturated acids have been used
extensively in photoaffinity labeling, however, their efficiency to cross-link to a
neighboring amino acid is not very high.13 Further optimization should be
performed to detennine whether it is the 0t,B unsaturated acid photo labeling
group that is hampering the cross linking process, or if it is the low concentration
of the protein.

As discussed previously in Chapter 2, attempts to express BCDOX in large

quantities in bacteria were not successful. The lack of large amounts of protein

has hampered further photoaffinity labeling studies.

101

3.5 Attempts to develop a rapid and efficient assay for BCDOX

As a consequence of the tedious extraction and the need for HPLC
analysis required for each BCDOX assay. the collection of kinetic data and
mechanistic studies have been difficult. There are two major problems with the
assays performed using HPLC to identify the products. First, for every assay the
carotenoids and retinoids have to be extracted from the aqueous solution and then
analyzed. Second, when the metabolites are extracted, non-selective oxidation
could occur resulting in false positives. Thus. an attempt to develop a new
sensitive and efficient assay for BCDOX was performed. The assay designed
exploits the difference in polarity of B-carotene and retinal. A partition assay
consists of the distribution of a known substance between two immiscible solvents
(partition coefficient K, is defined as the ratio of equilibrium concentration of the
substance in the octanol rich phase (A51) to the water rich phase (Asz)K=A31/Asz;
logP is the octanol-water partition coefficient). The assay will contain the
immiscible solvents water and isooctane. Due to differences in their polarity the
B-carotene and retinal will have different distributions in those solvents (Figure 3-
2). The partition coefficients (logP) of B-carotene, retinal, retinol and retinoic
acid are 17.62, 7.60, 5.68 and 6.30, respectively.27 Thus, fi-carotene has a strong
affinity (solubility) for isooctane and low affinity for water. However, solubility
of retinoids in water is also poor. To overcome this limitation, different co-
solvents were used (acetonitrile, dimethyl sulfoxide, 2-chloroethanol, 4-

chlorobutyronitrile and ethanolamine). These solvents are soluble in water and

102

insoluble in isooctane. They should increase the solubility of the desired
metabolites into the aqueous phase while at the same time not increase the
solubility of B-carotene into the aqueous phase. BCDOX oxidizes B-carotene into
retinal, thus, the partition assay would extract only retinal into the aqueous phase,

and B-carotene into the organic phase

T)
r
\/

 

 

 

 

 

 

 

9

Figure 3-2. Development of a facile assay to test BCDOX activity.
Partition assay consists of the distribution of a known substance between
two immiscible solvents. The assay will contain the immiscible solvents
water and isooctane, due to the differences in polarity of B-carotene and
retinal they will have different distribution in those solvents. B-Carotene has
a strong affinity (solubility) for isooctane and low affinity for water. A co-

solvent is added in the aqueous phase to increase the solubility of retinoids

in water.

103

To increase the sensitivity and efficiency of this assay, radioactive
materials were used. A small aliquot from the enzymatic assay was added to a
tube containing both aqueous and organic phases. The B—carotene and the retinal
would then distribute differently into the two phases and finally each phase would
be quantified by measuring the radioactivity. To establish the feasibility of this
assay first trans-[15,15’-3H]-B-carotene (21). all-trans-[lS-BI-fl-retinal (20) and
all-trans-[15-3H]-retinol were prepared.

Radioactive all-trans-[l5,15’-3H]-B-carotene (21) and all-trans-[15-3H]-
retinal (20) were synthesized as shown in Scheme 3-7. The all-trans-retinal (20)
was reduced with NaB3I-14 to produce all-trans-[15-3H]-retinol. The oxidation of
the labeled retinol with activated Mn0225 yielded all-trans-[15-3H]-retinal (20).
McMurry coupling?"30 of 20 afforded the radioactive all-trans-[15,15’-3H]-B-

carotene (21).

 

1. NaBaH4,MeOH 3
\ \ \ \ CH0 e \ \ \ \ Ho
19 2. Mnog (activated) 20

86%. 31.2 mCi/moi

1101,, mm,

Proton sponge
88%, 135.5 mCi/moi

 

Scheme 3-7. Synthesis of 3H labeled B-earotene and retinal.

104

The first step towards the development of the partition assay was to
develop a system that extracts all-trans-[lS-3H] retinal into the aqueous phase
without contamination of all-trans-[15,15’-3H]-B-carotene. After a series of
different extractions it was establish that the optimum volume to work with was 1
mL of isooctane (organic phase), 1 mL of the aqueous phase (different solvents)
and 0.1 mL of water or buffer. The optimal extractions of B—carotene, retinal and
retinol with different solvents in the aqueous phase are shown in Figure 3-3. The
graph shows that acetonitrile (Figure 3-3) is the best solvent to use in the aqueous
phase. Acetonitrile hardly extracts B-carotene in the aqueous phase (1.8%) while
extracting 43.5% of retinal in the aqueous phase. Unfortunately when the exact

assay conditions were used, due to the presence of Tween (detergent) the

 

 

 

 

% radioactivity in the aqueous phase

Figure 3-3. Extraction of fi-carotene( ), retinal (0). retlnol (o).
1.Water. 2.DMSO. 3.Elhanolamine, 4.Acetonitrlle. 5.2-Chloroethanol,

8.4-Chlorobutyronitrile.

105

extraction of retinal in the aqueous phase was decreased to only 18.3%. To
improve the percentage of retinal soluble in the aqueous phase, the assay was
quenched with hydroxylarrrine. Hydroxyiamine would form a Schiff Base with
the aldehyde, thus increasing its solubility. This indeed increased the percentage
of radioactivity in the aqueous phase to about 40%. However, this result is not
very accurate because addition of hydroxylamine resulted in the formation of
three phases. Thus an accurate measurement to quantify the concentration of
retinal was not possible.

The assay was tested using BCDOX and labeled [15,15’-3H] B-carotene.
As discussed in Chapter 2 the VMAX for the BCDOX used was l94pmol
retinal/mgomin. The final mixture contained B-carotene 7.1 11M, and of [15,15’-
3HHS-carotene (1 uL, 0.4 mM, 136 mCi/mmol). The assay was performed as
previously explained in Chapter 2. After the assay was quenched with
formaldehyde, 100 uL were added into 1 mL of isooctane and 1 mL of
acetonitrile. The amount of tritium in each phase was quantified, and only about 2
% of radioactivity was found in the aqueous phase. Unfortunately, the difference
between 2.0 and 1.8% was not sufficient to confirm a difference between the
retinal formed by the enzyme and the B-carotene extracted by the aqueous phase.
As mentioned previously, attempts to concentrate the protein were unsuccessful.
Thus further optimizations should be performed to optimize the extraction of
retinal to the aqueous phase. The activity of BCDOX is pretty low compared to
most enzymes. It is possible that such low activity of BCDOX originates from the

fact that the product obtained is retinal. As mentioned previously retinal is the

106

major source of retinoic acid in vivo. The concentration of retinoic acid is tightly
controlled”34 thus it is possible that BCDOX is the first step to control the
amount of retinal available in the cells. BCDOX is expressed in high amounts in
duodenal vile, lung, kidney, intestine, brain, stomach and testis.”38 All these
tissues require retinoids as cofactors to regulate a variety of functions.

Due to the fact that expression of large quantities of BCDOX was not
feasible using bacterial systems, progress in the development of a feasible assay
was slowed down. Until a successful system to express BCDOX in large scale is
established, a successful development of an efficient assay as well as a good

photolabile probe cannot be achieved.

107

3.6 Materials and Methods

A. Synthesis of Compounds

All reactions were carried under an atmosphere of nitrogen and removal of
solvents was performed under reduced pressure with a Buchi rotatory evaporator.
THF and EIZO were freshly distilled from sodium/benzophenone, and CH2C12 was
distilled over CaHz under a nitrogen atmosphere. Radioactive NaB3I-14 was
purchased from American Radiolabeled Chemicals, Inc (St. Louis, MO).
Analytical TLC was carried out using Merck 250 mm Silica gel 60 F254 and spots
were visualized under UV light. Column chromatography was conducted using
Silicycle silica gel (230-400 mesh). 300 MHz 1H-NMR and 75 MHz l3C-NMR
spectra were recorded on a Varian Gemini-300 or 500 instrument, and the residual
protic solvent (CDC13 or DMSO-d6) was used as internal reference. UV-visible
spectra were recorded on a Perkin-Elmer Lamda 40 spetrometer. A Waliac
WinSpectral 1414 liquid scintillation counter was used for quantification of
radioactivity. In a typical measurement 1-10 mL of sample was added to 3 mL of
Optipase ‘HiSafe’ 3 liquid scintillation cocktail (Wallac) and the solution was
counted for 1 rrrin. All the reactions were carried out in a darkroom under

minimal photographic safety red lights.

108

1. Synthesis of 2,7,1l-Trimethyl-14-(2,6,6-trimetyl-cyclohex-l-enyl)trideca-
2,4,6,8,10,12 hexaenal (l3)

0H(;/1\/\/\'/CHO WOW)
12

13

Sodium methoxide (121 mg, 2.250 mmol, 1.5 eq.) was added to a solution
of 12 2‘ (817 mg 1.5 mmoi) in dry methanol (8 mL) at 0 °C, and stirred until the
phosphonium salt was completely converted to the deep red phosphorane. A
solution of 7-dimethyl-2,4,6-trienedial 23 (164 mg, 1 mmol) in dry methanol (1
mL) was added and the mixture was stirred for 30 minutes. The reaction mixture
was diluted with water (10 mL) and extracted with diethyl ether (3 x 5 mL). The
organic layer was dried over Na2804 and the solvent was removed in vacuo to
give the Czs-aldehyde 13 (217 mg, 6 2%, 97:3 El), and B-carotene 1 (57 mg,
10%), which was purified by flash silica chromatography (2% ethyl acetate in
hexane).[lH NMR (300 MHz CDCI3): 8 9.4(1H, s), 7.03(1H, d, 1:12 Hz),
6.96(1H, dd, J=14.4, 12.3 Hz ), 6.77(1H, dd, J=14.4, 11.4 Hz), 6.70(1H, d, J=12.3
Hz), 6.50(1H, d, J=11.7 Hz), 6.41(1H, dd, J=14.4, 11.7 Hz), 6.12(3H, m),
2.00(3H, s, CH3), l.97(3H, s, CH3), 1.86(3H, s, CH3). 13‘0 NMR (300 MHz
CDCla): 8 194.38, 148.94, 141.72, 137. 88, 137.71, 137.42, 136.66, 136.22,
130.73, 130.30, 129.74, 127.74, 127.74, 127.63, 127.16, 39.51, 34.17, 33.03.

28.89, 26.81, 21.70, 19.13, 12.97, 12.77, 9.51. UV: hm, (hexanes) 414 nm

109

2. Synthesis of 4,9, l3-trimethyl-15-(2,6,6-trirnethyl-cyclohex-1-enyl)-

pentadeca-2,4,6,8,10,12,14-heptaenoic acid ethyl ester 14

Wcofit
1 4

A solution of 13 (110 mg, 0.31 mmol) and (carbethoxymethylene)-
tripehylphosphorane (323 mg, 0.93 mmol) in dry THF (4 mL) was heated at 95 °C
for 5 hours in a sealed tube. The reaction mixture was cooled to room
temperature and diluted with water (10 mL) and E120 (10 mL). Phases were
separated and the aqueous phase was extracted with 320 (2 x 10 mL). The
combined organics were dried over Nazsoa, filtered and concentrated under
reduced pressure. Crude product was purified by flash silica chromatography (5%
ethyl acetate in hexane) to yield 14 (93 mg, 71%, >97% E). The coupling
constant for the newly formed double bond confirms the E stereochemistry. The
Z isomer was not detected by NMR spectroscopy. [1H NMR (300 MHz CDC13):
5 7.35(1H, d, J=15.6 Hz), 6.79(1H, dd, J=14.7, 15.6 Hz), 6.78(1H, d, J=12.9 Hz),
6.65(1H. dd, J=13.5, 11.7 Hz), 6.45(1H, d, J=11.1 Hz), 6.33(1H, d, J=14.7 Hz),
6.23(1H, d, J=11.7 Hz), 5.96(3H, m), 5.84(1H, d, J=15.6 Hz), 4.20(2H, q, CH2, J:
7.2 Hz), 2.01(2H, m, CH2), l.97(3H, s, CH3), l.96(3H, s, CH3), 1.89(3H, s, CH3),
l.69(3H, s, CH3), l.60(2H, m, CH2). 1.45(2H, m, CH2), 1.28(3H, t, OCHZQH3, J:
7.2 Hz), l.00(6H, s, 2 x CH3). 13C NMR (300 MHz CDC13): 8 167.51, 148.72,
139.25, 138.93, 137.79, 137.58, 136.91, 136.73, 133.74, 133.36, 131.57, 130.56,
129.54, 128.60, 127.16, 126.28, 116.23, 60.17, 39.55, 34.20, 33.07, 28.92, 21.75,

19.18, 14.31, 12.90, 12.76, 12.53. UV: hm... (hexanes) 423 nm]

110

3. Synthesis of 4,9,13-trr'methyl-15-(2,6,6-trimethyl-cyclohex-1-enyl)-

pentadeca-2,4,6,8,10,12,14-heptaenal 15

\\\\\\\CH0 WOH
15 18

To a cold (0 °C) solution of 14 (90 mg 0.115 mmol) in dry EtzO (6 mL)
was added 1 M solution of DIBAL in cyclohexane (0.43 mL, 0.43 mmol) and
stirred for 30 minutes. Saturated sodium-potassium tartrate solution (3 mL) and
glycerol (5 drops) were added. The reaction mixture was stirred over night at
room temperature; phases were separated and the aqueous phase was extracted
with 320 (3 x 10 mL). The combined organic layers were washed with brine (3
mL), dried over Na2804, filtered and concentrated in vacuo to give the
corresponding allylic alcohol, 16 which was pure enough to be used without
further purification (80 mg, 90%). Spectra for 16 [1H NMR (300 MHz CDCI3):
8 6.67(3H, m), 6.22(7H, m), 5.84(1H, dt, J=12.0 J=6.0 Hz), 4.22(2H, t, J=6.0 Hz),
2.00(2H, m, CH2), l.94(6H, s, 2 x CH3), 1.89(3H, s, CH3), 1.70(3H, s, CH3),
l.60(2H, m, CH2), 1.45(2H, m, CH2), 1.00(6H, s, 2 x CH3). 13C NMR (300 MHz
CDC13): 6 137.82, 137.68, 137.06, 136.68, 136.27, 136.09, 134.62, 132.32.
131.96, 130.68, 130.35, 129.35, 127.28, 126.66, 125.14, 63.91, 39.55, 34.21.
33.05, 28.92, 21.73, 19.20, 12.82, 12.78, 12.71. UV: hm. (hexanes) 400, 422 nm]

MnOz (100 mg, 1.16 mmol, 20 eq.) was added to a stirred solution of the
aforementioned allylic alcohol 16 (22 mg 0.6 mmol) in CH2C12 (3 mL) at 0 °C and
stirred for 5 hours and stirred for 5 hours. The reaction mixture was filtered

through a pad of celite and washed with CHzClz (4 x 0.5 mL). Solvent was

111

removed under reduced pressure and the crude aldehyde was purified by flash
silica column chromatography (5% ethyl acetate in hexane) to yield 15 (15 mg,
68%). Spectra for 15 [‘H NMR (300 MHz ouch): 8 9.57(1H, d, J=7.8 Hz),
7.13(1H, d. 1:153 Hz), 6.82(1H, dd, J=13.8, 11.7.0 Hz), 6.73(1H, dd, J=14.7.
11.4 Hz), 6.53(2H, m), 6.35(1H, d, J=15 Hz), 6.25(1H, d, J=12.0 Hz), 6.14(4H,
m), 2.01(2H, m, CH2), 2.00(3H, s, CH3), l.96(3H, s, CH3), l.94(3H, s, CH3),
1.70(3H, s, CH3), l.60(2H, m, CH2), 145(211, m, CH2), 1.01(6H, s, 2 x CH3). 13C
NMR (300 MHz CDCI3): 5 193.74, 156.62, 141.21, 140.06, 137.81, 137.56.
137.44, 136.57, 135.36, 133.61, 131.43, 130.50, 129.73.128.36, 127.53, 127.08.
126.93, 36.61, 34.25, 33.11, 28.95. 21.76, 19.21, 12.99. 12.82, 12.70. UV: hmx

(hexanes) 436 nm]

4. Synthesis of 18 and reduction of 15 NaB3Ha.

SH SH

18 17

Aldehyde 15 .(15 0.04 mmol) was added to a cold (0 °C) solution of
NaB31-I, (1.8 mCi 255 mCi/mmol) in THF/MeOH (4 mL, 1:1). The reaction
mixture was stirred for 20 min at which time unlabeled NaBI-L (1.5 mg, 0.04
mmol) was added) to the reaction mixture (stirred for 15 minutes). The reaction
was quenched by addition of saturated NH4C1 and extracted with 320 (3 x 2 mL).

Phases were separated and the organic layer was washed with saturated NaCl, and

112

dried over Na2804. Removal of solvent under reduced pressure gave the
corresponding allylic alcohol 17 (13 mg, 93%). The allylic alcohol 17 was used
without further purification. After removal of solvent under reduced pressure, the
crude residue was dissolved in CHZClz (2 mL), freshly prepared W02 (58 mg, 20
eq. 0.68 mmol) was added at 0 °C, and the reaction was stirred for 5 h. The
suspension was filtered through a pad of celite, washed with CH2C12 (1 x 2 mL),
and concentrated under reduced pressure followed by chromatographic
purification over silica gel (5% ethyl acetate in hexanes) to yield pure aldehyde 18
(9.6 mg 73%). The latter yield refers to product obtained from non-radioactive
syntheses. The radioactive material was used without purification following
extraction and removal of CH2C12 in the next step and the amount isolated was

assumed to be the same as for the non-radioactive reaction.

5. Synthesis of 2,6,11,lS-Tetramethyl-17-(2,6,6-trimethyl-cyclohex-1-enyl)-

pentadeca-2,4,6,8,10,l2,14,16-octaenoic acid ethyl ester 19

3H '
W02Et
19

(Carbethoxyethylidene)triphenylphosphorane ((18 mg 0.05 mmol) was
added to a solution of 18 (9.6 mg, 0.025 mmol) in T1-IF(3.5 mL) and the reaction
was refluxed for 5 h. The reaction mixture was diluted with water (4 mL) and
extracted with EtZO (3 x 2 mL). The combined ether extracts were dried over

Na2804. The solvent was removed under reduced pressure and the crude product

113

was purified by flushing it through a small pad of silica in a Pasteur pipette (5%
ethyl acetate in hexane) to yield ethyl ester 19 (9.4 mg, 80%). During the
preparation of the radioactive material, the crude product was also flushed through
a small pad of silica, however, it was not isolated and characterized, but was used
directly in the next step. [1H NMR (300 MHz CDC13): 5 7.25(1H, d, J=1l.6 Hz),
6.59(4H, m), 6.40(2H, m), 6.25(1H, d, J=11.1 Hz), 4.21(2H, q, J=7.2 Hz),
2.00(2H, m, CH2), l.97(3H, s, CH3), l.95(3H, s, CH3), l.69(6H, s, 2 x CH3),
l.60(2H, m, CH2), 1.54(3H, s, CH3), 1.45(2H, m, CH2), 1.28(3H, t, ocmgi. J:
7.2 Hz), 1.00(6H, s, 2 x CH3). 13C NMR (300 MHz CDCl.): 8 168.52, 143.84,
138.74, 137.83, 137.80, 137.65, 136.94, 136.53, 135.71, 135.39, 131.99, 131.91.
130.67, 129.48, 129.32, 126.15, 125.72, 123.14, 60.46, 39.58, 34.22, 33.07, 29.65.
28.92, 21.73, 19.19, 14.32, 12.86, 12.81, 12.75, 12.68. UV: hm (hexanes) 440

nm]

6. Synthesis of [10’-3H]-8’-apo-B-carotenoic acid 20

3
H
W002i”.
20

KOH (25%, 2 mL) was added to a solution of 19 (9.4 mg, 0.02 mmol), in
iso-butyl alcohol (3 mL). The resulting mixture was heated at 95 °C for 2 h and
cooled to room temperature. H2804 (20%, 2 mL) was added and the reaction
mixture was again heated to 70 Tim 30 min. The organic layer was separated
and washed with hot water (3 x 1 mL). Dichloromethane (5 mL) was added to the

extracted iso-butyl alcohol resulting in two layers: the desired acid was extracted

114

into CH2Cl2. The CH2C12 layer was separated and dried over Na2SOa, and
concentrated under reduced pressure. The crude product was purified through a
small pad of silica in a Pasteur pipette (5-15% ethyl acetate in hexane) to yield
pure acid (7 mg, 80% 21 mCi/mmol, 0.34 mCi, 19% of total radionuclei
incorporated) based on comparative TLC, and UV with authentic cold material,

and all spectroscopic data were consistent with reported values.

7. Synthesis [15-3H]-retinol 21

3H
W011
21

Retinal (15 0.04 mmol) was added to a cold (0 °C) solution of NaB’H.
(1.8 mCi 265 mCi/mmol) in THF/MeOH (4 mL, 1:1). The reaction mixture was
stirred for 20 min at which time unlabeled NaBHa (1.5 mg, 0.04 mmol) was
added) to the reaction mixture (stirred for 15 rrrin). The reaction was quenched by
addition of saturated NH4C1 and extracted with 320 (3 x 2 mL). Phases were
separated and the organic layer was washed with sat. NaCl, and dried over
Na2SO4. Removal of solvent under reduced pressure gave the corresponding
retinol 21 (87%). 15-[3H]-retinol, 88 mCi/mmol. [1H NMR (300 MHz CDC13):
8 6.60(1H, dd, 1:15.38, 14.83 Hz), 6.27(1H, d, 1:15.38 Hz), 6.15(1H, d J=13.19
Hz), 6.10 (2H, m), 5.67(1H, t, J=7.14), 4.29(3H, d, J=6.59), 2.30(3H, s, CH3),
l.97(3H, s, CH3), l.67(3H, s, CH3), 1.85(3H, s, 2 x CH3). 13C NMR (300 MHz

CDC13): 8 137.82, 137.06, 136.27. 134.62. 132.32, 130.68, 129.35, 127.28,

115

126.66, 125.14, 64.51, 39.55, 34.21, 28.92, 21.73, 19.20, 12.82, 12.78, 12.71.

UV: hm (hexanes) 325 nm]

8. Synthesis [15-3H]-retinal 22

3H
We
22

The retinol 21 was used without further purification. After removal of
solvent under reduced pressure, the crude residue was dissolved in CH2C12 (2
mL), freshly prepared MnO2 (58 mg, 20 eq,. 0.68 mmol) was added at 0 °C, and
the reaction was stirred for 5 h. The suspension was filtered through a pad of
celite, washed with CH2Cl2 (1 x 2 mL), and concentrated under reduced pressure
followed by chromatographic purification over silica gel (5% ethyl acetate in
hexanes) to yield pure retinal 22 (50 mg, 86% yield). The radioactive material
was used without purification following extraction and removal of CH2Cl2 in the
next step and the amount isolated was assumed to be the same as for the non-
radioactive reaction. 15-[3H]-retinal, 31.2 mCi/mmol. [1H NMR (500 MHz
CDC13): 6 10.08(1H, d, J=8.34 Hz), 7.12(1H, dd, 1:15.34, 13.19 Hz), 6.37(1H, d,
1:15.34 Hz), 6.34(1H, d, J=13.19 Hz), 6.18 (1H, d, J=15.00 Hz), 6.16 (1H, d,
1:17.01 Hz), 5.95(1H, d, J=8.39), 2.30(3H, s, CH3), 2.00(3H, s, CH3), l.67(3H, s,
CH3), 1.10(3H, s, 2 x CH3). 13'C NMR (300 MHz CDC13): 8 190.5, 154.6,
140.8137.6, 137.0, 134.3, 132.2, 130.2, 129.8, 129.4, 129.0, 39.6,34.1, 33.3, 29.0,

21.7, 19.0, 13.0. UV: hm, (hexanes) 369 nm]

116

9. Synthesis [15-31-11-8-carotene 23

3H
\ \ \ \ 3H \ \ \ \
23

In the dark, LiAlHa (1 eq, 9 mmol) was added to a stirred slurry of TiCl3
under nitrogen in dry THF at 0 °C. The reaction was exothermic and a rapid
change in color was observed. Retinal (15 mg, 5 mmol) was added dropwise
and it was warmed to RT. Then the reaction was refluxed for four h. The
reaction was quenched slowly with NaHCO3 and exacted with hexanes 3x. The
hexane extract was dried under anhydrous Na2804 and concentrated under
vacuo A pipette silica column eluted with hexanes afforded 12 mg of 8-
carotene in a giving a 80% yield 15-[3H]- 8,8 carotene, 136.5 mCi/mmol. [1H
NMR (300 MHz CDC13): 5 7.25(1H, d, J=1 1.6 Hz), 6.59(4H, m), 6.40(2H, m),
6.25(1H, d, J=11.1 Hz), 4.21(2H, q, J=7.2 Hz), 2.00(2H, m, CH2), l.97(3H, s,
CH3), l.95(3H, s, CH3), l.69(6H, s, 2 x CH3), l.60(2H, m, CH2), 1.54(3H, s,
CH3), 1.45(2H, m, CH2), 1.28(3H, t, OCH2_C_I_-_I3, J: 7.2 Hz), 1.00(6H, s, 2 x
CH3). 13C NMR (300 MHz CD013): 8 168.52, 143.84, 138.74, 137.83, 137.80.
137.65, 136.94, 136.53, 135.71, 135.39, 131.99, 131.91, 130.67, 129.48,
129.32, 126.15, 125.72, 123.14, 60.46, 39.58, 34.22, 33.07, 29.65, 28.92, 21.73,

19.19, 14.32, 12.86, 12.81, 12.75, 12.68. UV: hm (hexanes) 440 nm]

117

Assay studies
All the reactions were carried out in a darkroom under minimal

photographic safety red lights.

B.1 Preparation of 8-carotene micelles

Five different stocks of 8-carotene were prepared. 8-carotene (81 11L of a
solution 875 mM) was combined with taurocholate (100 11L) and of Tween 40
(3340 111., 4% w/v acetone). The solution was dried under a stream of nitrogen

and redissolved in Tris- KOH 10 mM buffer (10 mL, pH 8.0, 6 mM sodium

taurocholate, 0.5 mM DTT).

8.2 Preparation of 8’ ape-carotenoic acid micein

Five different stocks of 8’apo-carotenoic acid micelles were prepared (A-
E); by mixing 8’-apo-carotenoic acid (41, 42, 47, 54.7 and 82 11L) taurocholate
(5.56 mM, 10 11L) and Tween 40 (0.5 M and 334 111., 4% w/v acetone). The
solutions were dried under a stream of nitrogen and dissolved in Tris- KOH buffer

(1 mL, pH 8.0 6 mM sodium taurocholate, 0.5 mM DTT).

13.3 Inhibition studies using all-truns-S’apo-B-carotenoic acid
The final reaction mixture for the cleavage reaction contained 8’-apo-
carotenoic acid (A-E, 1, 10, 50, 100, and 200 M) 8-carotene (7.1 11M), (0.068

mM), Tris-KOH buffer pH 8.0, (sodium taurocholate(4.05 mM), DTT (0.34 mM),

118

Tween 40 (0.9%), a—tocopherol (50 11M», FeSOa (1.38 mM), enzyme (20 pg of
BCDOX).

The assay was started by adding the protein to solution A (solution A: 540
11L Tris- KOH buffer pH 8.0, 6 mM sodium taurocholate, 0.5 mM DTT, and 20
pl. of FeSOa). The solution was pre-incubated for 5 min at 30 °C. Then 8’-apo-
carotenoic acid (A-E, 1, 10, 50, 100, and 200 11M) was added and the mixture was
incubated for 10 min at 30 °C. Then the 8-carotene micelles solution (160 11L)
was added and the mixture was then incubated for 2-3 h at 30 °C. The reaction
was stopped by addition of formaldehyde ((200 11L, 46% in water) and was
incubated for another 10 min. The carotenoids and products were extracted
sequentially with chloroformzmethanol (1.5 mL, of a 1:2, v/v, 0.01% pyrrogallol)
and hexane (1.5 mL) and with chloroformzhexane (0.5 mL:1.5 mL). After
addition of each organic solvent, the reaction mixture was mixed using a vortex
for 40 seconds. The combined extracts were then dried under a steam of nitrogen
gas, and dissolved in isooctane: toluene (200 11L of (8/2, vlv) containing 0.01%
butylhydroxytoluene).

The formation of retinal at a different concentration of apo-carotenoic acid

was monitored. Analysis of the results suggests that 3 ‘inhibits’ the oxidation of

8-carotene to retinal, thus suggesting that it binds to the active site.

119

B.4 Radioactive extractions
Each one of the extractions was repeated six times. Test tubes containing
an aqueous phase and the organic phase were prepared. A variety of solvents

were used. The best system will be used as an example

 

 

Organic Aqueous Water or buffer
solvent co-solvent
isooctane 1 mL acetonitrile 0.9 mL 0.1 mL

 

 

 

 

 

Then the radioactive substrate was added

0 For studies involving carotene extraction 1 ML of 3H-8-carotene 136

mCi/mmol was used.

9 For studies involving carotene extraction 1 [AL of 15-[3H1-retinal 31.2

mCi/mmol was used.

0 For studies involving carotene extraction 1 11L of 15-[3H]-retinol, 88

mCi/mmol was used.

Each tube was covered with parafilm and vortex for 40 sec (speed 4). The
tubes were centrifuged 5 min at 2000 RPM. An aliquot (100 11L) of each were
transferred out of each phase and were analyzed for its content of radioactivity.
To each aliquot (100 lLL) of solvent (1 mL) of Opti Phase "Hi Safe 3" cocktail
was added to measure radioactivity. To determine the radioactivity in DPM the

sample plus the cocktail were put in 8 Liquid Scintillation Counter, Protocol:

3Montse. The percentage of radioactivity in each phase was calculated.

120

Protocol: 3Montse

Wallac 1414 WinSpectral v1.40 S/N4140295

Counting mode:
Quench index:
Isotope:
Counting time:
Repeats:
Cycles:
Replicates:

2 sigma %:
Minimum cpm:

Sp. Library of isotope 3H:

Vial type:

Liquid system:
Advaced modes:
Output to display:
DPMl

Additon to display:
Header:
Spectrum:
Window 1:
Window 2:
Window 3:
Window 4:
Window 5:
Window 6:

DPM

SQP(E)

3H, 3H=,12.43 y

60 min

1

l

l

0.01

0.00 checking time: 10
Wallac

clear (for 10 mL vials),
HiSafe

Chemlum

POS, RACKPOS, CPM, CTIME, SQPI,

Listing, Spectrum
3H

Beta

1-1024/Beta
1-1024/Beta
1-1024/Beta
1-1024lBeta
1-1024/Beta
1-1024/Beta

121

Example:

211915591.
jg
can:

Retinol

3 Montse
P03118079
DEC-04-2000

Isooctane 1 ml
2-Cl Ethanol 0.9 ml
Buffer 0.1 ml

 

 

Average
Standard Deviation

Variance
Confidence Intervals 95%
Confidence intervals 99%

122

41 .52
2.689
7.1ch

41.52 3.09
41.52 4.85

58.48
2.689
7.228
58.48 3.09
58.48 4.85

B.5 Assay of BCDOX activity using 3H-8-Carotene

Preparation of 8-carotene micelles

The final reaction mixture for the cleavage reaction contained 8-carotene
(7.1 ttM, 3Hid-Carotene (1 1.1L of (136 mCi/mmol), Tris-KOH buffer (8 mM pH
8.0), sodium taurocholate (4.05 mM), DTT (0.34 mM), Tween 40 (0.3%), 0t-
tocopherol (50 11M), FeSOa (1.38 mM), and BCDOX (roughly 20 ug of protein).

The assay was started by adding the protein to solution A (solution A: 540
11L Tris- KOH buffer pH 8.0, 6 mM sodium taurocholate, 0.5 mM DTT, and 20
11L of FeSO4). The solution was pre—incubated for 5 min at 30 °C. Then the 8-
carotene micelles solution (160 11L) was added and the mixture was then
incubated for 2-3 h at 30 °C. The reaction was stopped by addition of
formaldehyde ((200 uL, 46% in water) and was incubated for another 10 min.
The carotenoids and products were extracted sequentially with
chloroformzmethanol (1.5 mL, of a 1:2, v/v, 0.01% pyrrogallol) and hexane (1.5
mL) and with chloroform:hexane (0.5 mL:1.5 mL). After addition of each organic
solvent, the reaction mixture was mixed using a vortex for 40 seconds. The
combined extracts were then dried under a steam of nitrogen gas, and dissolved in
isooctane: toluene (200 11L of (8/2, v/v) containing 0.01% butylhydroxytoluene).
Then the radioactivity was measured (100 11L of the assay mixture were

transfened to a tube containing lmL of isooctane and 1 mL of acetonitrile) each

test was repeated six times. Each tube was covered with parafilm and vortex for

123

40 sec (speed 4). The tubes were centrifuged 5 min at 2000 RPM. 100 11L of
each were aliquoted out of each phase and were analyzed for their content of
radioactivity. To each (100 uL) phase (1 mL) of Opti Phase "Hi Safe 3" cocktail

was added to measure radioactivity.

124

3.6

References

E. 1. Solomon; T. C. Brunold; M. 1. Davis; J. N. Kemsley; S. K. Lee; N.
Lehnert; F. Neese; A. J. Skulan; Y. S. Yang; J. Zhou, "Geometric and
electronic structure/function correlations in non-heme iron enzymes."
Chemical Reviews 2000, 100, (1), 235-349.

F. Takuzo, Oxygenases and model systems. ed.; Klwer Academic
Publishers: Norwell, Ma, 1997; 'Vol.' 1, p 1.

M. G. Leuenberger; C. Engeloch-Jarret; W.-D. Woggon, "The reaction
mechanism of the enzyme catalyzed central cleavage of beta-carotene to
retinal." Angewandte Chemie International Edition 2001, 40, 2614-2617.

H. 0. Olson J. A., "The enzymatic cleavage of 8-carotene into vitamin A
by soluble exymes of rat liver and intestine." Proceedings of the National
Academy of Sciences of the United States of America 1965, 54, 1364-1370.

A. Lindqvist; S. Anderson, "Biochemical properties of purified
recombinant human beta carotene 15-15'-monooxygenase." Journal of
Biological Chemistry 2002, 277, (26), 23942-23948.

W. Yan; G.-F. Jang; F. Haeseleer; N. Esumi; J. Chang; K. Palcewski; D.
Zack, "Cloning and characterization of a human beta-carotene." Genomics
2001, 72, 193-202.

M. R. Lakshman; J. L. Pope; A. O. J, "The specificity of a partially
purified carotenoid cleavage enzyme of rabbit intestine." Biochemical and
Biophysical Research Communications 1968, 33, (2), 347-352.

R. C. Mordi; J. C. Walton; G. W. Burton; L. Hughes; K. Ingold; D.
Lindsay; D. Moffatt, "Oxidative degradation of B-carotene and B-apo-8'-
carotenal." Tetrahedron 1993, 49, 911-928.

H. Singh; H. R. Cama, "Enzymatic cleavage of carotenoids." Biochimica
et Biophysica Acta 1974, 370, 49-61.

125

10.

11.

12.

13.

14.

15.

l6.

17.

18.

19.

20.

P. S. Bernstein; S.-Y. Choi; Y.-C. Ho; R. R. Rando, "Photoaffinity
Labeling of Retinoic Acid-Binding Proteins." Proceedings of the National
Academy of Sciences of the United States of America 1995, 92, 654-658.

G. Chen; A. Radominska-Pandya, "Direct photoaffinity labeling of cellular
retinoic acid bindign protein I with all trans retinoic acid: Identification of
amino acids in the ligand binding site." Biochemistry, 2000, 39, 12568-
12574.

W. B. Jackoby; M. Wilchek, "Affinity Labeling." Methods in Enzymology
1977, 66, 69-114.

F. Kotzyba-Hibert;1. Kapfer, M. Goeldner, "Recent trends in photoaffinity
labeling." Angewandte Chemie International Edition 1195, 34, 1296-1312.

D. Schuster; a. et., "Photoaffinity Labeling." Photochemistry and
Photobiology 1989, 49, 785-804.

A. E. Purcell; W. M. Walter, "Preparation of MOB-Carotene." Methods in
Enzymology 1971, I8, (3), 701-706.

A. A. Liebman; W. Burger; S. C. Choudhry; J. Cupano, "Synthesis of
carotenoids specifically labeled with isotopic carbon and tritium." Methods
in Enzymology 1992, 213, 42-49.

P. Tosukhowong; T. Supasiri, Journal of labelled Compounds and
Radiopharmaceuticals 1985, 22, 843-850.

8. w. Rhee; J. r. Degraw; H. H. Kaegi, Journal of Labelled Compounds
and Radiopharrnaceuticals 1985, 22, 843-850.

H. H. Kaegi; J. E. Bupp; J. I. Degraw, Journal of Labelled Compounds
and Radiopharmaceuticals 1982, 19, 745-752.

P. L. Chien; M. S. Sung; D. B. Bailey, Journal of Labelled Compounds
and Radiopharmaceuticals 1979, 16, 791-797.

126

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

P. V. Reddy; M. Rabago-Smith; B. Borhan, "Synthesis of all-trans-[lO '-H-
3]-8 '-apo-beta-carotenoic acid." Journal of Labelled Compounds and
Radiopharrnaceuticals 2002, 45, (1), 79-89.

G. Britton; S. Liaaen-Jenses; H. Pfander, Carotenoids-Synthesis. ed.;
Verlangoston, 1995; 'Vol.' 2, p.

J. H. Babler 107,030, 1992.

P. Horst, "Synthesen in der Vitamin-A-Reihe." Angewandte Chemie 1960,
72, 811-819.

I. M. Goldman, "Activation of manganese dioxide by azeotropic removal
of water." Journal of Organic Chemistry 1969, 34, 1979-1981.

R. Yamauchi; N. Miyake; K. Kato; Y. Ueno, "Reaction of alpha-
tocopherol with alkyl and alkylperoxyl radicals of methyl linoleate."
Lipids 1993, 12, 201.

C. Hansch; A. Leo, In Exploring QSAR. Fundamentals and applications in
chemistry and biology. ed.; American Chemical Society:: Washington, D.
C, 1995; 'Vol.' p 125-168.

Y. Handa; J. Inanaga, "A highly stereoselective pinacolizatin of aromatic
and alpha-beta unsaturated aldehydes mediated by titanium(III)-
magnesiumGI) complex." Tetrahedron Letters 1987, 28, (46), 5717-5718.

J. E. M. Murry; M. P. Fleming, "A new method for the reductive coupling
of carbonyls to olefins. Synthesis of beta-carotene." Journal of the
American Chemical Society 1974, 94, (17), 4708-4709.

J. E. M. Murry; M. P. Flenring; K. L. Kees: L. R. Krepski, "Titanium-
Induced reductive coupling of carbonyls to olefins." Journal of Organic
Chemistry 1978, 43, (17), 32553265.

D. E. Ong, "A novel retinol-binding protein form rat." The Journal of
Biological Chemistry 1984, 259, (10), 1476- 1482.

127

32.

33.

34.

35.

36.

37.

38.

J. E. Smith; D. S. Goodman, "Retinol-binding protein and the regulation of
vitamin A transport." Federal Proceedings 1979, 38, 2504-2509.

S. Futtelman, and Saari, J. C., "Occurrence of ll-cis-retinal-binding
protein restricted to the retina." Investigative Ophthalmology and Visual
Science 1977, 16, 768-771.

Y. L W. Lai, B., Liu, Y.P., and Chader, G. J ., "Interphotoreceptor retinol-
binding proteins: possible transport vehicles between compartments of the
retina." Nature 1982, (298), 848-859.

G. Wolf, "The enzymatic cleavage of beta-carotene: still controversial."
Journal of Nutritional Reviews 1995, 53, 134-137.

C. Duszka; P. Grolier; E. M. Azim; M. C. AlexandreGouabau; P. Borel; V.
AzaisBraesco, "Rat intestinal beta-carotene dioxygenase activity is located
primarily in the cytosol of mature jejunal enterocytes." Journal of
Nutrition 1996, 126, (10), 2550-2556.

A. During; A. Nagao; C. Hoshino; J. Terao, "Assay of beta-carotene
15,15'-dioxygenase activity by reverse-phase high-pressure liquid
chromatography." Analytical Biochemistry 1996, 241, (2), 199-205.

T. V. Vliet; R. V. Schaik; H. V. D. Berg; W. H. P. Schreurs, "Effect of
vitamin A and beta-carotene intake on dioxygenase activity in rat
intestine." Annals of the New York Academy of Sciences 1993, 691, 220-
222.

128

Chapter 4

Introduction

4.1 Vision

Vision is one of the most intriguing and fascinating processes in nature.
Vision is the act or sense of light which involves the capture of light by the eye,
followed by its transmission to the brain as electrical pulses.1 These pulses are
recognized as images and colors in the brain. The eye is composed by specialized
components; the cornea, lens and the vitreous fluid that focus the light waves onto
the retina (Figure 4-1).3 The retina is located in the inner part of the eye, and
contains the photoreceptors. The photoreceptors, the rod and the cone cells
(Figure 4-2) are responsible for capturing the image and sending the signal to the
brain via the optic nerve.‘ An eye contains approximately 100 million rod cells

and about 6 million cone cells.5 The

Pupil Cornea

cone cells are responsible for color
(photopic) vision and the rod cells
are responsible for dim (scotopic)

vision. Cone cells are both quicker

   
 

to adapt to dark environments (~10 Vitreous

fillid
min) and sensitive to a broader range Retina
of wavelengths. On the other hand,
rod cells take longer to adapt to dark Flame 4-1- The Human Eye.

129

environments (20 - 30 min), but once adapted they are 500 times more sensitive

than cone cells.

rod

disk

cytoplasm

nochondna

 

nucleus

 

—synapse —-
Figure 4-2. The rod and cone cells in retina.

4.1.1 Process of vision

Rod cells are more abundant than cone cells but both cells are very similar
and contain a nucleus. Both cells differ from normal cells because they contain
two unique features: a synaptic ending and an outer segment.6 The synaptic
ending is where the neural signals pass through to the brain for processing. The
outer segment is filled with disks made of a phospholipid bilayer. Spanning the
bilayers of these disks is a membrane bound protein called opsin, which is the one

responsible for the visual signal transduction (Figure 4-3). There are four

130

different opsins. In the rod cells the opsin protein is bound to a chromophore (11-
cis-retinal) as a Protonated Shiff Base (PSB). This complex is called rhodopsin.
Rhodopsin (Rh) is a hydrophobic trans-membrane 7-a-helical protein
from the family of G—protein coupled receptors (GPCR). GPCRs are membrane-
bound proteins that activate a wide variety of biological functions including signal

transduction and regulatory control.7 In the 1930s Wald discovered that

Photoreceptor

Q
s‘.
u
s
‘s
‘\
‘~
‘
\ ‘5‘
‘~
.
‘Q
‘~
‘s
‘s
‘5

Outer
Segment

~
“‘
‘s
‘s
“Q
‘s
.
‘s
‘Q
‘s
Q

Inner

Segment 1

 

 

1 1 -cr's-retinal

Figure 4-3. Schematic representation of rhodopsin in a red cell.

131

rhodopsin is formed by an apoprotein called opsin

(Op) and a chromophore, a retinoid derivative, 8

\

which later on was identified as the aldehydic \Nfio:
form, retina1.9 In the 50’s it was suggested by

Collins10 and Morton,n that the retinal was bound ”9“". 4’4- A Schiff base ‘3

’ - 1 9d b 11- 'S i' I
to the protein as a protonated Schiff base (PSB) orm etween cr 1° "‘3

and a lysine residue in
via a Lys residue (Figure 4-4). In humans and

Rhodopsin.
most animals the chromophore is ll-cis-retinal. Each nod cell can contain up to
1700 disks and each disk can contain up to 1.5 million molecules of rhodopsin.12
At one time, one uses only about 10% of the rhodopsin complexes, thus allowing
vision to be a continuous process.

The visual process begins in the dark state where the chromophore, ll-cis-
retinal, is bound as a protonated Schiff base via Lys 296 (inactive state). When a
photon is absorbed it results in the isomerization of 11-cis-retinal to all-trans-
retinal. The isomerization of retinal results in a conformational change that
catalyzes the activation of a G-protein (transducin),l3 which activates a cyclic
GMP phosphodiesterase”l7 leading to a decline in the free GMP that results in
membrane hyperpolalization.’8‘2°

The isomerization can be monitored by low temperature UV-vis and other
spectroscopic methods.”24 This isomerization goes through a series of discrete
intermediates (Figure 4-5). The first observed intermediate, formed in less than

200 femtoseconds after irradiation, is known as photo rhodopsin, and is

considered to be a highly distorted all-trans-PSB chromophore. Formation of the

132

bathorhodopsin leads to a more stable distorted trans intermediate.”26 Then in
nano seconds, the lumi rhodopsin is formed followed by meta I and meta 11 states.
The meta 11 state is the only one that forms an unprotonated Schiff base.”28
Finally, the imine bond is protonated and hydrolyzed to afford the free all-trans-
retinal and the opsin protein.

The initial stages of the transduction pathway are controlled by steric
factors,25 while the latter stages primarily involve proton transfer processes.28 The

initiation of the visual transduction occurs at the meta 11 state, and the all-trans-

retinal is re-isomerized to the ll-cis-retinal in the retinal pigment epithelium.

11

 

 

i 11 \ \ \ \ ‘0
[W12 Opsin ail trans 385 nm
t\/\ \ t._Y—/ r

11-crs, 498 nm §NH® To 00
O
hvl<20013 Meta-ll, o c. 380 nm
Unprotonated
Schiff Base
11 12
'1 a t +H‘l - I-i+
Photo. 570 nm
"all-trans“ highly distorted W. .15 °C,
1 480 nm
11 1 us
\ \ \ \ 4,? i... Luml.-40°C,
12 H 497 nm

Bathe, ~14O °C. 543 nm
'dlstorted trans’

Figure 4-5. Schematic representation of the intermediates observed in the

isomerization of 11-cr's-retinal to all-trans-retlnal in rhodopsin.

133

 

Figure 4-6. The isomerization of 11-cis-retinal to the ail-trans form causes
a change in the protein conformation, which allows for binding and

activation of the G-protein transducin.

The isomerization of ll-cis-retinal to the all-trans form causes a change in
the protein conformation (Figure 4-6), which allows for binding and activation of
the G-protein transducin.21 Transducin is formed by three different subunits 01, 8
and y, and a GDP molecule is bound to the a component. When the rhodopsin
conformation changes (R*), the affinity of GTP to transducin decreases, and the
affinity to GDP increases, resulting in the exchange of a GTP for a GDP
molecule. Binding of GTP results in the dissociation of the a-subunit (Ta-GTP).
The Ta-GTP subsequently binds to the y-subunit of PDE (Figure 4-7). Activation
of PDE results in the hydrolysis of cyclic GMP to GMP. This results in the
decline in free intracellular cyclic GMP. Hydrolysis of cyclic GMP to GMP
forces the Ca2+ ion channel in the rod/cone outer segment to close and halts the

flow of Ca2+ into the cell. Thus a change in Ca2+ concentration occurs (from 0.5

134

 

Figure 4-7. The activation of G-protein transducin
upon binding to the activated rhodopsin (R'). Once
activated, the a subunit of transducin activates the
enzyme phosphodiesterase (PDE) by binding and
dissociating the y subunit. Active PDE catalyzes

the hydrolysis of cGMP to GMP. which causes

 

changes in ion concentrations in the cell and
results in an electrical signal being sent to the

brain.

M to 0.1 MM).29 The hyperpolarization of the membrane results in an electric
signal that it is sent to the brain and results in a visual image.”30 The visual cycle
stops when rhodopsin is phosphorylated by a protein called arrestin. “'32 Finally,
the phosphorylated opsin binds a new molecule of ll-cis- retinal, to restart the
visual cycle.

