.wfi .s. u, . _,
.fi .. unarmwfimfia .s
./ Gr“ ”My” WW. .mu/r... N n J...
. an; Mina. . ...,
.an.§.«.,..%; .n H... aw: V... . ,.

,..._._._.»£z.n...,wm s... «we... 2.: ,

V v Iv. .
a .m. w... my.”
, . 0 19*». .
H . xx. ..
I‘U «Inna.§r
J
.

aways” 2a.”? ..

:1)

« .l.a 9 v (A. o
In... v 1.: a f. n‘ a. Run
a A no:
Vt; #
‘I

..
llb
A;

. J r «L. ._)v.w11 ‘

2V5! 11.1031:
. um?“ 31$: i t

53;;
”"do
‘

, l

o
I
v

'H'

.n,‘

I u an
{1-..}-.. n 2 . ;
7,: ....u..3\..)&mi.n....XJhlll!
Wm: -..lr.r.. . tunic...”th

(lo

 

. I'll-‘0 ,

. III'I - I lo

3.4.: I .| (.ltlu:.fll‘!t..)lovl\.‘lll ‘.
-.0Tl OOIJUFII” !|
l l O “l'-’ I..l.“|o

 

‘ '.V.~‘
llflpl

tI‘P‘§U‘L 1‘

THESlS

3

lllllllllllllllllllllllllllllllllllllllllllllllllllllIll

1293 01690 0684

This is to certify that the

dissertation entitled

Biochemical and NMR Studies of
Human Cellular Retinoic Acid Binding Proteins

presented by
Lincong Wang

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Biochemistry

 

Hm 27%

Major protessor

a!

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY

Mlchlgan State
University

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINE return on or before date due.

 

DATE DUE MTE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/96 alClRC/DalaDuopGS—p.“

BIOCHEMICAL AND NMR STUDIES OF
HUMAN CELLULAR RETINOIC ACID BINDING PROTEINS

By

Lincong Wang

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1998

Copyright by
LINCONG WANG
1 998

BIOCHEMICAL AND NMR STUDIES OF
HUMAN CELLULAR RETINOIC ACID BINDING PROTEINS

By

Lincong Wang

ABSTRACT

The structure-fimction relationships of cellular retinoic acid binding proteins
(CRABPS) have been investigated by site-directed mutagenesis, biochemical and NMR
methods. The major results and conclusions are summarized as follows.

Electrostatic interactions have been suggested to be critical for binding of retinoic
acid (RA) by CRABPs. However, the roles of two conserved arginine residues (Argl ll
and Arg13l in type-I CRABP; Argl 11 and Argl32 in type-II CRABP) that interact with
the carboxyl group of RA have not been evaluated. A novel competitive binding assay
was developed for measuring the relative dissociation constants of the site-directed
mutants including the two mutants R1 11M and R132M. Binding studies of the two
mutants showed that Argl 11 and Argl32 contribute to the binding energy by ~2.2 and
1.2 kcal/mol, respectively.

The crystal structure of holo-CRABP suggests that RA cannot enter or exit the
deep binding pocket without major conformational changes in the protein. Dimerization
has been proposed to be the mechanism by which CRABPs open their binding pockets.
The solution structure of apo-CRABPII has been determined by NMR spectroscopy.
Although the apo solution structure was similar to the holo crystal structure, the ligand

entrance was greatly enlarged in the former and readily accessible to RA. The

enlargement was mainly due to a concerted conformational change in three structural
elements, namely the second helix, the BC-BD loop and the BE-BF loop. The ligand-
binding pocket of apo-CRABPII was rather dynamic. CRABPII is predominately
monomeric in solution. Although the formation of transient dimers could not be ruled
out, dimerization apparently is not a prerequisite for entry of retinoic acid into the binding
pocket of CRABPII.

Preliminary NMR analysis of holo-CRABPII showed that RA binds to CRABPII
in solution in the same manner as determined by crystallography. NMR studies indicated
that the ligand entrance of holo-CRABPII is much less flexible than that of apo-
CRABPII. The results taken together suggested that RA binding indeed induces major
changes in the conformation and dynamics of CRABPII.

The solution structure of R11 1M has been determined by NMR spectroscopy.
Although apo-R1 11M had a structure similar to that of wild-type apo-CRABPII, there
were significant conformational differences between the two proteins, mainly localized to
three segments, clustered around the ligand entrance more than 17 A from the site of
point mutation. Furthermore, the ligand-binding pocket of apo-R1 1 1M, especially the
ligand entrance, was much less flexible than that of apo-CRABPII. The results suggested

that Argl 11 play a critical role in determining the structure and dynamics of CRABPII.

To my family

ACKNOWLEDGMENTS

I would like to express my gratitude to my thesis advisor, Dr. Honggao Yan, for
his financial support and for providing me the unique opportunity to learn both protein
biochemistry and NMR. I have learned a lot from him and become a better researcher.

I would like to thank my guidance committee members Dr. David McConnell, Dr.
John McCraken, Dr. Robert Hausinger, Dr. John Wilson and Dr. William Wells for their
criticism and advice, which truly guided me through the tough journey of graduate
school.

I would like to thank Mr. Kermit Johnson for teaching me NMR. His kindness
and patience made doing NMR experiments much more enjoyable. Thanks also go to Dr.
Yue Li for teaching me molecular biology. Special thanks go to Dr. Robert Cukier of
Chemistry Department for his advice on my thesis. His comment, “You have summarized
a great deal of (complicated) wor ”, encourages me to go further.