In summary as shown in Figure 4-8, one photon of light causes the
isomerization of the chromophore, leading to the activation of the photoreceptor.
Thus a single isomerization of a double bond triggers a cascade of events that lead

14.17.33.34

to vision. Each photon activates about 100 PDE molecules. This

135

activation leads to the hydrolysis of roughly 100,000 cyclic GMPs.35'36 The visual
transduction is one of the most efficient and sensitive processes known, with a

calculated quantum yield of ~ 0.67.37‘41

 

 

Rhodopsin \hv
Bathe
Luml
1 1-cr's-retinal M01!"
1L Ga'GDP'GpY 19015,,Y cGMP
all-trans-retinal
t_ + m =>
. Opsin GDP
ADP . ATP Ga'GTP PDEam f3?
Rhodopsrn Klnase '
1 photon -————> 100 = 100,000

Figure 4-8. A summary of the visual transduction process. One photon of light
causes the isomerization of the chromophore, leading to the activation of the
photoreceptor. A single isomerization of a double bond triggers a cascade of

events, which leads to vision.

4.1.2 Color vision
A very important aspect of vision is the fact that not only objects can be

identified, but also different colors can be perceived. Color vision is defined as

136

the capacity of detecting a variety of different wavelengths. In humans this
process is possible due to the existence of different receptors. The eye contains
three different colored cone opsins, red, blue and green and a rod opsin. Each of
these opsins binds the same chromophore, ll-cis-retinal as a protonated Schiff
base via a Lys residue. Even though they bind the same chromophore,
interestingly each one of them results in different wavelength absorption. The
human rod rhodopsin absorbs at 500 nm, the green cone opsin at 530 nm, the red
cone opsin 560 nm and the blue cone opsin 420 nrn."2"3 As shown in Figure 4-9

the absorption of the four rhodopsins cover the whole visible spectrum.

4. Color perceived

 

 

 

 

Blue Rhodopsin

80 t. 420 nm
Rod Rhodopsin

so - _ 500 nm

% A

Green Rhodopsin

40 " — 530 nm

4
20 _ Red Rhodopsin
_ 560 nm
1 1 1 1 1 1 1 1 l L 1 1 L 1_1
380 400 440 480 520 580 600

Figure 4.9. Each of the four rhodopsin complexes (blue, green, red and rod) absorb

at a different wavelength. comprising the entire visible spectrum.

137

1 1 -cr‘s-retinal

1 1 -cls-retlnal
Schiff base

1 1 -cls-retlnal
protonated Schiff base

 

hum-440 nm

Figure 4-10. UV-vis absorbances of 11-cis-retinal as a free
aldehyde (380 nm), a n-butyiamlne Schiff base (365 nm) and a
protonated Schiff base (440 nm) when ethanol is used as solvent.

The maximal absorption of the free 11-cis-retinal is ~380. When a Shiff
base is formed with n-butylamine, the spectrum blue shifts to ~365 nrn.

Protonation of the Schiff base causes a large red shift to ~440 nm (Figure 4-

Changes in the concentration and solvent conditions can shift the

wavelength up to ~500.“ From the comparison of the maximal absorption of the
protonated Shiff base (~500) and the absorptions of the opsins (~530, 560), it is

evident that the protonation of the Schiff base is not the only factor involved in the

138

red shifting of the wavelength. The difference in the absorbance observed for the
visual pigments and the PSB of ll-cis-retinal in solution is known as the opsin
shift. It is assumed that the opsin shift is due to interactions between the protein
and the retinylidene chromophore. Rhodopsin is a large, hydrophobic, membrane-
bound protein consisting of 348 amino acids and two oligosaccharide chains
(~39,000 Da). Many attempts to solve the 3D structure of rhodopsin were
performed.""’8 It was not until 2000 that the first crystal structure of a rod
rhodopsin was published."9 The structures of the cone opsins have not been
solved up to date. Different theories that attempt to explain what causes the opsin
shift have been proposed. Most of the proposed theories explain the wavelength
regulation by interfering with the degree of conjugation of the chromophore.

As mentioned previously, Morton and Goddwin were the first to propose
that the chromophore ll-cis-retinal binds as a Schiff base. But it was not until
about ten years later that Akhtar’0

(1968) and Bownds (1967)”

 

«A
independently provided evidence \9
for the imine bond formed between \

the 0 sin and a l sine residue. The
p y Figure 4-11. Distance of the counter anion

lysine identified to form the imine to the protonated Schiff base and positioning

bond via proteolytic fragmentation of charges or dipoles along the backbone of

was Lys 296.10 In an attempt to the ”We” may modulate "‘9 maximal

_ . wavelength of the chromophore.
explain the wavelength absorptron

of rhodopsin, in 1958 Hubbard

139

suggested that the Schiff base must be in the protonated form, and interactions
with the protein would cause a delocalization of the positive charge throughout
the polyene backbone (Figure 4-11).52 Krospf and Hubbard suggested that
wavelength regulation in rhodopsin may result from electrostatic interactions
between the SB and charged amino acids within the binding pocket.53 In 1974 the
protonated nature of the Schiff base was confirmed by Callender through Raman
studies.“ In 1979 Honi g introduced the idea that a negatively charged amino acid
might be used as a counter anion to stabilize the PSB, and that a second negatively
charged amino acid could be placed at another position along the polyene
backbone.” Later, NMR studies carried out by N akanishi and others verified the
existence of a second negative charged amino acid near carbon 12 of retinal.’“8
In 1989 Oprians9 and Saltmar60 identified the counter anion as Glul 13. When the
Glu113 was mutated to Gln the protein existed as a mixture of two different
species, one with an absorbance at 490 nm, and the other at 380 nm. These two
species existed in a pH-dependent equilibrium, thus they were identified as the
protonated form and unprotonated form of the Schiff base. The presence of the
counter anion is essential for the formation of a stable PSB, possibly due to the
unfavorable positive charge formed by the PSB. When high concentrations of
halide were used, the protonation was recovered.“ It was suggested that the
halide functioned as a counter anion, and therefore, stabilized the PSB. Addition
of halide to the wild type rhodopsin did not change the chromophore maximal
absorbance. In 1972 Blatz proposed the hypothesis that varying the distance

between the counter anion and the protonated Shiff base changes the maximum

140

wavelength obtained.62
When the counter anion is
close to the positively
charged nitrogen, the charge
is localized at the protonated
nitrogen. But when the
counter anion is further away,
the positive charge exists as a
resonance over the polyene.
causing a red shift in the AM.
A mutation of Glu113 for
Asp (shorter counter anion)
causes a 5 nm red shift in the
ham (505 nm) supporting
Blatz theory.60 Sheves and
co-workers prepared a series
of model compounds to probe

the effect of varying the

FEM

s
\inOH R‘

"6°00” RNXCOOH

 

 

 

1 2
"a "‘1 00H "1
kaOOH Lij NW
4 5 8
in... (rim)
Compound SB PSB pK.
1 340 450 12.32031
2 386 448 95:20.1
3 388 450 8.04:0.1
4 380 444 85:01
5 375 442 10010.1
6 360 440 7.54:0.1

Figure 4-12. Absorption maxima and apparent
pK. values of Schiff base (88) and protonated SB
(PSB) of retinal with various amino acids. The
relative position of the carboxylic acid dictates the

1...... and the pK. values.

distance and the angle between the counter anion and the PSB (Figure 4-12).‘53

The results showed that the largest shift obtained in the hm, of the PSB species

was 10 nm, the larger the distance between the counter anion, the larger the red

shift observed. This does not account for the opsin shift observed in the opsin

proteins. However the pK. varied significantly, demonstrating that positioning of

141

the counter anion has a significant influence in the stability of the PSB. Hoing
proposed that polar groups localized in the binding cavity could help to stabilize
the PSB formed in the protein.“

Along the same lines, Nathans suggested that regulation of the wavelength
could be accomplished by the positioning of a negatively charged amino acid.
Positioning of the negatively charged arrrino acid would result in a different
distribution of the conjugation of the positive charge along the polyene (Figure 4-
11).“

A different hypothesis suggests
that the wavelength is regulated by
twisting the planes about the single

bonds in the molecule to achieve

 

different levels of conjugation (Figure

4-13).66 Although retinal would prefer ”9"" 4'1 3' Twisting 0‘ the single W

. will reduce the degree of it orbital overlap,
to adopt a planar conforrnatron to

and thus will lead to different maximal
maximize p-orbital overlap, it is

wavelengths.
known that ll-cis-retinal is not a
planar molecule. The chromophore has two strong steric interactions, the first one
between the C8-H and the C5-CH3 and the second one between Clo-H and the
C13-CH3 (Figure 4-14). This leads to a twist of the C6-C7 and C12-C13 single
bonds to alleviate the steric hindrance."'“ It is possible that when the

chromophore binds the protein, it is forced to a more planar conformation. Thus

maximizing the p-orbital overlap.

142

Therefore, a red shift in the

absorbance would be observed. And

 

on the contrary, when a highly twisted

Fl ure 4-14. 11-cls-retlnal is not a nar
chrom0phore exists, a blue shift in the g pia

molecule due to steric interactions. The

absorbance would be observed.
first steric interaction is between the C8-H

Different degrees of twisting would and the 05.0.43 and the second one

result in different conjugation, which beMeen C10-H and the C13-CH3. This

would cause different maximal leads to a twist of the 06-07 and 012-013

sin le bonds to alleviate the steric
absorbances.66 Theoretical studies 9

hindrance.

provide evidence that twisting of the
chromophore alone does not account for the large observed opsin shifts.69

Many different studies such as NMR, cryoelectron microscopy,
photoaffinity labeling, etc, have been performed in an attempt to prove the
theories discussed previouslymu Some of those theories have been partially
confirmed by model compounds.66 However the real breakthrough occurred in
2000 when Palczewski et. al. reported the crystal structure of bovine rod
rhodopsin.7s In the binding pocket it was confirmed that the 11-cis-retinal was
bound to Lys296 as a Schiff base, and the counter anion, Glu113, was close to the
nitrogen (about 3.5 A). A second G1u181 was found, and could support Honig’s
theory that the shift is modulated by a two charge model.76

In 1979 Rafferty proposed that another Way to red shift the wavelength
would be via excitonic coupling which could occur between the Trp265 and the

77

chromophore. There is an important interaction between the Cl3-CH3 of the

143

chromophore and Trp265. When a rhodopsin is prepared using a retinal that does
not contain a C13 methyl group, the activity is only partially recovered.78 This
supports the proposed theory that different twisting in the chromophore (to
optimize p-orbital overlap) would cause a shift of the wavelength.

As mentioned previously, a large number of studies were performed to
understand the modes of wavelength regulation. Since all four rhodopsin proteins
are membrane proteins, their crystallization has been hampered for many years.

The human green and red cone proteins have a 96% homology sequence
and differ only in 15 amino acids.65 Interestingly, even though their sequence is
so similar, their wavelength absorption is significantly different, 570 nm and 530
nm for red and green rhodopsin, respectively. Oprian et. al. have performed
studies using red rhodopsin, and they were able to shift the wavelength from 570

79 This was accomplished by mutating only seven arrrino acids

nm to 530 nm.
Sl16Y::Sl80A::1230T::A233S::Y277F::T285A::Y309F. Mutation of three of
those amino acids would account for most of the shift.8°'8' All the substitutions
involve the conversion of a hydrophobic amino acid to a polar residue near the
ionone region, namely F277Y, A285T, and A1808. The same trend was observed
when site directed mutagenesis was performed in rod rhodopsin to mutate
hydrophobic residues to polar in analogous positions. The rod mutant
F261Y::A269T::A1648 has a very large red shift, from 500 nm to 700 nm.82
Studies in blue rhodopsin have been more difficult due to the difficulty to

express and obtain large protein quantities. Blue rhodopsin shares only 46% of

homology with the other pigments.“83 Because of this, most of the studies have

144

been performed using mutagenesis of other opsins to promote a blue shift in the
wavelengths. Sakmar found that by mutating nine residues in rod rhodopsin
M86L::G905::A117G::El22L::A124T::W265Y::A292S::A295S::A299C a shift
from 500 to 438 nm is observed.“ Also Farrens observed that the a triple
mutation in rod rhodopsin T118A::E122D::A2928, shows an absorption at 453
nm (47 nm blue shifted)” The single mutation W265Y produces a 15 nm shift,
and the single mutations G908 and A292S causes a 11 nm, and 10 nm blue shift,
respectively.86 Mutation of corresponding amino acids in the blue pigment seem
to force red shifting. Blue opsin mutant Y262W (8-ionone ring vicinity)
generates a 10 nm red shift, whereas S289A (vicinity of SB) results in no change,

and S876 (vicinity of SB) actually blue shifts 10 nm.

4.2 Designing of a rhodopsin surrogate

As previously discussed, several theories have been proposed to explain
the mechanism of wavelength regulation. In the last fifty years, our understanding
of the mode of wavelength regulation increased greatly. However, as mentioned
previously, rhodopsins are membrane-bound proteins and techniques such as
crystallization, mutagenesis, etc. are very challenging when with this group of
proteins. Therefore, most of the structural information obtained on membrane
bound proteins is the result of theoretical analysis and low resolution structural
data coupled with results of site directed mutagenesis. This has hampered

advances in the understanding of the mode of action of cone and rod rhod0psins.

145

We are interested in exploring the nature of protein/substrate interactions
at" the molecular level since these interactions are at the heart of biochemical
events that regulate biological systems. As a model for such studies we have
chosen the case of wavelength regulation in rhodopsin, i.e., the mechanism by
which we can see colors as a powerful example of the consequences of
protein/substrate interactions. Thus, we decided on engineering a protein that
mimics rhod0psin, but it does not hold the same difficulties as rhodopsin,
especially in regards of crystallographic studies and site directed mutagenesis.
The engineered rhodopsin mimic would allow us to test the proposed theories
more precisely and efficiently. In more detail, a rhodopsin mimic would provide
information about the important factors that determine retinal binding as a PSB
and how wavelength modulation occurs. In the last decade, protein design has
been widely used, and has lead to an increase in the understanding of protein
folding and structure/function relationships.3"""'99 Development of such proteins
offers a greater understanding of the natural proteins. It also represents the first
step toward a new generation of novel macromolecules that will have practical
applications in industry and biomedicine.‘°°"°2 After designing a protein that
mimics the PSB formation with retinal we will systematically introduce charges,
sterics, and dipoles at varying locations with respect to the bound chromOphore
with hopes of deconvoluting and understanding the possible reasons for the
altering opsin shifts.

A suitable protein mimic should have the following characteristics. The

protein should have a robust structure to tolerate multiple mutations. Different

146

amino acids will be mutated, first to accomplish the binding of retinal as a PSB,
and second to probe amino acids along the polyene and PSB region to probe the
effects in the absorption wavelength. The 3-D structure should be solved, mainly
because the design of the rhodopsin mimic would be based on the use of rational
mutagenesis. The protein cavity should be large, hydrophobic, and protected from
the solvent.

Also it is important that the protein can be expressed in large scale as a
water soluble protein and should be small in size. Preference for binding retinoid
structures will be advantageous in the engineering process. Two families of
proteins were considered as the platform protein for our engineering studies. The

‘03 The retinoid receptors have

first family were the Retinoid Nuclear Receptors.
been classified into two subfamilies,‘°‘“°7 the Retinoic Acid Receptors (RAR) that
bind to all-trans-retinoic acid and the Retinoid X Receptors (RXR) that bind to 9-
cis retinoic acids. The partial crystal structure of these proteins has been reported,
but unfortunately these proteins tend to form dimers and are membrane
associated. The second family is the Intracellular Lipid Binding Proteins (iLBP)
and it is believed that these proteins are involved in the transport and metabolism
of hydrophobic ligands in the cell.‘°8"°9 In particular a very interesting subgroup
of proteins (subcategory of the iLBP) are the retinoid binding proteins (RBPs).
The RBP proteins bind retinoids, which are very important in the control of a wide
variety of cellular processes like transduction,”° gene regulatory control,”1

cancer,”2 and disease preventionm'm The concentration of retinoids in the cell

is tightly controlled by the retinoid binding proteins, Cellular Retinol Binding

147

Proteins (CRBP) and Cellular Retinoic Acid Binding Proteins (CRABP). CRBP
control the concentration of retinol (retinoids in the alcoholic form) and CRABP
control the concentration of retinal (retinoid in the acid form).115 Retinoic Acid
Binding Proteins have been widely studied. There are two forms of CRABP
proteins present in most species, CRABPI and CRABPII, which are highly
conserved within vertebrates.”6"” Both proteins that bind all-trans-retinoic acid
through a series of electrostatic interactions, are small (15 KDa) and cytosolic
proteins.”8 NMR, crystallography and over-expression of both proteins have

ed.”5‘”9'122 In particular, it has been reported that

been previously perform
CRABPII can tolerate mutations of some
anrino acids in its structure while
maintaining its general 8-barlel structure
and satisfactory expression levels.123 The
crystal structure of both the apo-CRABPII
and CRABPH bound to retinoic acid has
been previously reportedm‘m Professor
Yan (Department of Biochemistry,
Michigan State University) graciously Figure 4.1 5. The crystal structure

donated to us a pET-17b vector containing of CRABPII bound to all-trans-

the human CRABPII gene (Figure 4-15). retinoic acid.

148

4.3 Wild Type CRABPII binding properties
Cellular Retinoic Acid Binding Protein II (CRABPH) was chosen as the
template for the engineered rhodopsin mimic. CRABPH is a small (137 amino
acids) cytosolic, B-barrel protein with a large (~600 A3) binding cavity made up of
two five-stranded B-sheets and a helix-tum-helix motif. All-trans—retinoic acid,
the natural substrate, is bound through a series of electrostatic interactions
between Argl32 and Tyr134 and a hydrogen bond between a water molecule
Arg] 11 and Thr54 (Figure 4.16).'25
The binding of all-trans-retinoic acid to CRABPII is stoichiometric, with a
dissociation constant of ~2 nM.118 Due to the structural homology between

retinoids we expected that retinal would occupy roughly the same place within the

    

Figure 4-16. Wild type CRABPII bound to all-trans-retinoic
acid. The ligand carboxylate is directly interacting with Arg132
and Tyr134, while it is forming a hydrogen bonds with Arg111.

149

CRABPH binding pocket.
This assumption, verified
by modeling studies, was
critical for the rational re-
designing of the binding
pocket of CRABPII in
order to achieve a
protonated Schiff base
forming protein. High
selectivity towards binding
of retinoic acid vs other
retinoids has been reported.
This observation was
verified (Figure 4-17),
since the aldehydic analog
of retinoic acid, all-trans-

retinal, exhibits a poor

 

 

1 ‘l—WHW
. O.8~
5 05 .
.5
,9 0.4~
.2 02 ~
g o
I I I I I
0 1 2 3 4
Equivalents retinoic acid/retinal
B

 

 

Absorbance

o , . j

300 400 500
nm

Figure 4-17. A. Fluorescence titration curves for the
retinoic acid and retinal Kd determination with wild-type
CRABPII. Retinoic acid shows a stoichiometric binding
relationship (K., = 2.0 :e 1.2 nM). Retinal shows lower
binding affinity, K, = 6600 a 360 nM.; B. UV-vis of
retinal incubated with wild-type CRABPII portrays a

maximal absorbance at 377 nm.2

dissociation constant for CRABPH, Kd=66001360 nM, as determined by

fluorescence quenching.”'126 Detailed description of all the spectroscopic assays

used for the evaluation of substrate binding to CRABPII is provided in Chrysoula

Vasileiou’s Thesis.2 Because CRABPH binds all-trans-retinoic acid as the native

substrate, the engineering of the rhodopsin mimic would be performed using all-

trans-retinal instead of the rhodopsin substrate ll-cis-retinal. Also the use of all-

150

trans-retinal was preferred because of the sensitive nature of ll-cis-retinal. The
work described next has been performed by Chrysoula Vasileiou and Rachel M.
Crist (detailed explanation can be found in their Theses).2

When retinal was incubated with WT CRABPH a km of 377 nm was
observed, a 3 nm shift as compared to the retina] absorbance in phosphate buffer
(380 nm, Figure 4-17). There are 13 Lys residues in WT CRABPH and all of
them are located on the surface of the protein. Therefore, none of those Lys
residues are nucleOphilic enough to form a Shiff base (SB) with retinal at pH 7.3
(pKa of Lys side chain in solution is about 10.7). MALDI-TOF analysis verified
that no covalent bond species between retinal and the protein is formed. Thus the
first step in the engineering of CRABPII is the positioning of a nucleophilic Lys
in the cavity. Assuming that the retinal binding site in the cavity will be similar to
that of the retinoic acid, mutation of Argl32 directly interacting with retinoic acid
(carboxylate) to a Lys residue seemed to allow formation of a SB. A minimized
model of the CRABPII-R132K mutant showed that the nitrogen atom of Lysl32
resides ~3 A away from the retinal’s carbonyl carbon, approaching from the side
of the chromophore (dihedral angle ~135°). This orientation is close to the
optimal angle for nucleophilic attack to the carbonyl (~107°), as defined by Burgi
and Dunitz.127 The position 134 was also considered as a possible site for the
introduction of a Lys residue, however, a series of proteins containing the Y134K
mutation did not yield the desired PSB formation (for details refer to Rachael M.
Crist’s Thesis). Mutagenesis and expression of CRABPII-R132K was performed

using standard molecular biology protocols. This mutation resulted in a

151

significant decrease in retinoic acid affinity, from 2 nM to 65 nM, as shown in
Figure 4-18. At the same time the retinal binding, as determined from
fluorescence quenching, increased dramatically with the dissociation constant K,
changing from 6600 nM to 280 nM. The UV spectrum of the CRABPII-R132K
mutant showed that the UV trace changed with time, shifting from ~392 nm, upon
addition of retinal, to ~376 nm 15 min later. Second derivative analysis of the
obtained spectra revealed the presence of a mixture of peaks, with a maximal at

~4l7 nm and ~376 nm, at a relative ratio of 1.2 to 1, respectively. The UV Am

 

 

 

 

 

 

0 i t f r i 0 T i
0 1 2 3 4 5 300 400 500
Equivalents retinoic acid/retinal nm

Figure 4-18. A. Fluorescence titration curves for the retinoic acid and retinal K.
determination with the single mutant CRABPII-8132K. The mutation shows a drop in
retinoic acid binding affinity (Kr-65:14 nM). which is concomitant with a large increase in
retinal binding affinity, Kd=280117 nM; B. UV-vis of retinal incubated with CRABPII R132K.
The spectrum changes with time, from 392 nm (red trace, 1 min) to 376 nm (magenta trace,

15 min)?

152

observed for all-trans-retinal bound to the R132K mutant (~417 nm) is ~40 nm
red shifted as compared to the km, of free retinal in solution (~380 nm). This
could be an indication of PSB formation between the engineered Lys residue and
the retinal chromophore. However, the red shifted species is not stable with time
and the system equilibrates into a state where the peak at 376 nm is maintained
Indication of Schiff base formation was provided by the MALDI-TOF
experiments. In particular, an adduct corresponding to the SB and reductively
aminated SB form is detected through mass spec analysis. The same experiments
with the WT-CRABPII had failed to detect any mass adduct. These results
suggested that the engineered Lys residue does yield a SE.

The next step to optimize the rhodopsin mimic was to increase the
hydrophobicity in the cavity, especially to increase the nucleophilicity of the
Lysl32. The two candidate residues to mutate were Ser12 and Tyr134,
considering that Ser12 is further away form the Lys residue of interest (5.01 A),
Tyr134 (3.39 A away) was the immediate target for mutagenesis. Keeping these
observations under consideration, Tyr134 was mutated into a Phe residue
(CRABPII-R132KzY134F). As shown in Figure 2-19, the second mutation
resulted in a significant increase of the retinal binding affinity, with the calculated
Kd value at 120 :l: 5 nM. In addition, the retinoic acid affinity has further declined
(Kd=100:t7 nM). consistent with the removal of one of the critical residues
responsible for binding. The UV titration of the new double mutant
R132K:Y134F with all-trans-retinal revealed the presence of a peak at 396 nm,

representing a ~15 nm red shifting as compared to the free retinal spectrum (Am

153

 

 

 

 

 

 

 

0 2 4 6 8
Equivalents Retinoic Acid I Retinal
C
16798.1
17067.4
14000 16000 16000 20000
Mesa (De)

 

 

 

 

300 400 500
nm
16788.1
.e JUL . ,
14000 16000 18000 20000
Mass (De)

Figure 4-19. A. Fluorescence titration curves for the retinoic acid and retinal Kg determination

with the double mutant CRABPII 6162K::Y134F. The retinoic acid binding affinity has dropped

(K. = 100 1 7.1 nM). while the retinal binding affinity increased (K. = 120 :i: 4.9 nM). B. UV-vis

of retinal incubated with CRABPII 8132K::Y134F portrays a maximal absorbance at 396 nm. C.

MALDl-TOF of the retinal incubated double mutant shows an M+ = 16798.1 Da corresponding to

the protein mass. calculated to be 16785.2 0a. A second peak, M‘ = 17067.4 De, 269.3 mass

units higher than the protein peak, represents a covalently linked retinal molecule (calculated

mass difference 266). D. Fieductive amination of retinal with CRABPII 8132K::Y134F results in

loss of the covalent complex and only the protein mass M+ = 16788.1 Da can be detected. The

small peak to the right of the major protein peak is due to matrix addition to the protein. and

does not correspond to the calculated retinal addition molecular weight.

154

~380 nm). However, the MALDI-TOF based binding assays did not support the
presence of a stable bond between the CRABPII-R132K:Y134F mutant and all-
trans-retinal (Figure 4-19C). The crystal structures obtained by our collaborators
(Soheila Vaezeslami, and Professor James Geiger, Michigan State University)
also support the above conclusion. The structures of CRABPII-R132KzY134F
bound to both all-trans-retinoic acid and all-trans-retinal were solved (Figure 4-
20). The bound retinoic acid is located at exactly the same position as in WT-
CRABPII. Two of the residues originally implicated in direct interactions with
the carboxylic acid have been removed (Argl32, Tyr134), however the interaction
with the water bound to Arglll and Thr54 is maintained (water #1, 2.7 A).
Double mutant R132KzY134F failed to produce evidence for SB formation with
retinal. A variety of different mutants were prepared in the process of designing a
rhodopsin surrogate. In the process it was observed that replacing Arglll with a
hydrophobic residue was a critical mutation for the formation of a PSB. It is
believed that the removal of the guanidine group results in the lost of the water
molecule that forms a hydrogen bond between the retina] and the Arglll. It was
suggested that by removing the interaction between the retinal and the water, the
carbonyl could rotate to a position in which favors the nucleophilic attack of Lys
(optimal Bilrgi and Dunitz angle of 107°, Figure 4-21).

Figure 4-21 shows the model structures of the CRABPII-R132KlellL
mutant with all—trans-retinal bound either non-covalently (A) or as a Schiff base
(B). Ly3132 is in close proximity to Tyr134 (2.8 A), suggesting that Tyr134 can

activate the primary amine and facilitate the nucleophilic attack. The necessary

155

 

Figure 4-20. A. crystal structure of CRABPII-R132K::Y134F with all-trans-retinal.

Two independent conformations of both retinal and Ly132 are found. The
chromophore exists in the free form, no SB formations is observed. 3. Overlay of the
crystal structures of CRABPII-R132K::Y134F bound to all-trans-retinal and all-trans-
retlnolc acid (represented by grey carbons and green carbons, respectively). c.
Overlay of the WT-CRABPII and CRABPII-R132K::Y134F mutant bound to all-trans-

retinoic acid (represented by orange carbons and green carbons, respectively).

156

 

R132K "
Nu 107.
Optimal
146°

. o BtirgI-Dunitz
\ angle

     
 

Figure 4-21. A. Closer look at the conformation of bound retinal to the
6132KzY134F mutant. The two conformations are represented by different colors.
6. Lys132 can attack the 1" conformation of the carbonyl from a ~70° angle. For
optimal attack a rotation of ~30° is required. C. Lys132 can attack the 2""
conformation of the carbonyl from a ~146° angle. For optimal attack a rotation of
~20° is required; D. BUrgi-Dunitz angle for optimal nucleophilic attack on a

carbonyl.

rotation of the carbonyl can occur since the water molecule associated with
Arglll has been removed. This could explain why the Schiff base formation
appears to be more successful when the R132K:R111L mutant is used. After the

SB is formed, the phenolic OH of Tyr134 can be close to the imine nitrogen on a

157

A B KFLE
(magenta)

 

 

 

 

 

0.7 r r l 1 .
0 0.5 1 1.5 2 2.5

Equivalents Retinoic Acid I Retinal

 

 

 

 

C D 15503.5
15511 .4 15776.3
15760.0
15000 15500 16000 16500 15000 15500 16000 16500
Mesa (Da) Maee (De)

Figure 4-22. A. Fluorescence titrations of retinoic acid (K. = 426 a: 47 nM) and retinal (K, =
, 1.4 a: 4.9 nM) with.the CRABPII mutant R132K::FH11L::L121E; B. UVNis spectrum of retinal
bound to WT CRABPII (blue trace) and the R132K::Y134F::R111L::L121E (magenta trace,
KFLE), R132K::R111L::L121E (red trace, KLE) mutant. Retinal bound with wild-type absorbs
at 377 nm. The UV of retinal with the KLE mutant reveals a 449 nm peak, indicating formation
of PSB. The UV for the KFLE mutant is a mixture of two peaks at 378 nm and 436 nm; C.
MALDI-TOF of the incubation of retinal with R132K::R111L::L121 E. in addition to the apo
protein peak (15511.4 Da, calculated value 15506.2 Da) a second peak at 15780.0 Be is
found, that corresponds to the covalent complex of the mutant with retinal; D. Reductlve
amination of retinal with the R132K::R111L::L121E mutant reveals a mixture of the apo-

protein (15508.5 Da) as well as the covalent complex of the protein with retinal (15776.3 0a).”

158

trajectory that can favor stabilization of a PSB. However, since the system is not
stable, positioning of a counter anion to further stabilize the protonated imine is
essential.

Research on the bovine rhodopsin suggests that the position of the counter
anion around the imine is critical for the PSB stabilization.”w"5“"128 This was
further proved by the recent crystal structure of the bovine rhodopsin which
reveals the presence of a Glull3 4.2 A away from the Shiff base nitrogen.”
Thus, the same principle was used and the model of the covalently bound retinal
suggested that the best position for introducing a counter anion is residue 121. In
the course of designing a rhodopsin mimic, it was found that Tyr134 participates

in the activation of both retinal and the PSB counter anion (GlulZl). The

 

Figure 4-23. Crystal structure of the binding cavity of CRABPII-
R132K:R111LzL121E bound to all-trans-retinal. The retinal is confirmed to be
bound as a PSB via Lys132. The important distances around the imine are

highlighted.

159

 

 

 

 

Figure 4-24. Proposed mechanism for the Schiff base formation between triple
mutant CRABPII-R132K: R1 1 1 L:L121 E and all-trans-retinal.

160

spectroscopic results for the CRABPH triple mutant R132K: RllleL121E are
presented in Figure 4-22. The protein has a remarkable affinity for retinal as
evident from the extremely low dissociation constant (Kd=1.36:l:4.9 nM) obtained
by fluorescence quenching. At the same time, the UV-vis trace in the presence of
all-trans-retinal reveals the presence of a single peak at 449 nm, a 72 nm red shift
as compared to the trace of retinal in the presence of WT-CRABPII. The crystal
structure of CRABPII-R132K:R111LzzL121E bound to all-trans-retinal was
obtained by our collaborators (Soheila Vaezeslami, and Professor James Geiger,
Michigan State University) is shown in Figure 4-23.

Based on the spectroscopic and crystallographic data gathered, the
following mechanism was proposed for the Schiff base formation between all-
trans-retinal and the CRABPII-R132K:R111LzL121E triple mutant. As shown in
Figure 4-24, Tyr134 activates the bound aldehyde for nucleophilic attack by
Ly8132, followed by elimination of water and Schiff base formation. The imine
remains protonated due to the close proximity of Glu121 resulting in a Glu121-

Schiff base salt bridge.

161

4.4

10.

11.

References

H. Shichi, Biochemistry of vision. ed.; Academic Press, Inc: 1983.; 'Vol.' p.

C. Vasileou. Protein Desiganeengineering Cellular Retinoic Acid Binding
Protein 11 into a Retinal Binding Protein. Michigan State University, East
Lansing, 2005.

J. M. Tiffany; P. Noorjahan; A. A. Abdel-Latif; J. J. Harding; R. Mayne;
R. G. Brewton; Z. X. Ren, Biochemistry of the Eye. ed.; Chapman and
Hall: London, UK, 1997; 'Vol.' p.

F. Crescitelli, Handbook of Sensory Physiology,. ed.; Springer-Verlag,:
New York , NY, 1972,; 'Vol.' Vol. VII, p 245-363.

B. Wissinger; L. T. Sharpe, "New aspects of an old theme: The genetic
basis of human color vision." American Journal of Human Genetics 1998,
63, (1257-1262).

B. N. Pugh, "Rods are rods and cones are cones, and (never) the twain
shall meet." Neuron 2001, 32, 375-380.

R. Hoffmann, The Same and Not the Same. ed.; Columbia University
Press:: New York, NY,, 1995; 'Vol.' p.

G. Wald, "Pigments of the retina." Journal of General Physiology 1935,
19, 781-795.

R. A. Morton; T. W. Goodwin, "Preparation of retinene in vitro." Nature
1944, 153, 405-406.

F. D. Collins, "Rhodopsin and indicator yellow." Nature 1953, 171, 469-
472.

R. A. Morton; G. A. J. Pitt, "Rhod0psin. IX. pH and the hydrolysis of
indicator yellow." Biochemical Journal 1955, 59, 128-134.

162

12.

13.

14.

15.

l6.

17.

18.

19.

20.

21.

R. R. Rando, "Polyenes and vision." Journal of Chemical Biology 1996, 3,
255-262.

J. Nathans, "Rhodopsin: Structure, function and gentics." Biochemical
Journal 1992, 31, (21), 4923-4931.

G. L. Wheeler; M. W. Bitensky, "A light-activated GTPase in vertebrate
photoreceptors: Regulation of light-activated cyclic GMP
phosphodiesterase." Proceedings of the National Academy of Sciences of
the United States of America 1977, 74, 4238-4242.

P. A. Liebman; E. N. Pugh, "The control of phosphodiesterase in rod disk
membranes: kinetics, possible mechanisms and significance for vision."
Vision Research 1979, 19, 375-380.

B. Fung; L. Stryer, "Photolyzed rhodopsin catalyzes the exchange of GTP
for bound GDP in retinal rod outer segments." Proceedings of the National
Academy of Sciences of the United States of America 1980, 77, 2500-2504.

B. Fung; J. B. Hurley; L Stryer, "Flow of information in the light-
triggered cyclic nucleotide cascade of vision." Proceedings of the National
Academy of Sciences of the United States of America 1981, 78, 152-156.

E. E. Fesenko; S. S. Kolesnikov; A. L. Lyubarsky, "Induction by cyclic
GMP of cationic conductance in plasma membrane of retinal rod outer
segment." Nature 1985, 313, 310-313.

L Haynes; K.-W. Yau, "Cyclic GMP-sensitive conductance in outer
segment membrane of catfish cones." Nature 1985, 317, 61-64.

T. P. Sakmar, "Rhodopsin: A prototypical G protein-coupled receptor."
Progress in Nucleic Acid Research 1998, 59, 1-34.

W. J. chl'ip; D. Gray; J. Gillespie; P. H. M. Bovee; E. M. M.
Vandenberg; J. Lugtenburg; K. J. Rothschild, "Photoexcitation of
rhod0psin -Conformation changes in the chromophore, protein and
associated lipids as determined by ftir difference spectroscopy."
Photochemistry and Photobiology 1988, 48, 497-504.

163

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

A. Cooper, "Energy uptake in the 1st step of visual excitation." Nature
1979, 282, 531-533.

R. R. Birge, "Photophysics of light transduction in rhodopsin and
bacteriorhodopsin." Annual Review in Biophysics and Biology 1981, 10,
315-354.

R. Mathies, "Resonance raman-spectroscopy of rhodopsin and
bacteriorhodopsin isotopic analogs." Methods in Enzymology 1982, 88,
633-643.

T. Shieh; M. Han; T. P. Sakmar; S. O. Smith, " The steric trigger in
rhod0psin activation." Journal of Molecular Biology 1997, 269, 373-384.

B. Yan; K. Nakanishi; J. L. Spudich, "Mechanism of activation of sensory
rhodopsin-I - evidence for a steric trigger." Proceedings of the National
Academy of Sciences of the United States of America 1991, 88, 9412-9416.

I. Szundi; T. L. Mah; l. W. Lewis; S. Jager; O. P. Ernst; K. P. Hofrnann;
D. S. Kliger, "Proton transfer reactions linked to rhodopsin activation."
Biochemistry 1998, 37, 14237-14244.

S. Amis; K. P. Hofmann, "2 Different forms of metarhodopsin-II-Schiff-
base deprotonation precedes proton uptake and signaling state."
Proceedings of the National Academy of Sciences of the United States of
America 1993, 90, 7 849-7853.

P. P. M. Schnetkamp, "How does the retinal rod Na-Ca + K exchanger
regulate cytosolic free Ca2+?" Journal of Biological Chemistry 1995, 270,
13231-13239.

R. R. Rando, "The chemistry of vitamin A and vision." Angewandte
Chemie International Edition 1990, 29, 461-480.

D. A. Moffet; M. A. Case; J. C. House; K. Vogel; R. D. Williams; T. G.
Spiro; G. L. McLendon; M. H. Hecht, "Carbon monoxide binding by de
novo heme proteins derived from designed combinatorial libraries."
Journal of the American Chemical Society 2001, 123, (10), 2109-2115.

164

32.

33.

34.

35.

36.

37.

38.

39.

41.

R. B. Penn; R. M. Pascual; Y. M. Kim; S. J. Mundell; V. P. Krymskaya;
R. A. Panettieri; J. L. Benovic, " Arrestin specificity for G protein-coupled
receptors in human airway smooth muscle." Journal of Biological
Chemistry 2001, 2 76, 32648-32656.

G. A. Schick; T. M. Cooper; R. A. Holloway; L. P. Murray; R. R. Birge,
"Energy-storage in the primary photochemical events of rhodopsin and
isorhodopsin." Biochemistry 1987, 26, (9), 2556-2562.

A. Cooper; C. A. Converse, "Energetics of primary processes in visual
excitation -photocalorimetry of rhodopsin in rod outer segment
membranes." Biochemistry 1976, 15, (14), 2970-2978.

N. Miki; J. J. Keims; F. R. Marcus; J. Freeman; M. W. Bitensky,
"Regulation of cyclic nucleotide concentrations in photoreceptors-ATP-
dependent stimulation of cyclic nucleotide phosphodiesterase by light."
Proceedings of the National Academy of Sciences of the United States of
America 1973, 70, (12), 3820-3824.

T. D. Lamb, "Gain and kinetics of activation in the G-protein cascade of
phototransduction." Proceedings of the National Academy of Sciences of
the United States of America 1996, 93, (2), 566-570.

S. Hecht; S. Shlaer; M. H. Pirenne, "Energy, quanta, and vision." Journal
of General Physiology 1942, 25, 819-840.

Schneider, "Spectral variation in the photosensitivity of visual purple."
Proceedings of the Royal Society A 1939, 170, 102-112.

J. E. Kim; M. J. Tauber; R. A. Mathies, "Wavelength dependent cis-trans
isomerization in vision." Biochemistry 2001, 40, (46), 13774-13778.

H. J. A. Dartnall, "The photosensitivities of visual pigments in the
presence of hydroxylamine." Vision Research 1968, 8, 339-358.

H. J. A. Dartnall; C. F. Goodeve; R. J. Lythgoe, "The quantitative analysis
of the photochemical bleaching of visual purple solutions in
monochromatic light." Proceedings of the Royal Society A 1936, 156, 158-
170.

165

42.

43.

45.

46.

47.

48.

49.

50.

51.

S. L. Merbs; J. Nathans, "Absorption-Spectra of Human Cone Pigments."
Nature 1992, 356, 433-435.

D. D. Oprian; A. B. Asenjo; N. Lee; S. L. Pelletier, "Design, chemical
synthesis, and expression of genes for the 3 human color-vision pigments."
Biochemistry 1991, 30, 11367-11372.

P. E. Blatz; N. Baumgartner; B. Balasubramaniyan; P. Balasubramaniyan;
F. Stedman, "Wavelength regulation in visual pigment chromophore.
Large induced bathochromic shifts in retinol and related polyenes."
Photochememistry and Photobiology 1971, 14, 531-549.

G. A. J. Pitt; F. D. Collins; R. A. Morton; P. Stok, "Rhodopsin. VIII. N-
Retinylidenemethylarnine, an indicator yellow analog.” Biochemical
Journal 1955, 59, 122-128.

P. E. Blatz; J. H. Mohler; W. Ahmed, "Spectroscopic observation of
solvent interaction with selected retinal Schiff-bases." Photochememistry
and Photobiology 1991, 54, 255-264.

P. A. Hargrave; J. H. McDowell; D. R. Curtis; J. K. Wang; E. Juszczak; S.
L. Pong; J. K. M. Rao; R. Argos, "The structure of bovine rhodopsin."
Biophysics of Structure and Mechanism 1983, 9, 235-244.

P. A. Hargrave, "The amino-terrninal tryptic peptide of bovine rhodopsin.
A glycopeptide containing two sites of oligosaccharide attachment."
Biochimica and Biophysics Acta 1977, 492, 83-94.

K. Palczewski; T. Kumasaka; T. Hori; C. A. Behnke; H. Motoshima; B. A.
Fox; 1. Le Trong; D. C. Teller; T. Okada; R. E. Stenkamp; M. Yamamoto;
M. Miyano, "Crystal structure of rhodopsin: A G protein-coupled
receptor." Science 2000, 289, 739-745.

M. Akhtar; D. Blosse; P. Dewhurst, "Studies on vision. The nature of the
retinal-opsin linkage." Biochemical Journal 1968, 110, 693-702.

D. Bownds, "Site of attachment of retinal in rhodopsin." Nature 1967, 216,
1 178-1 181.

166

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

R. E. Hubbard Proceedings of the Natural and Physics Laboratory
Symposium; London, 1958; pApp.

A. Kropf; R. Hubbard, "The mechanism of bleaching rhodopsin." Annals
of the New York Academy of Sciences 1958, 74, 266-280.

A. R. Oseroff; R. Callender, "Rh "Resonance Rarnan-Spectroscopy of
Rhodopsin in Retinal Disk Membranes"." Biochemistry 1974, 13, 4243-
4248.

B. Honig; U. Dinur; K.Nakanishi; V. Balogh-Nair; M. A. Gawinowicz; M.
Amaboldi; M. Motto, "An external point-charge model for wavelength
regulation in visual pigments." Journal of the American Chemical Society
1979, 101, 7084-7086.

K. Nakanishi, "Bioorganic studies with rhodopsin." Pure Applied
Chemistry 1985, 5 7, 769-776.

L. C. P. J. Mollevanger; A. P. M. Kentgens; J. A. Pardoen; J. M. L.
Courtin; W. S. Veeman; J. Lugtenburg; W. J. Degrip, "High-resolution
solid-state C-lB-NMR study of carbons C-5 and C-12 of the chromophore
of bovine rhodopsin - evidence for a 6-S-cis conformation with negative-
charge perturbation near 012." European Journal of Biochemistry 1987,
163, 9-14.

M. Amaboldi; M. G. Motto; K. Tsujimoto; V. Baloghnair; K. Nakanishi,
"Hydro-retinals and hydro-rhodopsins." Journal of the American Chemical
Society 1979, 101, 7082-7084.

B. A. Zhukovsky; D. D. Oprian, "Effect of carboxylic side chains on the
absorption maximum of visual pigments." Science 1989, 246, 928-930.

T. P. Sakmar; R. R. Franke; H. G. Khorana, "Glutamic acid-113 serves as
the retinylidene Schiff base counterion in bovine rhodopsin." Proceedings
of the National Academy of Sciences of the United States of America 1989,
86. 8309-8313.

J. Nathans, "Determinants of visual pigment absorbance: Role of charged
amino acids in the putative transmembrane segments." Biochemistry 1990,
29. 937-942.

167

62.

63.

65.

66.

67.

68.

69.

70.

P. E. Blatz; J. H. Mohler; H. V. Navangul, ""Anion-induced wavelength
regulation of absorption maxima of Schiff-bases of retinal." Biochemistry
1972, I I , 848-856.

Y. Gat; M.Sheves, "A mechanism for controlling the pKa of the retinal
protonated Schiff base in retinal proteins. A study with model
compounds." Journal of the American Chemical Society 1993, 115, 3772-
3773.

B. Honig; A. Greenberg; U. Dinur; T. Ebrey, "Visual pigment spectra:
Implications of the protonation of the retinal Schiff base." Biochemistry
1976, 15, 4593-4499.

J. Nathans; D. Thomas; D. S. Hogness, "Molecular-genetics of human
color-vision - the genes encoding blue, green, and red pigments." Science
1986, 232, 193-202.

P. E. Blatz; P. A. Liebman, "Wavelength regulation in visual pigments."
Experimental Eye Research 1973, 17, 573-580.

P. J. E. Verdegem; P. H. M. Bovee-Geurts; W. J. d. Grip; J. Lugtenburg;
H. J. M. d. Groot, "Retinylidene ligand structure in bovine rhodopsin,
metarhodopsin-I, and 10-methylrhodopsin from internuclear distance
measurements using C-13-labeling and 1-D rotational resonance MAS
NMR." Biochemistry 1999, 38, 11316-11324.

Q. Tan; J. H. Lou; B. Borhan; E. Kamaukhova; N. Berova; K. Nakanishi,
"Absolute sense of twist of the C12-Cl3 bond of the retinal chromophore
in bovine rhodopsin based on exciton-coupled CD spectra of 11,12-
dihydroretinal analogues." Angewandte Chemie International Edition
1997, 36, 2089-2093.

B. Honig; B. Hudson; B. D. Sykes; M. Karplus, "Ring orientation in beta-
ionone and retinals (Theoretical/Nmr/Semi-Empirical
Approach/Overhauser Effect/S -cis Conformation)" Proceedings of the
National Academy of Sciences of the United States of America 1971, 68,
1289-1293.

G. F. X. Schertler; P. A. Hargrave; P. C. Edwards; V. M. Unger, "The
arrangement of alpha helices in rhodopsin." Investigative Ophthalmology
and Visual. Science 1997, 38, 3379-3379.

168

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

V. M. Unger; P. A. Hargrave; J. M. Baldwin; G. F. X. Schertler,
"Arrangement of rhodopsin transmembrane alpha-helices." Nature 1997,
389, 203-206.

G. F. X. Schertler; P. A. Hargrave; V. M. Unger, "Three dimensional
structure of rhodopsin obtained by electron cryo-microscopy."
Investigative Ophthalmology and Visual. Science 1996, 3 7, 3701-3701.

J. M. Baldwin; G. F. X. Schertler; V. M. Unger, "An alpha-carbon
template for the transmembrane helices in the rhodopsin family of G-
protein-coupled receptors." Journal of Molecular Biology 1997, 272, 144-
164.

G. F. X. Schertler; C. Villa; R. Henderson, "Projection structure of
rhodopsin." Nature 1993, 362, 770-772.

K Palczewski; T. Kumasaka; T. Hor; C. A. Behnke; H. Motoshima; B. A.
Fox; I. L. Trong; D. C. Teller; T. Okada; R. E. Stenkamp; M. Yamamoto;
M. Miyano, "Crystal structure of rhodopsin: A G protein-coupled
receptor." Science 2000, 289, 739-745.

K. Nakanishi; V. Balogh-Nair; M. Amaboldi; K. Tsujimoto; B. Honig,
"An external point charge mode] for bacteriorhodopsin to account for its
purple color." Journal of the American Chemical Society 1980, 102, 7954-
7947.

C. N. Rafferty, "Light-induced perturbation of aromatic residues in bovine
rhodopsin and bacteriorhodopsin." Photochemistry and Photobiology
1979, 29, 109-120.

T. Ebrey; M. Tsuda; G. Sassenrath; J. L. West; W. H. Weddell, "Light
activation of bovine rod phosphodiesterase by non-physiological visual
pigments." FEBS Letters 1980, 116, 217-219.

A. B. Asenjo; J. Rim; D. D. Oprian, "Molecular determinants of human
red/green color discrimination." Neuron 1994, 12, 1131-1138.

M. Neitz; J. Neitz; G. H. Jacobs, "Spectral tuning of pigments underlying
red-green color vision." Science 1991, 252, 971-974.

169

81.

82.

83.

84.

85.

86.

87.

88.

89.

R. Yokoyama; S. Yokoyama, "Convergent evolution of the red- and green-
like visual pigment genes in fish, astyanax fasciatus, and human."
Proceedings of the National Academy of Sciences of the United States of
America 1990, 87, 9315-9318.

T. Chan; M. Lee; T. P. Sakmar, "Introduction of hydroxyl-bearing amino-
acids causes bathochromic spectral shifts in rhodopsin - arnino-acid
substitutions responsible for red-green color pigment spectral tuning."
Journal of Biological Chemistry 1992, 267, 9478-9480.

T. Yoshizawa, "The road to color-vision-structure, evolution and function
of chicken and gecko visual pigments." Photochemistry and Photobiology
1992, 56, 859-867.

S. W. Lin; G. G. Kochendoerfer; K. S. Carool; D. Wang; R. A. Mathies; T.
P. Sakmar, "Mechanisms of spectral tuning in blue cone visual pigments."
Journal of Biological Chemistry 1998, 273, 24583-24591.

J. M. Janz; D. L. Farrens, "Engineering a functional blue-wavelength-
shifted rhodopsin mutant." Biochemistry 2001, 40, 7219-7227.

J. I. Fasick; N. N. Lee; D. D. Oprian, "Spectral tuning in the human blue
cone pigment." Biochemistry 1999, 38, 11593-11596.

M. H. Hecht; M. W. West; 8. Roy; N. R. Rojas; S. Kamtekar, J. McLean;
K. J. Helmet; C. T. Simons; J. R. Beasley, "Design or libraries of de novo
proteins by binary patterning of polar and nonpolar amino acids." Protein
Engineering 1997, 10, 20-20.

C. Kiefer; S. Hesse]; J. M. Lampert; K. Vogt; M. 0. Lederer; D. E.
Breithaupt; J. von Lintig, "Identification and characterization of a
mammalian enzyme catalyzing the asymmetric oxidative cleavage of
provitamin A." Journal of Biological Chemistry 2001, 276, (17), 14110-
14116.

M. H. Hecht, "Structures and functions of de novo proteins from designed
combinatorial libraries." Abstracts of Papers of the American Chemical
Society 2003, 226, U27l-U271.

170

90.

91.

92.

93.

94.

95.

96.

97.

98.

M. H. Hecht, "De novo proteins from rationally designed combinatorial
libraries." Abstracts of Papers of the American Chemical Society 1999,
217, U603-U603.

M. H. Hecht, "De-Novo Design of Beta-Sheet Proteins." Proceedings of
the National Academy of Sciences of the United States of America 1994,
91, (19), 8729-8730.

M. Roth; A. Jeltsch, "Changing the target base specificity of the EcoRV
DNA methyltransferase by rational de novo protein-design." Nucleic Acids
Research 2001, 29, (15), 3137-3144.

U. T. Bomscheuer", M. Pohl, "Improved biocatalysts by directed evolution
and rational protein design." Current Opinion in Chemical Biology 2001,
5, (2). 137-143.

T. Lanio; A. Jeltsch; A. Pingoud. "On the possibilities and limitations of
rational protein design to expand the specificity of restriction enzymes: a
case study employing EcoRV as the target." Protein Engineering 2000, I3.
(4), 275-281.

J. P. Caradonna; A. L. Pinto; C. D. Coldren; Y. Wang, "Incorporation of
metal-based active sites into host proteins by rational protein design
methods." Journal of Inorganic Biochemistry 1999, 74, (1-4), 16-16.

H. W. Hellinga, "Construction of a blue copper analogue through iterative
rational protein design cycles demonstrates principles of molecular
recognition in metal center formation." Journal of the American Chemical
Society 1998, 120, (39), 10055-10066.

H. W. Hellinga, "Rational protein design: Combining theory and
experiment." Proceedings of the National Academy of Sciences of the
United States of America 1997, 94, (19), 10015-10017.

A. L. Pinto; H. W. Hellinga; J. P. Caradonna, "Construction of a
catalytically active iron superoxide dismutase by rational protein design."
Proceedings of the National Academy of Sciences of the United States of
America 1997, 94, (11), 5562-5567.

171

99.

100.

101.

102.

103.

104.

105.

106.

107.

G. Kakefuda; J. G. Kwagh; K. H. Ott; G. Stockton; V. Sidorov, "Rational,
protein design of herbicide resistant acetohydroxyacid synthese." Plant
Physiology 1996, III, (2), 49-49.

M. A. Dwyer; L. L. Looger; H. W. Hellinga, "Computational design of a
Zn2+ receptor that controls bacterial gene expression." Proceedings of the
National Academy of Sciences of the United States of America 2003, 100,
1 1255-1 1260.

A. J. Chirino; M. L. Ary; S. A. Marshall, "Minimizing the immunogenicity
of protein therapeutics." Drug Discovery Today 2004, 9, 82-90.

S. A. Marshall; G. A. Lazar; A. J. Chirino; J. R. Desjarlais, "Rational
design and engineering of therapeutic proteins." Drug Discovery Today
2003, 8, 212-221.

J. Bastien; C. Rochette-Egly, "Nuclear retinoid receptors and the
transcription of retinoid-target genes." Gene 2004, 328, 1-16.

S. M. Lipkin; T. L. Glider; R. A. Heyman; C. K. Glass; F. H. Gage,
"Constitutive retinoid receptors expressed from adenovirus vectors that
specifically activate chromosomal target genes required for differentiation
of promyelocytic leukemia and teratocarcinoma cells." Journal of Virology
1996, 70, (10), 7182-7189.

M. F. Boehm; L. Zhang; B. A. Badea; S. K. White; D. E. Mais; E. Berger;
C. M. Suto; M. E. Goldman; R. A. Heyman, "Synthesis and Structure-
Activity-Relationships of Novel Retinoid-X Receptor-Selective
Retinoids." Journal of Medicinal Chemistry 1994, 3 7, (18), 2930-2941.

D. J. Mangelsdorf; R. M. Evans; R. A. Heyman, "RXR - Retinoid
Receptor or Cofactor." Calcified Tissue International 1994, 54, (4), 331-
331.

D. J. Mangelsdorf; C. Thummel; M. Beato; P. Herrlich; G. Schutz; K.
Umesono; B. Blumberg; P. Kastner; M. Mark; P. Charnbon; R. M. Evans,
"The Nuclear Receptor Superfamily - the 2nd Decade." Cell 1995, 83, (6),
835-839.

172

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

P. Krieg; S. Feil; G. Furstenberger; G. T. Bowden, "Tumor-specific
verexprcssion of a novel keratinocyte lipid-binding protein - identification
and characterization of a cloned sequence activated during multistage

carcinogenesis in mouse skin." Journal of Biological Chemistry 1993, 268,
17362-17369.

M. Buelt; L. Banaszak; D. A. Bemlohr, "Regulation of adipocyte lipid-
binding protein-phosphorylation by fatty-acids." FASEB Journal 1991, 5,
A417-A4l7.

R. A. Morton, Photochemistry of Vision. ed.; Springer-Verlag: New York,
1972; 'Vol.' p 33.

C. Merlet-Benichou; J. Vilar; M. Lelievre-Pegorier; T. Gilbert, "Role of
retinoids in renal development: pathophysiological implication." Current
Opinion in Nephrology and Hypertension 1999, 8, (1), 3943.

P. Palozza; e. al., "Regulation of cell cycle progression and apoptosis by b-
carotene in undiferentiated and differentiated HL-60 leukemia cells:
Possible involvement of a redox mechanism." International Journal of
Cancer 2002, 97, 593-600.

L. J. Gudas, In The retinoids: Biology chemistry and medicine. ed.; New
York, 1994; 'Vol.' p 443-520.

A. C. Ross, "Symposium-retinoids-cellular-metabolism and activation-
introduction." Journal of Nutrition 1993, 123, (2), 344-345.

B. N. Chaudhuri; G. J. Kleywegt; I. Broutin-L'Hermite; T. Bergfors; H.
Senn; P. L. Motte; O. Partouche; T. A. Jones, "Structures of cellular
retinoic acid binding proteins I and II in complex with synthetic retinoids."
Acta Crystallography D 1999, 55, 1850-1857.

J. S. Bailey; C. H. Siu, "Purification and partial characterization of a novel
binding-protein for retinoic acid from neonatal rat." Journal of Biological
Chemistry 1988, 263, 9326-9332.

D. E. Ong; F. Chytil, "Cellular retinol-binding protein from rat-liver-
purification and characterization." Journal of Biological Chemistry 1978,
253, 828-832.

173

118.

119.

120.

121.

122.

123.

124.

125.

126.

L. C. Wang; Y. Li; H. G. Yan, "Structure-function relationships of cellular
retinoic acid-binding proteins-Quantitative analysis of the ligand binding
properties of the wild-type proteins and site-directed mutants." Journal of
Biological Chemistry 1997, 272, 1541-1547.

G. J. Kleywegt; T. Bergfors; H. Senn; P. L. Motte; B. Gsell; K. Shudo; T.
A. Jones, "Crystal structures of cellular retinoic acid binding proteins I and
II in complex with all-trans retinoic acid and a synthetic retinoid."
Structure 1994, 2, 1241-1258.

X. Chen; M. Tordova; G. L. Gilliland; L. C. Wang; Y. L. G. Yan; X. H. Ji,
"Crystal structure of apo-cellular retinoic acid-binding protein type II
(RlllM) suggests a mechanism of ligand entry." Journal of Molecular
Biology 1998, 278, 641-653.

J. Rizo; Z. P. Liu; L. M. Gierasch, "Structure and folding of cellular
retinoic acid-binding protein, a predominantly beta-sheet protein."
Biophysical Journal 1994, 66, A397-A397.

J. Rizo; Z. P. Liu; L. M. Gierasch, "H-1 and N—lS Resonance assignments
and secondary structure of cellular retinoic acid-binding protein with and
without bound ligand" Journal of Biomolecular NMR 1994, 4, 741-760.

L. C. Wang; H. G. Yan, "NMR study suggests a major role for Arglll in
maintaining the structure and dynamical properties of type 11 human
cellular retinoic acid binding protein." Biochemistry 1998, 37, 13021-
13032.

A. C. M. Young; G. Scapin; A. Kromminga; S. B. Patel; J. H. Veerkamp;
J. C. Sacchettini, "Structural studies on human muscle fatty-acid-binding
protein at 1.4-Angstrom resolution-binding interactions with 3 C18 fatty-
acids." Structure 1994, 2, 523-534.

G. J. Kleywegt; T. Bergfors; H. Senn; P. Le Motte; B. Gsell; K. Shudo; T.
A. Jones, "Crystal structures of cellular retinoic acid binding proteins I and
II in complex with all-trans retinoic acid and a synthetic retinoid."
Structure 1994, 2, 1241-1258.

A. W. Norris; L. Cheng; V. Giguere; M. Rosenberger; E. Li,
"Measurement of subnanomolar retinoic acid-binding affinities for cellular
retinoic acid-binding proteins by fluorometric titration." Biochimia

174

127.

128.

Biophysics Acta-Protein Structure and Molecular Enzymology 1994, 1209,
10-18.

M. M. Midland; Y. C. Kwon, "Stereochemistry of hydroboration of a-
chiral olefins and reduction of a-chiral ketones. An unusual anti-Cram
selectivity with dialkylboranes." Journal of the American Chemical
Society 1983, 105, 3725-3727.

A. Terakita; M. Koyanagi; H. Tsukamoto; T. Yamashita; T. Miyata; Y.
Shichida, "Counterion displacement in the molecular evolution of the
rhodopsin family. " Nature Structual Molecular Biology 2004, II , 284-289.

175

STUDIES ON 15-15’-B-CAROTENE DIOXYGENASE AND REEINGINEERING
CELLULAR RETINOIC ACID BINDING PROTEIN II INTO A RETINAL
BINDING PROTEIN AND ITS INTERACTION WITH RETINAL MIMICS
VOLUME II
By

Montserrat Rabago-Smith

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Chemistry

2006

Chapter 5
Attempts to label Lys132 to determine the pK. of CRABPII mutants '

5.1 Introduction

In the process of designing a rhodopsin mimic, it was apparent that the
environment around the lysine would have a great influence in the nucleophilicity
of the lysine, as well as the stability of the Schiff base. For example, was found
that Tyr134 participates in the activation of the bound aldehyde followed by
nucleophilic attack by Lys132. The effect of Tyr134 is evident when one
compares the UV traces of the R132K::R111L::L121E triple mutant (Max:449

nm) and that of the R132K::Y134F::R111L::L121E tetra mutant (Amx=378 nm

I

“K91 He *1 a
R \ O H2N/V\Lys —- R \ 0 HaNMLye

Amu=377nm Am=377nm

/

H 9 Me

I-vc pK.

 

Amu=360nm (7) I MSMSnm
PSB pK. UV-vle oboerved
Scheme 5-1. In the formation of a protonated Schiff base various

equilibrium are involved.

177

and hm=449 nm, figure 5-
1). Also, replacing Arglll
with a hydrophobic residue
was a critical mutation for
PSB formation, we believe
that the removal of the
guanidine moiety removes a
water molecule that forms a
H-bond with the retinal
molecule, thus it positions it
for an optimal nucleophilic
attack. A good example for

the stabilization of the PSB is

the presence of the counter

 

 

 

450 550
nm

350

Figure 5-1. UVle Spectrum of retinal bound to WT

CRABPII (blue trace) and the

R132K::Y134F::R111L::L121£ ~ (magenta trace,

KFLE), R132K::R111L::L121E (red trace, KLE)
mutant. Retinal bound with wild-type absorbs at 377
nm. The UV of retinal with the KLE mutant reveals a
449 nm peak. indicating formation of PSB. The UV
for the KFLE mutant is a mixture of two peaks at 378

nm and 438 nm.

anion Glu121 (L121E), which stabilizes the protonation of the Schiff, base

formed. The effect of the Glu121 is can be observed when one compares the UV

traces of the R132K::R111LzzL12lE triple mutant (Am=449 nm) and that of the

R132K::R111LzzL121Q triple mutant (Am=376 nm). We have realized that it is

important to tune the electronic environment to increase the nucleophilicity of

Lys, while at the same time retain a polar enough cavity to support PSB

formation. It was envisioned that determining the pK. values in an accurate and

quantitative manner for both the Lys residue of good and bad proteins and the

protonated Schiff base could be instructive for further designs. The best mutant,

178

triple mutant R132K::R111L::L121E exhibits stoichiometric binding in the
formation of the protonated Schiff base at pH=7.5. Scheme 5-1 shows the pH
dependent equilibrium between the non-bonded and the retinal-bound forms of
Ly3132. Calculation of the pK. values for the PSB of various proteins using UV-
vis spectroscopy has been possible. This was accomplished by a pH titration
between the PSB (Amx=445 nm) and the SB (Am=377 nm). It is important to
mention that measurement of this pK. is only possible when the equilibrium of the
protein favors the PSB. As shown in Scheme 5-1 UV-vis spectroscopy does not
differentiate between the unbound retinal and the SB between retinal and the
rhodOpsin mimic. Because of this, the measurement of the pK. of Ly3132 using
UV-vis assays cannot be measured directly. The apparent pK. value for Ly8132
could be estimated by first incubating the protein at different pI-Is, followed by the
addition of retinal. Second, addition of acid to push the equilibrium towards the
PSB. Then followed by plotting the amount of red shifting present at 441 nm
(Am) versus the different pH values. Following this method, the calculated value
for the R132K::R111L::L121E triple mutant was ~5.4 (Figure 5-2).2 In a similar
manner, the value calculated for the R132K::R111L::L121B::T54V mutant was
even lower (~4.9).2 These values compare well with the reported pK. for the
active site Lys residue for rhodopsin (~6.0).2 Because we are monitoring an
overall change of the UV-vis absorbance that it is indirectly connected to the
calculated pK., therefore the validity of these data has been questioned.
Irnportantly we cannot be confident that the change we observe after a change in

pH is due to the pK. of the nucleophilic Lysine and not due to other amino acid

179

0.16-J
0.14 -
0.12l

0.08 -l
0.06 -
0.04 d
0.02 ‘

 

 

 

 

Figure 5-2. A. UV titration of R132K::R111L::L121 E incubated at different pHs (4-
7.5) upon addition of all-trans-retinal. 8. Graph of the absorption change at 441 nm
(A4“) with different pHs.

residues that we have identified as essential for PSB formation. For example a
protonation of Glu121 would probably result in the loss of the UV red shifting
(PSB) but would necessarily prevent the SB formation. Because of the fact that
we cannot decisively distinguish between the UV-vis absorbance of free retinal
and that of a SB, a different method should be use to determine the pK. of the
Lys132. A direct measurement of the protonation state of the engineered Lys
could be performed using NMR spectroscopy.

NMR spectroscopy is one of the most common methods for the study of
biomolecules in solution. CRABPII has been extensively studied by NMR, and

the solution structures of both apo and hole protein have been solved. NMR

180

studies have been performed to determine the pKa of different essential amino
acids in proteins.‘ NMR could provide direct information about the electronic
environment of the retinal and the protein-imine formed. Measurement of a lysine
residue has been previously described for the case of 4-oxalocrotonate
tautomerase.“

Use of NMR for measurment of the pK. would entail labeling of the Lys
of interest with 15N, a change in the pH of the solution will lead to a change of the
1SN-NMR chemical shift of the specifically labeled lysine from unprotonated to
protonated state, affording directly the pK,,.6 Therefore, we could use the same
methodology to estimate an accurate pK. value for the free Lys and the PSB

nitrogen in the rhod0psin surrogate. Also, it has 0
'3“:

been shown that a free aldehyde, a Schiff base and a H3‘5NQ’V\/\002H
protonated Schiff base have very different IH-NMR,
13C-NMR and ”N-NMR signals.7 The pit. for the
protonated Schiff base in the rhodopsin family 2

proteins has been estimated by using labeled l3C-retinal and lsN-lysine.8 The 13C-
NMR chemical shifts for several iminium salts have been previously reported,
with the imine (~l60 ppm) and the iminium (~180 ppm), and free aldehyde (~200
ppm) are significantly different chemical shifts.9'” Along the same lines the use
of 15N NMR for determining the pK. values has been previously reported.6 In
the study of bacteriorhodopsin, addition of a 13C-retinal has been used to perform
13C-NMR studies. The results revealed a shifting of the retinal C-lS (the

aldehydic carbon) of ~30 ppm upon addition to bacteriorhodopsin (the shift is due

181

to the formation of protonated imine). The 013 was also altered by a downfield
~11 ppm shift.7'12"5 In addition, solid state NMR experiments showed that the
aldehyde proton in ‘H-NMR shifts from 9.7 ppm (free aldehyde) to 8.9 ppm
(protonated Schiff base)”17 As mentioned previously, if the side chain of
Ly5132 is labeled with 15N or 13C, a change in the protonation state of N would
directly translate into a change of the ”N-NMR/BC-NMR chemical shifts. Thus
by following the signal at different pHs a value for the pK. of Lys could be
calculated. Thus, we proposed to perform experiments that would involve the use
of 13C-labeled retinal (1) and/or 13C and 15N labeled lysine (2) (for synthesis of
l3C-retinal refer to Rachael Crist’s Thesis). A change in the pH of the solution
translates into a change in the protonation state of the imine nitrogen. The change
in the imine protonation would cause a change in the NMR signal (from the
unprotonated to the protonated state). Therefore it would be possible to monitor
the shift in the NMR signal to calculate the pK,. Since the 15N chemical shift
changes will be used to follow the protonation state of the amine, the e-amino
group of Lysl32 must be specifically labeled. Thus two different approaches
were designed to accomplish this goal. The first one involves the non specific
labeling of Ly5132 with 8—‘3C-lysine. And the second involves the specific

labeling of Lysl32 with e-BC-lysine via via Native Chemical Ligation.

182

5.2 Non selective labeling of CRABPII

5.2.1 Use of ”N-NMR to determine the pK. or Lys
The first attempt to determine pK. values using NMR was by preparation

of a fully labeled 8-1’N-Lys protein. This proteins was prepared by addition of

lsN-e-Lys (commercially available and the price is accessible) to the culture, and a
uniformly l5N labeled protein was obtained. MALDI-TOF and electrospray
ionization spectra of the labeled double mutant verified the incorporation of the
labeled Lys (For details refer to Chrysoula Vasileiou’s Thesis).2 Unfortunately,
when the l‘N-NMR experiments were performed, the "N signal could not be
identified. The signal for l"’N-lysine could not be identified, even when up to 20
mg of protein were used (> 1 mM). The major problem when 1"’N NMR is used is
that the natural sensitivity for 15N is 0.001104 as compared to ‘H (1.0) or 13'C
(0.0159). The natural sensitivity of each nuclei is determined by the
gyromagnetic ratio and the energy difference (AE=yhBa) between the spin states.
The larger the energy differences the more nuclei are present in the lower spin
state and hence are available to absorb energy."3 When lH-‘sN HMQC spectrums
was performed, still no signal was identified. Besides the fact that the natural
sensitivity for N is too low, the failure of the lH—‘sN HMQC experiment could be
due to the fact that the amine protons are exchangeable protons and were
exchanging with the D20 present in the sample. Therefore, to solve this problem,
instead of using 1s‘N-NMR we decided to use 13C-NMR. The first step was to

accomplish the synthesis of [6-‘3C]-L-lysine, express the protein in the presence

183

of the labeled lysine and finally use the l3C-Lys labeled protein for NMR

experiments for the determination of the Lys pK. value.

5.2.2 Use or 13C-NMR

Because of the fact that the performed 15N—NMR experiments resulted in
the lack of signal, we decided to perform a lH-BC HMQC instead using l3C—ClS
labeled retinal, (for its preparation refer to Rachel Crist Thesis). The first attempt
to take the N MR of this protein was using a protein concentration of about 1.5
mM. No 13C signal could be identified in the 13C-NMR spectrum of this sample,
suggesting that the labeled retinal had precipitated. When the 1H-‘3C HMQC
spectrum was recorded no signal was identified for either the aldehydic proton or
the PSB proton. The experiment was repeated with the final concentration of the
protein at 0.5 mM. This time, when the lH-13C HMQC spectrum was recorded a
signal at about 8.8 ppm, which corresponds to the PSB was identified. (24 hours
in the 600 NMR instrument). The lack of sensitivity is mainly attributed to the
fact that the retinal concentration was about 0.25 mM and protein instability.
However, as mentioned previously, when the concentration is increased, the
retinal precipitates. The next step would be the titration of the NMR signal at
different pHs. When HCl was added into the solution, the protein-retinal mixture
became unstable, and no signal could be identified in the IH—BC HMQC spectrum
even after 24 of hours acquisition. After 24 hours the NMR tube was colorless
and cloudy. It seems that the problem was again that when the pH was changed in

the presence of the protein complex (retinal-CRABPII mutant PSB) it precipitated

184

out of solution. We envision that this problem could be avoided if the protein
itself is labeled with 13C, because, to calculate the pK. of the Lys addition of
retinal would not be necessary. Therefore, the protein could be grown in media
supplemented with [6-‘3C]-L—lysine instead of 15N. Therefore, the synthesis of

[6-‘3C1-L-Lys was performed.

5.2.3 Synthesis of 6-‘3C-lysine

The preparation of [6-‘3C]-L-lysine would be accomplished by using
['3C]-sodium cyanide as the source of isotopic label. The retro synthetic strategy
is shown in Scheme 5-2, [6-‘3C]-L-Lysine (3) could be obtained from the
reduction of nitrile 4. Nitrile 4 would be derived from the treatment of tosyl 5
with cyanide. The tosyl 5 could be obtained from a chemoselective reduction of 7
and tosylation. The (rt-methyl ester 7 could be obtained from L-glutamic acid (8)

after a series of transformations (Scheme 5-2).

6
p . P .
13 QHS 6 =2 13 9 91H =>Na13CN g Q‘H

9
NHS Pg‘NH

? 9 4’:— e ? <= ?
902C/\:\002 02c’\7/\cozue ”°\/\6/‘co2~le

Scheme 54. Retrosynthetic analysis for the preparation of [6-‘3C]-L-Lyslne.

185

9 Ph/\0 N 0 (CH20)n

 

 

 

NHa /\ =
Hoch/‘CO? 9 ‘ > P“ oil?“ T°'“°"°-
8 H020MC°2H reflux
NaOH. dioxane 1o TsOi-I , 6h
90 °C. 6h
0
NaOMe,
rah/‘0JL f§’\ : PhAOfiNH
H0 0 . MeOH. rt. 5h ?
2 - 11 40% over 3 steps H02M°2M°
12

Scheme 5-3. Synthesis of N-CBZ-Glutamlc acid a—metyl ester 12.

The synthesis of lysine started by treatment of L-glutamic acid (8) with N-
benzyloxycarbonloxy)succinimide (9) to afford the N-CBZ protected amino acid
(10). A selective protection of the (ll—carboxylic acid was accomplished following
Scholtz and Barlett’s work.”21 The protected glutamic acid was used with out
further purification and was treated with paraformaldehyde to forrrr N-CBZ-S-
oxo—4-oxazolidinepropanoic acid (11) followed by a selective ring opening after
treatment with sodium methoxide in methanol, to yield the N-benzyloxycarbonyl-
L-glutamic acid or-methyl ester (12) (Scheme 5-3). Attempts to purify 12 using
column chromatography were unsuccessful. Thus, 12 was purified by

recrystallization from dicyclohexylammonomium as described in the literature.22

186

Compound 12 was obtained in a 40% overall yield from glutamic acid (8). The
overall yield for 12 was low due to the last treatment of the oxazolidine with
methoxide. Even when the reaction was performed using absolute methanol and
molecular sieves, no real change in the overall yield was obtained. All these
reactions can be scaled up to about 20 grams of starting material without problem
or any change in the yields.

The chemoselective reduction of the carboxylic acid of 12 to the alcohol
was next. The most common strategy to reduce carboxylic acids in presence of

esters is via the preparation of the mixed anhydride 13 follow by reduction using

Table 5-1. Attempts to selectively reduce N-CBZ-Glutamic acid u-methyl ester.

‘Depends on scale.

PhAoj‘NH clcoaer. EtaN i

Pit/‘0 NH P

hAOilSH
Hozc’E/‘cone THF,1h,-10°C 5‘0 Vow/re ”Moms
13 14

 

 

 

Entry Reaction conditions Yield of 14
1 20 eq. NaBHa, HzoITHF (10:2) 0°C -t RT, 2hts 12-60%'
2 3 eq. NaBH... HgozTHF (10:2) 0°C -9 RT. 2hre 16%
3 3 eq. NaBHJaluminaTHF, RT, 4hrs 80% (12)
4 3 eq. NaBH.,:THF 0°C —9 RT, 2hrs 68% (12)
5 3 eq. NaBH,, H20:THF (4:1) 0°C —> RT, 2hrs 60%
6 3 eq. NaBHl, in THF. MeOH added drop wise 0°C -> 25-89%'
RT, 1hrs
7 3 eq. NaBH... in H20: 0°C —-> RT, 1hrs 25%

 

187

NaBI-L (Table 5-1). ”’27 The first attempt was using 20 equivalent of the reducing
agent and addition of water (see details in experimental). The reaction works well
in small scale (30-100 mg) giving about 60% yield of the desired product 14.
However in a larger scale the yield decreased drastically to 16%. When a large-
scale reaction was performed only the alcohol could be isolated after column.
When NaBHa supported in alumina was used no reduction took place, only
starting material was recovered. Reduction using only NaBI-L provided the same
result. By increasing the amount of water the reaction seemed to proceed better,
although no transformation takes place when water is used as the sole solvent.
Conditions 4, 5 and 7 (Table 5-1) were only tried in a small scale (5 100 mg).
Since entry 6 was promising in small scale, (89% yield), the reaction was repeated
on a larger scale but again the yield deereased significantly (25%). All the
previous reactions were repeated using recrystalized NaBI-Ia, but no improvement
in the final yield was observed.

The chemoselective reduction of carboxylic acids using 8H3 has also been
reported. When BH3 (l M in THF) was used to reduce 4, the desired alcohol was
obtained. The yield for the desired alcohol varied from 44 to 48% (Scheme 54).

and was reproducible regardless of the scale performed.

188

fl 1. 8H3, (1.5 sq) THF, ofi
_ 0
PhAO UH 10 C-rt. 5-6 h fi_ PhA NH

. H :
HoeC’ja/‘coeMe 2. MeOH,rt.15 n °\’\/‘<><>aMe
414-48%

Scheme 5—4. Selective reduction of N-CBZ-Glutamic acid a metyl ester.

Tosylation of the alcohol 16 was accomplished by treatment of the alcohol
(16) with p-toluene sulfonyl chloride in triethylarrrine. (The use of pyridine as a
base did not provide consistent results, 30-90%, Scheme 5-5). The next step
would be the introduction of the 13C using Na‘3CN. The transformation of the
tosyl (17) to the nitrile (18) was uneventful yielding nitrile (18) in 46-50% yield
(when no label NaCN was used the same yield was obtained).

The next step toward the synthesis of l3C-labeled Lys would be the
hydrogenation of the nitrile to an amine. Hydrogenation of nitriles has been

previously reported and a variety of different catalyst have been used, including

189

p-TsCl (1 .Zeq)

 

 

 

 

r
Py, rt, 12-24 h
oi 30-90% oi

Ph" 13H < PhA 1ng
HONCOZMG Tchone

15 p-TsCl (1 .2eq). 17

cnzcrz. rt, 18 h
78-85%
13
Ph/‘ofiNl-i Na CN (129°) _ ghAOiNH
? a N1 C :

TE"’\/\/‘cone DMF, n, 24 h N002“?

1" 46-50% ‘3

Scheme 5-5. Tosylation of alcohol using triethyl amine provides more
consistent results. Introduction of the label 13C occurred in a 50% yield

using Na‘SCN.

Table 5-2. Attempts to hydrogenate nitrile 19.

 

 

PhAOfiNH 8H2
NCNcone HZNWCOQMO
19
Entry Conditions Product
1 PtOz, isopropanol HCI (50:1) 20 psi, RT, 3 hrs SM
2 Ptop, isopropanol HCI (50:1) 50 psi, RT. 3 hrs SM
3 Pd(OH)2 20% wt isoprOpanol HCI (50:1) 100 psi. RT, 3 hrs SM
4 Pd(OH)2 20% wt isopropanol HCI (50:1) 300 psi, RT. 3 hrs SM
5 Pd(OH)2 20% wt isopropanol HCI (50:1) 500 psi, RT, overnight Complex

mixture

190

Pd/C, Pt/C, Ru/C, Rh/C Raney Ni, Raney Co, Adams catalyst (R02), and
Pt(OH)2.28'35 The reaction have been performed at different pressures (1-500 psi).
The first attempt to reduce the nitrile was performed using PtOz (Table 5-2) at 20
psi and only starting material was recovered. The pressure was increased to 50 psi
without any change in the course of the reaction. When Pt(OH)2 was used, again,
only starting material was observed, and increasing the pressure to 300 psi did not
improve the results. When both the pressure and the reaction time were increased,

decomposition was observed.

 

RCHO
1”“3
RCH(OH)NHZ -H20
ll-Hzo ”2 RCHsNHa ”CH0 RCH(OH)NHCH2R \4-120
“2 3 H2
RCN —. HcH=NH H2 "“20 (RCHalzNH
A RCH=NH20H2R /H; c
'RCHgNHp [HcHo
RCH2(NH2)NHCH2R -NHs +12 0 RCHtonlNlcrleFtle
E H
, 2 (scream Hop, N /Hz
(I NHa / c ( ale \ (+120
H, D
RCH=NH20H2R “RCHaNH H2 RCH=CHN(CH2R)2

RCH2(NH2)N(CH28)2 -NH3

H2
I 'NHa (901193”
RCH=CHN(CH2R)2 He D

Scheme 5-6. Proposed mechanism for the hydrogenation of nitriles.1

It has been reported that the reduction of nitriles can afford primary (B),
secondary (C) and/or tertiary (D) amines via an intermediate imine (A). In the

proposed mechanism for reduction of nitriles, first, nitriles are hydrogenated to the

191

 

imine (A) an then to the primary amine (B). The primary amine could react with
the intermediate imine and form the secondary amine via a gem-diamine
intermediate (E) (Scheme 5-6).1 This explains the formation of different products.
It has been suggested that to avoid the formation of side products addition of acid
is required. However, other researchers have observed the contrary; an increase in
the rate of the hydrolysis instead the reduction in the presence of acid. At this
point the 6-nitrile-l-hexyne was used as a model until the reductions were
optimized. After quenching the reaction, the same decomposition was observed.
Different catalysts were tried (PtOz), Pt(OH)2 and Pt/C, but the same result was
observed.

Reductions of the nitrile under atmospheric pressure are not very common,
a few examples can be found in the literature.33'3'6 First the search for a good
reduction protocol to reduce the nitrile to primary amine was performed using the
model compound (6-nitrile-l-hexyne). The first attempt was using Raney nickel
and hydrazinium monoforrnate36 as the hydrogen donor at room temperature
resulting in 30—80% of recovered starting material, while the rest was a
combination of many different spots in the TLC (the discrepancy in the yield of
recovered starting material was due to the fact that different Raney Nickel bottles
gave different results, Table 5-3). This method of reduction for nitriles has been
previously used only for aromatic nitriles. Thus, it is possible that aliphatic
aromatics are not activated enough to be reduced by this method. Very few
protocols that reduce aliphatic nitriles at atmospheric pressure are reported in the

literature. One of these methods utilizes a palladium activated, Raney-Nickel in

192

presence of H2.33 The first time this reaction was performed using 3 eq. of Raney-
Ni, the crude ‘H—NMR showed the desired amine product (the reduced nitrile 21),
but also the reduced alkane 22 was obtained (Table 5-3). To allow the full
reduction of the nitrile and alkyne the reaction was repeated using 6 equivalents

and only the n-aminohexane was obtained in about 76% yield (crude).

Table 5-3. Reduction of nitrile using Raney-Ni. 'T he yield was calculated using NMR

and the final total weight, no attempt to purity 22 was performed.

 

 

 

20 21 22
Entry Reduction condltlons , Results
Raney Ni IN2H2'HCOOH / MeOH, RT 4 hours 20, 80%
Raney Ni mzHg‘HCOOl-l / MeOH. RT 8 hours 20. 40%
Raney Ni INzl-lz‘HCOOl-i / MeOH, RT overnight 20. 30%

Raney Ni (3 eq) / Pd/C, H2 1,4 dioxane. H20, LiOH 21 2096', 22 2096'
Raney Ni (6 eq) / Pd/C, Hz 1.4 dioxane, H20, LiOH 22 about 76%'I

(”#00104

 

Since the above conditions were successful for the reduction of the model
compound, the actual system (18) was treated under the same conditions. The
results showed that lysine was obtained, which indicated that the CBZ group came
off (as expected), but also the methyl ester was hydrolyzed The purity of the
lysine obtained was not good and at least three different spots were identified.
The most common procedure to purify lysine is the use of Cation Exchange

Resins. The use of a variety of resins have been reported, such as IR 120,

193

DOWEX N406, DOWEX MARATHON, DOWEX HCRS DOWEX N606.
Dowex 50Wx8—400 among others.”39 Due to the accessibility to a DOWEX-
50Wx8 resin it was chosen to be used for the purification of lysine.
Unfortunately, when the purification was performed the eluted lysine was not
pure. A detailed analysis of the mechanism of nitrile reduction (Scheme 5-7)
suggests that a variety of side products can be formed.1 In particular, if a
secondary amine were formed the separation of those two components using an
ion exchange column would not be possible. Also it is important to keep in mind
that if the side product results from the hydrolysis, separation using ion exchange

resin would not be possible.

 

fl Raney-Ni, Pd/C 3H
PhAO NH *- 1gc = 3
N130\/\/\002Me LlOH. H20. dioxane Hag M02“
is 50°C. 16 h. 3

Scheme 5-7. Reduction of nitrile using Raney-Ni, Pd/C.

Finally, the method used to purify the [6-13C]-L-lysine was a modification
of the procedure reported by Schuster in 1980”“ The lysine was dissolved in a
2 M HCl solution and was injected into an HPLC. The HPLC was fitted with a
uBondapak NH; column and the amino acid were detected at 200 nm. The
mobile phase used was 30% 0.01 M KH2P0J70%CH3CN/HZO (100/16). The [6-
13C]-I..-Lysine was purified in a 85% yield form nitrile 18.

Expression of the CRABPH mutants with the label Lys has not been

performed. However we are confident that by following the protocol used to

194

express the uniformly label 15N-Lys, the desired 13C-Lys labeled protein would be

obtained, and the lH-‘3C HMQC can be performed.
5.3 Specific labeling of Lysine

5.3.1 Chemical Ligation

Chemical synthesis has always been considered an attractive alternative to
biological methods for protein production. The use of chemical synthesis
provides the possibility of creating unlimited variation of polypeptide chains.
Total chemical synthesis of native proteins has made a number of important
contributions to biomedical research.“45 In polypeptides and proteins the size
and the need for correct folding of the peptides synthesized increases the
complexity of the synthesis. The challenge for a chemist was always to be able to
chemically synthesize active proteins by stitching various synthetic peptides

rm“ and chemical ligation seems to provide that tool. However, the

togethe
original ligation chemistries gave a nonpeptide bond at the site of ligation. The
first ligation chemistry consisted of a nonpeptide bond (oxime, thiazoline) at the
site of ligation, and although these unnatural structures were often well tolerated,
trials for the formation of a native peptide bond were perssisted.‘9'5' In the native
chemical ligation method, the ligation occurs at a unique N-terrninal Cys residue,

independent of any additional Cys residues. The side chain of the N-terrrrinus Cys

forms a thioester-linked product that undergoes spontaneous rearrangement, via

195

an irreversible intramolecular nucleophicic attack of the Cys amine nitrogen, to
provide the amide-linked product.

The first case of Native Chemical Ligation was reported by Muir and
Dawson in 199452 and since then a series of chemically and enzymatically active
proteins have been made following different variations of the same central idea.
According to the established protocols, proteins can be made in high yield and
good purity from unprotected peptide building blocks, allowing for the
incorporation of unnatural segments, essential for the study of protein structure

and function.53

5.3.2 Native chemical ligation

The essential feature of the Native Chemical Ligation is the formation of a
thioester-linked product that undergoes spontaneous rearrangement, via
intramolecular nucleophilic attack, to give the desired amide-linked product. This
method requires a N-terminal Cys that is responsible of the formation of the
thioester, which is followed by rearrangement to form the peptide bond. Since
Lysl32 that was to be labeled is only two amino acids away from Cysl30, we
strategize to use this method to selectively label Lysl32. Recently, this approach
has been extended by the use of self splicing proteins as a mediator for the
formation of a native amide bond for the synthesis of proteins. Protein splicing is
a post-translational process that includes the excision of an internal protein
segment, called intein from the primary translational product with associated

54-56

ligation of the flanking sequences, the exteins. Intein-mediated protein

196

Clone target gene into intein-vector
l express in E. coli

~-—<:38“

contaminants 04 affinity chromatography

N-jSfi M“ Ohltln

transecterlfication t/‘r synthetic peptide
~-4<°
(St)

.-....-..-

Semi-Synthetic Protein

  

 

    

 

 

 

 

Flgure 5-3. Schematic representation of the lntein-mediated native chemical ligation

process.

197

splicing has been proposed to occur via intein-mediated formation of thioester-
linked intermediates. This intermediate is cleaved by nucleophilic attack to form
the final amide-linked spliced polypeptide. Utilizing similar features, intein-
mediated splicing has been used for applying the native chemical ligation
methodology (Figure 5-3).57

The intein mediated chemical ligation, requires the gene of a target peptide
to be cloned in an intein-containing vector and expressed as a recombinant
protein. To facilitate its purification the protein contains an affinity tag, (a chitin
binding domain (CBD)). While the recombinant protein is attached to the
column, a second, synthetic peptide, that contains a Cys residue at the N-terminus,
is added and a trans-thioesterification occurs. This process is facilitated by the
intein conformation, that eventually results in elution of the semi-synthetic
protein, while the intein part remains bound on the affinity column (Figure 54).
Thus, this method allows for the incorporation of labeled amino acid or any
unnatural amino acid in the part of the protein originating from the synthetic
peptide.

As mentioned previously, CRABPII contains a Cys residue at position 130
that would allow for N-terminus ligation because the labeled Ly5132 can be
accommodated in a chemically synthesized peptide. Thus, expression of the
fusion protein of intein-CRABPII(1-129) can be combined with a synthetic
peptide that has the remaining 16 amino acids (His-tag protein) to complete the
CRABPII sequence. After intein splicing, the method should afford the whole

CRABPII protein (Figure 5-4).

198

menu 30w-1 5N-Lye(1 32w Glu(137)
<— Chemlcally Synthesized Peptide —>

CRABPII(1-128)~W Val (129W
<- E. coli expressed protein —->

-”V" "Gig“.

CRABPII-intein fusion protein

 

   

N. S acy
rearrangement

 

 

0
Q??? DTT Induced on-

 

OH
15
s N

 

  

15'11-‘132K rah:1:\o\
s... s... waiv-
NH,
1 5 N 3 8
¢_‘— reaarrannement
Spontaneous
—Js. m...

Seml-synthetlc CRABPII-‘5N-Lys132

Figure 5-4. Formation of ‘bN-Lys132 CRABPII via lntein Mediated Chemical

Ligation.

199

The system currently used is called IMPACT TM-CN System (lntein Mediated
Purification with an Affinity Chitin-binding Tag, New England Biolabs) and
utilizes the inducible self-cleavage activity of inteins to separate the target protein
from the affinity tag. IMPACT systems have been used in the literature with good
results for proteins up to 160 KDa”7o The vectors used contain an engineered
intein and are designed for fusing the C-terminus of a target protein to the intein-
Chitin Binding Domain tag. The Chitin binding domain is used to generate C-
terminal thio-ester tagged proteins, which would be used in further protein
ligations. Therefore, the desired protein will be expressed as an intein fused
protein. The overexpressed fusion protein would be loaded on a chitin column
and could separate from the rest of the bacterial proteins. After the recombinant
protein is bonded to the column, treatment with a thio-containing molecule
(thiophenol, or 2-mercaptoethanesulfonic acid (MESNA) N-
methylmercaptoacetamide (NMA)) induces protein splicing. Finally, the
synthetic peptide is added and the newly formed protein elutes from the column.
As discussed in Chrysoula Vasileiou’s Thesis, many mutations of CRABPII were
carried out to optimize the retinal binding as a protonated Schiff base (PSB). To
date, the R132K::R111L::L121E triple mutant is the best protein obtained This
protein contains two mutations within the first 130 amino acids that are proven
essential for the PSB formation. Therefore, one would expect that the gene of
CRABPII(1-l29, R111L:L121E) would be the obvious choice for the formation of
the recombinant protein. However, when our efforts to utilize the native chemical

ligation method began, double mutant R132K::Y134F was the most promising

200

protein in hand. Because of this, the native CRABPII( 1-129) gene was used for
the formation of the recombinant protein. We thought that if the process was
successful, it could then be repeated for any desired mutant, without major

modifications to the protocol.

5.3.3 Expression of CRABPII fused to a mini intein

For the cloning of the CRABPII(l-129) gene into the pTXBS (a mini-
intein (198 residues) derived from the gyrA gene of Mycobacterium xenopt), refer
to Chrysoula Vasileiou’s Thesis. CRABPII (1-129) and CRABPII (1-V129A)
were cloned into the pTXBS vector and were expressed as a fusion protein with a
mini-intein (Mxe GyrA). The band expected for the CRABPII(1-129)-Intein

fusion protein should show at ~43 KDa (327 amino acids total) and as expected

  

12345678

 

45 KDa _
36 KDa _

29 KDa
20 KDa _ --

CRABPII _
14 KDa_ : .

Figure 5-5. Overexpression of the fusion CRABPII(1-V129A)Mxe
GyrA in (BL21(DE3)pLysS. host 1mM IPTG. 30 ‘'C). 1. Marker.
2.emptylane, 3. Crude. 4. Supernatant 5.6 Unsoluble fraction. 7.
empty lane, 8. CRABPII. The expression uslng different IPTG

concentrations and different E. coli host gives similar results.

201

Table 5-4. Use of different E. coli. hosts to test solubility of the fusion inteln
protein. The results are shown as: S=Supernatant, I.B.= lncluslon bodies. is
CRABPII fusion present, x= No CRABPII fusion present. XaNo cleavage of the
lntein was observed. CRABPII corresponds to both CRABPII(1-129) Mxe GyrA and
CRABPII(1-A129V) Mxe GyrA, both provide same results.

 

16 9C 30 ac lnteln

E. coll host IPTG (hit!) I

0.0

0.01

ER2566 0.10
1 .00
0.0
0.01
0.10
1 .00
0.0
0.01
Tuner(DE3)pLacl O. 1 0
1 .00
0.0
0.01
0.10
1 .00

Origami

BL21 (DES)pLysS

xxxxxxxxxxxxxxxxgn
xxxxxxxxxxxxxxxxg

xxxxxxxxxxxxxxxxg’
\\\x\\\xxxxx\\\x§n

\\\X\\\xxxxx\\\xin

the desired protein band was overexpressed. Unfortunately, it was expressed in
the form of inclusion bodies as summarized in Table 5-4, (Figure 5-5 shows an

example for the expression of the fusion intein).

5.3.4 Attempts to solubilize fusion protein
When inteins are used as fusion proteins, the formation of inclusion bodies

is a common problem. Literature reports to solve this kind of problem mostly

202

consist of a trial and error approach.61 Refolding of proteins is not commonly
used due to the low yield in activity after refolding”71 At this point the challenge
was to express a fusion protein that showed high splicing activity. Because
refolding leads to a low yield of active protein, an attempt to express the fusion
protein in soluble form was performed.

The E. coli Origami cells have been used to improve the solubility of the
fusion proteins. These host cells promote disulfide bond formation due to the fact
that they express a mutated thioredoxin reductase (tpr) and glutathione reductase

”'73 Protein solubility can increase due to the Optimization of

(gar) genes.
disulfide bond formation?“ 75 The vector containing CRABPII(l-129)- Mxe
GyrA, or CRABPII(l-V129A)- Mxe GyrA, were transformed into Origami cells
and expression of the recombinant proteins was attempted. Unfortunately the
results showed no recombinant protein. Even when different concentrations of
IPTG were tried (1, 0.1 and 0.01 mM) and at different temperatures (16 °C and 32
°C) no overexpression was Observed (Table 5-4). As discussed in Chrysoula
Vasileiou’s thesis the Tuner cells provided only inclusion bodies. Lack of
overexpression has been previously reported by Walker and colleagues.76 They

suggest that the expression of mutated tpr and gor exhaust the E. coli

metabolism, thus affecting the overexpression of the fusion protein.
5.3.5 Refolding of Mxe GyrA intein system

As mentioned previously the formation of inclusion bodies when using

inteins is not surprising. In the literature, however not many refolding attempts

203

have been reported, and the reported data shows low activity after refolding (up to
50% activity is recovered).48'“ Formation of inclusion bodies could offer several
advantages for the production of recombinant proteins if they can be refolded.
The proteins produced in inclusion bodies might be unstable in the cytoplasm of
E. coli due to proteolysis and may be toxic for the host cell in the native
conformation. It has been reported that under appropriate conditions, the
recombinant proteins deposited in inclusion bodies amounts to about 50% or more
of the total cellular protein. And because inclusion bodies have a relatively high
density48 they can be isolated away from the cellular proteins, with a purity up to
90% under optimal condition.49 A number of methods have been reported for
solubilizing and refolding proteins from inclusion bodies. Denaturants such as
urea and guanidine hydrochloride, and the use of highly alkaline pHs, strong thiol-
containing reductants (dithiothreitol, dithioerythritol mercaptoethanol), chelating
reagents (EDTA), have been reported to solubilize a significant percentage of
insoluble proteins. Renaturation is usually accomplished by the removal of excess
denaturants by either dilution or a buffer-exchange step (dialysis, diafiltration,
gel-filtration chromatography or immobilization onto a solid support). Therefore,
different proteins may require different conditions for folding and searching for
folding conditions can be a daunting task. In general, refolding is initiated by
diluting the denatured protein into a refolding buffer. Once the denaturant
concentration is reduced the protein refolds at rate equivalent to first order
kinetics. The unfolded protein quickly folds to an intermediate form, then there is

a slow transition from the intermediate to the native form (limiting step) which

204

may take rrrinutes to days to complete. At this point nonproductive aggregation

reactions can occur, resulting in the formation of unstable aggregates. Therefore,

it is very important to avoid this pathway, which can generally be achieved by

having very low concentrations of protein.