In experimental aspect I would like to thank Dr. Yue Li for establishing the
expression systems for CRABPs and for generating the mutants, Dr. Frits Abildgaard at
the National Magnetic Resonance Facility at Madison, Wisconsin for recording some of
the triple resonance spectra of apo-CRABPII and Dr. Honggao Yan for acquiring the
NMR spectra of uniformly 15N-labeled apo-R11 1M.

Finally, I am grateful to my wife for her understanding and support.

vi

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... ix
LIST OF FIGURES ........................................................................................................ x
LIST OF ABBREVIATIONS ....................................................................................... xiv
CHAPTER 1
LITERATURE REVIEW .............................................................................................. 1
Part 1. Cellular retinoic acid binding protein-II (CRABPII) .................................... 1
Section 1. Biological functions ofCRABPII 1
Section 2. Biochemical characterizations of CRABPs ............................. 7
Section 3. Structural and dynamical characterizations of CRABPs ....... 13
References ............................................................................. 22

Part 2. Physical basis ofprotein NMR 28

Section 1. Theory of Fourier transform pulse NMR ............................... 28

Section 2. NMR experiments for protein studies .................................... 42

Section 3. Theories of NMR relaxation in liquid .................................... 47

References ............................................................................. 77

CHAPTER 2

STRUCTURE-FUNCTION RELATIONSHIPS OF CELLULAR RETINOIC ACID

BINDING PROTEINS ................................................................................................... 87

Abstract ................................................................................................................... 88

Experimental procedures ...................................................................................... . 89

Results ................................................................................................................... 90

Discussion ............................................................................................................... 91

References .............................................................................................................. 94

vii

CHAPTER 3
SOLUTION STRUCTURE OF TYPE-II HUMAN CELLULAR

RETINOIC ACID BINDING PROTEIN ....................................................................... 95
Introduction ............................................................................................................. 95
Experimental procedures ........................................................................................ 97
Results .................................................................................................................. 103
Discussion ............................................................................................................. 1 26
Conclusion ................................................................................... 1 36
References ............................................................................................................. 137

CHAPTER 4

SOLUTION STRUCTURE OF A MUTANT PROTEIN (R1 11M) OF HUMAN

CRABPII ...................................................................................................................... 141
Introduction ......................................................................................................... . 141
Experimental procedures ..................................................................................... 143
Results ................................................................................................................. 148
Discussion ............................................................................................................ 1 70
Conclusion .................................................................................. 1 85
References ............................................................................................................ 186

CHAPTER 5

NMR STUDIES OF HUMAN CRABPII IN COMPLEX WITH

ALL-TRANS—RETINOIC ACID ............................................................... 189
Introduction .......................................................................................................... l 89
Experimental procedures ...................................................................................... 191
Results and Discussion ........................................................................................ 193
Conclusion .................................................................................. 2 1 9
References ............................................................................................................ 220

viii

LIST OF TABLES

CHAPTER 2
Table 2.1. 1H Chemical shifts of the ring protons of the aromatic
residues of the wild-type CRABPII, R11 1M mutant, and
CHAPTER 3

Table 3.1. lH, 13C and 15N chemical shifis of apo-CRABPII at pH 7.3

Table 3.2. Restraint and structural statistics of apo-CRABPII ........................ 124
CHAPTER 4

Table 4.1. 'H and 15N chemical shifts of Apo-Rl 1 1M at pH 7.3
and 27 °C ............................................................................................ 155

Table 4.2. Restraint and structural statistics ofapo-Rl 1 1M 164
CHAPTER 5

Table 5.1. 1H and 15N chemical shifts of holo-CRABPII at pH 7.3
and 25 °C ......................................................................................... 194

Table 5.2. Summary of proton chemical shifts of CRABPII-bound RA
at pH 7.3 and 25 °C ............................................................................ 207

Table 5.3. Relative volumes of intramolecular NOEs of
CRBAPII-bound RA ......................................................................... 208

ix

CHAPTER 1

Figure 1.1.

Figure 1.2.

Figure 1.3.

Figure 1.4.

CHAPTER 2

Figure 2.1.

Figure 2.2.

Figure 2.3.

Figure 2.4.

Figure 2.5.

Figure 2.6.

CHAPTER 3

Figure 3.1.

LIST OF FIGURES

Chemical structures and the numbering conventions of
all-trans-retinol (vitamin A), all-trans—retinal and

all-trans-retinoic acid (RA) ................................................... 3

(A) The ligand binding pocket of human CRABPII.
(B) expanded view of the interactions between the carboxyl
group of RA and the side chains of Argl 11, Argl32 and

Tyrl34 ofthe protein 12
The backbone fold and the secondary structural elements of

Huma CRABPII in complex with all-trans-RA ............................ 16
The polypeptide backbone and side chains showing the

one—bond coupling constants through which the coherences

are transferred in double and triple resonance NMR
experiments 44
F luorometric titration of the wild type CRABPI (A) and

CRABPII (B) ....................................................................................... 90
Competitive binding assays for measuring the relative

dissociation constants of CRABPs ....................................................... 91
TOCSY spectrum of the aromatic protons of R132M at 32°

showing two sets of resonances for Trp-87 ......................................... 92
Parts of the 500-MHz NOESY spectrum with 150 ms

mixing time ofthe wild type CRABPII at 32°C 92
Parts of the SOO-MHz NOESY spectrum with 150 ms

mixing time ole 1 1M at 32°C .. 83
Parts of the 500-MHz NOESY spectrum with 150 ms

mixing time of R132M at 32°C .......................................................... 93

Strips of the HNCACB (a) and CBCA(CO)NH (b) spectra

showing sequential connectivies for residues

Gly68—Gln74 .