In the course of our experiments we came across a protocol used in Muir's

group on how to deal with the inclusion bodies problem (For details see Chrysoula

Vasileiou Thesis). The inclusion bodies were resuspended in the presence of 6 M

urea and then dialyzed. This was followed by incubation with DTT and the

results revealed the presence of a new protein band, at around 15 KDa, that can be

attributed to the CRABPII(l-129) protein. At the same time, there was another

new protein band, of almost
equal intensity, present on the
gel, at ~29 KDa, which was
believed to be the cleaved intein.
However, this band
corresponded to only a small
amount of the protein bound on
the chitin beads. Furthermore
when this intein was used to
carry out the actual chemical
ligation cleavage conditions, that
is the use of MESNA (2-

mercaptoethanesulfonic acid)

 

Figure 5-6. Protein Gal. 1. Marker. 2. Dialysis
supernatant loaded on Chitin Beads. 3.4. Chitin
beads after loading. 5.6. Chitin beads digested
with 10 mM DTT. 7. tst flow through. 8. 2nd flow

through. 9. Wash.

205

instead of D'IT to induce the protein splicing (the use of thiol reagents such as
MESNA provides a reactive thioester at the C-terminus of the target protein for
use in intein mediated Protein Ligation, while DTT does not), only traces of
protein at ~15 KDa were present (Figure 5-6). The problem with dialysis is that
the protein is exposed to a large amount of chaotrope, increasing the amount of a
partially formed intermediate in the folding pathway. Accumulation of this
intermediate shifts the equilibrium to the incorrect folding, thus producing
inactive protein.

Therefore, a different approach to attempt to refold the fusion protein
MxeGyrA-CRABPII (both recombinant protein were tested) was followed. This
time a 100 fold dilution was performed. This technique has the advantage that the
concentration of both chaotropes and protein is significantly decreased, and
therefore, the equilibrium should shift to the correct folding of the protein. The
disadvantage is that the final concentration of the protein is very small. After the
protein was diluted 100 times and stirred for at least 24 hours, it was directly
added into the chitin beads. When this small amount was treated with DTT or
MESNA no cleavage of the intein protein was observed. Because it has been
reported that inteins can cleave at high urea concentrations (up to 4 M), a 1/6
dilution was performed (decrease of the chaotropic from 6 M urea to 1 M urea).
Right after the solution was diluted, it became cloudy suggesting that probably the
protein was precipitating. Still the protein was stirred for 24 hours and the
solution was loaded onto the chitin beads. After addition of MESNA no cleavage

was observed. Because the inteins have been reported to be active in the presence

206

of 4 M urea the experiment was repeated at a 2/3 dilution, but again the cloudiness
was observed and after activity was tested no cleavage was observed.

The Hampton Research Foldlt Screen is a commercially available kit that
allows for determination, whether a protein of interest can be folded frOm
inclusion bodies. FoldIt is a fractional protein-folding screen which evaluates 10
factors in 16 unique solutions: protein concentration, polar additive, detergent,
pH, chaotrope, ionic strength, divalent cations, non-polar additive and
polyethylene glycol. The use of FoldIt could give significant information about
the feasibility of refolding the CRABPII-Mxe GyrA protein. After preparation of
inclusion bodies, (about 70% pure, determined by running an acrylamide gel,
stained with commasie blue, and comparing the CRABPII Mxe GyrA band
intensity with the rest of the proteins), the refolding experiments were started with
a final concentration of the proteins (all proteins) of 1 mg/mL. Then a 50-fold
dilution was performed. The resulting solution was added to a chitin bead column
and slowly eluted. Unfortunately after analysis of the chitin beads, no fused
protein was observed. Therefore, a second trial was performed, but this time a 10
times dilution, was used. The solution again was to chitin beads, and the slowly
eluted solution was resubmitted to the resin two more times. When DTT was
added to promote cleavage a weak band at 48 KDa corresponding to uncleaved
intein was observed.

The search for folding conditions can be an overwhelming task, proteins
require different conditions for folding, and no prediction for the appropriate

conditions for folding can be performed. The Hampton Research FoldIt Screen

207

allows determining with a reasonable degree of confidence whether the protein of
interest can be refolded from inclusion bodies. Foldlt is a screening kit designed
to be a rapid and statistically meaningful method for determining if a protein will
fold in vitro and what factors are important for folding. If the results of Foldlt
indicate that the folding from inclusion bodies is not feasible, then the results of
Foldlt suggests that the search for an appropriate system to fold the protein
properly, would be time consuming and expensive. The use of this kit consists in
mixing the protein (solubilized in 6 M urea) with 16 different prepared solutions.
Each sample was incubated at 4 °C forlO hours, and the intein activity was
determined. The results showed that none of the conditions resulted in an active
protein suggesting that in the folding of this protein many different factors are
involved and screening of the best conditions would not be a trivial screening. In
the literature, there are a few examples for screening of soluble and active inteins,
however, all of them involve engineering the protein by adding a fluorescent tag.
In this case, addition of a fluorescent tag would only increase the complexity of

our system.

5.3.6 Use different intein

In the intein literature it has been found that the final yield of the desired
recombinant protein can depend strongly on which intein-fusion system is
employed.“ There are two major contributions to low yield in the expression of
fusion proteins, the low levels of soluble expression (as observed with Mxe GyrA)

and in vivo cleavage. Up to now, no conclusive study on which intein works

208

better for which system has been performed, and therefore, the suggestion is that
if a system is problematic a different intein-mediator should be tried. As
previously discussed when the mini intein Mxe GyrA was used, inclusion bodies
were formed. Other available intein is See VMAl. This system utilizes an intein
(454 amino acid residues) from the Sacharomyces cerevisiae VMAl gene. A
chitin-binding domain is attached to the C terminal for easy purification. Because
this is a larger intein we thought that the proper intein folding would not be
affected with the CRABPII fusion. Therefore, the Obtained fusion protein would
be more soluble and more active. The pTYB3 vector, contains a larger intein (454
amino acid residues) from the Saccharomyces cerevisiae VMAl gene (Sce
VMA). Therefore, the CRABPII(l-V129A) gene was cloned into the pTYBB
vector. PCR afforded the CRABPII (l-V129A) gene with the cut sites Neal and
Sap]. The gene and the vector were purified and further digested. Ligations were
performed in a mole ratio of gene/vector of 10/1 and 15/1 with T4 Ligase at 16 °C
for 24 hours. After transformation into JM109 cells the correct sequence was
confirmed by sequencing. The gene was transformed and expressed in ER2566,
Tuner cells and Origami yielding the corresponding fusion protein.

The initial results showed the over expressed fusion protein CRABPII(1-
V129A)- Sce VMA band about 70 KDa (Figure 5-7) in the crude lysate. The
soluble fraction accounted for about 60% of the protein and the inclusion bodies
for about 40%. But when the soluble fraction was added to the chitin beads very
small amount of protein was found to bind the beads (shown in Figure 5-8). As

mentioned previously, three different strains were tested for the expression of the

209

123456789

29 KDa

 

Figure 5-7. Fused protein CRABPII-(1-V129A)
Sce VMA1 70KD expressed at 30 1’0 induced with
1 mM IPTG. 1. crude. 2. soluble fraction. 3.
inclusion bodies. 4. Wash 1. 5. Wash 2. 6. Chitin

beads. 7. Chitin beads. 8. empty lane 9. Marker.

recombinant intein protein, ER2566, Tuner cells and Origami. All the E. coli
strains gave very sirrrilar results. When different IPTG concentrations and
temperatures were used no improvement in the overexpression of the soluble
protein was found. Therefore, only the results using ERZSé6 are discussed. The
soluble fraction was added to the chitin beads, and the expected band at 70 [(1)3
was observed, however, two strong bands close to 50 KDa were also obtained.
The existence of the band at about 50 KDa suggests that the intein was cleaving in
vivo.

The same results were obtained when the protein expression was

performed at 16 °C overnight. When the beads were incubated with DTT to

210

_, —+16 KDa
.. — 97 KDa

‘ V—-66 KDa

—45 KDa

 

Figure 5-8. Binding of the
overexpressed fusion protein to the chitin
beads. A mayor band at 50 KDa is
observed. Expression at 30 “C induced
with 1 mM IPTG. Lanes 1,2,5.6. Shows
the binding of the super natant to the

chitin beads. 3. SceVMA1. 4. Marker.

induce intein cleavage the CRABPII(l-Vl29A)- Sce VMA band around 70 KDa
disappeared (Figure 5-9) and at the same time the band that corresponds to the
MW of CRABPII appeared in the eluted fraction (Figure 5-10). However, it can
be observed that a band at about 70 KDa also appeared in the eluted fraction. This
observation makes it difficult to conclude with certainty whether the absence of
the apparent fusion protein was due to cleavage or due to the fact that the
observed band was just a random protein, which had a relatively good binding to
the chitin beads. The problem was that if the intein was cleaving in vivo, the band

for the cleaved CRABPH could not be identified due to the fact that there are

211

many E. coli proteins that have similar molecular weight. Also it is possible that
the proteases would destroy the cleaved CRABPII. When a time study was
performed for the expression of the fusion protein an overexpressed band at the
same size as CRABPII was identified; however, it was present in the insoluble
fraction. CRABPII is known to be a soluble protein with a very robust structure,
thus it seems unlikely that this band corresponds to the cleaved CRABPII.

However it is important to mention that 17 amino acids were removed from

123456

45 KDa
36 KDa

29 KDa

20 KDa
RABPII
14 KDa

   

Figure 5-9. 10% acrylamide gel.
Cleavage of CRABPII-See VMA1 7o KDa. ”9"" 5‘10- 2°°/° “Mam“ 9°"
Expected bands after cleavage. See VMA1 Cleavage °t CRABPH'SCQ VMA1 70
so KDa, and CRABPII 20 KDa. 1. chitin KDa- Exp°°l°d bands am” °'°a""9°'
beads. 2. Marker. 3. Chitin beads after 809 VMA1 5° KDa, and CRABP" 2°

cleavage with DTT. 4_ Elution after KDa. 1. Elution after DTT cleavage. 2.

cleavage with MESNA. 5. Chitin beads “my 'ane' 3' Mm”- 4' CRABP"-

after cleavage with DTT. 6. Elutlon after 5' empty lane. 6' Elution after MESNA

cleavage with DTT. 7. See VMA1. cleavage.

CRABPII, and thus it is possible that the protein could collapse to form inclusion

bodies. An attempt to isolate this band using a Fast Q column resulted in no

212

protein, which suggests that the protein Observed is not CRABPII. As mentioned
previously cleavage in vivo is a frequent problem encountered by intein users.”'27
It has been reported that the cleavage in vivo depends mostly on the last amino
acid in the protein of interest. In our case, the amino acid at position 129, was a
valine. According literature reports, the last amino acid fused with the intein is

7 In fact, in a

important for the efficiency of the thiol induced cleavage.7
comprehensive study to evaluate the efficacy of the twenty amino acids as the C-
terminus residue, Val, residue 129 in the CRABPII sequence, is one of the worst
possible choices. In particular, ln model cases such as that of the expression and
purification of maltose-binding protein, if the C-terrninal residue is an Ala, the on
column cleavage with DTT occurs at 75-100% both at 4 °C and at 16 °C. Even
though results that have been reported about which amino acids promote cleavage
and which do not, they depend strongly on the protein used. Therefore in most of
the reports a variety of different amino acids are tested until a good result is
observed.37

At this point, although significant progress has been made towards
applying the intein-mediated native chemical ligation for our system, we were not

able to isolate enough thiol-activated CRABPH fragment to continue with the

final ligation step.

5.4 Labeling of Lys with c-"N-Lys or 6-13C-Lys
The downfall in the use of native chemical ligation is the necessity to form

a fused CRABPII-intein protein. As discussed previously this methodology was

213

tried but many different drawbacks were faced. Therefore, a different
methodology was designed. This would consist Of the conversion of a cysteine
residue to a lysine, which potentially will allow for the selective labeling of the

lysine mimic by forming a thiolysine adduct.

5.4.1 Conversion of Cys to Thio-Lys
In proteomics selective alkylation of Cys has been used Alkylation of
Cys has been accomplished using ethylenimine. iodoacetate, 2-bromoethylamine

"'83 In all the previous alkylation cases

and n-(iodoethyl)trifluoroacetamide.
(except iodoacetate), the alkylated product (23) resembles a lysine (24) (Figure 5-
11). The only difference between the lysine mimic and a lysine is a methylene vs.

a sulfur. It could be

envisioned that a \jfiAs/VNHQ Mm;

. INH as /NH 24
chemical transformation ‘5 R

Of a Cys into a Lys mimic Figure 5-11. Lysine 23 vs lysine mimic (thiolysine) 24.

The difference is that lysine contains a CH, and the
using a labeled alkylating .
thiolysine a 8.

agent would provide the
labeled thiolysine. The labeled thiolysine would behave as lysine. Assuming this
works the process would be very efficient, mostly because the full protein can be
expressed using bacterial expression and a small chemically synthesized molecule
will be added afterwards.

It has been previously reported that the lysine mimic 23 behaves as a real

lysine. For example, Ubiquitin is a protein that mediates proteolysis, which would

214

be involved in DNA repair, cell cycle passage, gene transcription among other.“
86 The experts in that area believed that a lysine residue was responsible for
Ubiquitin’s activity.84 To prove it, Chan and colleagues prepare ubiquitin with a
cysteine instead of lysine. It was observed that the cysteine containing protein
had no activity. But after treatment with n-(iodoethyl) trifluoroacetamide the
protein recovered its activity. The alkylating agent n-
(iodoethyl)trifluoroacetamide works via an S”; displacement of the iodide by the
sulfur, followed by a hydrolysis of the trifluoracetamide protecting group under

acidic conditions (Scheme 5-8). The fluoroacetamide group in 26 helps to avoid

A.
0 ii
YfifiAse INHTCFe Ws’v YCFa
NH 0 ,NH O

a.
o o
“QM WWW
Y 25 ' ‘< 39

Scheme 5-8. Conversion of Cys to Thiolysine via alkylatlon of
cysteine 25. Alkylating agent with lodoethyl-trifluoroacetamide 26
(A), or bromo ethylamlne 28(8) to form the thiolysine 29.

215

the polymerization of the reagent. It has been observed that in the case of
ethylenirrrine and 2-bromoethylamine sometimes the alkylation is not
accomplished due to polymerization of the alkylating agent.

Even though it has been reported that 23 could be a potential lysine mimic,
in order to selectively label CRABPII only one Cys should be present in the
protein structure. In the wild type CRABPII sequence three cysteines are present.
Two of those cysteines are located close to the surface and far away from the PSB
(Cys81 and Cys95) and one (CyslBO) inside of the pocket close to the active site

Lys residue (Figure 5-12). Because Cys81 and Cys95 are on the surface, therefore

 

Flgure 5-12. In the CRABPII sequence there are three
different Cys (089, 095 and 0130). Two of those Cys are

close to the surface and far from the P38.

216

most likely the rate of alkylation compared to Cys130 and L132C would be
slower and potentially not a problem. But if they do they could be mutated to
alanine. On the other hand Cys130 is in close proximity with Lysl32 and this
could pose some difficulties. Both Cys130 and Ly3132 are buried deep in the
binding pocket, the environment surrounding them is similar. Thus their rate of
alkylation would be very similar. If Cys130 is alkylated first it could interfere
with the formation of thiolysine 132. But if this indeed pose a major problem it
could be easily avoided by mutating C130A.

The triple mutant shows a shift of 447 nm in the UV as well as a 1:1 molar
ratio in the formation of protonated Schiff base of retinal and protein. Thus this
protein was used as a template to test this methodology. The first step then, would
be to prepare the R132C::R111L::L121E mutant. This mutant would retain the
C130, and C132 would be converted into the active thiolysine 132. If the
alkylation is successful the thyolysine would be formed. Thiolysine 132 would
act as a Lys and therefore after addition of retinal we would expect that the
thiolysine would form a PSB with retinal. The formation of PSB would cause a
red shift wavelength. To prepare the R132C::R111L::L121E mutant standard
mutagenesis protocols were used (for details see the Materials and Methods
section). The primers used were 5’- GACGWTGTUCACCTGCGTCTACGTCCGAGAG-
3’ and 5’- O‘CTCGGACGTAGACGCAGGTGCACACAACGTC- 3’. The PCR product
was transformed into XL1-Blue after digestion with Dpnl. The plasmid DNA
was isolated and sent for sequence to confirm the mutation. The DNA was

transformed into BL21(DE3)pLysS and the protein was expressed by addition of

217

IPTG and purified using the FPLC (for details see Materials and Methods). The
protein was expressed using the optimal conditions developed for CRABPII-
mutants. The expression was uneventful and provided about 25 mg/L of protein.
When this mutant was titrated using retinal, as expected, no shift in the
wavelength was observed (Figure 5-13). The e for R132C::R111L::L121E was

calculated to be 20,070.

 

 

 

300 350 400 450

Figure 5-13. Titration of R1320::H111L::L121E with all-trans-
retinal. Only a peak at 372 nm is present, no PSB formation was
observed. Each line corresponds to addition of 0.1 equivalent of

retinal.

To convert the cysteine to the lysine mimic the first alkylating agent used
was n-(iodoethyl)trifluoroacetamide. This alkylating agent was chosen for two
major reasons.

The first one was to avoid polymerization, due to the presence of the
protecting group on the amine. The second one is because the protecting group is
easily removed. According to the literature a variety of conditions have been used

for alkylation of cysteins using n—(iodoethyl)trifluoroacetamide. Most of these

218

used basic conditions (pH 9-10) and high temperatures (SO-90 °C); the proteins
are also denatured to assure that all Cys are alkylated.

To convert the cysteine to the lysine mimic the first alkylating agent used
was n-(iodoethyl)trifluoroacetamide. This alkylating agent was chosen for two
major reasons.

The first one was to avoid polymerization, due to the presence of the
protecting group on the amine. The second one is because the protecting group is
easily removed. According to the literature a variety of conditions have been used
for alkylation of cysteins using n-(iodoethyl)trifluoroacetamide. Most of these
used basic conditions (pH 9 - 10) and high temperatures (50 - 90 °C); the proteins
are also denatured to assure that all Cys are alkylated.

Because we wish to have some selectivity among the different Cys, and
keep a folded protein, relatively mild conditions were used. Thus the first attempt
was using 100 equivalent of 26 at pH 8.5 at 50 ‘C for 6 hours, and this was
followed by a 10 x dilution with buffer at pH 4.5. The reaction was left overnight
in order to remove the protecting group. The protein was concentrated (for details
see Materials and Methods) and the UV—vis was recorded. The UV of the protein
showed a very big signal around 250 nm and when the protein obtained was
titrated with retinal the results obtained were inconsistent. Thus, the next time the
method was modified. First 100 equivalents of 26 were added to about a 1.5 mL
of a 0.5 mg/mL protein solution (two tubes) at pH 8.5 at 50 °C for 6 hours. This
solution was then concentrated using mini filters, followed by a buffer exchange

to pH 7.5 (phosophate buffer), and then, the pH was adjusted to 4.0 (HCl 0.5 M).

219

After an overnight incubation the buffers were exchanged one more time to pH
7 .5.

The resulting protein did not have the extra peak at 250 nm, and therefore
it was titrated with retinal. The results were promising, since after addition of 0.1
equivalents of retinal a shift to 457 nm is observed (Figure 5-14, Scheme 5-9).
The peak at 457 nm indicates the formation of the PSB between the retinal and the

thiolysine (lysine mimic).

\ \ \ \ ‘0 H2N’Vsr/‘x

Am=377 nm E

WEDNSA

7km = 475 nm

Scheme 5-9. The thio-Lys acts as Lysine and it further forms a

PSB when treated with retinal.

Interestingly this peck is 8 nm red shifted form the Am observed in the
KLE mutant. Whether this shift is caused by the sulfur in the thio-lysine or not is
not clear. This observation proved the feasibility of this method. Unfortunately
the protein is active only at low equivalents of retinal.

There are many possible reasons why the overall ratio of equivalents of
retinal vs. protein is low. Most probably the problem was due to low alkylation
yield in the cysteine alkylation. Other problems could be over alkylation,

decomposition of the alkylating agent or denaturation of the protein.

220

0.065 A w=457 nm

0.055 0.1 equivalents.
0.045
0.035
0.025
0.015 r

 

 

A w=457 nm
0.12 0.3 equivalents.

 

 

 

 

Figure 5-14. Titration of the alkylated mutant 8132021811 1L::L121E treated with
iodoethyI-trifluoroacetamide 26, (alkylation with bromo ethyiamine 28 showed
similar results, therefore it is not shown). (A) Titration of the alkylated triple
mutant with 100 equivalents of the alkylating agents. The Am observed at 457
nm corresponds to the fon'natlon of the PSB of the thiolysine with retinal after
addition of 0.1 equivalents of retinal only the 380 Increased. (B) Titration of the
alkylated triple mutant with 300 equivalents of the alkylating agents. The 2...,
observed at 457 corresponds to the formation of the PSB of the thiolysine with
retinal after addition of 0.3 equivalents of retinal only the 380 nm absorption

increased.

221

To test the alkylation, MALDI-TOF analysis was used (Figure 5-15). After
incubation of alkylating agent with the protein an SN; displacement would occur.
The calculated mass protein for the triple mutant was W=15482.7 Da. After
alkylation occurs under basic conditions an W=15484.8 Da corresponding to the
mass [M+l4l]*=15623.7 Da, should be observed due to the addition of
CH2CH2NH2C(O)CF3 (141). Finally after acidic hydrolysis an [Mi-44]*=15484.8
Da would correspond to a covalently linked CH2CH2NH2 M=15523.7 Da. The
results are shown in Figure 5-15. Figure 5-15A shows the CRABPII
R132C::RlllL::L121E M=15484.8 Da corresponding to the protein mass,
calculated to be 15482.7 Da. Figure 5-ISB shows an M"=15484.8 Da
corresponding to the protein mass. A second peak, W=15528.M Da, 41 mass
units higher than the protein peak, represents a covalently linked CH2CH2NH2
(calculated mass difference 44) and a third peak W=15622.7 Da, 140 mass units
higher than the protein peak, could be three covalently linked CH2CH2NH2
(calculated mass difference 44x3 (132)) or one CH2CH2NH2C(O)CF3 (141).
Figure 5-15C shows M*=15484.8 Da corresponding to the protein mass, second
peak, W=15528.64 Da, 41 mass units higher than the protein peak, represents a
covalently linked CHzCHzNHz (calculated mass difference 44). Interestingly, the
results suggested that hydrolysis of the protecting group was occurring even at
basic pH.

The samples that were not treated with acid were titrated with retinal. The

results shows a shift in the wavelength (lm=457 nm). The shift in the UV

222

supports the observed formation of the thiolysine [M+41]+. The presence of M+

mass suggests that alkylation is not complete. The [M+l40]+ suggests that

 

 

A
\XAQNO CF3_' \/:!j/H\::No CF3 1'?) fi‘fif‘fSNNHz
NH
Y 25 23
M=15484.80a , [M+41]‘
k k
15500 _
Mass (De) MI“ (0!)

 

Figure 5-15. MALDl-TOF of the alkylated CRABPII R1320::R111L::L121E. A.
CRABPII R1320::R111L::L121E M+ = 15484.8 Da corresponding to the protein
mass. calculated to be 15482.7 Da. 8. M“ = 15484.8 Da corresponding to the
protein mass, a second peak, M+ = 15528.64 0a, 41 mass units higher than the
protein peak, represents a covalently linked mchzNHg (calculated mass
difference 44) and a third peak M‘ = 1562.7 Da, 140 mass units higher than the
protein peak, could be three covalently linked CHchzNHg (calculated mass
difference 44x3 (132)) or one CHZCHzNH2C(O)CF3 (141). C. M+ = 15484.8 Da
corresponding to the protein mass, second peak, M“ = 15528.64 0a, 41 mass
units higher than the protein peak, represents a covalently linked CH20H2NH2

(calculated mass difference 44).

223

multiple alkylation occurs (Figure 5-16). The rest of the experiments performed
were done only in basic solutions.

Different temperatures were tested for alkylation, and each sample was
treated with retinal. It was found that 50 °C was the temperature at which the
ratio of retinal vs. protein (to form PSB) was optimal. When the pH was changed,
no improvement was observed. When the equivalents of the alkylating agent 26
were increased to 300 equivalents at pH 8.6, with 50 °C overnight incubation,
alkylation was improved, the UV-vis titrations shown a 1:03 equivalents binding
of retinal vs. protein.

It has been reported that alkylation using 2-bromo ethylamine (28) results
in low alkylation yields, mainly due to the polymerization of the alkylating agent.
Interestingly when alkylation using 28 was performed the same results as with 26
were observed. Even though the protein obtained was only active at low
equivalents we believed that only the labeled l3C thiolysine would show a
distinguishable NMR signal. Thus we could monitor the change in the
protonation state of the labeled thiolysine, which will allow for the calculation of

the thiolysine pK..

5.4.2 Synthesis of the alkylating agent to convert Cys to Thio-Lys

The next step was the synthesis of the l3C-labeled alkylating agent. As
mentioned previously, formation of the thiolysine was attempted to insert the ”C
label at the e-position in the lysine mimic in order to use this labeled protein for

l3C-NMR experiments. However, the 2-'3C-Iodo-2-(trifluoracetylamino)ethane is

224

not commercially available, thus it

1icVOH
HzN 30
must be synthesized. For the j x
synthesis of 2-13C-1-iodo-2-
HZKFCVBF Faci 1§C\/l
(trifluoracetylamino)ethane we 31 32

decided to use the commercially ,
Scheme 5-10. The label ethanolamine

. 13 .
available N- C—ethanolamrne as the (30) could be used to prepare iodoethy‘.

source of the label l3C (Scheme 5- trifluoroacetamide (32). and bromo ethyl
10). The amine could be selectively amine (31)“
protected as the trifluoroacetamide
and the alcohol in the ethanolamine could be converted to the iodo compound.

The first attempt to selectively protect the amine was done using
trifluoroacetic anhydride. A di-protected product was obtained in 50% when 1
equivalent was used. When two equivalents were used, 100% of the diprotected
product was obtained This reaction was done in DCM and the ethanolamine was
not very soluble in it, thus it is possible that the monoprotected product was more
soluble in the DCM than the ethanolamine giving the diprotected product as the
major product. Thus the reaction was diluted (to increase the solubility of
ethanolamine in DCM) but again the same result was observed. Slow addition of
the trifluoroacetic acid anhydride did not improve the results. Also lower
temperatures were tried but a selective protection was not observed. Because a
selective protection was not accomplished, a selective deprotection was attempted.

Diprotected ethanolamine (30 mg) was stirred overnight with wet silica and a

selective deprotection of the alcohol was observed. However when a larger scale

225

was tried the reaction took much longer and at the same time the recovery of the
product was low (in a 1 g scale reaction only 20% of the mono protected alcohol
could be recovered). When a biphasic solution (CH2C12zH20 pH 4) was stirred
overnight the same result was observed. The first attempt was by mesylation of
the alcohol followed by treatment in situ with NaL This reaction never gave the
desired product, mostly starting material was observed but the recovery yield was
low. The next attempt was the tosylation of the alcohol to make it a good leaving
group, but unfortunately the product was not obtained. Also treatment of the
diprotected ethanolamine with 12 was tried, since the trifluoroacetate is a relatively
good leaving group. However, again no desired product was obtained. In the
literature,87 s-ethyl trifulorothioacetate (34) has been used to selectively protect
amines in the presence of alcohols. This reaction worked very well giving a 97%
yield of 35. Next the alcohol was converted to the tosyl by treatment with p-
toluene sulfonyl chloride. This reaction gave tosyl 36 in 75% yield. The
tolsylated compound 36 was refluxed with Na] in butanone to give the desired
iodo 26 in 50% yield . All the latter yields correspond to unlabelled reactions
(Scheme 5-11A).

Since we envisioned to use labeled ethanolamine as the source of 13C, we
would prefer to have less steps involved in the synthesis of the label thiolysis
alkylating agent.

Synthesis of 2-bromo ethyl amine (28) from ethanolamine has been
previously reportedw'gl When the ethanolamine was refluxed in the presence of

I-IBr a 87% yield of the desired bromo ethyl amine (28) was obtained.

226

A.

Cfi 1. Dry Pyridine, -10°C
’\ s4 0 I.

 

2, TsCl , ON, 4 °C

 

 

CHzciz, 1 h, RT
97% 35 75%
Nal, butanone
s s 2.
H 3‘ Reflux,60°C, H
50%
B.
H03:,NH2 40% HBr : H2N’\/B'
87% 28

Scheme 5-11. Synthesis of alkylating agents using ethanol amine as the
starting material. A. Synthesis of iodoethyl-trifluoroacetamide. 3. Synthesis

of bromo ethyl amine.

The mechanism of bromination of ethanolamine is not clear.92 Reports

that propose Sm and an SN; mechanism have been published, however none of
them provides conclusive results. The major concern was the fact that if the
bromination of the alcohol occurred via the formation of the intermediate aziridine
a mixture of 31 and 32 would be observed (Scheme 5-12A). If 30 is used as the
starting material, the 13c would be scrambled, and 50% of each isomer would be
obtained. Because the synthesis 31 from ethanol amine via bromination could
offer the label product after only one reaction this reaction was attempted first

regardless of the possibility of scrambling of the label l3C. Therefore, the reaction

227

”CV Br

H2N
A. Sm PE / 50% 31
13 130A VOH —
HeN'Cvo” '_ H N) \
U6 C2) 3.130va
Be Sm
‘30 H2 __. ‘écvsr
H V H;
H2N3 CV0 16211201 (8 N 31
Bre
C S
m Bra ’N
13

13

ma: H63 6) as

34

Scheme 5-12. Possible mechanisms involved in the bromination of the
label ethanolamine (34). A. Sm type mechanism in which aziridine 32 is
the intermediate. After bromination, then two products would be obtained in
50%, 31 and 33. 8. Sue type mechanism in which the bromide attacks
displaces the amine directly to afford only 31. The reaction is observed to
go via an Sm type mechanism in which the bromide attacks displaces the

amine directly to afford only 34.

using 30 was performed on small scale. The results showed that only one labeled
isomer was obtained. Initially this was very encouraging because it suggests that
only the desired product 31 was observed. But a closer look at the NMR obtained

showed that the product obtained was 33 not 31.

228

In order to obtain 100% of this isomer the only possible explanation was
that the label in the initial ethanol amine was in the opposite site (34), the 13C was
next to the 0 instead of the N. Thus the starting material was analyzed, by NMR
revealing that indeed the structure for the molecule was 34 not 30 (Scheme 5-12A,
128). The labeled ethanolamine was purchased form Cambridge Isotope
Laboratories, Inc. (Andover, MA) after the company was contacted, they accepted
that they had a mistake in their catalog, and that the desired 31 was not
commercially available.

Due to the lack of time the synthesis of 31 using a different source of 13C
label was not performed. Even though that advancement in formation of the
thiolysine was accomplished the final objective of using the labeled thyolisine to
measure the pKa could not be completed. However it is important to mention that
for the first time, we can finally prove that the bromination of 31 occurs via an SN;

type mechanism and not the proposed Sm type mechanism previously proposed.

229

5.5 Materials and Methods

A. Synthesis of Compounds

All reactions were carried under an atmosphere of nitrogen and removal of
solvents was performed under reduced pressure with a Buchi rotatory evaporator.
THF and H20 were freshly distilled from sodium/benzophenone, and CH2Cl2 was
distilled over CaH2 under a nitrogen atmosphere. Labeled H2NCH2‘3CH20H was
purchased form Cambridge Isotope Laboratories, Inc. (Andover, MA). Analytical
TLC was carried out using Merck 250 mm Silica gel 60 F254 and spots were
visualized under UV light. Column chromatography was conducted using
Silicycle silica gel (230400 mesh). 300 MHz 'H—NMR and 75 MHz 13c-NMR
spectra were recorded on a Varian Gemini-300 or 500 instrument, and the residual
protic solvent (CDC13, D20 or DMSO-d6) was used as internal reference. UV-

visible spectra were recorded on a Perkin-Elmer Lamda 40 spetrometer.

1. Synthesis of N-CBZ-Glutarnic acid (3)

MHz 1 ? OR
= Ph/‘ NH
HOgCMCOgH Pit/‘0 N ?
1 20 Hochs/‘cozn

A solution of L—glutamic acid 1 (0.59 g, 4 mmol) in a mixture of dioxane
(20 mL) and 0.4 M NaOH (20mL) was treated with n-(benzyloxycarbonyloxy)

succinimide 2 (1.2 g 5 mmol) at 90 °C for 1 h. The dioxane was evaporated under

230

reduced pressure and the product was partitioned between EtOAc and water. The
aqueous layer was acidified with HCl and extracted with EtOAc. Evaporation of
the organic layer yielded N-benzyloxycarbonyl)-L-glutamic acid 3 as a partially
crystallized gum (1.3 g crude 99%) which was used with no additional
purification. [1H NMR (300 MHz CD03): 5 9.89(2H, b, COOH), 7.23(5H, s,
ArH), 5.68(1H, d, J=7.41 Hz), 5.06(2H, s, PhCH2), 4.41(1H,dd, J=6.87, 7.41 Hz),
2.64(3H, s, (330), 2.45(2H, t, J=6.04 Hz CH2), 2.17(1H, m, CH2), 2.06 (1H, m,
CH2). ”C NMR (300 MHz C001,): 5 177.94, 172.7, 135.8, 128.51, 128.29.
128.07, 127.66, 127.05, 60.49, 25.24, 14.121. m/z (E.I): 281(M”), 262.9, 235.9,

192.0 (100)].

2. Synthesis of N-CBZ-Glutamic acid oxazolidinone (4)
PhAOfiN
140ch
4 0
A mixture of the diacid (0.4 g 1.42 mmol), paraformaldehyde (0.12 g 2.66
mmol) and p-toluene sulfonic acid*H2O (40 mg) in toluene (100 ml) was refluxed
in a Dean-Stark apparatus for 2 h. The cooled mixture was diluted with EtOAc
until it became clear, and then was washed with a solution of NaHC03 (16 mg) in
a few mL of water. Evaportation of the organic layer afforded the oxizolidinone
4, as an oil (0.32 g in 69% yield), which was used with no additional purification.
[‘11 NMR (300 MHz CDCl;): 5 9.63(1H, bs, COOH), 7.34(5H, s, ArH), 5.5(1H,
br, s, 1H, H-2), 5.19(1H, d, J=4.5 Hz, H-2), 5.16(2H, s, PhCH2), 4.37(1H, t, J=5.7

Hz), 2.46(2H, m), 2.27011, m), 2.16(1H, m). 13C NMR (300 MHz CDClg): 8

231

177.23, 171.67, 153.00, 135.16, 128.57, 128.51, 128.42. 128,21, 68.04, 53.78,

29.95, 25.58. m/z (ED: 2930“"), 274.9(100), 247.9,235, 204, 185.8

3. Synthesis of N-CBZ-Glutamic acid a methyl ester (5).

ii

PhAO NH
HO2CMCO2M0
5

A solution of oxazolidinone 4 (0.32 g) in MeOH (10 mL) was treated with
a NaOMe in MeOH (1.4 M, of 3.5 mL). The solution was kept at room
temperature for 40 min an then neutralized with 1 N HCl (6 mL). The MeOH was
evaporated and the residue was partitioned between ethylacetate and HCI (0.1 N).
The organic layer was evaporated under reduced pressure to obtain the monoacid
as 011 (0.27 g). The oil was re-dissolved in EtOAc N, N-dicyclohexylamine (360
11.1, 1 eq) was added and the solution was allowed to evaporate passively until
crystals of the dicyclohexylammonium salt were formed.

A solution of the previous crystals in 0.2 N HCl was extracted with EtOAc
and the organic layer was evaporated to dryness to obtain the free monoacid as an
oil (340 mg, 40% yield after three steps). The oil solidified to form white crystals
of s. [1H NMR (300 MHz CDC];): 5 10.72011, bs, coon),7.32(511, s, ArH),
5.56(1H, d, J=7.99 Hz), 5.07(2H, s, PhCH2), 4.41(1H, dd, J=5.2 Hz), 3.71(3H, s,
CH30), 2.40(2H, m), 2.16(1H, m), 1.97 (1H, m). 13C NMR (300 MHz CDClg): 5
177.91, 172.34, 56.02, 136.00, 128.475, 128.17, 128.03, 67.10, 53.09, 52.53.

29.79, 27.29. m/z (E.I): 2950(7), 277.0, 235.9. IR: 3331.490”). 2955.32,

232

2361.17, 2339.95, 1734.23, 1701.13. T113 (EtOAczHex, 1:1). Rr=0.27 (br). [Odo

=+7.5 (c 2.0, CHC13)]

4. Synthesis of N-CBZ-2-Amino—5-hydroxypentanoic Acid Methyl Ester (6).

The acid (0.5 g 1.69 mmol) of 5 were dissolved in dry THF (10 mL). The
solution was cooled down to -10 °C. BH3-THF (1 M, 2.5 mL, 2.5 mmol) were
added drop wise. After complete addition, the reaction was warmed slowly after 3
h the and when the reaction was completed MeOH (10 mL) was added and it was
stirred overnight. The reaction was neutralized (with 10% HC1) and the solvents
were evaporated. The residue was partitioned in ethylacetate and aqueous sodium
hydrogencarbonate. The organic phase was dried with anhydrous sodium sulfate.
A column 1:1 hexanes:ethylacetate gave the pure alcohol 6 in 48% yield (230 mg,
0.81 mmol) [Note a brand new bottle of 1M BHg-THF was used]. [1H NMR
(300 MHz CDC13): 8 7.32(5H, s, ArH), 5.61011, d, J=7.14 Hz), 5.07 (2H, s,
PhCH2), 4.37(1H, dd, J=4.95, 7.14 Hz), 3.70(3H, s, CH30), 3.6l(lH, t, 1:10.77
Hz), 2.15(1H, br 8, 01-1), 1.95-1.85(1H, m), 1.8-1.68(1H, m), 1.65-l.5 (2H, m).
13C NMR (300 MHz CDCI3): 6 172.91, 156.02, 136.13, 128.45, 128.11, 128.02.

66.94, 61.80, 53.50, 52.33, 29.19, 28.05. m/z (E.I): 281.0M), 263.0, 248(100).

IR: 3339.49(br), 3065.28, 3034.41, 2953.39, 2878.16, 17030, 1705.29. 1533.6.

1215.31. TLC (EtOAczHex, 1:1). RFOJS. [a]p=+6.4 (c 2.0, CHC13)]

233

5. Synthesis of N-CBZ-Z-Amino-S-(p-toluenesulfonyloxy)pentanoic Acid

Methyl Ester (7) i
PhAO NH

T"O\/?/\co,~le

(s)-N-CBZ-2-amine$-hydroxypentanoic acid 01-methyl ester (400 mg, 1.4
mmol) was dissolved in dry DCM. Trietylamine (1 eq.) and. p-toluenesulfonyl
chloride (1.4 mmol) were added to the mixture and it was stir overnight. After the
reaction was completed, the resulting solution was then quenched with distilled
water, acidified (2 M HCI). The organic layer was separated and the aqueous
phase was extracted with DCM and dried with ahydrous sodium carbonate .The
solvent was removed under vacuum and the product purified by flash
chromatography. Elution with 20% ethyl acetate in light petroleum gave (490
mg) the tosylate as a colourless oil in a 80% yield. [1H NMR (300 MHz CDCh):
8 7.74(2H, d, J=8.24, ArH), 7.31(5H, s, ArH), 7.30(2H, d, J=8.24 Hz, ArH),
5.30(1H, d, J=7.69 Hz), 5.06(2H, s, PhCH2), 4.30(1H, dd, J=7.69, 5.59 Hz),
3.99(1H, t, J=5.59 Hz), 3.70(3H, s, CH30), 2.41 (3H, s, CHgAr), 2.00-1.80(1H,
m), 1.72-1.60(3H, m). 13C NMR (300 MHz CDCI3): a 172.29, 156.00, 144.82,
136.02, 132.80, 129.83, 128.48, 128.18, 128.02, 127.81, 69.52, 67.12.53.12.
52.48, 28.77, 24.85, 21.56. m/z (E.I): 434.9(M), 375.9, 331.9, 299.9 204.0.

16.1.1279, 108.0, 90.9(100). IR:3373.93, 2955.32, 2878.16, 1722.65, 1527.82,

234

1358.06, 1176.73. TLC (EtOAczHex, 20:80). Rte-0.15. [Ot]D=-i-11.6 (c 2.0,

CHC13)]

6. Synthesis of [6]-(S)- N-CBZ-2-Amino-5-cyanopentanoic acid methyl ester (8)

The tosylate 7 (0.7g, 1.6 mmol) was dissolved in DMF (SmL) and
carefully added to sodium l3C-cyanide (0.17g, 4.4 mmol) in DMF (4mL) at room
temperature under nitrogen atmosphere. After 24 hours, the reaction was
quenched by the addition of aqueous sodium hydrogen carbonate, with the
resulting mixture extracted using diethyl ether. The solvent was dried, removed
under vacuum and the product purified by flash chromatography. Elution with
20% ethylacetate in hexanes gave nitrile 8 in a 48% yield (0.768 me], 220 mg)).
1H NMR (300 MHz ouch): 5 733(511, s, ArH), 5.39011, d, J=7.69 Hz),
5.09(2H, s, PhCH2), 4.38(1H, dd, J=7.4 7.69 Hz), 3.74(3H, s, CH30), 2.36(1H, t,
J=6.31 Hz), 240-190(111, m), l.80-l.60(3H, m). 13C NMR (300 MHz CDCh):
5 172.15, 155.86, 135.95, 128.53, 128.26, 128.10, 118.97, 67.14, 52.92, 52.64.
31.75, 21.39, 16.69. m/z (E.I): 290.4(M*), 231.1(100). IR: 3340.70, 2878.16,

2247.36, 1718.79. TLC (EtOAczHex, 20:80). R901.

235

7. Synthesis of 13c-s-Lysine

Nitrile 9 (114 mg 0.393 mmol) was dissolved in a mixture of dioxane (5
mL) and water (1.3 mL). Raney -nickel (120 mg) as a 50% suspension in water
and 10% palladium on charcoal (35 mg) were added together with litium
hydroxide monohydrate (35 mg) and the reaction mixture was stirred under
hydrogen atmosphere at 50 °C for 20 h. The catalyst were filtered off, the
solvents were removed in vacuo and the solution was dissolved in a 2 M HCl
solution.

Purification with the HPLC fitted with a uBondapak NH2 column was.
The amino acid was detected at 200 nm. The mobile phase was 30% 0.01 M
KH2POJ70%CH3CN/H2O (100/16). The [6-13C]-L-Lysine was purified in a 85%

yield.

8. Synthesis of (12) NH 0
10 11 12

2-aminoethanol 10 (600 1.1L, 600 mg, 9.8 mmol) was dissolved in IDCM (3
mL) and stirred as S-ethyl trifuorothioacetate 11 (1.26 mL) was added drop wised.
The mixture was stirred vigorously for 1 h. The solvent was boiled off (in a fume
hood), and several portions of DCM were added and evaporated in a N2 stream to
remove the last of the ethane thiol. The remainder of the solvent was removed in

vacuo, leaving a yellowish crystalline oil 1.5 g of 12 in 97% yield. 1H NMR (300

236

MHz (21101.): 5 7.68(1H, bs, NH), 3.79 (2H, t, J=4.94, CH2), 3.53 (2H, q, 1:543,
CH2). 1.8(1H, bs, OH). 13C NMR (300 MHz CDC13): 5 158.04 (d. J=37.8 Hz
com, 117.70 (1, J=287.45 Hz CF3). 60.59, 41.97. IR: 3310.70(bs), 2951.46,

1716.86. 1560.61. 912.45.

9. Synthesis of (13)

13
A solution of the alcohol (700 mg, 4 mmol) in dry pyridine (3 mL) was

cooled in an ice-salt bath with stirring, and reacted with p-toluenesulfonyl
chloride (1.715 g, 9 mmol). After 2 h, the solution was transferred to a
refrigerator (4 °C) and kept overnight. The suspension was poured into ice-water
(80 mL) and mixed vigorously. After a few minutes of stirring in the cold room,
according to the reference the product should solidify and then collected on a
buchner funnel. However, only an oily substance insoluble in water was
observed, thus I extracted added DCM and extracted the aqueous layer 4 times.
The organic extracts were combined and dried. After evaporation of the solvent
an oil was obtained and it was dried in vacuo to leave 1 g of the tosylated product
in a 75% yield. [‘11 NMR (300 MHz CDCl;): 5 7.76(2H, d, J=8.24 Hz),
7.34(2H, d, J=8.24 Hz), 4.13(2H, t, J=4.94 Hz, CH2), 3.60(2H, q, J=5.45 Hz,
CH2), 2.44(3H, s, CH3). 13C NMR (300 MHz ouch): 5 157.85(d, J=37.8 Hz
CCF3), 145.56, 131.95, 130.07, 128.21, 127.84, 117.45 (t, J=287.45 Hz CF3),
67.62, 38.97, 21.58. IR: 3343.06(bs), 2959.18, 1734.23, 1558.68. m/z (E.I):

311.1(M’), 91.7(100)

237

10. Synthesis of 14

H
14
The tosylate 13 (1 g) was dissolved in butanone (10 mL) of, this solution

Ffi:

was added to a NaI solution (0.839 g in 10 mL of butanone). The solution was
boiled (about 80 °C) under reflux for 3 hours. The suspension was filtered and the
precipitate washed with acetone. The combined filtrates were concentrated in
vacuo. The resultant residue was partitioned between ethyl ether and sodium
hydrogen carbonate. The aqueous phase was washed 2x with ether and the ether
extracts were combined and washed with sat NaCl, and dried. The solvent was
evaporated to provide a 50% yield. 1H NMR (300 MHz CDC13): 8 3.72(2H, q,
1:6.22 Hz, CH2), 3.29(2H, t, 1:6.31 Hz, CH2), 1.54(2H, s). 13C NMR (300 MHz
CDC13): 8 157.25(d, J=37.8 Hz CCF3), 117.45 (t, J=287.45 Hz CF3), 41.92, 1.91.
IR: 3304.00(bs), 3100.00, 2361.17, 1705.29, 1558.75. m/z (E.I): 266.6(M+),

140.0(100).

11. Synthesis of 18 or 19

1 1
Ho/VgHa BTA’gHa Hogcvgkb BraCVgHS
18 18 17 19

In a hood. concentrated HBr (40%) (2 eq. 0.072 mmol) was added
dropwise to ethanol amine (16 or 17) (5 mg, 0.036 mmol) in a distillation flask.
The solution was heated to boiling and then it was refluxed over night. The
solution was cooled for 20 min. and the water was evaporated using vacuum

distillation to afford bromo ethylamine salt 19 (6.3 mg 87% ). ‘H NMR (300

238

MHz H20): 5 3.58(2H, dt, 1:120, 5.5 Hz, 13930112), 3.31(2H, 1, 1:4.7 Hz,

CH2NH3*). l3c NMR (300 M112 1120): 5 57.66, 27.94.

Protein Mutation, Bacterial Expression and Purification

A. Protein Mutation

Mutagenesis of all CRABPII proteins was accomplished by utilizing
Stratagene‘s QuikChange‘D protocol. This method entails the use of
complimentary mutation encoding primers. Following a typical PCR process the
reaction mixture will contain circular parental plasmid DNA (unmutated) and
linear mutated DNA (Figure 2-32). The original template DNA is different from
the mutated DNA, not only in that it is circular, but also it has methylated adenine
residues. As the DNA is copied in a bacterial host, some of the adenine bases
become methylated. The restriction enzyme DpnI recognizes the specific
sequence GATC (blunt cut between A and T), but will only cut where methylated
adenine residues exist. The plasmids used in our studies contain several cut sites
(between 10-20, depending on the vector used), thus incubation of the PCR
product with the DpnI restriction enzyme effectively destroys the parental plasmid
and leaves only the linear, mutated DNA. The desired product DNA can then be

incorporated into the host strain of choice (JM109 or XL1-Blue E. coli).

B. Plasmid Purification
Following successful transformation into either JM109 or XL1-Blue

competent cells (Tet resistant), a colony was grown up in 500 mL LB (containing

239

appropriate antibiotics) and purified with via QiagenQ column purification. Upon
completion of the Qiagen‘ID purification, the recovered DNA was resuspended in
approximately 400 111.. sterile water, and analyzed by UV-vis for concentration
determination and purity.
Concentration of DNA sample (11g / 11L) =
Absm x (50 pg / 1000 11L) x (volume of DNA used / total volume UV

sample)
Purity of DNA sample = Abszm / Ab82go

= 1.8, pure

>1.8, RNA contamination

< 1.8, protein contamination

C. Sample preparation for sequencing DNA

In a sterilized eppendorf tube (0.5 mL), DNA (2 11g) and primer (30 pmol)
were mixed. When sequencing is performed, the largest amount of base pairs that
are sequenced with accuracy is about 400 bp. Therefore, to sequence the
complete gene, five different primers were used.

1. 5 ’-GCA GA CT GGAA TGCAGTGAAGC-3’ 3330052

The DNA must be purified by a Qiagen” column and resuspended in
sterile water prior to analysis. Failure to do this will result in poor sequence
data. In addition, the sequence sample should be diluted with sterile water,

not buffer.

240

D. Melting temperature calculation for primers

Primers should ideally have melting points 2 78 °C, and end in at least
one, if not more, GC base pair

Tm = 81.5 °C + (0.41)(% GC) — 675 IN - % Mismatch

where N = primer length

E. Site directed Mutagenesis

E.l Typical PCR Conditions for Stratagene’s QuikChange Site-Directed

 

 

 

 

 

 

Mutagenesis Protocol:
PCR recipe
Template DNA 10 ng
Primer 1 15 pMol
Primer 2 15 pMol
DNTP 200 11M
2 U
10foubuffer 511L(1x)
H20 50-RXI‘ULL
PCR grofiam
1 X 94 °C 5 min
94 °C 1 min
30 X 48 °C 1 min
72 °C 1 min
1 x 72 °C 10 min
1 x 25 °C 10 min
E.2. PCR Primers
Primere Sequence PCR Templete

5'-CGGACATATGGAGATAATATFTGGCCAGA
CRABPII-CLE ATAAGAAAGAACAGCTGG-a’ CRABPII-KLE
5’ -GACCCGGGTACCCTCTGGTGCTGTCGGATC-S’

241

Following the PCR reaction and agarose gel verification, 1 11L of DpnI is added to
the reaction tube and allowed to incubate at 37 °C for 1 h, after which time 1 - 5
1.1L of reaction is transformed into 100 111. of either JM109 or XL1-Blue

competent cells.

F. Protein Characterization and Binding Assays

F.l. Calculating Extinction Coefficients of CRABPII mutants

The absorption extinction coefficients (a) for the various CRABPII
mutants were determined according to the method first described by Gill and von
I-Iippel.93 The 8d“, (8 of denatured proteins in 6 M Guanidine HCl) is dependant
primarily on the number of Tyr, Trp and Cys residues present and can be
calculated according to equation:

em = as.” + be“, + csCys
where a, b, and 0 represent the number of respective amino acids per molecule of
protein, and their 8 values have been determined experimentally (en-9:569O M‘l
cm“, s,,:1280 M" cm", s,,,:120 M" cm", wild type CRABPII protein contains
3 Trp, 2 Tyr, and 3 Cys residues).

Two protein solutions are prepared, at identical concentrations, one in the
native buffer (4 mM NaH2P04, 16 mM Na2HPOa, 150 mM NaCl, pH=7.3) and
the other in a denaturing buffer (6 M guanidine HCI, 4 mM NaH2P04, 16 mM

Na2HPOa, 150 mM NaCl), and the Abs2go for each sample is measured.

242

The am, the extinction coefficient for the native protein, can be determined
according to Beer’s law:
Abs dcn+ 8dcn = C den
Abs m-I- em = c m.
where Abs is the UV absorbance at 280 run, under both native and denatured

conditions. Since the concentration of both samples is the same, we can equate

= (Absmxem)
m" (Absm)
the two equations and solve for the extinction coefficient by:

G. Typical preparation of competent cells

The E. coli strain of interest was grown (37 °C until ODooo of 0.4-0.6, the
media contained LB, the corresponding antibiotic). After about three h The cells
were harvested by centrifugation at 5,000 RPM for 10 min at 4 °C. The cells were
re-suspended (100 mL of 0.9% NaCl) and then centrifugated at 5,000 RPM for 10
nrin at 4 °C. The cells were re-suspended (50 mL of 100 mM CaCl2) and
incubated for 30 min at 0 °C. The cells were centrifuged at 5,000 RPM for 10
nrin at 4 °C, and the cells were re-suspended (4 mL of 100 mM CaCl2, 15%
glycerol). The cells in suspension were aliquoted (0.1 mL) and were quickly

frozen with liquid nitrogen.

H. Typical expression of CRABPII mutants
The DNA obtained form the site directed mutagenesis protocol was

transformed into BL21(DE3)pLysS cells using the standard protocol. The cells

243

were plated on an ampicilin-chloramphenicol LB plates. The number of colonies
obtained varied from 50-1000. A single colony was inoculated (100 or 200 mL of
LB ampicillin-Chloramphenicol) and grown at 37 °C overnight. This culture was
used to inoculate a fresh culture (30 ml per liter, LB with ampicillin and
Chloramphenicol) this culture was grown (ODooo = 0.6-1.0. approximately 3 h).
Expression was induced by addition of IPTG (1 mM) and the culture was
incubated at 30 °C for about 6 h. The cells were harvested by centrifugation
(6000 rpm, 30 min) and frozen at -20 °C overnight. The cell mass was thawed
and resuspended in Tris HCl (pH=8.0, 10 mM, 100 mL for 4 L expression). The
cells were sonicated (probe sonicator, 60% power, 3x1 min), the mixture was spun
down (20 rrrin, 4 °C, 7000 RPM) and the supernatant was collected. At this point
MgSOa (1mM, 300 11L) and benzonase (20 11L) were added to reduce viscosity.
The mixture was incubated on ice for 30 min. Then the cells were collected (6000
RPM, 30 min) and the supernatant was collected to be loaded on the Fast-Q

column.

I. Fast-Q column purification.

A detailed description of the purification of CRABPII mutants is provided
in Chapter 2, of Chrysoula Vasileiou’s Dissertation. In short, the crude CRABPII
expression was purified by affinity chromatography using Q Sepharose, Fast Flow
resin. This quaternary ammonium based resin is a strong anion exchanger. After
loading the crude protein, the column was washed with 10 m1. of 10 mM Tris

HCl, pH=8.0. CRABPII-mutants were eluted with 100 mL Buffer of 10 mM Tris

244

BC], 100 mM NaCl, pH=8.0. The collected fraction was concentrated and
desalted by a stirred Ultrafiltration Cell (Millipore). The concentrated sample was
dissolved in 10 mM Tris HCl, pH=8.0 to a final volume of 50 mL in order to be

further purified by FPLC.

J. FPLC protocol

A detailed description of the purification of CRABPII mutants is provided
in Chapter 2, of Chrysoula Vasileiou’s Dissertation. In short, the 50 mL obtained
from the Fast Q were applied into FPLC (Biol.ogics Duo Flow, BioRad). The
CRABPH proteins elute at ~4% NaCl. The purity of the protein analyzed by

SDS-PAGE electrophoresis, is greater than 90%.

K. UV-vis titrations of CRABPII with all-trans-retinal

A stock solution of the all-rrans-retinal was prepared in 95% spectroscopy
grade EtOH. All-trans-retinal, 8::48,000M"cm'1, Xm=380 nm. For a 5 M
protein solution in a buffer 4 mM NaH2P04, 16 mM Na2I-IP04, 150 mM NaCl,
pH=7 .3.

Additions from 0.1 -— 1.0 equivalents at 0.1 equivalent increments were
added and spectra recorded at room temperature from 200 - 700 nm (Cary
WinUV, Varian). The 2"" derivative of the spectra was calculated by using the
corresponding software provided with the UV instrument. The determination of
the Am of a UV peak using mathematical equations has been established as a

valid method, and has been used in a variety of applications.“'95 As previously

245

mentioned, all the UV data discussed and shown in tables refer to spectra recorded

in the presence of 0.1 — 0.2 equivalents of chromophore, unless otherwise noted.

L. Fluorescence and MALDI-TOF.
A detailed description of the determination of binding constant for the
CRABPII mutants using retinal, is provided in Chapter 2, of Chrysoula

Vasileiou’s Dissertation.

M. Molecular Modeling:

Computational results were obtained using software programs from
Accelrys. Dynamics and rninirnizations were done using the Discover 3Q
program, using the CVFF forcefield from within the InsightII 2000 molecular
modeling system. Protein figures were generated using PyMol Molecular

Graphics System, version 0.93, copyright by DeLano Scientific LLC.

N. lntein Mediated Native Chemical Ligation

 

 

PCR

CRABPH (1-129) (50 ng) 511.1
BBB24 (NcoI) 4 111
BBBS8 (SapI) 7 111
Mg804 (500M) 1 111
Deep vent l U 1 111
DNTP (20011M) 10 mM
Buffer 10x 10 111
H20 70 1.1.1

 

246

 

94 °C
94 °C
55 °C
72 °C
72 °C
23 °C

Program

0. QIAquick gel extraction kit protocol

The DNA fragment from the agarose gel was extracted with a clean sharp
scalpel. The silica gel was weight and sliced in a colorless tube. Buffer QG was
added (3 volumesto 1 volume of gel, 100 mg-lOO ill). The mixture was incubated
at 50 °C for 10 min until the gels slice were dissolved. After the gel slice was
dissolved completely the solution was placed in a QIAquick spin column in a
provided 2m] collection tube. To bind DNA, apply the sample to the QIAquick
column, and centrifuge for 1 min. The flow-through was discarded and washed
with Buffer PE (0.5 ml) and the centrifugation was repeated for l nrin Buffer PB.

The DNA was eluted with Buffer EB (50 111) and centrifuged for 1 min at

maximum speed. The average eluant volume is 48 111 from 50 111 elution buffer

volume.

P. Digestions

5 min
1 min
1 min
105 sec
10 min
10 min

} 30 cycles

 

 

Digestions
CRABPH (1-129) (from PCR) 48 ill
NcoI l 111
SapI 1 1.11
Buffer NEB 4 (10X) 6 pl
H20 4 ill

 

o Incubation at 37 °C for 2 hours

247

o NEBuffer 4 (20mM Tris-acetate, 10 mM magnesium acetate, 50mM

potassium acetate, 1mM dithiothreitol (DTT) and pH 7.9).

 

 

 

Q. Ligations
antions

1/10 1/15
CRABPH (1-129) 4.5xlo°"moles 6.8x10'“moles
pTYB3 45xlo"‘moles 4.5x10°“moles
T4 ligase 1 1.11 l 111
Buffer (10X) 1X 1 X
H2O

 

o Incubation at 16 °C for 24 hours

R. lntein-CRABPII fused protein expression

Transformation of the target gene (pTYB3-CRABPII) into E. coli
(ER2566, BL21(DE3)pLysS, Tuner(DE3)pI.acI) according to previous protocol
affords the expression system. A single colony was inoculated in 100 ml
LB/(ampicilin ER2566) (Ampicillin and Chloramphenicol in BL21(DE3)pLysS)
and grow at 37 °C overnight. A fresh LB/antibiotic resistant medium was
inoculated with 30 ml (for l L) of the overnight culture and it was grown at 37 °C
until ODsoo of about 1.0. Expression was induced by addition of IPTG (1 mM
final concentration) and incubated at 30 °C for 6 h or 16 °C overnight. Harvested
cells (centrifugation 7000 RPM for 20 min) and resuspended in Buffer A. The
cells were lysed by sonication for 1 min 3 times at 4 °C. After lysis (or not)
Triton X—100 was added (final concentration of 1%) and then it was incubated at

room temperature for 30 minutes. The insoluble fraction was collected by

248

centrifugation 7,000 RPM (1 h at 4 °C). The supernatant and unsoluble fractions

were analyzed by acrylamide gel.

S. Chitin binding and cleavage analysis

About 1 mL chitin resin was used for 2 mL of supernatant or solubilized
inclusion bodies. The solution was soaked with the resin for about 2 h. The
column was empty with a flow through at rate of 0.3 ml/min. The eluant was
collected and resubmitted to the column keeping the same flow rate. The column
was washed with 100 mL of 25 mM Tris, 200 mM NaCl, pH 7.5. Cleavage of the
intein protein was induced with 1 mM DTT or 10 mM MESNA (fresh solution
should be used) at RT overnight.
Expression of CRABPII mutants.

The CRABPH mutants were analyzed by sequencing. Followed by

expression of the mutant protein. The methods used were previously described

T. Refolding experiments

Preparation of the inclusion bodies was achieved by following the protocol
of expression of fusion intein-CRABPII proteins. However, no TritonX-IOO was
added and the pellet contained the inclusion bodies. The refolding experiments
were done in small scale (1.5 mL microcentrifuge tubes) and sixteen different
conditions were tested. 950 11L of Foldlt reagent (1-16) were added.
Approximately 0.1 mg of the protein was added to each tube. The tubes were

placed on a rocker and incubated at 4 °C overnight. The samples were diluted (50

249

times or. 10 times). Each one of the refolding conditions was tested for intein

cleavage (following the previously shown protocols).

The content of each one of the refolding systems is shown in the following table.

 

 

 

 

Solution 1 Solution 2 Solution 3 Solution 4
55 mM Tris pH 8.2 55 mM MES pH 6.5 55 mM MES pH 6.5 55 mM Tris pH 8.2
264 mM NaC] 10.56 mM NaC] 10.56 mM NaCl 264 mM NaCl
11 mM KC1 0.44 mM KC] 0.44 mM KC] 11 mM KC]
0.055% PEG 3350 550 mM Guanidine HC] 0.055% PEG 3350 2.2 mM MgC]
1.1 mM EDTA 2.2 mM MgCl 550 mM Guanidine HCl 2.2 mM CaCl
2.2mMCaC] leMEDTA 440mMSucrose
440 mM Sucrose 550 mM bArginine
550 mM L-Arginine
Solution 5 Solution 6 Solution 7 Solution 8
55 mM MES pH 6.5 55 mM Tris pH 8.2 55 mM Tris pH 8.2 55 mM MES pH 6.5
264 mM NaCl 10.56 mM NaCl 10.56 mM NaCl 264 mM NaCl
11 mM KC1 0.44 mM KC] 0.44 mM KCl 11 mM KC]
2.2 mM MgCl 0.055% PEG 3350 550 mM Guanidine HC] 0.055% PEG 3350
2.2 mM CaCl 550 mM Guanidine RC] 2.2 mM MgC] 1.] mM .TA
440 mM Sucrose 1.1 mM EDTA 2.2 mM CaCl 550 mM L-Arginine
440 mM Sucrose 550 mM L-Arginine
Solution 9 1. Solution 9 Solution 11 Solution 12
55 mM MES pH 6.5 55 mM Tris pH 8.2 55 mM Tris pH 8.2 55 mM MES pH 6.5
264 mM NaCl 10.56 mM NaC] 10.56 mM NaCl 264 mM NaCl
11 mM KC] 0.44 mM KC1 0.44 mM KC] 11 mM KCl
0.055% PEG 3350 1.] mM EDTA 0.055% PEG 3350 550 mM Guanidine HC]
550 mM Guanidine HC] 440 mM Sucrose 2.2 mM MgCl 1.1 mM EDTA
2.2 mM MgCl 2.2 mM CaCl 550 mM L-Arginine
2.2 mM CaCl 550 mM bArginine
440 mM Sucrose
Solution 13 Solution 14 Solution 15 Solution 16
55 mM Tris pH 8.2 55 mM MES pH 6.5 55 mM MES pH 6.5 55 mM Tris pH 8.2
264 mM NaCl 10.56 mM NaCl 10.56 mM NaCl 264 mM MC]
11 mM KC] 0.44 mM KC] 0.44 mM KC] 11 mM KC]
550 mM Guanidine HCl 0.055% PEG 3350 1.1 mM EDTA 0.055% PEG 3350
1.1 mM EDTA 2.2 mM MgCl 440 mM Sucrose 550 mM Guanidine HCl
2.2 mM CaCl 550 mM L-Arginine 2.2 mM MgCl
2.2 mM CaCl
440 mM Sucrose

 

 

 

550 mM L-Arginine

 

 

T. Conversion of Cys to Lys

To a protein (1-10 mg/mL in 25 mM Tris, 200mM NaCl pH 8.00) the
alkylating reagent was added in a 10 and 100 fold molar excess over the (number
of sulfhydryl groups present in the mutant CRABPH). This calculation should
include the sulfhydryl groups contributed by any DTT added. The reaction was
incubated at 37 °C for 4 h. A buffer exchange was performed via filtration. (4
mM NaH2P04, 16 mM Na2HPOa, 150 mM NaCl, pH=7.3). The obtained
concentrated solution was tested for retinal binding (the equivalents of retinal
were calculated based on the total amount of protein present). Retinal binding
was performed as previously described. In cases where the MALDI-TOF was

acquired the same protocol as previously discussed was used.

Note. The stock solution of alkylating agents should be 750 mM in MeOH, and

10 mL were added to a 1 mL of protein solution. The final concentration of

methanol in the reaction solution should always be below 10%.

251

5.6

References

S. Gomez; J. A. Peters; T. Maschmeyer, "The reductive amination of
aldehydes and ketones and the hydrogenation of nitrileszMechanistic
aspects and selectivity control." Advanced Synthesis and Catalysis 2002,
344, (10), 1037-1057.

C. Vasileou. Protein Desiganeengineering Cellular Retinoic Acid Binding
Protein 11 into a Retinal Binding Protein. Michigan State University, East
Lansing, 2005.

L. C. Wang; A. Y.; F. Li; J. L. Markley; H. G. Yan, "NMR solution
structure of type 11 human cellular retinoic acid binding protein:
Implications for ligand binding." Biochemistry 1998, 37, 12727-12736.

R. M. Czerwinski; T. K. Harris; W. H. Johnson; P. M. Legler; J. T.
Stivers; A. S. Mildvan; C. P. Whitman, "Effects of mutations of the active
site arginine residues in 4-oxalocrotonate tautomerase on the pK(a) values
of active site residues and on the pH dependence of catalysis."
Biochemistry 1999, 38, 12358-12366.

R. M. Czerwinski; T. K. Harris; M. A. Messiah; A. S. Mildvan; C. P.
Whitman, "The structural basis for the perturbed pKa of the catalytic base
in 4-oxalocrotonate tautomerase: Kinetic and structural effects of
mutations of Phe-SO." Biochemistry 2001, 40, (1984-1995).

A. F. L. Creemers; C. H. W. Klaassen; P. H. M. Bovee-Geurts; R. Kelle;
U. Kragl; J. Raap; W. I. d. Grip; J. Lugtenburg; H. J. M. d. Groot, "Solid
state N-15 NMR evidence for a complex Schiff base counterion in the
visual G-protein-coupled receptor rhod0psin." Biochemistry 1999, 38,
(7195-7199).

Y. Inoue; Y. Tokito; S. Tomonoh; R. Chujo, "C-13 Cherrrical-shifts of
retinal isomers and their Schiff-bases as models of visual chromophores."
Bulletin of the Chemical Society of Japan 1979, 52, 265-266.

M. Eilers; P. J. Reeves; W. Ying; H. G. Khorana; S. O. Srrrith, "Magic
angle spinning NMR of the protonated retinylidene Schiff base nitrogen in
rhodopsin: expression of 15N-lysine- and l3C-glycine-labeled opsin in a

252

10.

11.

12.

13.

14.

15.

16.

stable cell line." Proceedings of the National Academy of Sciences of the
United States of America 1999, 96, 487-492.

J. Lugtenburg; A. F. Creemers; M. A. Verhoeven; A. A. C. Wijk; P. J. E.
Verdegem; M. C. Monnee; F. Jansen, "Synthesis of C-13-labeled
carotenoids retinoids." Pure Applied Chemistry 1999, 71, 2245-2251.

J. A. Pardoen; P. P. J. Mulder; E. M. M. Vanderberg; J. Lugtenburg,
"Synthesis of 8-mono-C-l3-retinal, 9-mono-C-13-retinal, 12-mono-C-13-
retinal, lB-mono-C-l3-retinal." Canadian Journal of Chemistry -Reviews
1985, 63, (1431-1435).

J. Lugtenburg, "The synthesis of C-13-labeled retinals." Pure Applied
Chemistry 1985, I985, (57), 753-762.

R. R. Birge; L. P. Murray; R. Zidovetzki; H. M. Knapp, "2-Photon, C-13
and two-dimensional H-l-NMR spectroscopic studies of retinyl Schiff-
bases, protonated Schiff-bases, and Schiff-base salts- evidence for a
protonation induced pi -pi-star excited-state level ordering reversal."
Journal of the American Chemical Society 1987, 109, 2090-2101.

A. Albeck; N. Livnah; H. Gottlieb; M. Sheves, "C-13 NMR-Studies of
model compounds for bacteriorhodopsin-factors affecting the retinal
chromophore chemical-shifts and absorption maximum." Journal of the
American Chemical Society 1992, 114, 2400-241 1.

Y. Inoue; Y. Takahash; Y. T. A; R. Chujo; Y. Miyoshi, "C-13 NMR-
spectra of retinal-l and its related compounds." Organic Magnetic
Research 1974, 6, 487-491.

Y. Inoue; Y. Tokito; R. Chujo; Y. Miyoshi, "Study of pi-electron
delocalization in model compounds of visual pigment by UV and C-13
NMR-spectra"." Journal of the American Chemical Society 1977, 99,
5592-5596.

G. S. Harbison; S. O. Smith; J. A. Pardoen; J. M. I... Courtin; J.
Lugtenburg; J. Herzfeld; R. A. Mathies; R. G. Griffin, "Solid-state C-l3
NMR detection of a perturbed 6-S-trans chromophore in
bacteriorhodopsin." Biochemistry 1985, 24, 6955-6962.

253

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

S. 0. Smith; H. J. M. Degroot; R. Gebhard; J. M. L. Courtin; J.
Lugtenburg; J. Herzfeld; R. G. Griffin, "Structure and protein environment
of the retinal chromophore in light-adapted and dark-adapted
bacteriorhodopsin studied by solid-state NMR." Biochemistry 1989, 28,
8897-8904.

J. B. Lambert; H. F. Shurvell; D. A. Lightner; R. G. Cooks, Organic
Structural Spectroscopy. ed.; Prentice-Hall, Inc: New Jersey, 1998; 'Vol.' p
25-27.

L. Mou; G. Singh, "Synthesis of (s)-2-amino-8-oxodecanoic acid and
apicidin A." Tetrahedron Letters 2001, 42, 6603-6606.

J. M. Scholtz; P. A. Barlett, "A convenient differential protection strategy
for functional group maniulation of aspartic and glutamic acids." Synthesis
1989, 542-544.

P. D. Baley; J. S. Bryans, "Chiral synthesis of 5-hydroxy-(L)-pipecolic
acids from (L)-glutamic acid." Tetrahedron Letters 1988, 29, (18), 2231-
2234.

R. Forch; A. Rosowsky, "Synthesis of y-[ISN]-L-glutamyl derivatives of
5,10didezantetrahydrofolate." Journal of Labelled Compounds and
Radiopharmaceuticals 1999, 42, 1 103-1117.

A. Sutherland; C. L. Willis, "Synthesis of [6-13C]-L-Lysine." Journal of
Labelled Compounds and Radipharmaceuticals 1996, 38, (1), 95-102.

R. K. Olsen; K. Ramasamy; T. Emery, "Synthesis of N,N-Protected N-
hydroxy-L—ornithine from L-glutamic acid." Journal of Organic Chemistry
1984, 49, 3527-3534.

M. Rodriguez; M. Linares; S. Doulut; A. Heitz; J. Martinez, "A facile
synthesis of chiral n-protected amino alcohols." Tetrahedron Letters 1991,
32. (7). 923-926.

K. Soai; S. Yokoyama; K. Mochida, "Reduction of symmetric and mixed
anhydrides of carboxylic acids by sodium borohydride with dropwise
addition of methanol." Synthesis 1987, 647-648.

254

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

E. R. Civitello; H. Rapoport, "Synthesis of the enantiomeric
furobenzofurans, late precursors for the synthesis of (+)- and (-)-aflatoxins
Bl, B2, 01, and G2." Journal of Organic Chemistry 1994, 59, (14), 3775-
3782.

H. Walter, "Catalytic reduction of nitriles and oximes." Journal of the
American Chemical Society 1928, 50, 3370-3374.

R. Lusskin; J. Ritter, "A new reaction of nitriles. V. Preparation of N-(2-
halo-l-ethyl)-amides." Journal of the American Chemical Society 1950,
72, 5577-5578.

F. Benson; J. Ritter, "A new reaction nitriles. III. Amides from dinitiles."
Journal of the American Chemical Society 1949, 71, 4128-4129.

H. Ritter; P. Minieri, "A new reaction of nitriles. I Amides form alkenes
and mononitriles." Journal of the American Chemical Society 1948, 70,
4045-4048.

W. Huber, "Hydrogenation of basic nitriles with Raney nickel." Journal of
the American Chemical Society 1944, 66, 876-879.

B. Klenke; I. H. Gilbert, "Nitrile reduction in the presence of Boc-
protected amino groups by catalytic hydrogenation over palladiu activated
raney nickel." Journal of Organic Chemistry 2001, 66, 2480-2383.

C. Stephen; J. Duncan; L. Alexandra; R. K. de K; M. Williams; R. V.
Meredith, "A generic approach for the catalytic reduction of nitriles."
Tetrahedron 2003, 59, 5417-5423.

H. Ritter; P. Minieri, "A new reaction of nitriles. I. Amides form alkenes
and mononitriles." Journal of the American Chemical Society 1948, 70,
4045-4048.

S. Gwoda; C. Gowda, "Application of hydrazinium monoformate as new
hydrogen donor with Raney nickelzfacile reduction of nitro and nitrile
moieties." Tetrahedron 2002, 58, 2211-2213.

255

37.

38.

39.

40.

41.

42.

43.

45.

M. D. Jaffari; J. T. Mahar; R. L. Bachert Ion Exchange recovery of L-
Lysine. 1989.

E. Takahashi; M. Furui; H. Seko; T. Shibatani, "D-Lysine production from
L-Lysine by successive chemical racemization and microbial asymmetric
degradation." Applied Microbiology and Biotechnology 1997, 4 7, 347-351.

R. N. Costilow; O. M. Fochovansky; H. A. Barker, "Isolation and
identification of b-lysine as an intermediate in lysine fermentation."
Journal of Biochemical Chemistry 1966, 241, (7), 1573-1580.

R. Schuster, "Determination of free amino acids by high performance
chromatography." Analytical Chemistry 1980, 52, 617-620.

J. D. Aberhart; S. J. .Gould; H.-J. Lin; T. K. Thiruvengadam; B. H.
Weiller, "Stereochemistry of lysine 2,3aminomutase isolated form
Clostridium subtenninale strain 8B4." Journal of the American Chemical
Society 1983, 105, 5461-5470.

A. Wlodawer; M. Miller; M. J askolski; B. K. Sathyanarayana; E. Baldwin;
1. T. Weber; L. M. Selk; L. Clawson; J. Schneider; S. B. H. Kent,
"Conserved folding in retroviral proteases crystal-structure of a synthetic
HIV-1 protease." Science 1989, 245, 616-621.

M. Miller; J. Schneider; B. K. Sathyanarayana; M. V. Toth; G. R.
Marshall; L. Clawson; L. Selk; S. B. H. Kent; A. Wlodawer, "Structure of
complex of synthetic HIV -1 protease with a substrate-based inhibitor at
2.3-A resolution." Science 1989, 246, 1149-1152.

J. Schneider; 8. B. H. Kent, "Enzymatic-activity of a synthetic-99 residue
protein corresponding to the putative HIV -1 protease." Cell 1988, 54, (3),
363-368.

M. Jaskolski; A. G. Tomasselli; T. K. Sawyer; D. G. Staples; R. L.
Heinrikson; J. Schneider; 8. B. H. Kent; A. Wlodawer, "Structure at 2.5-A
resolution of chemically synthesized human-immunodeficiency-virus type-
1 protease complexed with a hydroxyethylene-based inhibitor."
Biochemistry 1991, 30, (6), 1600-1609.

256

46.

47.

48.

49.

50.

51.

52.

53.

54.

T. M. Hackeng; P. E. Dawson, "Protein synthesis by native chemical
ligation: expanded scope by using straightforward methodology."
Proceedings of the National Academy of Sciences of the United States of
America 1999, 96, 10068-10073.

P. E. Dawson; M. J. Churchill; M. R. Ghadiri; S. B. Kent, "Modulation of
reactivity in native chemical ligation through the use of thiol." Journal of
the American Chemical Society 1997, 119, (19), 4325-4329.

T. M. Hackeng; C. M. Mounier; C. Bon; P. E. Dawson; J. H. Griffin; S. B.
H. Kent, "Total chemical synthesis of enzymatically active human type II
secretory phospholipase A2." Proceedings of the National Academy of
Sciences of the United States of America 1997, 94, 7845-7850.

P. E. Dawson; M. J. Churchill; M. R. Ghadiri; S. B. H. Kent, "Modulation
of reactivity in native chemical ligation through the use of thiol additives."
Journal of the American Chemical Society 1997, 119, (19), 4325-4329.

T. M. Hackeng; C. M. Mounier; C. Bon; P. E. Dawson; J. H. Griffin; S. B.
H. Kent, "Total chemical synthesis of enzymatically active human type II
secretory phospholipase A(2)." Proceedings of the National Academy of
Sciences of the United States of America 1997, 94, (15), 7845-7850.

T. M. Hackeng; J. H. Griffin; P. E. Dawson, "Protein synthesis by native
chemical ligation: Expanded scope by using straightforward
methodology." Proceedings of the National Academy of Sciences of the
United States of America 1999, 96, (18), 10068-10073.

P. E. Dawson; T. W. Muir; I. Clarklewis; S. B. H. Kent, "Synthesis of
proteins by native chemical ligation." Science 1994, 266, 776-779.

P. E. Dawson; S. B. H. Kent, "Synthesis of native proteins by chemical
ligation." Annual Review of Biochemistry 2000, 69, 923-960.

S. R. Chong; F. B. Mersha; D. G. Comb; M. E. Scott; D. Landry; L. M.
Vence; F. B. Perler; J. Benner; R. B. Kucera; C. A. Hirvonen; J. J.
Pelletier; H. Paulus; M Q. Xu, "Single—column purification of free
recombinant proteins using a self-cleavable affinity tag derived from a
protein splicing element." Gene 1997, 192, (2), 271-281.

257

55.

56.

57.

58.

59.

60.

61.

62.

63.

T. T. Hoang; Y. F. Ma; R. J. Stern; M. R. McNeil; H. P. Schweizer,
"Construction and use of low-copy number T7 expression vectors for
purification of problem proteins: purification of Mycobactcrium
tuberculosis leD and Pseudomonas aeruginosa LasI and Rhll proteins,
and functional analysis of purified Rhll." Gene 1999, 237, (2), 361-371.

S. Mathys; T. C. Evans; 1. C. Chute; H. Wu; S. R. Chong; J. Benner; X. Q.
Liu; M. Q. Xu, "Characterization of a self-splicing mini-intein and its
conversion into autocatalytic N- and C-terminal cleavage elements: facile
production of protein building blocks for protein ligation." Gene 1999,
231, (1-2), 1-13.

G. J. Cotton; T. W. Muir, "Peptide ligation and its application to protein
engineering. " Chemistry and Biology 1999, 6, (9), R247-R256.

T. Otomo; N. Ito; Y. Kyogoku; T. Yamazaki, "th observation of
selected segments in a larger protein: Central-segrnent isotope labeling
through intein-mediated ligation." Biochemistry 1999, 38, 16040-16044.

K. Alexandrov; I. Heinemann; T. Dunek; V. Sidorovitch; R. S. R. S.
Goody; H. J. Waldrnann, "lntein-mediated synthesis of geranylgeranylated
Rab7 protein in vitro." Journal of the American Chemical Society 2002,
124, (20), 5648-5649.

U. Arnold; M. P. Hinderaker; B. L. Nilsson; B. R. Huck; S. H. Gellman;
R. T. Raines, "Protein prosthesis: A semisynthetic enzyme with a -peptide
reverse turn." Journal of the American Chemical Society 2002, 124, 8522—
8523.

B. Ayers; U. K.Blaschke; J. A. Camarero; G. J. Cotton; M. Holford; T. W.
Muir, "Introduction of unnatural amino acids into proteins using expressed
protein ligation." Biopolymers 1999, 51, (5), 343-354.

U. K. Blaschke; J. Silberstein; T. W. Muir, "Protein engineering by
expressed protein ligation." Methods in Enzymology 2000, 328, (478-496).

U. K. Blaschke; G. J. Cotton; T. W. Muir, "Synthesis of multi-domain
proteins using expressed protein ligation: Strategies for segmental isotopic
labeling of internal regions." Tetrahedron 2000, 56, (48), 9461-9470.

258

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

J. A. Borgia; G. B. Fields, Chemical Synthesis of Proteins. Trends
Biotechnology 2000, I8, 6.

J. A. Camarero; A. Shekhtman; E. A. Campbell; M. Chlenov; T. M.
Gruber; D. A. Bryant; S. A. Darst; D. Cowbum; T. W. Muir,
"Autonegulation of a bacterial sigma factor explored by using segmental
isotopic labeling and NMR." Proceedings of the National Academy of
Sciences of the United States of America 2002, 99, (13), 8536-8541.

H. D. Mootz; T. W. Muir, "Protein splicing triggered by a small
molecule." Journal of the American Chemical Society 2002, 124, (31),
9044-9045.

T. Otomo; K. Teruya; U. K.; T. Yamazaki; Y. Kyogoku, "Improved
segmental isotOpe labeling of proteins and application to a larger protein."
Journal of Biomolecular NMR 1999, 14, 105-114.

T. J. Talbert; C. H. Wong, "lntein-mediated synthesis of proteins
containing carbohydrates and other molecular probes." Journal of the
American Chemical Society 2000, 122, 5421-5428.

M. Q. Xu; T. C. J. Evans, "Purification of recombinant proteins from
Escherichia coli by engineered inteins." Methods in Molecular Biology
2003, 205, 43-68.

G. Amitai; S. Pietrokovski, "Fine-tuning an engineered intein." Nature
Biotechnology 1999, 17, (9), 854 - 855.

Z. Y. Li; J. H. Fan; J. M. Yuan, "Single-column purification of
recombinant human neurotrophin-3 (hNT3) by using the intein mediated
self-cleaving system." Biotechnology Letters 2002, 24, 1723-1727.

w. A. Prinz; F. Aslund; A. Holmgren; J. J. Beckwith, "The role of the
thioredoxin and glutaredoxin pathways in reducing protein disulfide bonds
in the Escherichia coli cytoplasm." Joumal of Biological Chemistry 1997,
272, (25), 15661-15667.

E. LaVallie; E. A. Diblasio; S. Kovacic; K. L. Grant: P. F. Schendel; J. M.
McCoy, "A thioredoxin gene fusion expression system that circumvents

259

74.

75.

76.

77.

78.

79.

80.

81.

82.

inclusion body formation in the Escherichia coli cytoplasm."
Bio/1‘ echnology 1993, 11, 187-193.

L. Lobel; S. Pollak; B. Lustbader; J. Klein; J. W. Lustbader, "Bacterial
expression of a natively folded extracellular Domain fusion protein of the
hFSH receptor in the cytoplasm of Escherichia coli." Protein Expression
and Purification 2002, 25, (1), 124-133.

L. I. Lobe]; S. Pollak; J. Klein; J. W. Lustbader, "High-level bacterial
expression of a natively folded, soluble extracellular domain fusion protein
of the human luteinizing hormone/chorionic gonadotropin receptor in the
cytoplasm of Escherichia coli." Endocrine 2001, I4, (2), 205-212.

M. Albrecht: N. Misawa; G. Sandman, "Metabolic engineering of the
terpenoid biosynthetic pathway of Escherichia coli for production of the
carotenoids -carotene and zeaxanthin." Biotechnology Letters 1999, 21,
791-795.

In IMPACTIM-CN System Instruction manual, New England Biolabs, ed.;
'Ed.""Eds.' 'Vol.' pApp.

L. Gregori; M. S. Poosch; G. Cousins; V. Chaw, "A uniform isopeptide-
linked multiubiquitin chain is sufficient to target substrate for degradation
in ubiquitin mediated proteolysis." Journal of Biological Chemistry 1990,
265, (15), 8354-8357.

S. Hartrnann; J. Hofsteenge, "Properdin, the positive regulator of
complement is highly C. mannosylated." Journal of Biological Chemistry
2000, 275, (37), 28569-28574.

J. Piotrowski; e. al, "Inhibition of the 26 S proteasome by polyubiquitin
chains synthesized to have different lengths." Journal of Biological
Chemistry 1997, 272, (38), 23712-23721.

C. E. Hopkins; e. al, "Chemical modification rescue assessed by mass
spectrometry demonstrates that thia-lysine yields the same activity as
lysine in aldolase." Protein Science 2002, 11, 1591-1599.

S. N. Gorlatov; T. C. Stadtman, "Human thioredoxin reductase from HeLa
cells: Selective alkylation of selenocysteine in the protein inhibits enzyme

260

83.

84.

85.

86.

87.

88.

89.

90.

91.

activity and reduction with NADPH influences affinity to heparin."
Proceedings of the National Academy of Sciences of the United States of
America 1998, 95, 8520-8525.

M. A. Raftery; D. Cole, "On the aminoethylation of proteins." Journal of
Biological Chemistry 1966, 241, (15), 3457-3461.

M. Rechsteiner, "Ubiquitin-mediated pathways for intracellular
proteolysis." Annual Reviews in Cellular Biology 1987, 3, 1-30.

A. Hershko, "Ubiquitin-mediated protein degradation." Journal of
Biological Chemistry 1988, 263, 15237-15240.

M. G. Goebl; e. al, "Yeast cell cycle gene CDC34 encodes a ubiquitin-
conjugating enzyme." Science 1988, 241, 1331-1335.

J. Slama; R. R. Rando, "The synthesis of glycolipids containing a
hydrophilic spacer-group." Carbohydrate Research 1981, 88, 213-221.

I. Bird; P. B. Farmer, "The synthesis of deuterium labelled 2-
bromoethanol, acrylonitrile, acrylamide, 2-aminoethanol, 2-
bromoethylamine hydrobromide and 1-bromo-2-cloroethane." Journal of
Labelled Compounds and Radipharmaceuticals 1988, 27, (2), 1999-216.

P. H. Bach; J. Bridges, "Synthesis of 2-Bromo[1-14C]ethanamine
hdrobromide." Journal of Labelled Compounds and Radipharmaceuticals
1981, 19, (3), 425-431.

V. P. Wystrach; D. W. Kaiser; F. C. Schaefer, "Preparation of
ethylenimine and tarieghlenemelamine." Journal of Organic Chemistry
1955, 77, 5915-5918.

B. Schutte; R. J. R. Weakle; D. R. Tyler, "Radical cage effects in the
photochemical degradation of polymers: Effect of radical size and mass on
the cage recombination efficiency of radical cage pairs generated
photochemically from the
(CpCH2CH2N(CH3)C(O)(CH2)nCH3)2M02(CO)6 (n =3)." Journal of the
American Chemical Society 2003, 125, 10319-10326.

261

92.

93.

94.

95.

A. S. Nagle; e. al., "Efficient snthesis of beta-amino bromides."
Tetrahedron Letters 2000, 41, 3011—3014.

S. C. Gill; P. H. Vonhippel, "Calculation of protein extinction coefficients
from amino-acid sequence data." Analytical Biochemistry 1989, 182, (2),
319-326.

D. C. Lee; J. A. Hayward; C. J. Restall; D. Chapman, "2nd-Derivative
infrared spectroscopic studies of the secondary structures of
bacteriorhodopsin and Ca-2+-ATPase." Biochemistry 1985, 24, (16),
4364-4373.

M. P. Horvath; R. A. Copeland; M. W. Makinen, "The second derivative
electronic absorption spectrum of cytochrome c oxidase in the Soret
region." Biophysical Journal 1999, 77, (3), 1694-1711.

262

Chapter 6

Interaction of 1 l-cis retinal and other chromophores with the protein

rhodopsin mimics.

6.1 Introduction

We are interested in exploring the nature of protein—substrate interactions
at the molecular level since they are at the heart or biochemical events that
regulate biological systems. Wavelength regulation in rhodopsin is a remarkable
example of such interaction in which different proteins elicit different properties
from the same substrate, ll-cis-retinal 1. Visual transduction is initiated by
photo-isomerization of the ll-cis-retinal chromophore to all-trans-retinal 2, which
in turn leads to a series of conformational changes in rhodopsin, and eventually
results in the enzymatic cascade responsible for vision (Figure 6-1)."2 As
discussed previously, we chose to use wavelength regulation as a model to study
protein-substrate interactions and due to the
difficulty of the study of rhodopsin proteins, we
engineer a protein mimic of rhodopsin, that can

act as a surrogate.

 

The design of the engineered rhod0psin
was performed usmg all-trans-reunal 21nstead of Flour. 6-1. Representation of
ll-cis-retinal due to the difficulty in working with 11-cIs-retlnal and all-trans-

ll-cis-retinal. Other group members have retinal.

263

successfully developed a number of successful rhodopsin surrogates that bind all-
trans-retinal as a PBS (See Crist and Vasilieou’s Thesis). Therefore, we have
shifted our interested to optimizing our system for binding of ll-cis-retinal. At
the same time, we were intrigued in how the engineered proteins will interact with
different aldehydes that mimic the all-trans-retinal.

As mentioned previously we were interested in deconvoluting the factors
involved in wavelength regulation. Several theories have been proposed for the
mechanism of wavelength regulation?"4 Generally, an increase in the conjugation
results in a bathochromic shift, while decreasing the conjugation yields a blue
shifted pigment. Twisting about retinal’s single bonds decrease de-localization of
the PSB, which should result in a blue-shift.3'7 Using the triple mutant
R132K::R111L::L121E we developed a series of mutants with the hope of
deconvoluting some of the factors involved in wavelength regulation.

As discussed previously, ll-cis-retinal is bound as a Schiff base via Lys296,
in rhodopsin and Glull3 acts as the counter anion. The binding pocket of rod
rhodopsin and the residues directly surrounding the retinal molecule are shown in
Figure 6-2A. As can be seen, the ll-cis-retinal is twisted due to sterics, which
supports the suggestion that twisting around the single bonds dictates different
level of conjugation, therefore, changing the observed maximal absorption. The
indole ring of Trp265 is close to the C13-Me (3.8 A) causing some steric
interactions. Because deletion of this methyl group is known to cause partial
constitutive activity of rhodopsin in the dark,8 loss of its interaction with Trp265

may be a possible mechanism of this activity. Also the position of Trp265 can

264

suggest possible excitonic coupling with the polyene, as originally proposed by
Rafferty in 1979.9 The breakthrough in the crystallization of bovine rhodopsin
presented the opportunity for a better understanding of how the different pigments
perform spectral tuning over a wide range of wavelengths. Although the crystal
structures of the cone opsins have not been solved yet, their theoretical models
have been updated based upon the rhodopsin structure.10 Figure 6-2A Figure 6-
2B and Figure 6-2C shows the models for the three pigments, red, green, and blue

opsin.ll It can be observed that in bovine rhodopsin, the environment around the

Y178 M207
E1 81

?"\- 0.1-3

\ vzes

M.Ss

Flgure 6-2A. Binding pocket of the rod rhodopsin from the crystal

F261

structure. The chromophore, 11-cis-retinal is shown bound as a Schiff
base to Ly3296. Selected Important amino acids around the polyene
have been labeled.

265

M

 

 

er 1
Met 207 TW ‘2‘ (4Y0 )
(3.4 A) (4.3 ) , Tyr1 e
“269 PM 8: (30 A) .(3.8 ) x ’1 ’4’ (4-4 )
(3.7%)\ (4.8 ): \‘ : ’r it I! ’I' Tyr2
Ala272 x x : \ixmrw,’ ,x (3.5 )
<56“ 939‘s ‘i’ «err/24'": in:
His 211 ' ""’ ""' '
(4.0 A)
'3'
Phe 212-'
(3.9 A)
"""" o ’ I’ ‘ 3‘9 )

L901 'l ” ’ 0' \ l A\ (

(5,0 ).”I’ ”,v (Galualg', Gly120 21,85”): I”: r~flz- ..... All 2
P119261 ’ ' .' (5'8” ° .’ i “ (3'6 l
(3;, x) Trp 265 Gly121 ' ' PM

(3.9 A) (3.6 A) Thr Phe 1 (5-0 )
(5.9 ) (6.0 )
1KPN1blug)
Gly114
Na 269 (3.71) Phe15
. .v (5.9 )
(5.1x) ,. Glyi
Phe 5 (4.8 )
(4.7 ) ~ ,x' Glyi 1
(4.1 )
AI3266
(3.93) -~~.. (381‘;
c 208 --- Pnerrz
(:87 A) __________ (5.7 A)
Leu 213 """" ’ 1, ‘ ....... ((31:31 ;)
(5.9x) ‘\ N922
Pnezoe -' ,’ , :3 (3.6 )
(3.4x) ““1 ' . '. Seri s
’ . GI 13 ‘. ‘ (3'2 l
7333235? Tyr 262 Alarirl3-4 ) ($.19 ) ‘ Ser2 9
' (4,0 A) (5.83) Ph9290 (315 l
(5.4 A)

Figure 6-28. This figure shows the amino acid residues within 6 A of the
polyene for rhodopsin (Rod). and the colored pigment blue (blue). The
rhodopsin data is extrapolated form the crystal structure. The data for the blue

were obtained from analysis of theoretical models (RCSB proteln data bank

accession no 1KPN (blue).