Figure 3.2. Gradient- and sensitivity-enhanced 2D lH—‘SN HSQC

spectrum ofuniformly ISN-labeled apo-CRABPII . . .

Figure 3.3. The fingerprint region of the homonuclear 2D TOCSY

spectrum with40 ms mixing t1me

Figure 3.4. Summary of the sequential and medium-range NOEs

involving HN and H“ atoms, slow-exchange backbone amide

protons and the deduced secondary structures of

apo-CRABPII

Figure 3.5. The distribution of NOEs along the amino sequence of

Figure 3.6. (A) Stereoview of the C“ traces of the superimposed 25

refined structures of apo-CRABPII. (B) Stereoview of the
C“ trace of the restrained minimized mean structure of
apo-CRABPII (thin line) superimposed with the C“ trace
of the crystal structure of holo-CRABPII (thick line). (C)
Stereoview of the C“ trace of the restrained minimized
mean structure of apo-CRABPII (thin line) superimposed
with the C“ trace of the crystal structure of apo-CRABPI

(molecule A) (thickline)

Figure 3.7. The relative peak intensity of the 2D lH—‘sN HSQC
spectrum of apo-CRABPII color-coded along the C“
trace of the solution structure. The strong peaks are in
white, the weak peaks in pink and the missing peaks in
cyan. RA is positioned on the basis of the superposition
of the solution structure of apo-CRABPII and the crystal

structure ofholo-CRABPII

CHAPTER 4

Figure 4.1. Gradient- and sensitivity-enhanced 2D 'H—‘SN HSQC

spectrum ofuniforrnly 15N-labeled of apo-R1 1 1M

Figure 4.2. 2D lH-‘SN HMQC spectrum of apo-R1 1 1M selectively
labeled with l5N-leucine

Figure 4.3. Strip plots extracted from the 3D 15N-edited NOESY-

HMQC spectrum of apo-R1 11M showing the sequential
and medium-range NOE connectivies of or-helix A (A)

xi

105

107

111

118

121

123

131

150

......152

and B-strand J (B) .............................................................................. 159

Figure 4.4. Summary of the sequential and medium-range NOEs
involving HN and H“ atoms, slow-exchange backbone
amide protons and the deduced secondary structures
ofapo-R111M162

Figure 4.5. The distribution of NOEs along the amino sequence
of CRABPII ...................................................................................... 166

Figure 4.6. (A) Stereoview of the C“ traces of the superimposed
28 refined solution structures of apo-R1 1 1M. (B)
Stereoview of the superimposed C“ traces of the restrained
minimized mean structures of apo-R1 1 1M (thin line) and
wild type apo-CRABPII (thick line). (C) Stereoview of the
C“ trace of the restrained minimized mean structure of apo-
R1 1 1M (thin line) superimposed with the C“ trace of the
crystal structure ofholo-CRABPII (thick line) 168

Figure 4.7. C“ atom deviations between apo-R11 1M and wild-type
holo-CRABPII (A) and between apo-R1 1 1M and wild-type
apo-CRABPII (B) 173

Figure 4.8. The fingerprint regions of the homonuclear 2D TOCSY
spectra of apo-CRABPII (A) and apo-R1 11M (B) recorded
with 40 ms mixing time . ...............176

Figure 4.9. Relative peak intensities of the 2D lH—‘SN HSQC spectra
of apo-R1 1 IM (A) and wild type apo-CRABPII (B) as a
function of the residue number. (C) was obtained by
subtracting (B) from (A) . ............180

CHAPTER 5
Figure 5.1. Summary of the sequential and medium-range NOEs
involving HN and H“ atoms and the deduced secondary
structures of holo—CRABPII .............................................................. 198
Figure 5.2. Strip plots extracted from the 3D lsN-edited NOESY-HSQC
spectrum (150 ms mixing time)of holo-CRABPII showing
the NOE connectivies for residues Met27—Lys38 200

Figure 5.3. Gradient- and sensitivity-enhanced 2D lH—‘SN HSQC
spectrum of uniformly 'SN-labeled of holo-CRABPII ......................... 203

xii

Figure 5.4. Relative peak intensities of the 2D lH—‘SN HSQC spectra
of apo-CRABPII (A) and holo-CRABPII (B) as a function
of the residue number. (C) was obtained by subtracting
(B) from (A) ...................................................................................... 205

Figure 5.5. Part of the 2D T OCSY spectrum of holo-CRABPII with 33
ms mixing time at 25°C ..................................................................... 210

Figure 5.6. Part of the NOESY spectrum of holo-CRABPII with 100 ms
mixing time at 25°C ........................................................................... 212

Figure 5.7. Relative volumes (V "”6) of some of the NOEs observed for
bound RA versus the internuclear distance (r) measured from
the crystal structure of holo-CRABPII .............................................. 214

xiii

BPP theory:
CD:
CMPG:
COSY:
CRABPI:
CRABPII:
CRBPI:
CRBPII:
CSA:
DEPT:
DGSA:

DMSO:

DQF-COSY:

E.COSY:
FABP:
FID:
HMQC:
HSQC:
iLBP:
INEPT:

IPTG:

LIST OF ABBREVIATIONS

Bloembergen-Purcell-Pound theory
circular dichroism
Carr-Purcell-Meiboom-Gill
correlated spectroscopy
type-I cellular retinoic acid binding protein
type II-cellular retinoic acid binding protein
type-II cellular retinol binding protein
type-II cellular retinol binding protein
chemical-shielding anisotropy
distortionless enhancement by polarization transfer
distance geometry-simulated annealing
dimethyl sulfoxide
double-quantum-filtered correlated spectroscopy
exclusive correlated spectroscopy
fatty acid binding protein
free-induction decay
heteronuclear multiple-quantum coherence
heteronuclear single-quantum coherence
intracellular lipid binding protein
insensitive nuclei enhanced by polarization transfer

isopropylthiogalactoside

xiv

NMR:

NOE:

NOESY:

P.E.COSY:

RA:

RAR:

RMSD:

ROESY:

TOCSY:

TPPI:

2D:

3D:

4D:

nuclear magnetic resonance

nuclear Overhauser effect

nuclear Overhauser enhancement and exchange spectroscopy
primitive exclusive correlated spectroscopy
retinoic acid

retinoic acid receptor

root-mean-square deviation

rotating frame Overhauser effect spectroscopy
retinoid X receptor

total correlation spectroscopy

time proportional phase incrementation
two-dimensional

three-dimensional

four-dimensional

XV

CHAPTER 1

Literature Review

Part 1. Cellular Retinoic Acid Binding Proteins

Section 1. Biological Functions of CRABPII

Vitamin A (retinol) and its metabolic derivatives (naturally occurring retinoids)
such as retinal and retinoic acid (RA) (Figure 1.1) are essential for vertebrates, with
profound effects on a wide variety of biological processes such as vision, cell
differentiation, embryonic development and homeostasis (Spom et al., 1994; Chambon et
al., 1996). Both the naturally occurring retinoids and their synthetic analogs have been
used to treat several kinds of human diseases including skin disorders (Peck &
DiGiovanna, 1994), and certain cancers, such as cervical cancer and acute promyelocytic
leukemia (Hong & Itri, 1994). A unified picture of the apparent pleiotropic effects of
retinoids is gradually emerging, and some key components of the retinoid signal
transduction pathway have been discovered over the past five decades. It has been shown

that retinal is a key component of visual signal transduction (Wald, 1968), while RAs,

Figure 1.1 Chemical structures and the numbering conventions of all-trans-retinol

(vitamin A), all-trans-retinal and all-trans-retinoic acid (RA).

 

All-trans-retinoic acid (RA)

especially all-trans-RA and 9-cis-RA, are the active forms of vitamin A in most of the
other biological processes. RA administration can prevent or reverse the majority of the
defects originating from post-natal vitamin A deficiency. Two distinct classes of RA-
binding proteins have been found to be involved in retinoid signal transduction: nuclear
retinoid receptors and cellular RA binding proteins (CRABPs).

Nuclear retinoid receptors. There exist in cells two retinoid receptor families: the
RA receptor (RAR) family with three subtypes RAR-0t, ~13 and -y, and the retinoid X
receptor (RXR) family also with three subtypes RXR-0t, -B and -y (Mangelsdorf et al.,
1994; Mangelsdorf & Evans, 1995; Chambon, 1996). The nuclear retinoid receptors play
critical roles in retinoid signal transduction. They have been shown to mediate many of
the biological effects observed with administration of vitamin A. For example, all
abnormalities exhibited by fetuses of vitamin A deficient dams are recapitulated in RAR
or RXR single mutants and RAR-RXR double mutants of mouse (Chambon, 1996).

Amino acid sequence comparison suggests that retinoid receptors belong to a
protein superfarnily of ligand-activated transcriptional factors (Evans, 1988; Tsai &
O’Malley, 1994) that includes also the receptors for steroid and thyroid hormones and
vitamin D3. The RAR proteins are activated by both all-trans-RA (K, = ~1-5 nM) and 9-
cis-RA with similar effectiveness, whereas the RXR proteins bind exclusively 9-cis-RA
(K, = ~10 nM) (Heyman et al., 1992; Levin et al., 1992). In vitro studies have
demonstrated that the functional units of the retinoid receptors are either the RAR-RXR
heterodimers consisting of one RXR subunit and one RAR subunit, or the homodimers of
two RXR subunits. These two forms of dimer appear to have different functions. The

heterodimers bind the DNA response elements with direct repeats of the hexamer (purine-

G(G/T)TCA) separated by one, two, or five base pairs, while the RXR homodimers prefer
the DNA sequence with direct repeats of the hexamer separated by only one base pair.
For the same response element the heterodimers have higher affinity than the
homodimers. Interestingly, RXR proteins also form heterodimers with other nuclear
receptors such as thyroid hormone and vitamin D3 receptors, and are required for the
biological functions of these receptors (see, for example, Yu et al., 1991). The diversity
of the biological effects of retinoids can be partly explained by the existence of two
families of receptors and the interactions between the two families of receptors
themselves and with other receptors. It appears that the availability of RA in cells is
regulated by other molecules such as the two isoforrns of CRABP.

Biological significance of CRABPs. There are two types of cellular retinoic acid
binding protein (CRABPI and CRABPII) (Sani & Hill, 1974; Ong & Chytil, 1975; Bailey
& Siu, 1988; Kitamoto et al., 1988). They have been found in all vertebrates relying on
vitamin A (Ong et al., 1994), and are highly conserved proteins. For example, rat, mouse
and cow CRABPI (Nilsson et al., 1988; Stoner & Gudas, 1989; Shubeita et al., 1987) all
have identical 136 residues, and human CRABPI has only one residue difference (Astrom
et al., 1991). Of 137 residues of mouse CRABPII (Giguere et al., 1990) and human
CRABPII (Astrom etal., 1991; Eller et al., 1992) 130 are identical.