266

1KPNjgreen)

Ala 288
(4'9 A) Pro 205 Tyr 1 1
Leu 223 . (3 ,8 2i) (5.5 )
I Try r194
“‘4 A) ' er 1E4Hp183 (50A)
I r
. ’ Giy204
Met 2 4 . 3 6 (5 o ,'
(47 : ( ' / A) '7, ’64530
‘ ‘ \t: r ’l’,:l’ Gly130
\ l V [‘4' ,.-(4.2 A)

   
     

    
  

12392;? .. 2’. Cys 293 Glu1§9
' (3.1 A) , (3.4 )
Leu 141 __ '.-'
(3‘7 ) ...... y ______ Val 133
Cys 228 ------- , 1 (3.4 A)
(3.9 A) ........ Ser 2 2
Cys 227 I. . N“. (3.3 )
(4-9 A) ' net 8' JJW
Ph 277 Trp 281(3 5 i ‘, G'" ‘02
e (4.0 A) ' . (3.7 A) . \ (A)
(4-5 A) - Ph 309
Cys 1,36 Ala 306 9
(5.9 A) (4.0 A) (5.8 11)
Leu 223
1 KPX 1’ d (40 A) Pro 5 T 1 4
Cys 227- (3.5 ) (5'0 )
(40A) - ‘
, Ser1Tyr219 .’
Met2 4 ‘. ; (3.0 ); (5.0A) .'
(4.7 ) 4‘ ‘t I “t r, ’7'
“‘ \ I ‘ O I ” G'Y
\ V 3’ x (4.1 l Glytio
Thr 285 ,’ .' (4.2 A)
(3.3 A) ..2, ..s’Cys 293 Glu 1E9
Leu141-’ (3.1 A), (3.5 )
(3.7 A) ‘ \‘131', , -- Val 13
(3.3A)
“3'," . \ ,,,,, Sen? 2
Leu141 ..v:,/ x , (40 ) ’4 Ntfi.‘ (33 l
(3.7 A) 'l’ '1’ .I' :“91 8 . ” |t I ‘\‘ T“ 6'” 102
@0823? T 2” Trp 221 :(3 5 ) Pm i1 i. W T 305‘)”
- yr 77 4. 5.9 W
(3.1 A) ( 0 ) Gil/137 ( )A“! 6 (5.5 A)

Figure 6-20. This figure shows the amino acid residues within 6 A of
the polyene for the colored pigment green (green) and the colored
pigment red (red). The rhod0psin data is extrapolated form the crystal
structure. The data for the blue were obtained from analysis of
theoretical models (RCSB protein data bank accession no 1KPW
(green) and MW (red), respectively).

267

retinal in all three cone pigments is strongly hydrophobic. Red and green cone
pigments share nearly the same sequence (96% homologous, differ by only 15
amino acids). However, their UV-vis absorbance differs by 40 nm (570 nm and
530 nm, respectively).12 Through site directed mutagenesis Oprian has proven by
changing only seven amino acids of the red cone protein, namely
Sl16Y::S180A::1230T::A233S::Y277F::T285A::Y309F mutant, a pigment
resembling the green opsin is obtained.”M As discussed previously, hydrophobic
to polar mutations at analogous positions in red rhodopsin can also produce a red
shifted pigment. Rod mutant F261Y::A169T::A1648 red shifts to ~700 run, an
amazing 200 nm shift.ls

Although the latter residues are important, as can be seen in Figure 1-13B
and C, they are not the residues that the model places in the vicinity of the
chromophore. According to the latter, the central residue forming the cavity is
Trp281, analogous to Trp265 in bovine rhodopsin. The counter anion to the SB is
Glu129, while a second carboxylate, G1u102, is located close to Ly5312, the Lys
covalently linked to retinal.

In the case of the blue cone pigment, which only bears a 46% homology
with the other cone pigments};16 the central residue forming the cavity is Tyr262
(Figure l-13A). Because of the proximity of the retinal B-ionone ring, Tyr262 is
considered to be the major factor in the blue shifting of this pigment. This
explanation agrees with the experimental observation that single mutant Y262W

generates a 10 nm red shifted absorption spectrum." In addition, the second

268

glutamic acid present in red and green pigments is absent in this protein, which is
believed to cause an additional blue shift.

Rhodopsin and the related cone pigments are members of a broader family
of membrane bound light transducing proteins.‘8 This group of proteins contain a
seven transmembrane helices motif with a binding site imbedded in the interior of
the protein. All of them have a photoactive chromophore (retinal) attached as a
Schiff base to a conserved Lys residue, which upon absorption of light isomerizes,
triggering conformational changes translating into either light-driven ion transport
or photosensory signaling.”"20

Bacterial and archaea rhodopsins which also belong to this group of
proteins, perform an ion pump action and significant body of work has been
carried out in an attempt to elucidate their mode of action. Bacteriorhodopsin
(th), the purple light-harvesting membrane protein of Halobacterium halobium,
is responsible of the convertion of light energy into a proton gradient across the
bacterial membrane.” Bacteriorhodopsin binds all-trans-retinal through Lys2l6
as a PSB absorbing at 570 nm.3 Similar to rhodopsin, several theories have been
proposed to account for the 70 nm opsin shift observed (relative to rod rhodopsin),
most of them involving the electronic-environment around the polyene chain.”

The crystal structure of th bound to all-trans-retinal revealed a binding site
were (Figure 6-3) a group of aromatic amino acids surround the SB bound
chromophore, joined by several Met and Thr residues.6 The pK. of the th PSB
has been calculated to be >12, 5 units higher than what is typically observed for

protonated SB model compounds."22’23 This has been attributed to a hydrogen

269

M145

K216

 

W189

Figure 6-3. All trans-retinal binding site of ion pump
bacteriorhodopsin (th).

bond between the retinal PSB and either a carboxylate or a tyrosinate residue. As
can be seen in Figure 6-3, the protonated form of the SB could be stabilized by
either of two Asp residues (D112, D85, 3.74 A and 3.79 A away from the SB
nitrogen, respectively), which have also been proposed to accommodate the
proton transfer across the membrane,21 or by the nearby Tyr85, although the
distance of 4.77 A between the phenolic oxygen and the SB nitrogen does not
support the latter. Interestingly, there is a clear interaction, in the form of a
hydrogen bond, between the carboxylate of Asp85 and the tyrosine oxygen, which
suggests that one of them can exist in the unprotonated form, to act as a
counterion. Previous studies in this area have failed to detect the presence of a
tyrosinate ion, suggesting that it is a carboxylate moiety that is responsible for the

PSB stabilization?” In addition, Arg82 (Figure 6-4) has been identified as a

270

K218

1 u was 0212(0)-wr:2.9A
055(0)-W1:2.eA
0212 :- K216(N)-W1:2.9A

2 82
was 0 - D212 (0 : 2.6 A
E194 D212 ((0))- W2: 2.7
as: (N) - W2: 2.5 A
* n92 (N) - we: 2.9 A
Y57 3

4 * é V57 (0) - we: 3.0 A
5204 (c) - £194 (a): 3.0 A

Figure 6-4. Crystal structures of the retinal binding sites of bacteriorhodopsin The

A. Bacteriorhodopsin
A... a 570 nm

important amino acids surrounding the polyene are highlighted. Red crosses
indicate ordered water molecules. Selected distances between residues are shown

on the right.

principal counterion, important for the stabilization of the two carboxylates in the
vicinity of the SB, while it is believed that the imine proton is strongly hydrogen
bonded to a structured water molecule in the binding pocket?” As previously
mentioned, the counterion structure in th is quite unusual, involving Asp85,
Asp212 and Arg82. At the same time, there is an ordered water molecule in close
proximity to the SB, which acts as the bridge for the interaction with the two
aspartates. The importance of Arg82 for the protonation of the th SB has been
pointed out, although the residue is placed relatively far away form the imine
moiety. As can be seen in Figure 6-4, in the th case Arg82 is pointing towards
Asp212, interacting with it, as well as with Tyr57 through bridged water

molecules.

271

Table 6-1. Spectroscopic characteristics of rhodopsin proteins of different species.

 

Protein Retinal UV of Free UV 0‘ 35 UV 9‘ PSB PSB

 

Retinal (hm) (nm) mm) pig.__
Rh 1 1-cis 380 360 500 >16
Octopus Rh 11-cr's 380 376 475 10.6
Octopus Rh 9-cis 380 362 468 10.4
Gecko 1 1-cis 380 376 521 9.9
Gecko 9-cis 380 350 487 9.3

 

Vertebrates, also use rhodopsin as visual pigments, and in has been observed
that all cases the Glul 13, is a conserved residue. Glull3 is considered to be the
counterion responsible for stabilizing the PSB formed between retinal and the
opsin protein. For invertebrates, retinal remains the chromophore present in all
pigments, although for most systems in the all-trans-retinal instead the ll-cis-
retinai.‘9'2° Table 6-1 describes the spectroscopic characteristics of Octopus and
Gecko rhod0psin proteins, as compared to the bovine Rhfr"29 The opsin shift, as
well as the pK. value for the PSB can differ significantly within different systems,
or in the presence of a different retinal isomer. In the presence of the ll-cis-
retinal chromophore, the UV-vis absorbance of octopus rhodopsin is blue shifted
while that of the gecko pigment red shifts around 20 nm, as compared to bovine
rhodopsin. It is believed that the different placement of the PSB counterion is
what determines the spectral behavior, however, there is no crystal structure

available to prove the proposed explanation.

272

The above observations verify the suggestion that the mechanism of
wavelength regulation is much more complicated than originally thought, and is
probably due to a combination of the factors previously mentioned It is therefore
apparent that since no crystal structures of the cone pigments are available, an
alternative approach for the study of the wavelength regulation, involving systems

easier to study and manipulate, could be of great use.

6.2 Preparation of mutants to study wavelength regulation

Using CRABPII we have developed a series of mutants with the hope of
deconvoluting some of the factors involved in wavelength regulation.
Mutagenesis of CRABPII was performed using standard molecular biology
protocols. When the CRABPII-mutant plasmids were obtained, the protein was
overexpressed in E. coli and purified following established protocols (for all
procedures mentioned above see Materials
and Methods).

We were interested to first explore

the theory proposed by Blatz,30 which

 

hypothesizes that the linear distance of the

. . . , , . , , Figure 6-5. Distance of the counter
rrmnrum ion conjugation wrthrn the

anion to the protonated Schiff base and

polyene can be modulated by placement of
positioning of charges or dipoles along

an overall negative dipole or point charge the backbone of the pom“ may

at the vicinity of the polyene (Figure 6-5). modulate the maximal wavelength of

For example a short conjugation can be the chromophore.

273

promoted by placing a negative point charge close to the iminium nitrogen. On

the other hand, positioning of the negative point charge

/ / /
in the vicinity of the polyene or the ionone ring extends N\
3
the iminium conjugation and should yield a red-shifted
pigment. To explore this point charge theory, a set of / / ’
4

mutants were prepared and their spectroscopic properties
were studied using all-trans-retinal (2), ll-cis-retinal (1), merocyanine (3) and
azulene (4). Because the triple mutant has been the optimal rhodopsin surrogate
obtained so far, this mutant would be used as the standard protein to test the
effects of different negative point charges placed along the polyene. Figure 6-6
shows the amino acids that are in proximity to the all-trans-retinal in the binding

site of CRABPII-R 132K::R1 11L::L121E. The mutant

 

Figure 8-8. This figure represents the amino acid residues within 7 A of
the polyene for the triple mutant CRABPII-R132K::R1 11L::L121 E. The data

is extrapolated from the crystal structure.

274

R132K::Rll1L::L121E::R59E was prepared to explore the influence of a point
charge in the vicinity of the polyene. Arg59 is positioned 6.2 A away from C-10,
right in the middle of the polyene. Therefore, if a Glu is present at this position a
negative point charge would be present in the middle of the polyene, thus the
delocalization of the positive iminium would be more extensive than the triple
mutant R132K::R111L::L121E, thus leading to a red shift in the wavelength.
However, when the mutant was incubated with all-trans-retinal the wavelength
observed was blue shifted 2 nm (Amx=445 nm).

The next mutant prepared was R132K::R111L::L121E::A36E. Ala 36 is
positioned 4.0 A away from C-11, and it is also located in the middle of the
polyene. Because the Glu is closer to C-ll than the N of the iminium we would
expect a larger delocalization of the positive charge leading to a red shifted
wavelength compared to the triple mutant R132K::R111L::L121E.

Also some of the mutants that were prepared in the process of developing
the triple mutant R132K::R111L::L121E were studied using ll-cis-retinal 1 and
the retinal mimics (3) and (4). In Chrysoula Vassileou’s Thesis the spectroscopic
data of these mutants with all-trans-retinal are discussed in detail. Therefore in
this chapter we would discuss the properties between the different ligands and
their wavelengths observed when they interact with different mutants. In
particular we were interested in studying the single mutant R132K, the double
mutant R132K::R111L, and the triple mutant R132K::R111L::L121Q. These
mutants lack the counter anion L121E, so the positive charge would be forced to

be delocalized along the polyene to increase its stability. We would thus expect to

275

observe a red-shifted wavelength. Finally the double mutant R132K:L121E was
also studied.

The tetra mutant R132K::Rl11L::L121E::R59W was prepared to test the
possibility of an excitonic coupling between the polyene and the trypthophan, and
as previously discussed it has been proposed that excitonic coupling between
Trp265 in rod rhod0psin and the polyene can contribute significantly to the
observed opsin shift.

Also the CRABPII-WT and the triple mutant R132L::R1 l 1L::L121E were
studied and both proteins were used as controls to assure that the observed results
were due to a PSB formed between the chormophbres and Lysl32 and not any

other Lys present in the sequence.

6.3 Studies using ll-cis-retinal

The major sources of ll-cis-retinal (1), used in rhodopsin studies, come
from light induced isomerization of all-trans-retinal (2).3 1'36 However, its
isomerization leads to a complex mixture of isomers, which are hard to separate,
and ll-cis-retinal is only one of the minor isomers obtained. It is important to
mention that even though isomerization of all-trans-retinal seems to be cheaper
and faster, it was already attempted in this laboratory with out success. The
chemical synthesis of 11-cis-retinal (1) has been previously accomplished.
However due to the extreme tendency of the ll-cis- double bond to isomeraze to
the more stable all-trans-retinal, its synthesis is very challenging (2).”39 It is

believed that the unusual difficulty in generating the ll-cis double bond is a result

276

of the steric hindrance between lO-H an the
13-CH3, (Figure 6-7). A few elegant synthesis
have been reported in which different
strategies have been used as the key steps/‘0'“
The synthesis proposed by Nakanishi41 and co
workers appeared to be one of the most
promising, due to the high final yield and the
mild conditions used. The key step in the
proposed synthesis involves the zinc-mediated

semi-hydrogenation of the ll-yne-retinoid to

the ll-cis-retinal.

6.3.1 Synthesis using ll-cis-retinal

 

Figure 6-7. 11-cls-retinal is not a
planar molecule due to steric
interactions. The first steric

interaction is between the CB-H and
the C5-CH, and the second one
between C10-H and the C13-CH3.
This leads to a twist of the 06-07
and 012-C13 single bonds to

alleviate the steric hindrance.

The synthesis of ll-cis-retinal (I) started by the preparation of ylide

diethyl (3-trimetylsilyl-2-propnyl) phosphonate (6), (Scheme 6-1). The ylide was

prepared from commercially available protected propargyl bromide 5 (75% yield).

Subsequent HWE coupling“2 of B-ionone (7) with (3-trimethylsilyl-2-

propynyl)phosphonate (6) afforded the alkyne 8 in 85% yield, with an isomeric

ratio of El (86:14) at C9.

Deprotection of the silyl group with

tetrabutylammonium fluoride (TBAF) gave the terminal acetylene 9, El at a ratio

of 86:14.

277

HP(O)(OEt)2 LnBuLWHF, 0 °QL15 min ;

NaHMDS/THF .9
Br\_-_—___—__TMS : (OEI)2':‘ : TMS 2 7
5 '10 00' 2 h n 5 h 35%

75%

TMS - - ~
\ / neu,Ni=.THr-; W
8 E:Z(86:14) 11. 3h. 84% 9 E:Z(86:14)

Scheme 64. Synthesis of acetyline (9) intermediate in the synthesis of 11-cls-

 

 

retinal.

Hydromagnesiation43 (Scheme 6-2), of prop-2-vinylic alcohol (10),
followed by treatment with iodine at --78 °C, afforded the vinyl iodide 11 in 44%
yield after column purification as a single isomer (the column was performed in
the dark). Protection of the alcohol using tert-butyldiethylsilyl chloride

(TBDMSCl) afforded vinyl iodide 12 in 78% yield. Palladium coupling of 9 with

1. BuMgCl.Ether, 0 °c
2. szTlCIZ, o °c

OH TBDMS. lmldazole OTBDMS
HQ _ 4, V : V

T0 3.|2,-78°C,1h I 11 DMF.rt.16h.78% l 12
rtSOmin, 44%

W (PhaP)Pd,Cul,PrNH2
9 s, RT,4h.84% 7

E22 (86:14 )

 

 

 

 

    

[BU4NF, THF
rt, 3 h, 90%

 

14 5:2 (93:17)

Scheme 6-2. Synthesis of acetyline (14) precursor of the Zn mediated reduction.

278

vinyl iodide 12 proceeded smoothly with complete stereochemical retention to
yield retinoid 13 as a mixture of El isomers (83:17) in 84% yield. Deprotection
with TBAF afforded the allyl alcohol 14 in 90% as a mixture of 13:2 isomers
(83:17).

As previously mentioned, the acetilyne is only 86% pure, it contains 14%
of the isomer 17 (Figure 6-8). Therefore, to simplify the separation of isomers,
after the Zn-mediated reduction, the acetilyne was purified using HPLC (see

Materials and Methods for detail (Figure 6-8)).

 

 

 

\ ¢ OH
14 14\
\ \
17
1., H \H )
\
OH 'o.‘s"”'"""Tfifi""fi”'”'si.9

Figure 6-8. Separation of the isomers alkyne 14 and 17. Using HPLC
ZORBAX-SIL column mobile phase of 80:20 hexanes:ethyl acetate, 3

mUmin, at room temperature and a UV detection at 317 nm.

The semi-hydrogenation of acetylenes via zinc mediated in our case the
reaction was more challenging than expected. The first attempt was performed
following Feigel’s protocol, and the mo; oxidation was performed immediately
because it is known that the ll-cis-retinol (1) is not very stable (Table 6-2).“

Home made MnOz” was prepared (from previous experiences, old MnOz is too

279

Table 6-2. Reduction of alkyne 14 to obtain 11-cls-retinal (1 ).

    

O I V
14 5.2 (83.17)
«mairmnrrS‘rtxu-. .n:-..".'>'.-‘:i.F-.f' 4412412933“. 2311: tW'c’H’ih‘ef‘T'l' Le-tw-w.’r---r.’flt-» '.'.“;é‘:~".hl-‘ 1 ;'.'ri.“:'.tt‘r.’-.t:~‘.=Kn.k1.ZtWMWm‘C-Wflifflnflfuvfl'mviMilli-

Trlele Conditions Products

1";1"1"43111-£l.l3£3*3;7 1:1":3514’34131 H " -'. left: " "-t"'“~."'1'..‘\’l'-h ""V ’bL‘J- 1'7‘5' if 141% f» '14 . ' ' .- vt‘i' t.» ‘ l ." 31"": "LI-.1391: -"'3w‘nfitW"‘HNfilfiL'r‘r-Jd Alvin's-ml 33’8“M‘P§4':‘~3~ ‘34"

Jim
oFiltration for Zn work up,

1 oRoom temperature, 16 hours, 750/
olsopropanol °
OM00: oxidation.

 

oZn (dry stored)
oFlltration for Zn work up,
2 oRoom temperature, 16 hours,
oMethanol
eNO MnOa oxidation.

85%

 

oPurified 14 only E

oZn (dry stored)

oCentrifugation for Zn work up. ~50? \ \ \
3 oRoom temperature, 16 hours, crudeo

oMethanol

eNO MnOz oxidation. 01-)

oPurified 14 only E

oZn (dry stored)

oCentrifugation for Zn work up,
4 oRoom temperature, 16 hours, after

oMethanol . .
thOz oxidation 9“""°3“°" o

~10°/o \ \ \

slow and the major product is the decomposition of retinoids). After oxidation the
crude product was run through a small silica column, however, 1H-NMR showed
no reduction of the alkyne, only the expected oxidation of the alcohol to the
aldehyde. The fact that reduction did not occurr suggested that the Zn was not

activated enough to perform the reduction.

280

This time only the reduction step was performed and the crude lHNMR
(CDClg) showed starting material, some unidentified products, but no ll-cis-
retinol 1 was observed. When zinc is activated the last step involves the filtration
of the powder zinc. But every time the filtration was performed it was very slow
and messy. And it has been reported that activated zinc deactivatcs in contact
with air. The next time, the Zn was centrifuged instead of filtrated (see
experimental for details) and MeOH was used as solvent. Gratifyingly the crude
lHNMR (CDC13) showed the desired product 16. When the oxidation using
MnOz was performed it was not as clean as expected (several spots were formed
before all starting material was consumed). The MnOz oxidation was performed
at various temperatures, to minimize the formation of undesired products. But
decreasing the temperature to —78 °C, --50 °C and -20 °C did not show any
improvement in the formation of pure ll-cis-rctinal. It was also observed that
decreasing the temperature seriously slowed down the reaction to a great extent.
Therefore, oxidation at 0 °C was adapted even though other products were
formed. The crude was filtered through a celite pad and a micro column was
performed to separate the polar products before HPLC purification. The lHNMR
(CDC13) of the eluted fractions showed about 50% of the desired ll-cis-retinal
and some isomers.

The HPLC isolation of 11-cis-retinal (1) has been previously reported
using either normal phase or reverse phase. The first attempt to isolate ll-cis-
retinal was made using a normal phase ZORBAX-SlL semi-prep column (4.6 mm

ID x 25cm (5 pm). mobile phase 97:3 Hexanes:Ethyl acetate, 3 mllmin. at room

281

temperature and a UV detection at 362 nm. A small injection of the crude
showed four major peaks in the HPLC but the ratios did not match with the
lI-INMR data. Isolation of each one of those peaks and analysis showed that none
of those fractions was the ll-cis-retinal, suggesting that the retinal was isomerized
in the column. To avoid the isomerization of the ll-cis-retinal catalyzed by the
presence of acid, 0.05% of triethylarnine was added to the HPLC solvent. When
the reaction was repeated the separation seemed better but again no pure ll-cis
retinal could be isolated.

When pure acetylene 14 was used in the reduction, and was followed by
oxidation with freshly prepared MnOz, the ll-cis-retinal obtained was obtained in
80% yield. The ll-cis-retinal was the major isomer 94%. The obtained retinal
was very pure, about 94% and two other different isomers, one 4% and other 2%.

An attempt to further purify the retinal, via HPLC was performed. But

0.6 —.
0.5 -

3 0.4 ~
0.3 a
0.2 4

‘0.1-1

 

 

0 T T—fiTTTfi—T TT T T T 1

200 300 400 500
Wavelength (nm)

Figure 8-9. UV spectra of the isolated 11-cis-retinal. which

matches with the reported Aw=362.

282

unfortunately the isolated ll-cis-retinal was only 92% pure. No more attempts to
purify the ll-cis-retinal were performed. Therefore, all the studies performed
were done using are ll-cis-retinal that was 94% pure. The obtained ll-cis-retinal

matched the reported spectroscopic data (Figure 6-9).“

6.3.2 Protein substrate interactions using ll-cis-retinal

As discussed previously the binding pocket in CRABPII mutants is large,
therefore it can be envision that 11-cis-retinal can accommodate into a variety of
different configurations. The model of the covalently bound 11-cis-retinal in
Figure 6-10A and Figure 6-IOB suggests that the chromophore could exist in two
different conformations. The conformation shown in Figure 6-7 has less steric
interactions, therefore the following analysis will be based on this conformation.
When all-trans-retinal binds the triple mutant R132K::R111L::L121E the Glu121
is close enough to form a strong H bonding (3.4 A) away from the nitrogen of the
iminium ll-cis-retinal, on the other hand the ll-cis-retinal model shows that
Glu121 is further from the nitrogen of the iminium ion (4.2 A). The increase in
the distance of the counteranion (based in our models) is reflected in the
wavelength observed. When the ll-cis-retinal was incubated with the rhodopsin
surrogate R132K::R111L::L121E a broad absorption spectra was observed. The
second derivate was used to deconvolute the individual peaks that were observed
in the wavelength spectra (369 nm and another one at 464 nm). The first peak
(464 nm) corresponds to the formation of the PSB and the second peak (369 nm)

could be the SB or the free retinal. As shown in Figure 6-10 the counter anion is

283

Y134
H111L

.0'
‘k hills.

2.5 A for
trans-ref

“34.0’

R111L

L121 E
3.4 A for
11-cI9-ret
2.5 A to:
trans-ref

 

Figure 6-10. Using the crystal structure of CRABPII-R123K::R111L::L121E (blue)
two possible conformations were found for the 11-cis-retinal. A. The counter anion
Glu121 is 4.5 A from the PSB and B. The counter anion Glu121 is 3.4 A from the
PSB.

further away, thus it could destabilize the protonation of the imine. Also, because
of the different geometry it is possible that the nucleophilic Ly5132 cannot access
the aldehyde to form the imine. It is interesting to notice that the Am observed is
15 nm red shifted as compared to the Am of the all-trans-retinal. This could be
partially explained due to the fact that the counter anion is further away, thus

increasing the delocalization of the positive iminium. The red shifted peak

284

observed (Am, of 464 nm) is blue shifted compared to the absorptions of the

opsins (~500, 530 and 560).“ But when this value is compared to the maximum

wavelength of the Protonated Shiff base formed with n-butylamine, and ll-cis-

retinal (~440 nm) (Figure 6-11)“'48 it is evident that binding of the ligand to the

rhodopsin surrogate causes a 24 nm shift of the protonated Schiff base. This is the

first example in which a designed protein binds ll-cis-retinal as a PSB. The

double mutant
R132K::L121E
shows a similar
behavior as the
triple mutant,
(Figure 6-12, 6-
13). The pK.
was estimated to
be 7.2, (Figure
6-14) which is
lower when
compared to the
pK. with all
trans retinal 8.0,
which could
explain the

difference in the

 

 

11-cls-retlnal
11-cis-retlnal
Schiff base
\ \
\
\ 11-cls-retinai
EM protonated Schiff base

Figure 6-11. UV-vis absorbances of 11-cie-retlnal as a free
aldehyde (380 nm), a n-butylamine Schiff base (365 nm) and a

protonated Schiff base (440 nm), in ethanol as solvent.

285

Y134

 

Figure 8-12. Model structures of the binding cavity of the triple mutant and 11-

cis-retinal and the mutated amino acids are also shown.

  
 
  

 

  

 

0.1 1|
0.09 -
§ 0'08 R132K:L1210
0-07 ‘ 375 "m (”'1") R132K:R111L:L121E
0.06 369 nm, 484 nm
0.05 1 ”m"
0.04 4,
0.03 \
0.02 l R132K:L121E
380 nm, 463 nm

0.01 "l mam" ...

o I f 1 ‘T— T —._:

 

300 350 400 450 500 550 600
Wavelength (nm)

Figure 8-13. UV-vis absorbances of the ligand 11-cis-retinal with different

rhodopsin surrogates.

286

>
m

 

 

 

 

 

Si

dl " . pK.=7.8

cl

8;

3::

c1

3‘

of g.

.____.-..e...,-f-..._..---_.--.._.._---- ‘6 2-.---. __ ---1..._.--_-.___e-
350 400 450 500 550 6.4 as 7.2 7.6
Wavelength (nm) P"

Figure 8-14. A. UV titration of R132K::L121E at different pHs. 8. Graph of the

absorption change at 454 nm at different pHs.

pK.. Based on the model shown in figure 6-10 the counter anion seems to be
farther in the ll-cis-retinal than the all-trans-retinal.

The absorbance of ll-cis-retinal using different proteins is summarized in
Table 6-3. As expected the CRABPII-WT (Max:387 nm) and CRABPH-
R132L::R111L::L121E (Am=383 nm) do not show any red shift. When the
counter anion is placed in the opposite site (compared to L121E) the observed
wavelength is blue shifted, which suggests that a counter anion at that position
does not stabilize the protonation of the imine. When a second counter anion is
present at position R59E (close to the polyene) the wavelength observed did not
change significantly, but when the temperature of incubation was decreased, the
PSB was more stabilized. However, interestingly when the counter anion is close

to the polyene region (R132L::R1 1 1L::L121E::A836E,

287

Table 6-3. Titration of different CRABPII mutants with 11-cls-retinal (1) and all-

trans-retinal.

.. .. . Pgmn ,. . . . . "lbw-“Rommel?” allftl'an?fiw§lfln‘erw

WT 387 377
R132K:R111L::L121E 369, 464 449
R132K::L121 E 360, 463 457
R132K::L121Q 375 380
R132L::R111L::L121E 383 377

F11 32K::8111K3L121E 362, 468 365. 455
R132K::R111K2L121EzzF15Y 382 373
R132K::R111K::L121E::A36E 374 394
R132K::R111K::C130D 383 377
R132K::R111K::L121E::013OD 395 440
R132K::R111K::L121E::R59E 367, 460 448

R132L::R1llL::L12lE::Cl3OD) it results in the destabilization of the PSB and

the observed peak corresponds to either the SB or binding of retinal.

6.4 Retinal analogs
Since CRABPII contains a large binding pocket (~600 A) we expect that a
variety of different aldehydic chromophores could fit and form a PSB or SB.

Distefano an co-workers have performed studies on a related binding protein,
Adipocyte Lipid Binding Protein (ALBP).‘° ALBP has a B-barrel topology with a

remarkable structural similarity to CRABPH. Distefano has managed to modify

288

the pocket in order to incorporate a series of cofactors and organic molecules,
which are poised to perform chemistry on a variety of molecules that enter the
binding pocket.5052 As previously discussed we are especially interested in the
study of protein-ligand interactions. The study of the interactions between
rhodopsin surrogates and different chromophores that potentially could bind as
PSB would help us to understand those interactions in more depth.

Also it is possible that the rhodopsin surrogates could be ideally suited as
colorimetric protein fusion tags similarly to Green Fluorescent protein. Green
Fluorescent Protein (GFP) is a protein isolated from jellyfish Aquorea Victoria,53
and it shows fluorescence a couple of hours after it is transcribed. The inherent
fluorescence of GFP has been used as a tool in molecular and cellular biology. In
more detail, a fusion protein of GFP and the protein of interest can be formed. The
fusion protein would be fluorescent, and therefore, the fluorescence can be
monitored in real-time. The use of GFP has lead to a better understanding of
cellular biology and processes such as signal transduction, apoptosis, RNA
localization, monitoring ion channels among others.”58 Even though GFP has
been a great tool in the understanding of molecular and cell biology, they dohave
a few limitations. The first one is that the formation of the GFP chromophore is
slow (up to 2 hours) and this could be problematic for the proteins with short half-

lives.”60

Also, molecular oxygen is required for the formation of the fluorescent
chromophore, and the by-product formed is H202 that could have significant
harmful effects in the system that produces the GFP tag.’3 Also some variants

form homo-oligomers and inclusion bodies.‘51 A variety of GFPs that show

289

different excitation and emission spectra have been produced, in general green,
cyan and yellow pigments. The blue GFP variant photobleaches very rapidly and
it is highly unstablefmm’62M The rhodopsin mimics prepared are smaller in size,
(136 aa vs. 240 a for GFP) and thus its use as a protein tag would impose less of
a change to the overall size of the protein fusion tag.

As mentioned previously we thought that the engineered proteins could be
used successfully as spectroscopically active fusion protein tags, that may offer an
alternative to the use of GFP. Especially since their wavelengths can be tunable
and they could be switchable. The use of different chromophores as ligands could
change the color of the engineer proteins. Furthermore, binding of a fluorophore
could produce fluorescent tags if necessary. Also we could tune the color
observed using the same chromophore and different proteins. The proteins would
regulate the color based on the principles observed in wavelength regulation. In
particular we were interested in two groups of chromophores, azulene and
merocyanine derivatives, both which show a large red-shifted absorption relative
to the retinal.

The synthesis, design, and reactivity of cyanines dyes have been
intensively studied in the last decades.‘8‘65'66 The interest is mostly due to their
wide applications. For example they are important materials for non-linear optics
(NLO), laser technology, data storage, etc.65 Depending on the charge of the
polymethine unit, cyanine dyes can be classified into three main groups: cationic
(cyanine and hemicyanine) dyes, anionic (oxonol) dyes and neutral (merocyanine)

dyes. We were interested in the synthesis of merocyanines that contained an

290

indolylidene end group for two main
reasons: the first was that the UV
spectra of the dye would red shift

upon addition to the protein.

 

Figure 6-15. Merocyanines forms a very

Second, if the PSB is formed With stable PSB due to the last that the positive

the rhodopsin surrogate the positive charge is stabilized all around the polyene.

charge would be strongly delocalized

(Figure 6-15).

Azulene has intrigued scientists since it’s discovery. Azulene (18) has a

very beautiful blue color. The unusual color of azulene is attributed to its very

low first excited state 81.6769 Replacement of the cyclohexene ring in retinal by

an azulene group results in a molecule with a red shifted wavelength. An

interesting feature of azulene is that it can stabilize cations as well as anions, due

to its remarkable polarizability (Figure 6-16). When an azulenic aldehyde

derivative forms a PSB the cation formed can be in resonance, the positive charge

can be stabilized either at the iminium ion
21, at the tropylium ion 19 or all along the
molecule 20 (Figure 6-17). This could be
helpful in the study of wavelength regulation
in our rhodopsin mimics. Similar azulenes
were used in the study of bacteriorhodopsin.

After the retinal analogs were incubated in

291

”~

Figure 6-16. Azulene (18) has a
remarkable polarizability. Therefore
It can stabilize cations as well as

anions.

the presence of the protein, they formed PSB that had maximal absorption in the

in near infrared region (NIR).70

(G\\\$

19

 

Figure 6-17. Azulenes can stabilize the positive charge either at the iminium ion

21, at the tropylium ion 19 or in resonance all along the molecule 20.

6.4.1 Synthesis of different chromophores

The initial approach for the synthesis of retinoid mimics consisted of the
synthesis of a large ylide (22) that resembles the retinal polyene structure. This
ylide could be coupled with a variety of different aldehydes that have different
electronic and spectrosc0pic properties. (Scheme 6-3). The ylide would be
prepared from the available isoprene (23) and dimethyl butyric acid (24).

The synthesis of 22 started by the preparation of aldehyde 2‘7 and a ylide
30. The synthesis aldehyde 27 started with the preparation of (E)-4—chloro-3-
methyl-Z-buten-l-ol acetate 25 (Scheme 6-4). The chloro-acetate 25 was prepared
by the 1,4 acetoxychlorination of isoprene (23) in presence of t-butyl hypochlorite
and acetic acid in a 34% yield. Hydrolysis under basic conditions at 0 °C afforded
99.8% of the chloro alcohol 26. Oxidation of the chloro alcohol was achieved in

85% yield. In order for the reaction to give a good yield dry pyridinium

292

C5 vk» fin UJL

Scheme 6-3. Fletrosynthesis oi ylide 22.

chlorochromate (PCC) must be used as well as molecular sieves and celite in
dichloromethane (Scheme 6-4) The resulting aldehyde 27 is very unstable, and it

cannot be stored for more than a couple of days.

 

 

Nazcoa.
lBuOCI. AcOH; W l l :Hl H20
fi' CI \ z
>/ \\ 2 n 0 °c OAc ,
23 54% 25 0 °C, 5 h
99.8 %

 

PCC. 0 °C
27

26 Mel sieves 4 Al celite
CH2CI2, 85°50 96 Very Uflfllblfi"

Scheme 6-4. Synthesis of aldehyde 27.

293

The ylide 30 was prepared first by the bromination of 3,3 dimethyl acrylic
acid (24) with N-bromo succinimide (Scheme 6—5). The 'H-NMR showed a
mixture of E and Z isomers. The residue was picked up in MeOH and excess
K2C03 anhydrous was added to promote the cyclization of the Z isomer. The pure
(E)-4-bromo-3-methyl-2-butene acid (28) was obtained in a 42% overall yield
from 24. Esterification of the acid 28 afforded 82% of the (E)-4-bromo-3-methyl—

2-butene methyl ester 29. An Arbuzov reaction afforded the ylide 30 in a 95%

yield (Scheme 6-5).

 

 

 

o
(DE/Y Strong Light K200; We: H2804
H ' MeOH
Reflux. MeOH. 2 h OH
2‘ ON. cor. 42 as 20 82 9"
P(0Et)3
W e A... e
/ 29 125 °C, 95 % so

Scheme 6-5. Synthesis of ylide 30 starting from 3,3, dimethyl acrylic
acid (24).

Coupling of the ylide 30 and the aldehyde 27 using sodium methoxide as a
base did not provide any desired product. When n—butyl lithium was used as base
instead only a 15% of the desired chloro ester was obtained. On the other hand,
when NaI-IMDS was used as a base, an average 40% yield was obtained as a

mixture of E:Z isomers (2:1) (Table 6-4). When 32 was used to prepare ylide 33

294

Table 6-4. HWE between ylide 30 and aldehyde 27 to obtained

 

chloro ester 31.
”0'3 \ O/ ; M
r C' \ \ \ ’
30 31
Trials Conditions Product yield

 

ii‘.’.l-li?'~f‘ i‘W-“l‘fiflr'd‘l'ifliru‘ .‘ “.5- a, mIA---\_.- .

' ‘ ' ‘ neon. Nadine. o 0c as thin ‘
24 hours

1 then aldehyde 27 ° °/°
n-BuLi, O “C 30 min
6 hours . .
2 then aldehyde 2., 14.5 % (52) (1.5.2)
NaHMDS, THF. 0 “C 30 min
4 hours. 30-45 %
3 then aldehyde 27 (E2) (2:1)

the expected ylide was obtained in a 40% yield, but the product was not pure and
it could not be isolated. 0n the other hand, when the ylide 33 was prepared via an
Arbuzov reaction the desired product was obtained in a 91% yield (Scheme 6—6).

A few different aldehydes were prepared to be coupled with the ylide.
Azulene (18) was forrnylated using DMF and P0013 to afford 34 in 76%. The
porphyrin 36 was prepared via oxidation of the alcohol 35 with pyridinium
chlorochromate (PCC) in 20% yield. Indole 37 was formylated using DMF and
(CO)2C12 to afford 38 in about 45% (Scheme 6-7).

The HWE condensation using ylide 33 and NaHMDS as a base provided

very low or no yield of the expected ester product. Different temperatures and

29S

1. NaHMDS, THF

/\ E o °C.30min MEL/wry
0' ‘H a 6 \ \
2.
r 32 C'Wo’ r Nofiure

 

 

”0503 . W
OW / t “(0 \ :3 0’

o
31 5:2 2:1 10° °°' 0" 91% 52 2:1

Scheme 6-6. Synthesis of ylide 33.

bases were tried but no improvement was observed. When phorphyrin 36 and 40
were used the crude lH-NMR showed some product, but nothing could be
isolated. On the other hand when azulene 34 and merocyanine 38 were used 30
and 40% yields were obtained, respectively.

Unfortunately, the yield of the HWE reaction was not very consistent
(Scheme 6-8, Figure 6-18). When the product ester was reduced with DIBAL no
desired product could be recovered. Again, different temperatures and
equivalents of the reducing agent were utilized but no improvement was observed.

Because the use of ylide 33 did not provided consistent results a different
approach for the synthesis of the chromophores was taken. We started with the
synthesis of the merocyanine 4 (Scheme 6-9). This new methodology utilized a
smaller ylide (30) instead. The synthesis of 45 via HWE using potassium :-
butoxide and KHMDS as bases afforded no product. When NaHMDS was used
instead about 36% yield of 45 was obtained. If the amount of ylide was increased

to 10 equivalents the yield of 45 was increased to 50%.

296

 

DMF 1. 0 °C,30min: Om
P0013 2. O 3‘ HO
'1' 76%
2h rt

 

 

DMF 1. 0 °C. 30 mint —0
(CO)20|2 2. —
(3?; .
\
N

37‘

Scheme 6-7. Synthesis 01 various aldehydes.

Figure 6-18. Commercially available
aldehydes used.

297

1. NaHMDS1eq,

 

 

THF 0 °C, 30 min; Fl
2- s. ' 013‘

43 hrs 0 0c. rt Br Could not be

isolated pure

1. NaHMDS 1 eq,

THFO°C,30min q l R 1 13

> . 5- %

2. 34 Q 42

48 hrs 0 °C- rt

1. NaHMDS 1 eq,

0
”0.5%/kn THFO°C,30min: %
r 33 2' 38 43, R 15-30%
a
at“; “

 

48 hrs 0 °C- rt
1. NaHMDS1 eq,
THF 0 °C, 30 min

‘
7

Could not be

2. 40 isolated pure

Ft \ \ o’

1. NaHMDS1 eq,

33 5:2 2” THF 0 °C- 3° min: Nothing could be isolated
2. 33
48 hrs 0 °C- rt

 

 

 

 

 

Scheme 6-8. HWE using ylide 33 and various aldehydes.

When two equivalents of DIBAL were used at -78 °C to reduce the ester
to the alcohol followed by oxidation with MnOz a mixture of unidentified
products was obtained. When only one equivalent of DIBAL was used at -78 °C
and the reaction was quenched with ethyl acetate, 3 7% yield of the desired
aldehyde 3 was obtained (Scheme 6-9). Even though the yield was low it gave us

enough product to test the binding of the cyanine with the different proteins.

298

H 1. NaHMDS1 eq,

 

 

O‘F‘w THF 0 °C, 30 min / / °
/ :
2. % El (1 :1)
N\ ‘0 20-50%
as h 3‘
0 °C- rt
1. DIBAL, 1 eq
-73 °c 2 h / / / ’0
; N\
2. Ethyl acetate 3
1 h 7 %

Scheme 6-9. Synthesis of merocyanine 3, an HWE condensation between
ylide 30 and aldehyde 38 followed by DIBAL reduction.

An HWE reaction of 34 whith ylide 35 afforded the ester 46 in a 5% yield.
But when 10 equivalent of ylide was used instead, the yield was enhanced to 41%.
Reduction was accomplished using 2 equivalents of DIBAL at —78 °C, and
quenched with ethyl acetate. The quenched reaction was stirred with Na-K
tartrate for one hour. Then the crude was treated with fresh Mn02, and after
purification using column chromatography only 18% of the desired aldehyde 4 A
was isolated (Scheme 6-10). Even though the yield was low it provided enough
material to perform a study of this chromophore with different proteins. If the
reaction is not quenched with EtOAc or stirred with NaK tartrate no product can

be isolated.

299

1. NaHMDS 1 eq,
\ THF 00c. so min Cl /.
0 My O O\ E:Z 1:0.1
/’ 2' O 34 ‘3
3° 0 CH0 1eq ylide 5%
36 h 10 sq ylide 41%

 

ll

 

0 °C- rt
1. DIBAL, 1 eq
-78 °C 2 h
2. Ethylaeetate, 1 h 0 ‘ E:Z 5.1
3. NaK-tartrate, 1h 18 %

4. MflOz, 0 °C. 3 h

Scheme 6-10. Synthesis of azulene 4, an HWE condensation between ylide 30
and aldehyde 34 iollowed by DIBAL reduction.

6.4.2 Binding studies wit merocyanine 3 and azulene 4

As previously mentioned, binding of bacteriorhodopsin and merocyanine 3
has been studied. Hans-Dieter observed that merocyanine 3 bound th as a PSB
(km of 610 nm) with an opsin shift of 48 nm (Table 6-5) (Am of the PSB =578
nm).71 When merocyanine 3 was incubated with the CRABPII
R132L::R111L::L121E a single peak was observed with a km of 587 nm Figure
6-18A show comparison of the all-trans-retinal bond structure with the modeled
structure of merocyanine 3. It was assumed that when the merocyanine forms a
PSB it binds similar to the retinal. Interestingly merocyanine 3 shows a different
behavior than all-trans-retinal. The major difference is that when all-trans-retinal

binds the triple mutant CRABPH, binding and formation of the PSB occurs very

300

 

V76
L121E Y 1 "59

.l

“756
[1136‘ A32

ll

H132K

F15
Figure 6-18. A. Fill in model of binding of merocyanine 3 (red) to the triple mutant
CRABPII-R132K::R111L:L121E. The ionone ring is buried in the cavity. it was
assumed that the merocyanine would bind in a similar manner to all-trans-retinal. B
This model represents a comparison of the binding of all-trans-retinal (blue) and

merocyanine (red) the binding site show important amino acids in the binding pocket.

301

Table 8-5. Spectroscopic data of merocyanine 3 and the

bacteriorhodOpsin (bFlh).
(11mm l'll'l'l)
Aldehyde SB PSB th
463 422 578 610

u'»: wr-zllr-“z-n-r- ---: -.F-. .. we. .- .. m , -.>. ;-..- ~;-.-\’-_‘ as. qua-p... .-. 3'4. -‘r~.;-~.-.a;-. l -' al.-5.513111. ,.- ..ffll.’ m.~:.s.~.en..,--, --n -zsg—~v?.>m-3~nl"-sr-:as

rapidly, it usually occurs in about 5 minutes. On the other hand it was observed
that the merocyanine forms a PSB more slowly. When 0.2 eq. of merocyanine
were added at time 0 no shift in the wavelength was observed (“#80 nm).
After 5 minutes a small peak was observed (Am=587 nm) and with time the blue
shifted peak disappeared and the red shifted peak increased. The transition
between one peak to the other showed a clear isosbestic point indicating the
transition of one species to the another. After one hour complete transition was
observed (Figure 6-20). The formation of the PSB is much slower than that of
retinal. As shown in figure 6-19 all-trans-retinal and merocianyne 3 have a
similar 3D shape. thus we cold expect that they would have similar binding
properties (for retinal Kd=l.36:t:4.9 nM CRABPII triple mutant R132K:
R111L:L121E). However so far we cannot prove this because the method used to
determine the dissociation constant for retinal cannot be use to determine the
dissociation constant of merocyanine 3. In brief, the binding of retinal to the
hydrophobic pocket can be followed by fluorescence quenching."’73 The Trp

residues of the protein absorb at 280 nm and fluoresce at ~340 nm. When the

302

A

 

 

400 450 500 550 600 650
Wavelength (nm)

Figure 8-20. UV-vis spectrum of the binding of merocyanine 3 and CRABPII-
R132K::R111L::L121E. Complete shift of the wavelength from A“, = 480 to

1..."... = 587 nm takes about 60 minutes.

chromophore is within the CRABPII cavity, in close proximity to the Trp
residues, it can absorb the light emitted by the protein. The free chromophore 3
absorbs at ~465 nm in an ethanolic solution, therefore quenching of fluorescence
will not occur. Thus it is believed that the rate limiting step is the formation of the
Schiff base, such decrease in the rate of SB formation could be due to the fact that
the aldehydic carbonyl of merocyanine 3 is less electrophilic when compared to
that of retinal, or it could be due to steric factors. We yet have to design a method

that would allow us to measure the dissociation constant accurately.

303

The maximal absorbance observed is red shifted by 9 nm when compared
to the PSB with butylarnine in ethanol (Am=578 nm), however it is not as red
shifted as the maximal absorption observed when merocyanine binds to th.
Upon incubation of chromophore 3 with CRABPII-WT and CRABPII-
Rl32L::RlllL::L12lE (Lys control mutant) a small peak at 580 nm appeared
after 6 h. Appearance of the signal at 580 nm was probably due to the formation
of non specific PSB (a different Lys in the CRABPII sequence could have

reacted) (Figure 6-21).

 

 

0.1 —

0.08 a

g 0.06 -

g 0.04 ~
<

0.02 n

o ‘7— —7 i i j
400 450 500 550 600 650

Wavelength (nm)

Figure 8-21. Comparison 01 the UV-vis spectrum obtained after incubation
of merocyanine 3 and CRABPII-8132K3R111L::L121E Am=587 (blue),
CRABPII-R132K::R11 1L::L121E::R59W W599 (green) and WT-CRABPII

(magenta) km = 480.

304

Table 6-6 summarizes the results observed between the titration of the

rhodopsin surrogates and merocyanine 3. As expected the double mutant

R132K::R111L shows red-shift wavelength by 12 nm (Am=599 nm) as compared
to the triple mutant CRABPII-R132K::Rll1L::L121E, the lack of the counter
anion forces the positive charged formed to be stabilized all along the polyene.
However this protein takes much longer to reach equilibrium (6 hours). The triple
mutant CRABPII-R132K::R111L::L121Q shows similar results, a red shift of 12

nm (Km=597 nm) as compared to the triple mutant CRABPII-

R132K::R111L::L121E and also it takes a long time (10 hours) to stabilize(Figure

6-22).

Table 8-8. Titration of different CRABPII mutants with merocyanine 3 and all-

trans-retinal.

zine-r'lo'az;t.~-.4r«.-;;sl.t Sam-:11. gemsmz-Lnsam;-l'l.:.:.-:-;:_.t1.;y. _. ;-'53:.»earn-hearsewt“ ‘Zv‘lafies' IA .11.; 1.23312? spawn meutAWtie? 30519033” Mim

Protein “Merocyanlne all-bens-Fietinal
.. Lhaxtfl“) .........(hr:.!1"‘ m...