Although CRABPs have been suggested to play important roles in mediating the
biological functions of RA, their physiological functions have not been well defined.
They may be involved in directing the spatial organization of cells during development,
especially at the time of the limb generation (Maden, 1991; Hofinannn & Eichele, 1994;

Morriss-Kay & Sokolova, 1996). Experimental studies have shown that both CRABPI

and CRABPII are widely expressed at all stages of mouse embryo development. In
particular, the mRNAs of CRABPs form an anteroposterior gradient opposite to the
gradient of all-trans-RA in the budding of a chick limb (Maden et al., 1988). It has also
been proposed that CRABPs may protect cells by preventing RA from being incorporated
into cell membranes. The solubility of RA in aqueous solution is about 200 nM (Szuts &
Harosi, 1991), which is higher than its endogenous concentration (~50 nM). However,
RA molecules prefer the membrane to the cytoplasm with ~6 x105 fold higher affinity.
Indeed, in tissues that express CRABPs, RA is predominately in the protein-bound forrrrs.
Both CRABPI and CRABPII have very high affinity for all-trans-RA with an apparent
dissociation constant (K) less than 2 nM (Norris et al., 1994; Wang et al., 1997).
CRABPs, CRABPI in particular, may participate in the catabolism of all-trans-RA
(Fiorella & Napoli, 1991, 1994; Boylan & Gudas, 1992). The rate of the breakdown of
RA into inactive metabolites catalyzed by cytochrome P450 increases more than twenty-
fold when it binds CRABPI. CRABPs may also be directly involved in the transport of
RA to the nucleus (Takase et al., 1986; Blomhoff et al., 1992). Taken together, these in
vitro studies lead to the proposal that CRABPs may act as ‘buffer’ to control the actual
level of ‘free’ intracellular RA available for binding to the nuclear retinoid receptors.
However, the physiological functions of CRABPs remain controversial. Genetic
experiments have demonstrated that the mutant mice lacking CRABPI and/or CRABPII
are essentially indistinguishable from the wild-type (de Bruijn et al., 1994; Gorry et al.,
1994; Fawcett et al., 1995; Lampron et al., 1995) with the exception of a minor limb

malformation. These results suggest that their actual function may be to control

physiological levels of intracellular RA when the supply of vitamin A from diet is
insufficient.

The biological functions of CRABPI and II seem to be different. The homology
within either CRABPI or CRABPII sequences across different species is much higher
than the homology between CRABPI and CRABPII from the same species. For example,
human CRABPI and CRABPII have only 74% identity. Moreover, CRABPI and
CRABPII exhibit different expression patterns during development or in adults. In mouse
embryo they are expressed in distinct, non-overlapping patterns. In adults CRABPI is
expressed in a variety of tissues, while CRABPII is mainly restricted to the epidermis
(Astrom et al., 1991; Elder et al., 1992). The CRABPII gene, not CRABPI, has a retinoid
response element in its promoter region so that CRABPII can be induced markedly in
cultured cells and human skin by 9-cis-RA and, to a lesser extent, by all-trans-RA
(Giguere etal., 1990; Astrom et al., 1991; Durand et al., 1992; Melhus et al., 1994). It has
been shown that production of RA is correlated with the expression of CRABPII but not
CRABPI (Bucco et al., 1997). Overexpression of CRABPII enhances cellular response to
RA in breast cancer cells (Jing et a., 1997) but overexpression of CRABPI decreases the
biological potency of RA in F9 teratocarcinoma cells (Boylan & Gudas, 1991). However,
no difference in binding specificity has been reported between CRABPI and CRABPII,

and their affinities for retinoids are very similar.

Section 2. Biochemical Characterizations of CRABPs

Extensive biochemical studies have been performed on CRABPs. These studies
have been expedited by cloning and overexpression of the genes encoding CRABPs in E.
coli (Giguere et al., 1990; Astrom et al., 1991). The in vitro biochemical functions of
CRABPs have been established by ligand binding assays, and the residues critical for
such functions have been investigated by site-directed mutagenesis.

Binding assay. The ligand-binding properties of CRABPs have been studied by
several methods, which could be classified into two categories: non-equilibrium methods
and fluorescence titration methods. The non-equilibrium methods remove the unbound
ligand either by the use of dextran-coated charcoal (Daly & Redfem, 1988; Fogh et al.,
1993) or by chromatography (Siegenthaler et al., 1992). A potential problem with such
methods is that the bound ligand can come off the protein during the separation process.
The de measured by these methods depend on the off-rate of the ligand, and may be an
overestimation of its true value. The fluorescence methods (Cogan et al., 1976) measure
the concentration of either the bound ligand or the bound protein at equilibrium.
However, reported fluorescence measurements have been plagued by poor experimental
designs and data analysis. The problems have been recognized recently (Norris et al.,
1994; Wang et al., 1997). Furthermore, because of the low solubility of RA in aqueous
solution, the fluorescence methods are unsuitable for measuring the dissociation constants
of site-directed mutants with greatly reduced affinity for RA. The reported Kd of CRABPI
for all-trans-RA ranges from 0.4 nM to 10 nM (Ong & Chytil, 1978; Fiorella & Napoli,
1991; Norris, et al., 1994; Fogh et al., 1993), and the K, of CRABPII from 1.8 nM to 65
nM (Bailey & Siu, 1988; Fiorella & Napoli, 1993; Fogh et al., 1993; Norris et al., 1994;

Wang et al., 1997). These discrepancies may be caused by the different methods used, the

assay conditions and the sources of the proteins. The reported relative Kd of CRABPI vs.
CRABPII ranges from three-fold (Fiorella, et al., 1993; Wang et al., 1997), as measured
by competitive binding assay, to more than ten-fold based on their individual absolute
values. The value measured by competitive binding assay appears to be more reliable.
The stoichiometry of either CRABPI or II for all-trans-RA has been found to be one-to-
one.

Binding specificity. No significant differences in binding specificity for retinoids
have been reported between CRABPI and CRABPII. Both proteins show no apparent
affinities for all-trans-retinol, all-trans-retinal and esters of RA, and have relatively low
affinities for the cis-isomers of RA. For example, their affinities for 9-cis-RA range from
~50 nM (Fiorella etal., 1993) to ~200 nM (Norris etal., 1994), and their affinities for 13-
cis-RA are at least 3-4 fold lower than that for 9-cis-RA (Fiorella et al., 1993) or too low
to be measured (Redfem & Wilson, 1993; Fogh et al., 1993; Norris et al., 1994).
However, CRABPs have high affinity for retinoids with a modified B-ionone ring such as
3,4-didehydroretinoic acid (Torma et al., 1991), 4-hydroxyretinoic acid, 4-oxo-retinoic
acid, 18-hydroxyretinoic acid, 16-hydroxy-4-oxo-retinoic acid (Fiorella et al., 1993) and
acitretin (Norris et al., 1994) (an aromatic retinoid used in the treatment of psoriasis).
Both proteins also have high affinity for retinoids with modifications at C-7 and C-8
positions of the isoprene chain (Kleywegt et al., 1994). Although 3,4-didehydroretinoic
acid exists in cells and binds CRABPs as tightly as all-trans-RA does, the latter has been
shown to be the endogenous ligand for CRABPI and possibly CRABPII (Ross et al.,

1980; Sarri et al., 1982). Taken together, these experimental data indicate that the

10

carboxylic group of RA is critical for binding to CRABPs. Modifications at the B-ionone
ring and, C-7 and C-8 can be tolerated.

Site-directed mutagenesis studies. The interactions between CRABPs and
retinoids have been investigated by site-directed mutagenesis. The single mutants R1 1 IO
and R131Q, double mutant RllleR13lQ and triple mutant RlllQ/R131Q/Yl33F of
bovine/murine CRABPI all have much lower affinity for all-trans-RA than the wild-type
(Zhang et al., 1992). The two arginines and one tyrosine are conserved residues in
CRABPs and interact with the carboxyl group of RA in the crystal structures of hole
CRABPs (Figure 1.2A, B) (Kleywegt et al., 1994). Although no quantitative K, has been
reported for these mutants, a CD signal induced by the binding of RA has been observed
in the R131Q mutant as in the wild type. No such signal has been detected in the mutant
R1 1 IO even with a five-fold excess of all-trans-RA. It appears that in CRABPI Argl 11
is more critical for binding of RA than Argl31. By contrast, it has been reported that the
R1 11A mutant of mouse CRABPII has a slightly lower affinity for all-trans-RA than the
wild type, while either R132A or R132Q mutant has much lower affinity (Chen et al.,
1995), suggesting that only Arg132 is important for binding of RA by CRABPII.
However, no quantitative K, has been reported for either R132A or R132Q mutant.

Interestingly, there is no change in the retinoid specificity of CRABPI or
CRABPII because of the mutations. In particular, the mutants show no increase in affinity
for all-trans-retinol and all-trans-retinal compared to their wild-types. In other words,
neither CRABPI nor CRABPII could be converted into a retinol-binding protein by the
single, double or triple mutations. Similarly, type-II cellular retinol binding protein

(CRBPII), a close relative of CRABPs, could not be converted into a RA-binding protein

11

Figure 1.2 (A) The ligand binding pocket of human CRABPII showing the residues
(white) interacting with all-trans-RA (green). The two hydrogen atoms (yellow) of a
water molecule, the nitrogen atoms (blue) of the side chains, and the oxygen atoms (red)
of the side chains and the RA molecule and the water are colored. (B) expanded View of
the interactions between the carboxyl group of RA and the side chains of Argl 1 1, Arg132

and Tyr134 of the protein. Dotted line indicates the possible hydrogen bond interaction.

 

13

by site-directed mutagenesis. The two single mutants Q109R and Q129R (Gln109 and
Gln129 of CRBPII are structurally equivalent to Arglll and Argl32 of CRABPII,
respectively) display reduced affinity for all-trans-retinol but no apparent affinity for all-
trans-RA (Cheng et al., 1991). Surprisingly, the mutant protein Q108R of rat CRBPI
(Gln108 in rat CRBPI is structurally equivalent to Arglll of human CRABPII) has
similar affinity for all-trans-retinol, all-trans-RA and 13-cis-RA. The mutation reduces
the affinity for all-trans-retinol by only three-fold compared to the wild-type (Stump et

al., 1991).

Section 3. Structural and Dynamical Characterizations of CRABPs

The atomic details of the interactions between CRABPs and retinoids, and their
internal motions have been analyzed by physical methods, primarily the x-ray diffraction
of single crystals and solution NMR spectroscopy. The structural and dynamical
information provides a framework for understanding the nature of the interactions and the
ligand binding process, and for identifying the factors contributing to the binding affinity
and specificity.