WT 480 377
F1132K 597 (small) 375
R132K::R111L 599 (6 hours) 377
R132K:: R111L::L121E 587 (1 hour) 449
R132L::R111L::L121E 479 377
R132K::H111L::L121E::859E 599 (1 hour) 447
R132K::F111 1K::L121E::F159W 596 (40 minutes) 443
8132K::R11 1K::L121E::A36E 590 (4 hours) 382
H132K::R111K::L121Q 597 (10 hours) 379
R132K::L121E 591 (2 hours) 457

305

0.08 -

 
   

l ____KLE Ann-587 hm
0-07 _____Kl.o W7 nm
0.06 . ____KL M8599 rim

 

 

§ 0.05 «
0.04 ~
g .-
0.02 1
0.01 1
400 450 500 550 600 850
Wavelength (nm)

Figure 8-22. Comparison of the UV-vis spectra obtained after incubation
oi merocyanine 3 and CRABPII-F1132K::Fl111L::L121E Anus-587
(magenta), CRABPII-R132K::Ri11L::L1210 1“,:597 (green) and

CRABPII-R132K::R111l. (blue) it... = 599.

When a second counter anion is placed in the polyene the wavelength

changed as expected, the closer the second counter anion to the Lysine nitrogen
(iminium) the more blue shift is observed. Arg59 is about 6.2 A from the C-9 and
the maximal wavelength observed is 599 nm, and Ala36 is about 4.0 A from C-10
and the maximal wavelength blue shift is observed. Arg59 is about 6.2 A from
the C-9 and the maximal wavelength observed is 599 nm, and Ala36 is about 4.0
A from 010 and the maximal wavelength observed is 590 nm. When compared

with the triple mutant CRABPII-R132K:le11L::L121E, a 12 and 3 nm

306

wavelength red shift is observed for CRABPII-R132K:le11L::L121E::R59E and
CRABPII-R132K: :Rl 11L::L121EzzA36E correspondingly.

Also as discussed previously it has been suggested that a different mode of
regulation is excitonic coupling. The results observed when the tetra mutant
CRABPII-R132K::Rl11L::L121E::RS9W was incubated with merocyanine 3
support this proposed theory, a red shift of 9 nm is observed when a Trp is close
to the polyene region.

Interestingly the presence of a Trp59 also increases the rate at which the

PSB is formed between the rhodopsin surrogate merocyanine 3 (Figure 6-23).

 

 

 

400 450 500 550 600 650
Wavelength (nm)

Figure 6-23. UV-vis spectrum of the titration of merocyanine 3 and CRABPII-
8132KzzFl111L::L121EzzFi59W. Complete shift of the wavelength from W

to W599 nm takes about 40 minutes.

307

 

;./

 

Figure 6-24. A. Fill in model of binding of azulene 4 (red) to the triple mutant
CRABPII-R132K::R111L:L121E. The ionone ring is buried in the cavity. It was
assumed that the azulene would bind in a similar manner to all-trans-retlnal. B This
model represents a comparison of the binding of all-trans-retinai (blue) and azulene

(red) the binding site show important amino acids in the binding pocket.

308

Figure 6-24 compares the all-trans-retinal model with those of azulene 4.
Chromophore 4 is smaller than the retinal. Incubation of CRABPH-
Kl32EzlellezL121E with azulene 4 reveals a wide wavelength absorption,
which consists of by two different peaks. When the peaks were deconvoluted one
peak had a maximal absorption of Amx=426 nm and the second peak had a
maximal absorption of Am=527 run while the ratio between those peaks was
(2:1) nm. No red shifting was observed when the chromophore 4 was incubated
with the CRABPII-WT (km=440 nm) and CRABPII-R132LzlellezL121E
(Am=439 nm) which suggests that the peak observed at 527 nm is due to the

formation of the PSB with Lysl32 (Figure 6-25). The double mutant CRABPH—

A

 

 

 

Wavelength (nm)

Figure 6-25. Comparison of the UV-vis spectrum obtained after
incubation of azulene 4 and CRABPii- R132K::R111uzL121E W434.
511 (green), CRABPII- R132K::R111L: 7e... =418. 540 (blue) and WT-

CFiABPli (magenta) A"... = 440.

309

R132K::L121E shows very similar results to the triple mutant. Whether the
second peak (Amz426 nm) corresponds to free chromophore or the SB is not
clear. But when the pH is decreased the Ame-453 nm and when the pH is basic
the Km=415 predominates (Figure 6-26). This results suggest that a there is a
formation of 3 SB and there is an equilibrium between the PSB and SB.
Incubation of CRABPII-K132E::R111L::L121Q with azulene 4 showed an
unexpected result. Two peaks were observed, one with a maximal absorbance of
AWflZI nm and Amx=538 nm and a ratio of 2:3, which suggests that the PSB is
more stable than when the triple mutant that has the counter anion. The double
mutant CRABPII-R132K::R111L also shows two peaks one at Amx=418 nm and

other at Amax=540 nm with a ration of 1:2 respectively (Figure 6-25). The tetra

pHIG.2
_ ,sz7.3
[”144

   

o i l i l i Y I
350 400 450 500 550 600 650
Wavelength (nm)

 

Figure 6-26. UV-vis spectrum of the titration of azulene 4 and CRABPII-
8132K3R121E. At basic pH the Am = 415 species predominates,
presumably the SB. And acidic pH the am = 534 nm predominates,

presumably the PSB.

310

mutant CRABPII-R132Lzle11L::L121E::R59W shows very similar results, two
peaks with a 1:2 ratio (Am-:439 nm Amx=528 nm). These observations are
unexpected because it seems that the presence of a Glu121 does not promote the
formation of a PSB. Table 6-7 summarizes all the different proteins treated with

azulene 4.

Table 6-7. Titration of different CRABPII mutants with azulene aldehyde 4 and all-

 

trans-retinal.
prom... 3:21-27“; E13223 'i'a'zi'ifi'i.’
.‘ . .1 zewvwvgrw um

WT 440 - 438 377
R132K::R111L 418. 540 (1:2) 421. 547 377
R132K:: F1111L::L121E 434. 511 (2:1) 426. 527 449
R132L::H111L::L121E 439 439 377
R132K::R111K::L121E::859W 431, 528 (1 :2) 424. 537 443
R132K::R111K::L121E::A36E No binding 401 382
R132K::R111K::L121Q 421 , 534 (2:3) 421,534 379
R132K::L121 E 429, 522 429,522 457

In conclusion, different chromophores (3, 4)can form a PSB with the
rhodopsin surrogates developed. It is clear that because the protein surrogates
were developed to bind all-trans-retinal, when ll-cis-retinal is incubated, the

formation of the PSB is not as optimal as with all-trans-retinal. However this is

311

the first example in which an artificial protein binds ll-cis-retinal as a PSB. The
binding of different chromophores was confirmed by incubation of the rhodopsin
surrogates with 3 and 4. Both chromophores form PSB with the designed
proteins. Merocyanine 3 is particularly interesting because it provides a system to
be used to study wavelength regulation. Also when it is incubated with
CRABPII-R132K::Rll1L::L121E::R59W a strongly red shifted chromophore is
obtained (Am=599 nm) compared to the triple mutant CRABPII-

R132K::R111L::L121E.

312

6.5 Materials and Methods.

A. Synthesis of Compounds

All reactions were carried under an atmosphere of nitrogen and removal of
solvents was performed under reduced pressure with a Buchi rotatory evaporator.
THF and Et20 were freshly distilled from sodium/benzophenone, and 04sz was
distilled over Cal-I; under a nitrogen atmosphere. Radioactive NaB’IL was
purchased form American Radiolabeled Chemicals, Inc (St. Louis, MO).
Analytical TLC was carried out using Merck 250 mm Silica gel 60 F254 and spots
were visualized under UV light. Column chromatography was conducted using
Silicycle silica gel (230400 mesh). 300 MHz 'H-NMR and 75 MHz ”c-NMR
spectra were recorded on a Varian Gemini-300 or 500 instruments, and the
residual protic solvent (CDC13 or DMSO-d6) was used as internal reference. UV-

visible spectra were recorded on a Perkin-Elmer Lamda 4O spetrometer.

1. Synthesis of (6) diethyl(3-trimethylsilyl-2-propynyl)phosphonate

B - I
r\ : ms (0502 P. _ ms
5 e

To a stirred solution of NaHMDS in THF (1 M, 10 mL, 10 mol) at -10 °C is
added a solution of diethyi phosphonate in THF (3 mL). The solution was stirred
for 15 min at -10 °C, and a solution of 3 (1.7 g 0.9 mmol) in THF (3 mL) was
added at —10 °C. The reaction mixture was stirred for 2.5 h when the TLC showed

that the reaction was complete. The solution was diluted with water (6 mL) and

313

extracted with ether (2x). The organic layer was dried with NaZSO4. The crude
product was pure enough to continue to the next reaction. [ll-I NMR (300 MHz
CDCl3): 8 4.14(4H, q, J=7 Hz), 8 2.75(2H, d, J=22.2 Hz, CHzP), 5 l.30(6H, t,
J=7.0 Hz,), 50.09(s. 9H, 3 x SiCH3). ”C NMR (300 MHz CDCl3): 5 95.59,

87.81, 63.00, l9.22(d, J=l44.3 Hz, CHZP), 16.31, -0.27.].

2. Synthesis of (8)

TMS
\ ¢

8 E:Z( 86:14)

The ylide 6 was dissolved in anhydrous THF (1.6 g, 6.66 mmol in 30 mL).
The solution was chilled to 0 °C under nitrogen, in the dark room, and it was
treated with n-butyl lithium (1.6 M in hexanes, 4.2 mL, 6.66 mmol). The reaction
was stirred at room temperature for 15 min after which time B-ionone (640 mg in
5 mL was added). The mixture was stirred at room temperature for 5 h, after
which the TLC showed no starting material. The reaction was quenched with
aqueous ammonium chloride and extracted 2 x with ether. The organic phases
were washed with saturated N aCl and dried over Na2804. The crude product was
purified by column chromatography, hexanes:ethyl acetate 95:5 to yield 85% of 8
(13:2, 86:14). [1H NMR (300 MHz CDC13):6 6.24(1H, d, J=15.9), 6.05(1H, d,
J=15.9 Hz), 5.54(1H, s), 2.04(3H, s), l.98(2I-I, t, I: 6.59), l.66(3H, s, CH3),
l.59(2H, m), l.43(2H, m), O.98(6H, s, 2 x CH3), 0.18(9H, s 3 x SiCHa). 13C
NMR (300 MHz CDCls): 8148.55, 137.14, 135.15, 129.80, 129.74, 108.19,

103.76, 100.89, 39.24, 33.93, 32.75. 28.64. 21.39, 18.94, 14.88, —0.17].

314

3. Synthesis of (9)

W
9 E:Z(88:14)

Acetylene 8 (2.78 mmol, 0.798 g) in 10 mL of THF was treated with 10
mL of TBAF (1M, 10 mmol) and stirred at room temperature. After two hours a
small aliquot was quenched and a lI-INMR was taken. No starting material was
observed. Therefore the reaction was quenched with NI-I4Cl and extracted with 2
x ether. The combined organic layers were washed with saturated NaCl and dried
over Nazso... The product was purified by column chromatography, hexanes to
yield 84% of 9. [1H NMR (300 MHz cams 6.l4(lH, d, J=15.9), 5.95(1H, d,
J=15.9 Hz), 5.25(lH, s), 3.14(1H, s), 1.93(3H, s), 1.87(2H, t, J=6.03 Hz),
l.55(3H, s, CH3), l.49(2H, m), l.33(2H, m), 0.88(6H, s, 2 x CH3). 13C NMR
(300 MHz C‘Dc): 5 149.16, 137.07, 135.08, 130.10, 107.32, 107.26, 83.49, 82.35.

39.46, 34.13, 32.95, 28.84, 21.61, 19.15, 15.01].

4. Synthesis of (11)
Hoga— >=/‘OH

10 I 11

Alcohol 10 (1.5 g, 21.40 mmol) was dissolved in dry ether (30 mL)and
chilled to 0 °C under nitrogen. 2.4 equivalents of i-BuMgCl (51.36 mol, 2 M
solution, 25.7 mL) were added drop wised and the solution was stirred for 15 min
at 0 °C. 5% mol of catalytic CpTiClz was added (1.07 mrnol, 261.8 mg). The
solution was warmed up to room temperature and stirred for 5 h. The reaction

was diluted with 30 mL of dry ether and a solution of 1.65 equivalents of 12,

315

(35.31 moi, 8.82 g in 65 mL of dry ether) were added drop wised. The resulting
suspension was stirred at -78 °C for 1 hour and 30 min at room temperature. To
afford 11 mg in a 44% yield. [1H NMR (300 MHz CDCI3): 8 6.31 (1H, t, J=7.14
Hz), 3.99(2H, d, J=6.59 Hz), 2.98 (1H, s), 2.38 (3H, s, CH3). 13c NMR (300

MHz CDCI3): 8 139.55, 98.11, 59.43, 27.87].

5. Synthesis of (12)

WTBDMS

l 12
Alcohol 11 (9.1 mmol, 1.8 g) was dissolved in dry DMF (30 mL) followed

by addition of TBDMSCl (13.6 mmol, 2.04 g) and imidazole (18.2 mmol, 1.24 g).
The reaction was stirred under nitrogen at room temperature overnight. The
reaction was pick up with ether and water (100 mL, and 100 mL). The organic
layer was extracted 2x with ether. The combined organic layers were washed
with 2x water, rinsed with saturated NaCl, and dried over Na2804. The product
was purified via column chromatography using 1% ethyl acetate in hexanes giving
78% yield of 12. ‘H NMR (300 MHz ouch): 5 6.25(1H, l, J=6.04), 4.08(2I-I, d
J=6.59 Hz), 2.37(s, 3H, CH3), 8 0.86 (9H, s, 3 x CH3), 8 0.032 (6H, s, 2 x SiCH3).

13c NMR (300 MHz CDCls): 5 140.32. 95.72. 60.40. 27.86. 25.65, 18.06, -542].

6. Synthesrs of(l3) Womb”
13 E:Z(83:17)

Vinyl iodide 12 (268 mg, 0.86 mmol) was dissolved in isopropyl amine (3

mL), followed by addition of tetrakis(triphenylphospine)palladium (1% mol, 8.2

316

mg, 0.007 mmol) was added. The solution was stirred at room temperature for 5
min and CuI ( 1% mol) was added (1.4 mg, 0.007mrnol). After 5 min acetylene 9
was added (149 mg, 0.7 mmol, dissolved in isopropyl amine) and the reaction was
stirred at room temperature for 4 h. The reaction was quenched by removal of
solvent under reduced pressure followed by addition of ether. The organic layer
was extracted with aqueous NILCI. The organic phase was washed with water,
rinsed with saturated NaCl and dried over Nazsoa. The product was purified by
column chromatography using 2% ethyl acetate in hexanes to afford 13 in 87%
yield. in NMR (300 MHz ouch): 8 6.22(1H, d, J=16.0 Hz), 6.07(1H, d, J=15.9
Hz), 5.90(1H, t, J=6.59 Hz), 5.50 (1H, s), 4.25(2H, (1, 1:6 Hz), 1.98(2H, t,
J=6.59), 1.66(3H, s, CH3), 1.59 (2H, m), 1.43 (2H, m), O.98(9H, s, 3 x CH3),
80.18(6H, s, 2 x SiCHg). 13C NMR (300 MHz C001,): 5 147.07, 137.43.
135.99, 135.55, 129.95, 129.23, 119.29, 98.77, 60.09, 39.51, 34.18, 33.01, 28.99.

28.89, 25.91, 21.66, 19.19, 18.34, 17.72, 15.03, -5.l7].

7. Synthesis of (14)

\\/ OH

14 E:Z( 83:17)

Acetylene 13 (231 mg, 0.58 mmol) was dissolved in dry THF (5 mL) and
was slowly treated with TBAF (3 ml, 1M in THF, 3 mmol). The reaction was
stirred at room temperature for 3 h until the TLC showed that all starting material
was consumed. Water was added to quench the reaction and it was extracted 2x

with ether. The organic layers were washed with saturated NaCl and dried over

317

anhydrous Nazso... The product 14 was purified by column chromatography
(90% hexanes, 10% ethyl acetate) in a 90% yield as a mixture of (13:2, 83:17)

isomers.

Separation of isomers l4 and 17

A ALTEX UltraShereTM-Si (C8) semi-prep column (410 mm ID x 25cm
(5 pm) was used. As a mobile phase 80:20 HexaneszEthyl acetate was used, and a
3 mL/min flow. The peaks were detected at 317 run under dim red light. A
solution of 14 was prepared (1 mg/IOO uL 80:20 Hx:EtOAc). 100 u]. was
injected at every run and the two different fractions were collected (RF22.9 and
Rte-26.3). About 94% of the total weight was recovered every time. Each fraction
was concentrated under rotovap and dried under vacuum, and further analyzed by
‘H—NMR ((2.1).). [‘H NMR (500 MHz C60,): 8 6.28(1H, d, 1:15.91 Hz),
6.16(1H, d, 1:16.34 Hz), 6.02(1H, t, J=6.60 Hz), 5.68(1H, s), 3.83(2H, t, J=6.19
Hz), 2.12(3H, s, CH3), 1.91(2H, t, J=6.19), 1.68(3H, s), 1.66(3H, s, CH3),
1.55(2H, m), 1.43 (2H, m), 1.043 (6H, s, 2 x CH3). 13C NMR (300 MHz CDClg):
8147.07, 137.43, 135.99, 135.55, 129.95, 129.23, 119.29, 98.77, 60.09, 39.51,

34.18, 33.01, 28.99, 28.89, 25.91. 21.66, 19.19, 18.34, 17.72, 15.03, -5.17].

8. Synthesis of (16)

 

Activated Zn dust was prepared as described by Boland with a few

modifications. Argon was bubbled through a suspension of Zn dust (1 g) in

318

distilled water (6 mL) for 15 min. Cu(OAc)2 (100 mg) was added and the flask
(exothermic reaction) was sealed immediately. The mixture was stirred
vigorously for 15 min. AgNO3 (1 g) was then added and the solution was stirred
for 30 min. The activated Zn was centrifuged (400 RPM, rotor 243, 90 sec) and
washed successively with water, methanol, acetone and ether. The moist
activated Zn (important) was transferred immediately to a flask of the reaction
solvents (water 2mL and methanol 2 mL). Compound 14 (15 mg, 0.053 mmol)
was added to this mixture and was then stirred at room temperature in the dark
overnight. After the TLC showed no starting material, the Zn dust was filtered
through celite with ether and water. The organic phases were separated, washed
with saturated NaCl, and dried over anhydrous Nazso. The solvent was removed
under reduced pressure to yield 16 plus some isomers. Purification at this stage
was not performed. [1H NMR (500 MHz CDC13)28 6.53(1H, d, 1:11.93 Hz),
6.35(1H, t, 1:11.93, 11.93 Hz), 6.41(1H, d, J=16.13 Hz), 6.10(1H, d, 1:15.90),
5.87(1H, d, J=11.71), 5.72(1H, t, J=6.45), 4.26(2H, d, J=7.00 Hz), 2.01(2H, t,
J=6.19), 1.93(3H, s, CH3). 1.89 (3H, s, CH3), 1.71 (3H, s, CH3), 8 1.62 (2H, m),

1.47 (2H, m), l.03(6H, s, 2 x CH3)].

9. Synthesis of (l)

\
O

The crude reaction from 16 was dried under vacuum, dissolved in CH2C12

(3 mL) and chilled to 0 °C. MnOz was added (1.30 mmol, 120 mg) and the

319

reaction was stirred for 4 h. The reaction was passed through a celite pad to filter
M1102 and washed with CH2C12. The solvent was dried over anhydrous Na2804.
The crude was passed through a micro column to separate the polar impurities
(Silica was stabilized with hexanes and eluted with 5% EtOAc in hexanes). The
fraction containing the ll-cis-retinal 14 was collected and purified using HPLC,
(for details see below). The 1H-NMR spectrum matches the reported data. [1H
NMR (500 MHz CDCI3)316 10.07 (1H, d, J=8.17 Hz), 6.66 (1H, t, 1:12.15 Hz),
6.51 (1H, t, 1:12.59 Hz), 6.31 (1H, d, J=16.13 Hz), 6.12 (1H, d, J=16.13 Hz),
6.06 (1H, d, J=8.17 Hz), 5.91(1H, t, 1:11.49 Hz), 8 2.33(3H, s, CH3), 2.00(2H, t,
J=5.29 Hz), l.97(3H, s, CH3), l.69(3H, s, CH3), l.60(2H, m), 1.45(2H, m),

l.02(6H, s, 2 x CH3)].

Purification of 11-cis-retinal

A colum ALTEX UltraShereTM-Si (C8) semi-prep (410 mm ID x 25cm
(5 pm) was used, and the column was rinsed with 98:2 hexanes:ethyl acetate 0.05
v/v Et3N (to neutralize any acid present). All procedure were performed under
dim red light. All samples were kept at 0 °C or less, and in the dark when
possible. A mobile phase of 98:2 hexanes:ethyl acetate, was used, 3 mllmin. at
room temperature and a UV detection at 362 nm. The crude (10-15 mg) was
dissolved in 1 m1. of 2% EtOAc in hexanes. 100 11L was injected at every run and
three different fractions were collected (Re-18.39, 17.33%; Rr=22.64, 27%; and

R;39.74, 24%).

320

Each fraction was concentrated under rotovap and dried under vacuum.
Further analysis of the ‘H-NMR (CDCl3) of each fraction showed that fraction 2

(1111:2264) was ll-cis-retinal. 1.5 mg of 11-cis-retinal were recovered 10%.

Preparation of tert-butyl hypochlorite t-BuOCl

In a 1 L round-bottomed flask, provided with mechanical stirring and
protected from light, bleach (500mL) was cooled in an ice/water bath until the
temperature of the bleach was below 10 °C. A mixture containing tert-butanol
(35.25 mL) and acetic acid (25 mL) was then added in one portion and the whole
mixture was vigorously stirred for 3 min. The contents of the flask were poured
in a extraction funnel and the lower aqueous layer was discarded. The organic
fraction was washed with 10% solution of N82C03 (50 mL) and then with water
(50 mL). The t-butyl hypochlorite thus formed was dried over CaClz, filtered and
used subsequently without further purification. About 70 g were obtained; the

ten-butyl hypochlorite was used without further purification.“

10. Synthesis of (E)-1-O-acetyl-4-chloro-3-methyl-2-butene (25)
2] Cl\/j\/‘0lle
23 25
Isoprene 23 (12.5 mL, 12.5 mol) was dissolved in acetic acid (72.5 mL,
12.5 mol) and cooled in an ice/salt bath. tert-Butyl hypochlorite (12.5 g, 12.5
mol) was added in small portions, and the solution was stirred for 1 h. The excess

acid was carefully neutralized with a solution of sodium hydroxide and extracted

321

with ether (20 N NaOH was used). The organic phase was dried over Nazsoa, the
solvent evaporated and the crude product was purified by column chromatography
(pentane/ethyl ether 10:1) to afford 25 in 33% yield. [‘H NMR (300 MHz
CDC];): 8 5.64(1H, t. J=5.49 Hz), 4.54(d, 2H J=6.59 Hz), 3.95(2H, s, CH2Cl),
2.01(3H, s, CH3), 1.77(3H, s, CH3). 13C NMR (300 MHz ouch): 5 170.63,

136.81, 123.62, 60.57, 50.65, 20.71, 14.38]. V

11. Synthesis of (E)-1-O-acetyl-4-chloro-3-methyl-2-buten-1-ol (26)

C'VKAOH

tert-Butyl hypochlorite (8.65 g, 0.8 mol), was added at 0 °C to a solution
of i80prene (6.8 g, 0.1 mol) in acetic acid (29 mL, 0.48 mol). The reaction
mixture was stirred for 1 h, and extracted with water. The combined extracts were
successively washed with sat. NaHC03 and sat NaCl, and dried with Nazso.
The filtrate was concentrated under reduced pressure, dissolved in acetic acid (30
mL) and treated with both copper sulfate (0.20 g, 1 mmol) and sulfuric acid (0.2 g,
2 mmol). The resultant mixture was stirred for 4 days at ambient temperature
quenched by the addition of water and extracted with ether.75 The separation of
this isomers was impossible. 3.4 g of the crude product was dissolved in
methanol (60 mL) and stirred for 5 h at 0 °C in the presence of an aqueous
solution (20 mL) of sodium carbonate (3.6 g in 20 mL).75 The mixture was
filtered, concentrated and poured onto ice/water. The water was extracted with

chloroform (3x), and dried with anh. Nazso. The crude product was purified by

322

chromatography to afford the alcohol as a mixture of isomers. The crude product
(5 g) was dissolved in methanol (60 mL) and stirred for 5 h at 0 °C in the presence
of an aqueous solution (20 mL) of sodium carbonate (3.6 g in 20 mL).2 The
mixture was filtered, concentrated and poured onto ice/water. The water was
extracted with chloroform (3x), and dried with anh. N82304. The crude product
was purified by column chromatography usining 7:3 pentanezethyl ether. The

chloro-alcohol was isolated in (3.7 g) a 99.8%. [1H NMR (300 MHz CDCI3)I 8
5.67(1H, t, J=7 .14 Hz), 8 4.11(2H, d, J=6.04 Hz), 3.95 (2H, s, CHZCl), 1.72(3H, s,

CH3). 13C NMR (300 MHz (Inch): 5 134.17, 128.83, 58.77, 51.24, 14.23].

12. Synthesis of (D-l-O-acetyl-4-chloro- 3-methyl-2-butenal (27)

C'No

27
The chloro alcohol 26 (1.8 g, 13.2 mmol) in dichloromethane (20 mL) was

added at 0 °C to a suspension of pyridinium chlorochromate (4.4 g, 21 mmol) in
dichloromethane (40 mL). The solution was kept at 0 °C for an additional 90 min
then filtered through a short alumina column. Elution with ether yield 1.5 g 4-
chloro-3-methyl-2-butenal (85%) (two aldehyde peaks could be observed, in
87:13 ratio). After this material is prepared it should be used within one week to
perform the next reaction, otherwise it decomposes. 1H NMR (300 MHz
CDCI3): 8 9.95(1H, d, J=7.69 Hz), 6.03(2H, d, J=7.69 Hz), 4.04(2H, s, CHZCl),
2.19(3H, s, CH3). 13C NMR (300 MHz (Inch): 8190.82, 155.19, 49.06, 15.26.

m/z (E.I): 118.01(M*), 53000)].

323

13. Synthesis of (E)-4-bromo—3~methyl-2-butene acid (28)

BrvK/COOH
28

3,3 dimethyl acrylic acid (10 g, 0.1 mol) was dissolved in CC14 (70 mL).
Then of N-bromo succinimide (NBS) (0.11 mol 20 g) was added and the solution
was illuminated with a iarnp and heated to a reflux for about 4 h. After 4 h the
succinamide had precipitated out. The solution was stirred for extra 2 h without
the lamp. To the solution, hexane (30 mL) was added and the precipitate was
filtered out under reduced pressure. The filtrate was concentrated under vacuo.
The residue was picked up in MeOH and excess K2C03 anhydrous was added.
Then, the solution was stirred for 2 h resulting in lactonization of the cis-bromo-
alkene. Most of the MeOH was removed under reduced pressure and the residue
was picked up with ether and water. The aqueous phase was the acidified to
pH=l with concentrated H2804 and extracted three times with ether. The ether
was washed with saturated NaCl and dried over Na2804. The crude product gave
from 20 to 42% yield of the pure E-isomer. The highest yield was obtained when
fresh recrystallized NBS was used. [‘11 NMR (300 MHz ouch): 5 810011, b,
COOH), 5.96(1H, s), 3.94(2H, s, CHzBr), (3H, s, CH3). 13C NMR (300 MHz

CDC13): 8 171.08, 155.27, 118.81,37.91, 17.52].

324

14. Synthesis of (29)

ervkcooMe

The product 28 was used without further purification and was dissolved in
MeOH (25 mL) and it was acidified with concentrated H2804 (2 mL). The
reaction was refluxed overnight. The reaction was picked up with ether and
water. The aqueous layer was extracted three times with ether. The organic layer
was washed with aqueous K2C03 two times and with saturated NaCl. The
product was purified using column chromatography using eluant 2% ethyl acetate
98% hexanes to afford 82% yield of the bromo ester 29. [‘H NMR (300 MHz
CDCh): 8 5.94(1H, s), 3.92(2H, s, CHzBr), 3.69(3H, s, OCH3), 2.26(3H, s, CH3).
13C NMR (300 MHz CDC13): 5 166.15, 152.67, 118.94, 51.18. 38.10, 17.07. m/z

(E.I):191.9, 193.9 (1:1) M), 113.1(100)].

15. Synthesis of (30)
Etc-g OOMe
E10 30

The bromo ester 29 (1 g, 5 mmol) was combined with ethyl phosphite
(0.86 g, 5 mmol) in a sealed tube. The mixture was heated overnight at 125 °C.
The reaction yielded 2.2 g of crude product (93% yield). The product was used
without further purification. [1H NMR (300 MHz CDC13): 5 5.70(1H, s),
4.02(2H, q, J=8.14 Hz, OQHgCHg), 3.60(3H, s, 0%), 2.60(2H, (1, 1:23.35 Hz.
8911;). 2.22(3H, s), 1.23(6H, t, J=4.14 Hz, 2 x OCHLQHJ. ”C NMR (300 MHz
CDC13): 8166.27, 149.96, 119.44, 62.13, 50.84, 38.33(d, J=134.33 Hz, m,

19.87, 16.27. m/z (13.1): 250.0(M*). 1337000)].

325

16. Synthesis of (47)

CICHZCN P(OEt)3 (EtO)2P(O)CH20N
4s 4s 47

To a sealed tube one equivalent of and triethylphosphite 49 (11 g, 66
mmol) was added followed by a drop wise addition of one equivalent of
chloroacetonitrile 48 (5 g, 66 mmol). The mixture was heated to 125 °C over
night. The reaction was cooled down to room temperature, and solid formed was
resuSpended in a minimal amount of dichloromethane. The product was
precipitated by addition of ether, and the crystals obtained were re-crystallized
using dichloromethane and ether. The ylide 47 was obtained in a 98% yield (11.5
g as a yellowish oil). [1H NMR (300 MHz CD03): 8 4.45(4H, q, J=7.14 Hz,
OQHZCHg), 2.60(2H, d, 1:20.88 Hz, P_C_H;), l.48(6H, t, J=7.14 Hz, 2 x

OCH2CH3].

17. Synthesis of (31)

ClMCOOMe

5:2 (2:1) 81
The ylide 30 (0.4 g) were dissolved in dry TIIF (3 mL) and NaHMDS (1.6

mL of 1 M) were added at 0 °C. The solution was stirred at 0 °C for 30 min. The
aldehyde 27 (0.19 g) was dissolved in THF (5 mL) and was added slowly to the
ylide solution at 0 °C. Then, the reaction was stirred for 4 h. The reaction
quenched with water and the aqueous layer was extracted with ether three times.
The combined organic layers were washed with 10% aqueous HCI, followed by
saturated NaHC03, and saturated NaCl. Then the organic layer was dried N 82504

(anh) and was concentrated to give the chloro ester 31. The compound was

326

purified using chromatography gel hexzether 92:8 and 100 mg of chloro ester 31
were obtained in a 30% yield. [‘II NMR (300 MHz CDCl;): 5 7.75(1H, d,
J=16.0 Hz), 6.73(1H, m), 6.23(2I1, m), 5.65(1H, s), 4.06(2H, s, CH2C1), 3.66(3H,
s, OC_H;), 2.00(3H, s, CH3), 1.90(3H, s, CH3). 13C NMR (300 MHz CDCl3):
5166.27, 150.52, 136.96, 130.51, 129.68, 119.34, 117.37, 51.75, 50.91, 22.55.

20.72].

18. Synthesis of (33)

Eggdwkcooue

5:2 (2:1) 33

The chloro ester (0.63 g, 2.75 mmol) was combined with ethyl phosphite
(0.46 g, 2.75 mmol) of in a sealed tube. The mixture was heated for 16 hours at
100 °C. After the reaction the crude gave 33 (0.78 g) in a 90% yield The product
was used without further purification. [1H NMR (300 MHz CDC13): 8 7.64(1H,
d, J=16.0 Hz), 6.76(1H, m), 6.14(2H, m), 5.66(1H, d, 1:14.10 Hz), 4.05(2H, q,
J=8.14 Hz, OQECHg), 3.66(3H, s, 0(113), 2.62(2H, d, 1:23.45 Hz, P2112).

2.29(3H, s, CH3), 2.00(3H, s, CH3), 1.23(4H, t, J=4.14 Hz, OCH2CH3 )].

19. Synthesis of (34) wHo

34
POC13 (394 mg, 2.57 mmol) was added to a flask and cooled down to 0

°C, then DMF (6 mL) was added drop wised under N2. The mixture was stirred at
0 °C for 30 rrrin. Azulene (300 mg, 2.0 mmol) was dissolved in DMF (4 mL,

anh.) and were cooled down to 0 °C, then the azulene solution was added drop

327

wised to the mixture. The reaction was stirred for 2 h. Water was added to
quench the reaction and it was let to warm slowly (to avoid sudden reaction with
water and excess POC13). Saturated solution of NaHCO; (3-5 mL) was added to
basify the solution. The aqueous layer was extracted with ether five times. The
combined organic layer were washed with saturated NaCl and dried over
anhydrous sodium sulfate. The crude was purified using flash chromatography
hexanes:ether 4:1 to afford 75.5% of the aldehyde and about 10% of SM. TLC:
HexaneszEther 4:1, Starting Material Rf=0.64 blue color, product RF: 0.23 red
color. [‘H NMR (300 MHz CD03): 5 10.33(1H, s), 9.54(1H, d, J=9.89 Hz),
8.45(1H, d, J=9.6 Hz), 8.23(1H, d, 1:94. 12 Hz), 7.81(1H, t, J=9.9 Hz), 7.57(1H. t,
J=9.9 Hz), 7.48(1H, t, J=9.9 Hz), 7.29 (1H, d, J=5.1 Hz). 13C NMR (300 MHz
CDC13): 8 186.51, 146.02, 141.88, 139.69, 138.96, 137.44, 129.43, 128.20.

125.89, 119.00].

20. S thesis of 46
yr. ( ) WCOOMO
C 46 E:Z(2:1)

Yiide 30 (0.96 g of the 3.8 mmol) was dissolved in dry TI1F(10 mL) and
of NaHMDS (3.84 mL, 1 M) were added slowly at 0 °C. The solution was
stirred for 30 min. and then the solution was cooled to 0 °C. The aldehyde 34
(0.06 g) was dissolved in THF (5 mL) and cooled down to 0 °C and then it was
added slowly to the ylide solution. The reaction was stirred for 1 h and then it
was warmed slowly to 0 °C. After 6 h the reaction was stopped by addition of

water and it was warmed to room temperature. The aqueous layer was extracted

328

three times with ether. The combined organic layers were washed with 10%
aqueous HCl, saturated NaHCO3, saturated NaCl and dried over Na2804 (anh).
The organic layer was concentrated to give the ester 46. The compound was
purified via flash chromatography using hexane:ethylacetate 92:8 to yield 40 mg
of product 46 were obtained in a (41%) yield as a mixture of two isomers. TLC
90:10 HesztOAc, RF0.48 greenish and 20% of starting material 34 was
recovered [in NMR (300 MHz CD013): 5 8.43(1H, d, J=9.61 Hz), 8.21(1H. d.
J=9.06 Hz), 8.12(1H, d, J=4.12 Hz), 7.53(2H, d, J=13.18 Hz), 7.38(1H, d,
J=4.12 Hz), 7.13(2H, q, 1:10.44 Hz), 6.91(1H, d, 1:15.66 Hz), 5.89 (1H, s),
3.74(3H, s, 0%), 2.51(3H, s, CH3). 13(2 NMR (300 MHz ouch): 8167.82,
138.58, 137.11, 136.63, 133.69, 133.57, 129.44, 126.08, 124.67, 123.63, 119.59,

117.03, 67.96, 50.95, 13.82].

21. Synthesis of (50)
WW
50 E:Z(221)

Ester 46 (40 mg, 0.16 mmol) were dissolved in dry THF (5 mL). The
solution was cooled down to -78 °C and DIBAL-H (0.130 mL 2.5 M in hexanes)
was added drop wise, and then it was stirred for 4 h. The reaction was quenched
with Na,K tartrate. The solution was warmed to room temperature and stirred for
1 hour. The aqueous solution was extracted five times with ether. The organic
layers were combined and dried over Na2SOa (anh), then the organic layer was
concentrated to give the alcohol 50. The crude product was used directly for the

next reaction. ['H NMR (300 MHz CDC13): 5 8.38(1H, d, J=9.99 Hz), 8.17(2H,

329

m), 7.51(2H, m), 7.36(1H, t, J=4.54 Hz), 7.13-7.03(3H, m), 6.91(1H, (1, 1:15.66

Hz), 5.61(1H, t, J=6.87 Hz), 4.43 (2H, (1, 1:17.41 Hz), 2.09(3H, s, CH3)].

22. Synthesis of (4) C
We.

The crude mixture of 50 (6.5 mg, 0.16 mmol) was dissolved in CH2C12 (5
mL) and chilled to 0 °C. Mn02 was added (3.2 mmol, 300 mg) and the reaction
was stirred for 3 h. The reaction was passed through a celite pad to filter Mn02
and washed with CH2Cl2. The solvent was dried over anhydrous Na2SOr. The
aldehyde 4 was purified via flash chromatography using exane:EtOAc 92:8 to
yield 6.5 mg of 4 18%. [‘11 NMR (300 MHz CDC];): 5 10.13(1H. d. J=8.24 Hz).
8.46(1H, d, J=9.82 Hz), 8.23(2H, m), 7.68(1H, d, J=15.9 Hz), 7.63(2H, d,
J=13.18 Hz), 7.38(1H, d, J=3.82 Hz), 7.2(2H, q, J=9.88 Hz), 6.96(1H, (1, 1:15.96
Hz), 5.89 (1H, d, J=8.2 Hz), 2.47(3H, s, CH3). 13C NMR (300 MHz CDCh):
8 167.82, 138.58, 137.11, 136.63, 133.69, 133.57, 129.44, 126.08, 124.67, 123.63,
119.59, 117.03, 67.96, 50.95, 13.82. UV: 71...... (EtOH) 435 nm; e = 14,1918 M

lcm"].

23. Synthesis of (45)

N: \:/ \‘o

’ 45
(COCl)2 (1.175 g, 9.3 mmol) were added to a cooled flask (0 °C)

containing DMF (30 mL) and CCla (25 mL) and the mixture was stirred at 0 °C

for 30 min. Indolin was added dropwised (2 ml, 1.958 g, 11.3 mmol) were

330

dissolved in DMF (10 mL) and were cooled down to 0 °C. The reaction was then
warm up to room temperature and to 50-60 °C for 3 h. Cold water was added to
quench the reaction and the pH was adjusted to >10. The reaction was extracted
with ether five times. The organic layer was washed with saturated NaCl and
dried over anhydrous sodium sulfate. The crude product was purified using flash
chromatography hexanes:ether 4:1 to afford 93% of the aldehyde. [1H NMR (500
MHz CDCI3): 8 9.94(1H, d, J=9.07 Hz), 7.23(1H, t, J=7.42 Hz), 7.19(1H, d,
J=6.31 Hz), 6.99(1H, d, J=7.41 Hz), 6.80(1H, d, J=6.97 Hz), 5.33(1H, d, J=9.16
Hz), 3.17(3H, s, NCH3), 1.59(6H,s, 2 x CH3). 13'c NMR (300 MHz CDCl.):
8 186.32, 173.49, 143.26, 139.19, 127.90, 122.30, 121.63, 117.00, 107.87,

98.77.4724, 29.37].

24. Synthesis of (45)

Ont/W028“

46 E:Z(1:1)

\

The ylide 30 (1.37 g 5.50 mmol) was dissolved in TIE (20 mL, dry) and
NaHMDS (5.5 mL 1 M) were added at 0 °C. Then, the solution was stirred for 30
min and it was cooled to 0 °C. Aldehyde 38 (0.11 g, 1 mmol) was dissolved in
TIE (5 mL) and cooled down to 0 °C and then it was added slowly to the ylide
solution. The reaction was stirred for 1 h and then it was warmed slowly to room
temperature. After 6 h the reaction was stopped by addition of water and the
reaction was extracted three times with ether. The combined organic layers were

then washed with saturated NaHC03, saturated NaCi and dried with Na2SOa

331

(anh). The organic layer was concentrated to give the ester 38. The compound
was purified via flash chromatography using hexane: EtOAc 9:1 and 40 mg of
product 38 were obtained in a 50% yield as a mixture of two isomers. TLC 85:15
Hex:EtOAc, Rr=035. [1H NMR (500 MHz CDCI3): 5 7.29-7.16(2H, m,),
6.89(1H, t, J=7.42 Hz), 6.66(1H, d, J=8.06 Hz), 6.10(1H, d, 1:15.65 Hz),
5.69(1H, s), 5.54(1H, d, 1:15.14 Hz), 5.41(1H, d, J=11.96 Hz), 3.73(3H, s,
NCH3), 3.17(3H, s, OCH3), 1.63(3H,s, CH3), 1.62(6H, s, CH3). 13C NMR (300
MHz CDCla):8 168.10, 159.43, 154.33, 144.88, 138.69, 133.39, 127.74, 121.52.

119.79, 113.59, 111.42, 106.04, 95.96, 50.67, 45.66, 28.4, 21.09, 14.40, 13.87].

25. Synthesis of (3)
WC

’ s
The ester 45 (40 mg, 0.13 mmol) was dissolved in dry TIE (5 mL). The

solution was cooled down to -78 °C. DIBAL-H (0.130 mL of l M in hexane) was
added drop wise. After 4 hours the reaction was quenched with Na,K tartrate and
then it was warmed to room temperature and stirred for 4 h. The aqueous solution
was extracted five times with ether. The organic layers were combined and dried
over Na2804 (anh). The organic layer was concentrated under reduced pressure.
The crude product was used directly for the next reaction. The crude mixture was
dissolved in CH2Ci2 (5 mL) and chilled to 0°C. Mn02 freshly homemade was
added (2.6 mmol, 244 mg) and the reaction was stirred for 3 hours. The reaction
was passed through a celite pad to filter Mn02 and washed with CH2Cl2. The

solvent was dried over anhydrous N82304. The aldehyde 3 was purified via flash

332

chromatography using Hexane:EtOAc 85:15 to yield 1.5 mg (5% yield) starting
form the ester 45. (‘H NMR (500 MHz C1303): 5 10.01011. d, J=8.80 Hz),
7.20-7.00(2H, d, 1:14.28), 6.80—6.45(2H, m), 6.10(1H, d, J=15.93 Hz), 5.87(1H,
d, 1:14.83 Hz), 5.66(1H, d, J=9.21 Hz), 5.44.(1H, d, 1:14.28 Hz), , 3.20(3H, s,
NCH3), 3.17(3H, s, OCH3), 2.29(3H, s, CH3), 1.58(6H, s, 2 x CH3). UV: Am

(EtOH) 468 nm].

B. Analysis of CRABPII mutants with chromophores.

UV-vis titrations of CRABPH with chromophores

A stock solution of the chromophores was prepared in 95% spectroscopy
grade EtOH. For a 5 11M protein solution in PBS (4 mM NaH2P04, 16 mM
Na2IIP04, 150 mM NaCl, pH:7 .3).

Additions from 0.1-1.0 equivalents at 0.1 equivalent increments were
added and spectra recorded at room temperature from 200-700 nm (Cary WinUV,
Varian). The 2““ derivative of the spectra was calculated by using the
corresponding software provided with the UV instrument. The determination of
the km of a UV peak using mathematical equations has been established as a
valid method, and has been used in a variety of applications?“8 As previously
mentioned, all the UV data discussed and shown in tables refer to spectra recorded

in the presence of 0.1—0.2 equivalents of chromophore, unless otherwise noted.

333

C. Fluorescence and MALDI-‘I‘OF

A detailed description of the determination of binding constant for the

CRABPII mutants using retinal, is provided in Chapter 2, of Chrysoula

Vasileiou’s Dissertation.

In short, the cuvette is allowed to sit with 3 mL of a 0.01% gelatin

containing PBS (4 mM NaH2P04, 16 mM
Nazi-IP04, 150 mM NaCl, pH=7.3) for 30-
60 min. The sample is excited at 283 nm
with a slit width about 1.5 run (this varies
in order to ensure that the intensity of each
sample remains below one million counts).
The fluorescence is measured at the peak
maximum, about 345 nm (varies with
different proteins). The chromophore is
added at varying equivalents from a ~1.5
mM stock (in 95% spectroscopy grade
EtOH) sample maintained in the dark. A
measurement is taken after each addition
at the same wavelength. The results are
plotted as concentration of chromophore
versus relative fluorescence intensity.
When the curve levels off, the titration is

complete. The dissociation constant is

334

\\\\\o

i

Haiti/Y Lye132

i m

\ \ \ \ \N/ymm

[M-r

 

a"! Lye132

[M+

 

/
.__/.__,

/

/

\ \ \ \ ”Wham

i

[M +288]

 

Figure G-A Covalent bond formation
between retinal and the protein can be
detected by MALDI-TOF. Schiff base or
PSB formation can be seen as an [M+266]’
or [Ml-267]“ peak, respectively. Both
species can be reductively aminated and
trapped as an [M+268]’ species.

calculated with the program SigmaPlot through non-linear least square regression

analysis.72

D. MALDI-TOF

A detailed description of the determination of binding constant for the

R132K:Y134F:R111L:L121E
M’s: 15,490.?

WWW Hafiw 15’491'5

1 . F881 0 mlumn MALD|_TOF
2. NaCNBH3 w

 

 

15000 15500 16000 16500

 

G
\ \ \ \ \N Mass De
“W 15 4901 ( )
2. Fast Q column MALDI-TOF ’ -

\\\\ aw

 

15000 15500 18000 16500

Mace (De)
Figure 6-8. Schematic representation of the reductive amination process using
R132K:Y134F:R111L:L121E mutant. ii the Fast 0 column is run before the reductive
amination no retinal adduct is detected MALDI-TOF (top). On the contrary, when the
column follows the reductive amination, the characteristic adduct of (M+288)* is observed
(bottom). Fast 0 column can successfully separate the free retinal from the CRABPII

mutants.

CRABPII mutants using retinal, is provided in Chapter 2, of Chrysoula

Vasileiou’s Dissertation.

335

An additional binding assay performed on each CRABPII mutant involves
the use of two MALDI-TOF experiments. The first of these experiments is the
incubation of retinal with the protein for Schiff base formation. The sample is
then washed with ethanol and concentrated. The covalent bond between retinal
and the protein, if formed, can be detected as an [M+266]+ peak in the MALDI-
TOF spectrum. In addition, reductive amination of this incubated sample with
NaCNBH; is performed to covalently trap the retinal molecule with the protein
(Figure 6-B). The necessity for both MALDI-TOF experiments arises from the
fact that some Schiff bases, protonated or not, may not be stable enough to be
trapped under the reductive amination reaction conditions, but will survive the
work-up from the incubation long enough to be detected.

Since an excess of retinal is used for the reductive amination reaction, the
non-covalently bound chromophore must be separated using a Fast Q column
(quaternary ammonium resin).

The MALDI-TOF spectra for either the incubation or the reductive

amination reactions can only be evaluated qualitatively and not quantitatively.

E. Molecular Modeling:

Computational results were obtained using software programs from
Accelrys. Dynamics and rninimizations were done using the Discover 3"
program, using the CVFF forcefield from within the InsightII 2000 molecular

modeling system.

336

Protein figures were generated using PyMol Molecular Graphics System,

version 0.93, copyright by DeLano Scientific LLC.