The structural features of CRABPs. The crystal structures of apo-CRABPI
(Thompson et al., 1995), holo-CRABPI and holo-CRABPII (both in complex with all-
trans-RA) (Kleywegt et al., 1994) have been determined at 2.7 A, 2.9 A and 1.8 A

resolutions, respectively. These studies have demonstrated that the apo- and holo-

l4

CRABPs have very similar secondary and tertiary structures. Moreover, they all have
structural features typical of a protein family called intracellular lipid binding proteins
(iLBPs) (Banaszak et al., 1994). These proteins all bind hydrophobic ligands such as fatty
acid, lipids and retinoids. In addition to CRABPs, the crystal structures of more than
twenty-five proteins of the iLBP family have been determined at various resolutions with
or without bound ligands. The structures reported include rat CRBPI (Cowan et al., 1993)
and CRBPII (Winter et al., 1993), several fatty acid binding proteins (FABPs) including
those from chicken liver (Scapin et al., 1990), bovine heart (Muller-Fahmow et al., 1991),
human muscle (Zanotti et al., 1992; Young et al., 1994) and rat intestine (Sacchettini et
al., 1989a,b; Sacchettini et al., 1992; Eads et al., 1993), bovine myelin, (Jones et al.,
1988; Cowan et al., 1993) and adipocyte lipid binding protein (ALBP) (Xu et al., 1992;
1993). Although the sequence identities among the proteins are rather low (~20%), they
all have similar tertiary structures. All the proteins adopt a compact, single domain
structure consisting of a helix-tum-helix motif and two antiparallel B-sheets (Figure 1.3).
Both sheets are composed of five B-strands, the first BA, BB, BC, BD and BE, and the
second BF, BG, BH, BI and BJ. They are twisted into a nearly orthogonal flattened barrel
with a simple up-down topology. The regular hydrogen bond ladders between adjacent B-
strands are broken between BD and BE. One B-bulge occurs in the middle of the first
strand, resulting in a change of its direction so that it forms parts of the two sheets. BF is
also shared by the two sheets. A binding cavity is formed between the two B-sheets by the
side chains from eight of the ten B-strands. The surface of the cavity could be divided
into two regions: a hydrophobic region and a more centrally located polar region (Figure

1.2A). The former is formed by the side chains from the helix-tum-helix, the BC-BD loop

15

Figure 1.3. The backbone fold and the secondary structural elements of human CRABPII
in complex with all-trans-RA (green) showing the structural motif of iLBP family: ten
antiparallel B—strands (yellow) and two a. helices (violet) fortning a flattened barrel. The

loops or turns are colored in cyan.

 

17

and the BE-BF loop. The latter is formed by the side chains of the residues deep inside the
binding pocket such as Argl l l, Arg132 and Tyrl34 (CRABPII numbering). Generally,
only a small opening to the external solvent is observed in the crystal structures, which is
supposed to be the ligand entrance. The rear portion of the pocket, opposite to the
opening, is completely blocked and inaccessible to solvent. The ligand is either totally
buried inside the protein as observed in the majority of the crystal structures or its
hydrophobic tail extends to the surface through the ligand entrance as in the crystal
structures of CRABPs and ALBP (Xu et al., 1992). The polar group of the ligand is
bound in the innermost part of the pocket. Interestingly, in most structures, the volume of
the binding pocket is much larger than what is required by the ligand. The extra volume
may provide room for the fluctuation of ligand in the bound state.

The structural basis for affinity and specificity of CRABPs. The high-resolution
structures of the holo-CRABPs provide a basis for the interpretation of their binding
affinity and specificity in terms of the relative positions of atoms. The binding specificity
and affinity of iLBPs can be partially explained by variation in the positions and the
nature of amino acid residues, and by the tightly bound solvent molecules in the binding
pocket. Particularly, the interaction between the polar end of the ligand and the protein,
(Banazak et al., 1994), seems to be most critical. In the crystal structure of holo-
CRABPII, one of the carboxylate oxygens of RA forms hydrogen bonds with the
hydroxyl of Tyrl34 and the guanidinium group of Arg132. The other interacts with the
guanidinium group of Arglll mediated by a water molecule (Figure 1.2B). Similar
interactions between the carboxyl group and the two arginine and one tyrosine resiudes

have been observed in other iLBPs. The importance of such interactions for binding of

18

the ligands has been verified by site-directed mutagenesis studies. As described earlier,
mutation of these residues to other amino acids reduces greatly the binding affinities of
iLBPs.

In the structures of holo-CRABPs extensive hydrophobic interactions have been
observed between the isoprene chain of RA and CRABPs (Figure 1.2A). Furthermore, a
network of water molecules in the binding pocket is often detected in the high-resolution
crystal structures. For example, fourteen water molecules have been identified in the
crystal structure of holo-CRABPII (Kleywegt et al., 1994). Eight water molecules have
been identified in the crystal structure of holo-CRBPI (Cowan et al., 1993) and holo-
CRBPII (Winter et al., 1993). They have been suggested to be an integral part of the
binding pockets (LaLonde et al., 1994) and to be related to the binding affinity and
specificity of these proteins.

Some clues on the binding specificity of CRABPs can be drawn from the
comparison of the amino acid sequences and the structures of iLBPs. For example, the
two arginines and one tyrosine of CRABPs and FABPs interacting with the carboxyl
groups of their ligands are replaced, respectively, by two glutamines and one
phenylalanine in CRBPs. And, the arginine-carboxyl interactions in CRABPs and FABPs
are substituted by the interactions between the two glutamines and the hydroxyl group of
retinol. Indeed, replacement of Gln108 of rat CRBPI with arginine has changed the
specificity of the protein so that it binds all-trans-RA and 13-cis-RA in addition to all-
trans-retinol (Stump et al., 1991).