337

6.6

References

D. D. Oprian; A. B. Asenjo; N. Lee; S. L. Pelletier, "Design, chemical
synthesis, and expression of genes for the 3 human color-vision pigments."
Biochemistry 1991, 30, 11367-11372.

S. L. Merbs; J. Nathans, "Absorption-spectra of human cone pigments."
Nature 1992, 356, 433-435.

H. J. M. Degroot; G. S. Harbison; J. Herzfeld; R. G. Griffin, "Nuclear
magnetic-resonance study of the Schiff-base in bacteriorhodopsin-
counterion effects on the N-15 shift anisotropy." Biochemistry 1989, 28,
(8), 3346-3353.

J. Hu; R. G. Griffin; J. Herzfeld, "Synergy in the spectral tuning of retinal
pigments: Complete accounting of the opsin shift in bacteriorhodopsin."
Proceedings of the National Academy of Sciences of the United States of
America 1994, 91, 8880-8884.

A. G. Doukas; A. Pande; T. Suzuki; R. H. Callender; B. Honig; M.
Ottolenghi, "On the mechanism of hydrogen-deuterium exchange in
bacteriorhodopsin." Biophysical Journal 1981, 33, (2), 275-279.

H. Luecke; B. Schobert; H. T. Richter; J. P. Cartailler; J. K. Lanyi,
"Structure of bacteriorhodopsin at 1.55 angstrom resolution." Journal of
Molecular Biology 1999, 291, (4), 899-911.

A. Kusnetzow; D. L. Singh; C. H. Martin; I. J. Barani; R. R. Birge,
"Nature of the chromophore binding site of bacteriorhodopsin: The
potential role of Arg(82) as a principal counterion." Biophysical Journal
1999, 76, (5), 2370-2389.

T. Ebrey; M. Tsuda; G. Sassenrath; J. L. West; W. H. Waddell, "Light
activation of bovine rod ph08phodiesterase by non-physiological visual
pigments." FEBS Letters 1980, 116, 217-219.

C. N. Rafferty, "Light-induced perturbation of aromatic residues in bovine
rhodopsin and bacteriorhodopsin." Photochemistry and Photobiology
1979, 29. 109-120.

338

10.

11.

12.

13.

14.

15.

16.

17.

18.

T. Okada; M. Sugihara; A. N. Bondar; M. Elstner; P. Entei; V. Buss, "The
retinal conformation and its environment in rhodopsin in light of a new 2.2
angstrom crystal structure." Journal of Molecular Biology 2004, 342, (2),
571-583.

R. E. Stenkamp; S. Filipek; C. A. G. G. Driessen; D. C. Teller; K.
Palczewski, "Crystal structure of rhodopsin: a template for cone visual
pi grnents and other G protein-coupled receptors." Biochimica Er
Biophysica Acta-Biomembranes 2002, 1565, (2), 168-182.

J. Nathans; D. Thomas; D. S. Hogness, "Molecular-genetics of human
color-vision-the genes encoding blue, green, and red pigments." Science
1986, 232, 193-202.

R. Yokoyama; S. Yokoyama, "Convergent evolution of the red- and green-
like visual pigment genes in fish, astyanax fasciatus, and human."
Proceedings of the National Academy of Sciences of the United States of
America 1990, 87, 9315-9318.

A. B. Asenjo; J. Rim; D. D. Oprian, "Molecular determinants of human
red/green color discrimination." Neuron 1994, 12, 1131-1138.

T. Chan; M. Lee; T. P. Sakmar, "Introduction of hydroxyl-bearing arrrino-
acids causes bathochromic spectral shifts in rhodopsin - amino-acid
substitutions responsible for red-green color pigment spectral tuning."
Journal of Biological Chemistry 1992, 267, 9478-9480.

T. Yoshizawa, "The road to color-vision-structure, evolution and function
of chicken and gecko visual pigments." Photochemistry and Photobiology
1992, 56, 859-867.

J. I. Fasick; N. Lee; D. D. Oprian, "Spectral tuning in the human blue cone
pigment." Biochemistry 1999, 38, (36), 11593-11596.

J. L. Spudich; C. S. Yang; K. H. Jung; E. N. Spudich, "Retinylidene
proteins: Structures and functions from archaea to humans." Annual
Review of Cell and Developmental Biology 2000, 16, 365-392.

339

19.

20.

21.

22.

23.

24.

25.

26.

27.

C. Pande; A. Pande; K. T. Yue; R. Callender; T. G. Ebrey; M. Tsuda,
"Resonance raman-spectroscopy of octopus rhodopsin and its
photoproducts." Biochemistry 1987, 26, 4941-4947.

T. Kitagawa; M. Tsuda, "Resonance raman-spectra of octopus acid and
alkaline metarhodopsins." Biochimica Et Biophysica Acta 1980, 624, 211-
217.

W. Stoeckenius; R. A. Bogomolni, "Bacteriorhodopsin and Related
Pigments of Halobacteria." Annual Review of Biochemistry 1982, 5], 587-
616.

B. Ehrenberg; A. Lewis; T. K. Porta; J. F. Nagle; W. Stoeckenius,
"Exchange kinetics of the Schiff-base proton in bacteriorhodopsin."
Proceedings of the National Academy of Sciences of the United States of
America-Biological Sciences 1980, 7 7, (11), 6571-6573.

J. Favrot; C. Sandorfy; D. Vocelle, "Infrared study of basicity of non-
aromatic imines-relation to visual pigments." Photochemistry and
Photobiology 1978, 28, (2), 271-272.

J. Herzfeld; S. K. Dasgupta; M. R. Farrar; G. S. Harbison; A. E.
Mcderrnott; S. L. Peiletier; D. P. Raleigh; S. 0. Smith; C. Winkel; J.
Lugtenburg; R. G. Griffin, "Solid-state C-13 NMR-study of tyrosine
protonation in dark-adapted bacteriorhodopsin." Biochemistry 1990, 29,
(23), 5567-5574.

G. S. Harbison; S. O. Smith; D. P. Raleigh; J. E. Roberts; J. A. Pardoen; S.
K. D. Gupta; J. Lugtenburg; R. A. Mathies; J. Herzfeld; R. G. Griffin,
"High-resolution solid-state NMR-studies of bacteriorhodopsin."
Biophysical Journal 1986, 49, (2), A180-A180.

L. Ren; C. H. Martin; K. J. Wise; N. B. Gillespie; H. Luecke; J. K. Lanyi;
J. L. Spudich; R. R. Birge, "Molecular mechanism of spectral tuning in
sensory rhodopsin II." Biochemistry 2001, 40, (46), 13906-13914.

C. Yuan; O. Kuwata; J. Liang; S. Misra; S. P. Balashov; T. G. Ebrey,
"Chloride binding regulates the Schiff base pKa in gecko P521 cone-type
visual pigment." Biochemistry 1999, 38, 4649-4654.

340

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

M. Nakagawa; T. lwasa; S. Kikkawa; M. Tsuda; T. G. Ebrey, "How
vertebrate and invertebrate visual pigments differ in their mechanism of

photoactivation." Proceedings of the National Academy of Sciences of the
United States of America 1999, 96, 6189—6192.

S. Nishimur; H. Kandori; M. Nakagawa; M. Tsuda; A. Maeda, "Structural
dynamics of water and the peptide backbone around the Schiff base
associated with the light-activated process of octopus rhodopsin."
Biochemistry 1997, 36, 864-870.

P. E. Blatz; J. H. Mohler; H. V. Navangul, "Anion-induced wavelength
regulation of absorption maxima of Schiff-bases of retinal." Biochemistry
1972, I I , 848-856.

M. Sigh; H. R. Cama, "Enzymatic cleavage of carotenoids." Biochimia Et
a. Biophysics Acta 1974, 370, 49-61.

R. C. Mordi; J. C. Walton; G. W. Burton; L. Hughes; K. U. Ingold; D. A.
Lindsay: D. J. Moffatt, "Oxidative degradation of B-carotene and B-apo-8'-
carotenal." Tetrahedron 1993, 49, 911-928.

A. C. Ross, "Cellular metabolism and activation of retinoids: mics of
cellular retinoid-binding proteins." FASEB Journal 1993, 7, 317-327.

T. R. Evans; S. B. Kaye, "Retinoids: present role and future potential."
British Journal of Cancer 1999, 80, 1-8.

8. A. Kliewer; J. M. Lehmann; T. M. Willson, "Orphan nuclear receptors:
shifting endocrinology into reverse." Science 1999, 284, 757-760.

G. M. Morris-Kay; S. J. Ward, "Retinoids and mammalian development."
International Reviews in Cytology 1999, 188, 73-131.

D. Koch; W. Gartner, "Steric hindrance between chromophore substituents
as the driving force of rhodopsin photoisomerization: 10-methyl-13-
demethyl retinal containing rhodopsin." Photochemistry and Photobiology
1997, 65, 181-186.

341

 

38.

39.

40.

41.

42.

43.

45.

46.

47.

G. G. Kochendoerfer; P. J. E. Verdegem; I. v. d. Hoef; J. Lugtenburg; R.
A. Mathies, "Retinal analog study of the role of steric interactions in the

excited state isomerization dynamics of rhodopsin." Biochemistry 1996,
35, 16230-16240.

R. S. H. Liu; A. E. Asato; M. Denny; D. Mead, "The nature of restrictions
in the binding site of rhodopsin. A model study." Journal of the American
Chemical Society 1984, 106, 8298-8300.

J. Uenishi; R. Kawahma; O. Yonemitsu; A. Wada; M. Ito,
"Stereocontrolled synthesis of (112)-retinal and its analogues."
Angewandte Chemie International Edition 1998, 37, (3), 320-323.

B. Borhan; M. L. Souto; J. M. Um; B. Zhou; K. Nakanishi, "Efficient
synthesis of ll-cis-retinoids." Chemical European Journal 1999, 5, (4),
1172-1175.

A. W. Gibson; G. R. Humphrey; D. J. Kennedy; H. Stanley; 8. H. B.
Wright, "Diethyl (3-Trimethylsilyl-2-propynyl)phosphonate, a new
reagent for the preparation of terminal conjugated enynes." Synthesis
1991, 414-416.

F. Sato; Y. Kobayashi, "Hydromagnesiation reaction of propargylic
alcohols: (E)-3-pentyl-2-nonene-l,4—dioi from 2-octyn-1-ol." Organic
Synthesis 1990, 69, 106.

W. Boland; N. Schroer; C. Sieler; M. Feigel, Helvetica Chimia Acta 1987,
70, 1025-1046.

Journal of the Chemical Society 1952, 1094-1111.

P. E. Blatz; J. H. Mohler: W. Ahmed, "Spectroscopic observation of
solvent Interaction with selected retinal Schiff-bases." Photochemistry and
Photobiology 1991, 54, 255-264.

G. A. J. Pitt; F. D. Collins; R. A. Morton; P. Stok, "Rhodopsin. VIII. N-
Retinylidenemethyiarrrine, an indicator yellow analog." Biochemical
Journal 1955, 59, 122-128.

342

48.

49.

50.

51.

52.

53.

54.

55.

56.

P. E. Blatz; N. Baumgartner; B. Balasubramaniyan; P. Balasubramaniyan;
F. Stedman, "Wavelength regulation in visual pigment chromophore.
Large induced bathochromic shifts in retinol and related polyenes." '
Photochemistry and Photobiology 1971, 14, 531-549.

M. D. Distefano; H. Kuang; D. F. Qi; A. Mazhary, "The design of protein-
based catalysis using semisynthetic methods." Current Opinion in
Structure Biology 1998, 8, 459-465.

H. Kuang; D. Haring; D. F. Qi; A. Mazhary; M. D. Distefano, "Synthesis
of a cationic pyridoxamine conjugation ragent and application to the

mechanistic analysis of an artificial transaminase." Bioorganic Medicinal
Chemical Letters 2000, 10, 2091-2095.

H. Kuang; M. D. Distefano, "Catalytic enantioselective reductive

amination in a host-guest system based on a protein cavity." Journal of the
American Chemical Society 1998, 120, 1072-1073.

R. R. Davies; M. D. Distefano, "A semisynthetic metalloenzyme based on
a protein cavity that catalyzes the enantioselective hydrolysis of ester and
amide substrates." Journal of the American Chemical Society 1997, 119,
1 1643-1 1652.

R. Y. Tsien, "The green fluorescent protein." Annuals Reviews in
Biochemistry 1998, 67, 509-544.

J. M. Tavare; L. M. Fletcher; G. I. Welsch, "Review-Using green
fluorescent protein to study intracelullar signalling." Journal of
Endocrinology 2001, I 70, 297-306.

N. Billinton; A. W. Knight, "Seeing the wood through the trees: A review
of techniques for distinguishing green fluorescent protein from
endogenous autofluorescence." Analytical Biochemistry 2001, 291, 175-
197.

P. Roessel; A. H. Brand, "Imaging into the futurezvisualizing gene
expression and protein interactions with fluorescent proteins." Nature
Cellular Biology 2002, 4, E15-E20.

343

57.

58.

59.

61.

62.

63.

65.

66.

A. Chiesa; E. Rapizzi; V. Tosello; P. Pinion; M. d. Virgilio; K. E. Fogarty;
R. Rizzuto, "Recombinant aquorin and green fluorescent protein as
valuable tools in the study of cell signalling." Biochemical Journal 2001,
355. 1-12.

B. P. Cormack; R. H. Valdivia; S. Falkow, "FACS-optimized mutants of
the green fluorescent protein (GFP)." Gene 1996, 173, 33-38.

G. J. Phillips, "Green fluorescent protein-a bright idea for the study of
bacterial protein localization." FEMS Microbiology Letters 2001, 204, 9-
18.

R. Heim; D. D. Prasher; R. Y. Tsien, "Wavelenght mutations and
postranslational autoxidation of green fluorescent protein." Proceedings of
the National Academy of Sciences of the United States of America 1994,
91, 11984-11989.

G. S. Baird; D. A. Zacharias; R. Y. Tsien, "Biochemistry, mutagenesis ,
and oligomerization of DsRed, a red fluorescent protein from coral."
Proceedings of the National Academy of Sciences of the United States of
America 2000, 97, 11984-11989.

C. T. Baumann; C. S. Lim; G. L. Hager, "Simultaneous visualization of
the yellow and green forms of the green fluorescent protein in living
cells." Journal of Histochemistry and Cytochemistry 1998, 46, 1073-1076.

M. Chalfie; Y. Tu; G. Euskirchen; W. W. Ward; D. C. Prasher, "Green
fluorescent protein as a marker for gene-expression." Science 1994, 263,
802-805.

R. Hein; R. Y. Tsien, "Engineering green fluorescent protein for improved
brightness, longer wavelengths and fluorescence resonance energy
transfer." Current Biology 1996, 6, (178-182).

A. Mishra; R. K. Behera; P. K. Behera; B. K. Mishra; G. B. Behera,
"Cyanines during the 1990s: A Review." Chemical Reviews 2000, 100,
1973-2001.

L. Viteva; e. al, "Organimetallics in cyanine chemistry-synthesis,
reactivity and photophysical properties for some heptamethine

344

67.

68.

69.

70.

71.

72.

73.

74.

merocyanine dyes." European Journal of Organic Chemistry 2004, 385-
394.

R. S. H. Liu; A. E. Asato, "Tuning the color and excited state properties of
the azulenic chromophore: NIR absorbing pigments and materials."
Journal of Photochemistry and Photobiology, C: Photochemistry Reviews
2003, 4, (3), 179-194.

R. S. H. Liu; R. S. Muthyala; X.-s. Wang; A. E. Asato; P. Wang; C. Ye,
"Correlation of substituent effects and energy levels of the two lowest
excited states of the azulenic chromophore." Organic Letters 2000, 2, (3),
269-271.

S. Ito; A. Nomura; N. Morita; C. Kabuto; H. Kobayashi; S. Maejima; K.
Fujimori; M. Yasunami, "Synthesis and two-electron redox behavior of
diazuleno[ 2,1-a1,2-c]naphthalenes." Journal of Organic Chemistry 2002,
67, 7295-7302.

R. Muthyala; D. Watanabe; A. E. Asato; R. S. H. Liu, "The nature of the
delocalized cations in azulenic bacteriorhodopsin analogs."
Photochemistry and Photobiology 2001, 76, (6), 837-845.

H. Dorothee; S. Siegfried; G. Wolfgang; B. Volker; M. Hans-Dieter,
"Merocyanines as extremely bathochromically absorbing chromophores in
the halobacteriai membrane protein bacteriorhodopsin." Angewandte
Chemie International Edition 1997, 36, (15), 1630-1633.

L. C. Wang; Y. Li; H. G. Yan, "Structure-function relationships of cellular
retinoic acid-binding proteins-Quantitative analysis of the ligand binding
properties of the wild-type proteins and site-directed mutants." Journal of
Biological Chemistry 1997, 272, 1541-1547.

A. W. Norris; L. Cheng; V. Giguere; M. Rosenberger", E. Li,
"Measurement of subnanomolar retinoic acid-binding affinities for cellular
retinoic acid-binding proteins by fluorometric titration." Biochimia
Biophysics Acta-Protein Structure and Molecular Enzymology 1994, 1209,
10-18.

M. J. Mintz; C. Walling, "t-Butyl hypochlorite." Organic Synthesis
Collection 5, 184-187.

345

75.

76.

77.

78.

F. Lambertin; M. Wende; M. J. Quirin; M. Taran; B. Delmond, "New
retinoid analogs from y—pyronene a natural synthon." European Journal of
Organic Chemistry 1999, 1489-1494.

D. C. Lee; J. A. Hayward; C. J. Restall; D. Chapman, "2nd-Derivative
infrared spectroscopic studies of the secondary structures of
bacteriorhodopsin and Ca-2+-ATPase." Biochemistry 1985, 24, (16),
4364-4373.

A. C. Sehgal; R. Tompson; J. Cavanagh; R. M. Kelly, "Structural and

catalytic response to temperature and cosolvents of carboxylesterase ESTI
from the extremely therrnoacidophilic archaeon Sulfolobus solfataricus
Pl." Biotechnology and Bioengineering 2002, 80, (7), 784-793.

M. P. Horvath; R. A. Copeland; M. W. Makinen, "The second derivative
electronic absorption spectrum of cytochrome c oxidase in the Soret
region." Bi0physical Journal 1999, 77, (3), 1694-1711.

346

Chapter 7

Attempts to develop an assay to identify rhodopsin surrogates

7.1 Introduction

As discussed in Chapter 4 development of a rhodopsin mimic that formed
a protonated Schiff Base with all-trans-retinal was accomplished. In order to re-
engineer CRABPII into a rhodopsin protein mimic, an active site Lys residue was
necessary for the formation of a Schiff base. In silica mutagenesisl'2 and
minimizations based on the published crystal structure of Cellular Retinoic Acid
Binding Protein H (CRABPII) with retinoic acid led to position 132 as the ideal
choice for the installation of the Lys residue. As discussed in detail by Rachel M.
Crist (Michigan State University, Thesis). The use of rational design has provided
a number of good protein mimics.
However, the rate at which
successful mutants were obtained
was time-consuming. For example,
in addition to incorporating Lys at

position 132, which provided the

 

A
Figure 7-1. Different positions to possibly
introduction Of a Lys at position 134 insert the nucleophilic lysine CRABPII wild

best rhodopsin surrogates.

seems promising given that type 0111132, 0Y134 and oT54.

347

this residue is the closest to the retinal W
\ \ \ \ ‘0
carbonyl carbon. Also Lys at position 134 Retinal

Amax:380nm
will approach the carbonyl in a different
tra'ecto than L s at sition 132. L s 134
J 0’ Y P0 Y \ \ \ \ \fim
. . 1'1
approaches 1n a more head-on trajectory. ,
Protonated Shift base of
. . . retinal with but lamine
Thus, 1t 18 possrble that at least a few Max: “gym

different positions could be potentially good

. Figure 7-2. Retinal A“, a: 360 in
to posrtion the nucleophilic lysine. Also, in

ethanol, yellow. Protonated Schiff
the progress of this project it was observed

base of retinal with butyiamine has a
that a counter anion was important to km= 440nm ln EtOH.

stabilize the PSB, and the position Glu121

Therefore, we would like to use random mutagenesis3's to increase the number of
different mutants. In more detail, error-prone PCR is a random mutagenesis
technique for introducing amino acid changes into proteins. Mutations are
deliberately introduced during PCR through the use of error-prone DNA

6 Randomized DNA sequences are cloned

polymerases and reaction conditions.
into expression vectors and a large mutant library is obtained. When random
mutagenesis is used a vast number of different mutants can be obtained. Because
of the generation of such large number of different mutants with this technique,
the necessity of an assay to efficiently identify the desired mutants (in our case,
mutants that form a protonated Schiff base vs. the mutants that do not) is essential.

As a result, we attempted to develop an assay for the identification of CRABPII

mutants that form a protonated Schiff base with retinal. The assay design exploits

348

the difference in UV-vis absorptions of retinal and the protonated Schiff base of
retinal. The wavelength absorption for retinal in ethanol is 360 nm, and the

protonated Schiff base is > 440 nm (Figure 7-2).7

7.2 In vitro test by using pORANGE, pBCDOX-l and pCRABPII

As previously discussed in Chapter 2, we have in hand a system that can
produce retinal in E. coli. This system consists a of XL1-Blue host that contains a
plasmid producing B-carotene (pORANGE)8'9 and a plasmid that can oxidize 8-

carotene to retinal (pBCDOX-l)10 (Figure 7-3). When no IPTG is added the

 

Figure 7-3. pORANGE plasmid contains a gene cluster with crtE, crtB crtl
and crtY. These genes produce B-carotene from dimefl'lyl allyl
perphosphate and isopentyl pyrophosphate. pBCDOX-1 is a vector
contains a mice gene for B-carotene-15-15’ dioxygenase cloned in a

pQE30.

349

phenotype of this bacteria is an orange color. When the BCDOX is expressed
(addition of IPTG, see details in Chapter 2) the phenotype is yellow due to the
conversion of B-carotene to retinal, (Figure 7-4). Thus, a host that contains a
pORANGE, pBCDOX and pCRABPlI could be used as a system to identify good
mutants (Figure 7-5). When retinal is produced in the presence of a mutant that
forms a protonated Schiff base a shift in the color would be observed, from yellow
(380 nm) to red (> 440 nm). Thus, by the difference in colors (yellow vs. red) the
mutants that are successful in forming a protonated Schiff base could be identified

(Figure 7-6).

, z, \ V\ \x \ \i/VW
l
K/’\ li-Carotene

Figure 7-4. Production of retinal in viva. 1. XL1-Blue

 

transformed with pORANGE and pBCDOX-t with out
induction (addition of lPTG). Orange color due to formation
of B-carotene 2. XL1-Blue transformed with pORANGE and
pBCDOX-1 after addition of lPTG. The yellow coloration is

due to oxidation of B-carotene to form retinal.

350

 

 

 

\\\\\\\\\
l

! i
\A [l-Caroiene

I 15-15' BCDOX

Formation of a protonated
shlff base between CRABPII
mutants and retinal

G
\ \ \ \ ‘N
Retinal

Figure 7-5. Production of retinal in vivo could be accomplished by cloning

 

pORANGE and pBCDOX-2 in a host E. coil. When the host contains a
plasmid that expressed a mutant that forms a Schiff base a change in the

color should be observed.

351

multitude-notionnei’ss
O O WMdoabfl'haPSB

Figure 7-8. The colonies that are yellow would
correspond to mutants that do not form a
protonated Schiff base. On the other hand the red

colonies should contain mutants that form a

protonated Schiff base with retinal.

The expression of CRABPII is under the control of a T7 promoter and
under the selection of Chloramphenicol. The E. coli system described in Chapter 2
uses XL1-Blue as the host. In the development of this assay we could not use this
host because the CRABPII gene expression is under the control of a T7 promoter.
XL1-Blue does not have the capacity of producing the T7 polymerase therefore
the CRABPH gene would not be expressed. Thus a different host, ERZS66 was
used, that contains a copy of the T7 polymerase gene in its genome (Figure 7-7).
After the E. coli was transformed with pORANGE, it was grown under red safe
light. The obtained colonies did not show the expected orange coloration due to

the production of B-carotene in the E. coli. The reason why the ER2566 E. coli

352

does not produce B-carotene is unclear. As

discussed previously when XL1-Blue host was

 

used the coloration of the cells indicated the

production of B-carotene, and a comparison of

Am!)

both genotypes (XL1-Blue and ER2566) does
not provide an obvious reason of why the Figure 7-7. CRABPII is aimed

production of B-carotene in ER2566 is so low "“0 3 957479 plasmid ""991 8 T7

_ polymerase control.

or non-exrstent.

This is not the first time that this kind
of results are obtained. Wurtzel and colleagues had observed that many un-
pigmented colonies where obtained among the different hosts.8 Also, they noticed
that no obvious correlation between genetic markers and carotenoid pigmentation
exist. But they did suggest that it is likely that unidentified markers affect pigment
accumulation. On the other hand, Sandmannll observed that when zeaxanthin
was produced uSing a plasmid that was under the control of a T7 promoter in
BL21(DE3) E coli, no expression was observed. But when a JM101 strain (that
contained a mGP1—2 plasmid that expressed T7 RNA polymerase) was used,
production of the carotenoid was observed.

Different groups have expressed carotenoids in a variety of E. coli and the
production of carotenoids was also variable (Table 7-1).8'9"2'18 In general it has

been observed that the presence of a high copy plasmid (puck) the production of

carotenoids decreases. The deprived production of carotenoid could be due to the

353

Table 7-1. Expression of carotenoid genes from
E. uredovora in different E. coli. strain. Lyeopene

(L), zeaxanthin (Z), B-carotene (BC).

 

 

E. coli strain Expression
TOP10F' High (L) (Z)
AB2463 High (L) (Z)
XL1-Blue High (BC)

JM101 High(BC)
JM109 Medium (BC)
H8101 Medium (L) (Z)
SOLR Medium (L) (Z)
V73 Medium (L), high (2)
A81884 Medium (L) (2)
AB2480 Medium (L) (2)
A81886 Medium (L) (Z)
SK2267 Medium/unstable (L) (Z)
V71 Medium/unstable (L) (2)
KL168 Medium/ very unstable (L) (Z)
SK3451 Medium/ very unstable
Y1088 Low (L) (2)
SURE Low (L) (Z)
BL21(DE3) Low (BC)

 

, fact that a high-COpy number plasmid mediates the synthesis of large amounts of

B-iactamase, thus it might exhaust the E. coli metabolism in a way that negatively

affects the synthesis of carotenoids.

In vivo test for protonated PBS formation using pORANGE pRET-

CRABPII

It seems evident that to develop this assay, a gene that is not under the
control of a T7 promoter would be optimal for the expression of B—carotene. The
new strategy consisted of cloning the CRABPII gene next to the BCDOX gene

(pBCDOX-l). As discussed in Chapter 2 the pBCDOX plasmid is under the

354

control of a T5 promoter, and expression B-carotene and BCDOX were performed
using XL1-Blue E. coli. After the CRABPII mutant was cloned, the protein
expression would also be under the control of T5 promoter allowing the use of
XL1-Blue E. coli as the host. To prove the validly of this assay the CRABPII-
tetra mutant, R132K::Y134F::R111L::L121E, (this was the best mutant developed
so far, for details refer to Rachel Crist’s Thesis) was chosen to be cloned next to
the BCDOX. As a control experiment the CRABPII wild type was to be cloned
next to BCDOX.

A comparison of the spectroscopic data of the KLFE and WT is shown in
Figure 7-8. The UV-vis spectra of retinal incubated with the CRABPII-tetra
mutant shows maximal absorbances at 378 nm and 438 nm, a 61 nm red shift as
compared to the CRABPH wild type at 377 nm. The peak at 438 nm suggests the
formation of the protonated Schiff base. We decided that initially we would
prefer to express both proteins (BCODX and CRABPII) in their native form, and
not as fused proteins. The reasoning behind this was that BCDOX is a protein
that tends to aggregate (see Chapter 2 for details). Therefore addition of
CRABPII as a fusion protein would increase the possibility of the expression of an
insoluble protein. When the primers to clone CRABPII were designed, a stop
codon after the BCDOX gene was added, and a different ribosome-binding site for
CRABPII was introduced. The same RBS that was present in the previous
plasmids 8 bp far away from the starting ATG was used. The same RBS was used
for the CRABPH-wild type and CRABPH-tetra mutant . The CRABPII-wild type

and CRABPII-tetra mutant were PCR out from the original pET-17b vector. The

355

Relative lnbnslly

 

 

 

 

 

 

 

 

1 1m
0.81
0.6-
0.4- A
02-
0 r r r l '
o 1 2 3 4 0 l ’ ' ' '
300 350 400 450 500 550
Equivalents Retinoic Acid! Retinal nm
C D
E 1 0.03 ~
0
f . ...l
5 0.8 .
g 0.01 -
E
g 0.6 r 1 7 r o ' —j i r 1
0 2 4 6 3 10 300 350 400 450 500 550
Equivalents Retinoic Aeldl Retinal nm

Figure 7-8.. A. Fluorescence titration curves for the retinal. K566003860 nM.; B. UV-vls
of retinal incubated with wild-type CRABPII portrays a maximal absorbance at 377 nm. C
Fluorescence titration curves for the retinoic acid and retinal Kg determination with the tetra
mutant CRABPII R132K::Y134F::R1 11L::L121 E. The retinal K¢=224 re 18 nM. D. UV-vis of
retinal incubated with the CRABPII protein shows maximal absorbances at 378 nm and 438
nm, an 61 nm red shift as compared to the wild type CRABPII (377 nm), suggesting

protonated Schiff base formation.

356

primers introduced KpnI and Sail restriction sites (Primer 1 (KpnI), 5'-
CCCTCTAGAAATAATITI‘ GGGTACCTTAAGAAGG-3' and Primer 2 (Sall) 5 ’-
GTGATGGATATCI‘GCAGTCGACTCACTCTC-3’). The cloning plan consisted of three
steps: PCR, followed by digestion of the genes and plasmid, and finally, ligation
of the gene into the desired vector (pBCDOX-l).

The PCR was performed 'uneventfully and provided the RBS-CRABPII-wt
and RBS-CRABPII-tetra mutant gene with the corresponding Kpnl and SalI cut
sites. A sequential digestion (with KpnI and Sall) of the plasmid pBCDOX-l and
the PCR products (RBS-CRABPII and RBS-CRABPII-tetra mutant) were
performed. According to the manufacturer a sequential digestion is
recommended. The first enzyme used was SalI because it requires the highest
number of bp next to the cut site. After digestion, purification (using QIAquick
gel extraction kit) of the correspondent vector and genes was performed. To
avoid re-ligation, the digested vector and genes were treated with Calf Intestine
Phosphatase. The DNAs were quantified and the ligations were performed at
different molar ratios. The ligation products were transformed in XL1-Blue. A
few colonies were obtained, and DNA was isolated to identify the cloned
products. After electrophoretic analysis of the plasmid DNA confirmed the
existence of a new gene in CRABPII, the isolated plasmids were sent for
sequencing. The sequences of both plasmids indeed confirmed the presence of
BCDOX and CRABPII genes in the new vector (Figure 7-9). The orange cells
were prepared by transformation of E. coli XL1-Blue with pORANGE (cells that

produce B-carotene), followed by preparation of the competent cells. These

357

 

Figure 7-9. The CRABPII genes were cloned into pBCDOX-1. The
plasmid pRTE-CRABPlI-1 contains the mice BCDOX in frame with the
CRABPII wild type. pRTE-CRABPll-Z contains the mice BCDOX in
frame with the CRABPII R132K::Y134F::R111L::L121 E. Both plasmids

are under the control of T5 promoter.

competent cells were transformed with pRET-CRABPII-l and pRET-CRABPII—2.
Digestion of the isolated plasmids was performed to confirm the existence of

pRET-CRABPII vectors and pORANGE (Figure 7-10).

 

 

 

 

Figure 7-10. Transformation of pRET-CRABP1 or 2 in to the E.

coli host that produced B-carotene (XL1-Blue-pORANGE).

358

The cells were grown at 37 °C for about three hours under red safe light,
and the production of retinal was induced by addition of IPTG. The results
(Table 7 .2) showed no obvious difference in the coloration between the cells that
contained pBCDOXl and CRABPII-wt and the cells that contained pBCDOXl-
CRABPII-tetra mutant. The cells that contained the pORANGE plasmid did
produce the orange coloration due to the presence of B-carotene. When the

BCDOX and CRABPII wild type were induced the expected shift in color is

Table 7.2. The developed assay to identify CRABPII mutants that loan PSB
with retinal did not show the expected results. XL1-Blue was transformed with
pORANGE and pRET-CRABPll-1 or pORANGE and pREr-CRABPll-z. The
orange coloration was due to the accumulation of B-carotene (pORANGE). The
yellow coloration is due to the presence of retinal and but with out the iorrnation

of a PBS. The coloration between the wild type and the tetra mutant was the

 

 

same.
Expression system IPTG Color
1 XL1-Blue/pORANGE 8 orange
2 XL1-Blue/ pORANGE /pRET-CRABPii-1 / yellow
3 XL1-Blue/ pORANGE /pRET-CRABPli-1 1/ yellow
4 XL1-Blue/pORANGE X orange
5 XL1-Blue/ pORANGE /pRET-CRABPli-2 ~/ yellow
6 XL1-BIue/pORANGE lpRET-CRABP 1 yellow

 

 

359

observed (a change from orange to yellow). The yellow coloration was due to
the formation of retinal not binding as a PSB. However, when the BCDOX and
CRABPII-tetra mutant were induced, the expected red shift coloration was not
observed. Instead a yellowish color was obtained. Thus no obvious difference in
the coloration between the cells that contained pRET-CRABPII-l and pRET-
CRABPII-2 was found.

Since BCDOX is a very sensitive enzyme an could become inactive very
easily, extraction of the retinoids was performed. Analysis of the extracted
retinoids and carotenoids verified the presence of retinal and B-carotene. The
presence of retinal confirms the activity of BCDOX. Due to the fact that the T5
promoter is such a weak promoter, it is possible that the CRABPII protein was
not expressed. The strength at which a protein is expressed is proportional to
how far the promoter is located form the gene’s RBS. In this case the second
RBS for CRABPII is about 1.7 KDa bp from the promoter, therefore it is
reasonable to believe that probably CRABP was not expressed as expected. To
test whether the CRABPH mutants were expressed, a large-scale expression of
the XL1-Blue pORANGE-lpRBT-CRABPII-l or p0RANGE-lpRET-CRABPII-
2 was performed. An SDS gel showed no over expressed protein at about 15
KDa (the CRABPH approximate molecular weight). This could be explained
due to the fact that T5 promoter is not as strong as a T7 promoter, thus its
expression could be low. Next, an attempt to isolate the expressed CRABPII
using Q SepharoseTM was performed. CRABPII has a very high affinity to Q

SepharoseTM Fast Flow resin, but no binding to the resin was observed. This was

360

probably due to the fact that if CRABPII was expressed it was in very low
concentrations. To improve the sensitivity of this test, isolation using the FPLC
was performed. Conditions to selectively isolate CRABPII mutants using the
FPLC have been previously establish by Chryssoula. The FPLC elution showed
a protein eluting about the same time as CRABPII. However when an
acrylamide gel of the eluted protein was performed, only a high molecular
protein was identified around 66 KDa. Some times in the purification process of
CRABPII mutants some times a protein is eluted at the same time as the
expected CRABPH, thus it could be possible that the eluted protein was not
CRABPII, instead it might have been just a contaminant. Even though the T5
promoter is a weak promoter, in bacteria, the expression of several genes under
the control of a single promoter is very common in the so-called clusters, thus it
is a little puzzling why CRABPII is not produced. The lack of difference in
color between the CRABPII-native plasmid and the CRABPII-mutant could also
be due to the fact that retinoids, as well as carotenoids are stored in the
membrane (hydrophobic regions of the cell), and therefore the chromophore does
not have the opportunity of binding to CRABPII protein. It is important to keep
in mind that as shown in Figure 7-7 there is an equilibrium between the PSB
(440 nm) and the non protonated form. Thus, it could be that the amount of PSB
is not enough to cause a high number of CRABPII mutants that form a
protonated Schiff base, therefore the coloration caused by these mutants would
be almost negligible. It is clear, however, that the use of a pQE30 plasmid to

express BCDOX and CRABPII does not provide the desired results.

361

Unfortunately, as previously discussed, the expression of B-carotene in E. coli
cells that produce the T7 polymerase was impossible. Only in XL1-Blue or
JM109 E. coli hosts the B-carotene was accumulated in significant amounts.
Therefore, we considered inserting the T7 polymerase gene into XL1-Blue
genome. But the commercially available systems to infect cells with T7
polymerase cannot be used in XL1-Blue hosts. Bacteriophage CE6 can be used
to provide a source of T7 RNA polymerase to susceptible host cells carrying pET
recombinant plasmids.‘9 But usually after infection with the bacteriophage the
cells go to lysogenic growth, thus are destroyed in about 3-4 hours. In order to
avoid the lysogenic growth the E. coli should express supF and the XL1-Blue
strains doesn’t express supF, and neither does JM109. As a result, this
experiment cannot 'be used as an assay for the screening of mutants that form

protonated Schiff base with retinal.

362

7.4 Materials and Methods.

A. Plasmid Purification
Bacteria form a single colony was grown (500 mL, LB, containing
appropriate antibiotics) and purified with via Qiagenm column purification. Upon
completion of the QiagenO purification, the recovered DNA was resuspended in
(400 1.1L) sterile water, and analyzed by UV-visible for concentration
determination and purity. The concentration of DNA obtained is ug/uL
Concentration of DNA sample (11g / ltL) =
Absm x (50 ug/ 1000 11.1.) x (volume of DNA used / total volume UV
sample)
Purity of DNA sample : Abszso/ Absm
= 1.8, pure
>1.8, RNA contamination

< 1.8, protein contamination

B. Sample preparation for sequencing DNA

In a sterilized eppendorf tube (0.5 mL), DNA (2 ug) and primer (30 pmol)
were mixed. When sequencing is performed, the largest amount of base pairs that
are sequenced with accuracy is about 400 bp. Therefore, to sequence the
complete gene, five different primers were used

1. 5 '-GA TCTCGA TCCCGCGAAATI‘AATA CGAC-3 '

2. 5 ’-CCAGACCCTAGAGACCITGGAGAAGG-3 ’

363

5 '-CGAGGAGAAGTCCAGGCTGACC-3'

5’-GATCGA TCTCGATCCCGCG-3'

5 '-GCAGACTGGAA TGCAGTGAA GC-3 '

9‘99?”

5 '-GCA GACTGGAA TGCAGTGAA 603’

C. Melting temperature calculation for primers
Primers should ideally have melting points 2 78 °C, and end in at least
one, if not more, GC base pair
Tm = 81.5 °C + (0.41)(% GC) — 675 / N - % Mismatch

where N = primer length

Optimal PCR conditions. A small scale gradient PCR was performed to optimize

the extension temperature.

 

 

 

 

 

PCR recipe
Template DNA 100 ng
Primer 1 1 mM
Primer 2 1 mM
DNTP 200 W
Deep vent 2 U
10 x deep vent buffer 5 11L (1x)
M9304 5 MM
H20 SO'RXH BL
PCR mm
1 X 94 °C 5 min
94 °C 1 min
30 X 4‘: 48 °C 3 min
72 °C 3 min
1 X 72 °C 10 min
1 X 25 °C 10 min

364

D. PCR Primers

Primers Sequence Pen Template

pRET-CRABPII- S'CCCTCTAGAAATAATTTTGGGTACCTTAAGAAGGB' ETl 7-CFlAPil-WT
1 5' -5'-GTGATGGATATCTGCAGTCGACTCACTCTC-3' p

pRET-CRABPII- S'CCCTCTAGAAATAA TTTTGGGTACCTTAAGAAGG-

2 sore/17664771rcromorcencrmcrcro-a- ”3‘7'CRAP"'KFLE

E. Purification of DNA from agarose gel

GENCLEAN Turbo Qiagenm Protocol: The desired bad was excised from
the agarose gel. The gel was cut of in smaller pieces and transferred into an
eppendorf tube. Then a GENCLEAN Turbo solution (100 11L per 100 ug of
agarose) was added and the mixture was melted at 55 °C for 5 min. The melted
solution was transferred into GENCLEAN Turbo cartridge. The filter was spined
for 5 seconds. The filter was washed twice by addition of GENCLEAN Turbo
Wash (500 11L). The DNA was eluted using GENCLEAN Turbo solution (30

al.). The DNA obtained was used directly in the enzyme digestions.

F. Digestions and ligation reactions

As mention in previously the double digestion did not afford the desired
product therefore, subsequent single digestions were performed. Same protocols
were followed for the gene (PCR product) and the pET29b(+). The first digestion
was performed using Sal] followed by purification using GENECLEAN. The

second digestion was performed using Kpnl.

365

F.l Digestions

 

 

 

 

 

 

 

 

 

Digestions
CRABPII pQE30-BCDOX
Kpnl 1 11L - 1 pl. -
Sail - 1 pl. - 1 :11.
DNA 30 at 30 at 30 pl. 30 pt 30 uL
Bufferi or2(10X) 1 X 1X 1X 1X
“20 6 11L 6 at 6 11L 6 11L
0 Incubation at 37 °C for 2 hours
F.2 Calf Intestine Phosphatase treatment
CIP treatment
Digested Digested
pBCDOX1 CRABMne
DNA obtained from DNA purification 30 ill- 30 pl.
CIP 1 pl 1U 10
Butler (1 OX) 1 X 1 X
H20 4 pr. 4 pl.

 

o Incubation at 37 °C for 2 hours
0 The DNA was purified following the same procedure and

quantified before the ligations were performed

 

 

 

F.3 Ligations
Ligations

1/10
Gene 4.5x10"3moles
Vector 4.5x10’“moles
T4 ligaset pi 1U
Buffer (10X) 1 X
H30

 

0 Gene 1: CRABPII-WT + KpnI and Sal!
0 Vector :pET30-BCDOX + KpnI and Sal!
o Incubation at 16 °C for 24 hours

0 JM109 E. coli were transformed with 5 11L of the ligation reactions

366

G. Typical preparation of competent cells

The E. coli strain of interest was grown (37 °C until ODeoo of 0.4-0.6, the
media contained LB, the corresponding antibiotic). After about three h The cells
were harvested by centrifugation at 5,000 RPM for 10 min at 4 °C. The cells were
re-suspended (100 mL of 0.9% NaCl) and then centrifugated at 5,000 RPM for 10
min at 4 °C. The cells were re-suspended (50 mL of 100 mM CaCl2) and
incubated for 30 min at 0 °C. The cells were centrifuged at 5,000 RPM for 10
min at 4 °C, and the cells were re-suspended (4 mL of 100 mM CaCl2, 15%
glycerol). The cells in suspension were aliquoted (0.1 mL) and were quickly

frozen with liquid nitrogen.

H. Typical transformation of competent cells

Sterile conditions were maintained throughout the entire protocol. The
competent cells were incubated at 0 °C for 5 min and the plasmid DNA (10
ng/uL) was added. The mixture was incubated for 30 min and the cells were heat
shocked at 42 °C for 2 min at 0 °C. LB (antibiotic if necessary) was added and
the cells were incubated at 37 °C for 1 h. Then the cells were transfered to a plate

containing LB/antibiotic and were grown overnight at 37 °C.

1. Expression of BCDOX and CRABPH in viva
The expression of BCDOX and CRABPII in E. coli was monitored in viva

by the colorimetric change of the oxidation of B-carotene (orange) to retinal

367

(yellow) to retinal-CRABPII red. In detail an E. coli XL1-Blue was transformed
with pORANGE (plasnud that contains the genes necessary for the synthesis of
carotene). The transformed cells were grown and they showed an orange
phenotype. These cells were converted into competent cells and then transformed
with the pRET-CRABPII—l or pRET-CRABPII-2 plasmid. These transformed
cells (Cells that contained both plasmids) were grown under LB/chloramphenicol,
carbenicillin and 1% glucose (5 mL), to an OD of 1.00. Then expression of
BCDOX and CRABPII was induced by addition of IPTG (1 mM final
concentration). This culture was grown for 6 more hours. The cells were spun
down at 10000 RPM and rinsed with water. Analysis of the phenotype of the cells
showed that no clear difference could be observed between the cells that

expressed the BCDOX and CRABPII to the ones that did not.

368

7.5

References

C. Papworth; J. C. Bauer; J. Braman; D. A. Wright, Strategies 1996, 9, (3),
3-4.

Stratagene, "QuikChange Site -Directed Mutagenesis Kit."

R. Fujii; M. Kitaoka: K. Hayashi, "One-step random mutagenesis by error
prone rolling circle amplification." Nucleic Acid Research 2004, 32, (19),
145-150.

D. W. Leung; B. Chen; D. W. Goeddei, "A method for random
mutagenesis of a defined DBA segment using a modified polymerase
chain reaction." Techniques 1989, I, 11-15.

A. Parikh; F. P. Guengerich, "Random mutagenesis by whole-plasrrrid
PCR amplification." Biotechniques 1998, 24, (3), 428-431.

S. Shafkhani; R. A. Siege]; E. Ferrari; V. Schellenberger, "Generation of a
large iibraris of random mutants in Bacillus subtilis by PCR-based plasmid
multimerization." BioTechniques 1997, 23, 304-310.

Y. Gat; M.Sheves, "A mechanism for controlling the pKa of the retinal
protonated Schiff base in retinal proteins. A study with model
compounds." Journal of the American Chemical Society 1993, 115, 3772-
3773.

E. T. Wurtzel; G. Valdez; P. Matthews, "Variation in expression of
carotenoid genes in transformed Escherichia coli strains." Bioresearch
Journal 1997, 1.1-ll.

N. Misawa; Y. Stomi; K. Kondo; A. Yokoyama; S. Kajiwara; T. Saito; T.
Ohtani; W. Miki, "Structure and functional analysis of a marine bacterial
carotenoid biosynthesis gene cluster and astaxanthin biosynthetic pathway
proposed at the gene level." Journal of Bacteriology 1995, 1995, 6575-
6584.

369

10.

11.

12.

l3.

14.

15.

16.

17.

18.

J. von Lintig; K. Vogt, "Filling the gap in vitamin A research - Molecular
identification of an enzyme cleaving beta-carotene to retinal." Journal of
Biological Chemistry 2000, 275, (16), 11915-11920.

A. Ruther; N. Misawa; P. Bbger; G. Sandman, "Production of zeaxanthin
in Escherichia coli transformed with different carotenogenic plasmids."
Applied Microbiology and Biotechnology 1997, 48, 162-167.

G. Sandmann, "Carotenoid biosynthesis and biotechnological application."
Archives of Biochemistry and Biophysics 2001, 385 , 4-12.

P. D. Matthews; E. T. Wrtzel, "Metabolic engineering of carotenoid
accumulation in Escherichia coli by modulation of the isoprenoid
precursor pool with expression of deoxyxylulose phosphate synthase."
Applied Microbiology and Biotechnology 2000, 53, 396-400.

M. Albrecht; N. Misawa; G. Sandmann, "Metabolic engineering of the
terpenoid biosynthetic pathway of Escherichia coli for production of the
carotenoids carotene and xeaxanthin." Biotechnology Letters 1999, 21,
791-795.

M. R. Farmer; J. C. Liao, "Precursor balancing for metabolic engineering
of lycopene production in Escherichia coli." Biotechnology Progress

2001, 17, (1), 57-61.

C. Schmidt-Dannet; D. Umeno; F. Arnold, "Molecular breeding of
carotenoid biosynthetic pathways." Nature Biotechnology 2000, 18, 750-
753.

M. Albercht; S. T akachi; S. Steiger; Z. Wang; G. Sandman, "Novel
hydroxycarotenoids with improved antioxidative properties produced by
gene combination in Escherichia coli." Nature Biotechnology 2000, 18,
843-846.

C. Wang; M. Oh; J. Liao, "Directed evolution of metabolically engineered
Escherichia coli for carotenoid production." Biotechnology Progress 2000,
16, 922-926.

370

19. F. Miao; S. K. Drake; D. S. Kompaia, "Characterization of gene
expression in recombinant Escherichia coli cells infected with phage
lambda." Biotechnology Progress 1993, 9, 153-159.

371