Mechanisms of ligand entry or exit. Previous structural studies, mainly by means

of x-ray diffraction, have shown that most of the proteins of the iLBP family undergo

19

only small changes upon ligand binding, and there exists no entrance large enough for the
entry or exit of the ligand. In the holo crystal structures of CRBPI, CRBPII, intestinal
FABP and ALBP, the ligands are completely buried. The ligand entrances of the
unliganded forms are almost identical to those of the holo-forms. The crystal structures of
CRABPs show that RA is not completely buried (Kleywegt et al., 1994). The edge of the
B-ionone ring of bound RA is exposed to solvent. It appears, however, that significant
conformational changes are required for release of the ligand. The ligand entrance of apo-
CRABPI is slightly more open than that of holo-CRABPI, and appears to be large enough
to admit RA (Thompson et al., 1995). Surprisingly, apo-CRABPI is dirneric in the
crystalline state, held together by an intermolecular B-sheet. The slight opening of the
ligand entrance has been suggested to be caused by formation of the intermolecular B-
sheet.

Biochemical studies, by contrast, have suggested that ligand binding induces
ccnforrnational and dynamical changes in iLBPs. It has been shown that a microsomal
retinol esterifying enzyme, lecithin-retinol acyltransferase, can discriminate between the
apo- and the holo-CRBPs (Herr & Ong, 1992). It has also been demonstrated that ligand
binding reduces the protease susceptibility of the ligand entrances of CRBPI, CRBPII,
CRABPI and heart FABP (Jamison et al., 1994). The results suggest that the apo proteins
have more open binding pockets or their entrances are more flexible in solution. These
biochemical results are consistent with the physiological functions of these proteins,
which require that there exist an entrance large enough for ligand admittance, at least
temporarily. Furthermore, if holo-CRABPs and holo—CRBPs are the substrates for

metabolic enzymes (Herr & Ong, 1992; Fiorella & Napoli, 1991, 1994; Boylan & Gudas,

20

1992), there should exist structural features that can be used by the enzyme to distinguish
the apo-forms from the hole-forms. Indeed, lecithin-retinol acyltransferase can
distinguish the apo and holo forms of CRBPs (Herr & Ong, 1992).

NMR studies of the iLBP family. NMR spectroscopy has the advantage over x-
ray diffraction in providing dynamical information. Such information is critical for
understanding the ligand binding process. 1H and 15N resonance assignments have been
reported for both apo-CRABPI (pH 7.5) and holo-CRABPI (pH 3.8) (Rizo et al., 1994).
Secondary structures derived from NOE and chemical shift indices indicate that both the
apo- and the holo-CRABPI share the backbone fold of the iLBP family: ten B—strands and
a helix-tum-helix. However, significant differences in internal motions between the apo-
and holo-forms have been observed in the ligand entrance region. The ligand entrance of
apo-CRABPI is rather flexible in solution, and the flexibility is reduced greatly after the
binding of RA. Remarkable motional differences in the ligand entrance region have been
reported between rat apo and holo intestinal FABP at pH 7.2 (Hodsdon et al., 1995;
Hodsdon et al., 1996; Hodsdon & Cistola, 1997a, b). Furthermore, it has been proposed
that the decrease in the mobility of the ligand entrance in the holo form is brought about
by a series of long-range cooperative interactions that cap and stabilize the C-terminal
half of the second helix. In contrast to what was observed in the crystal structure, the
ligand entrance of bovine heart FABP is more mobile than the other parts of the
molecule, even in the holo form (Lucke et al., 1992; Lassen et al., 1995). Proton
resonance heterogeneity has been observed for Ser22-Thr36 in the helix-tum-helix,
Thr53-Thr60 in the BC—BD loop and His] l9-Ala122, indicating that these residues have

two or more slowly exchanging conformations in solution.

2]

NMR studies of the bound RA. The conformation of CRABP-bound RA has been
studied by NMR (Norris et al., 1995). The NMR results indicate that the bound RA in
CRABPII adopts a conformation with the 6-3 dihedral angle being -60° skewed from a cis
conformation, in contrast to the -33° observed in the crystal structure of holo-CRABPII
(Kleywegt et al., 1994). However, RA in holo-CRABPI appears to be quite flexible about
the 6-8 bond. The 8-3 and lO-s dihedral angles are 180° i 30° in both holo-CRABPI and

holo-CRABPII in solution as well as in the crystalline state.

22

References

Astrom, A., Tavakkol, A., Petersson., U., Cromie, M., Elder, J. T., & Voorhees, J. J.
(1991).]. Biol. Chem. 266, 17662-17666.

Bailey, J. S., & Siu, C. H. (1988) J. Biol. Chem. 263, 9326-9332.

Banaszak, L., Winter, N., Xu, 2., Bernlohr, D. A., Cowan, S. W., & Jones, T. A. (1994)
Adv. Protein Chem. 45, 89-151.

Blomhoff, R., Green, M. H., Berg, T., & Norum, K. R. (1992) Science 250, 399-404.

Boylan, J. F ., & Gudas, L. J. (1991) J. Cell. Biol. 112, 965-978.

Boylan, J. F., & Gudas, L. J. (1992) J. Biol. Chem. 267, 21486-21491.

Bucco, R. A., Zheng, W., Davis, J. T., Sierra-Rivera, E., Osteen, K. G., Chaudhary, A.
K., & Ong, D. E. (1997) Biochemistry 36, 4009-4014.

Chambon, P. (1996) FASEB. J 10, 940-954.

Chambon, P. (1994) Sem. Cell Biol. 5, 115-125.

Chambon, P., Olson, J. A., & Ross, A. C. (coordinators) (1996) The retinoid revolution.
FASEB. J 10, 938-1048.

Chen, L. X., Zhang, Z. -P., Scafonas, A., Cavalli, R. C., Gabriel, J. L., Soprano, K. J ., &
Soprano, D. R. (1995) J. Biol. Chem. 270, 4518-4525.

Cheng, L., Qian, S-J., Rotthschild, C., d’Avignon, A., Lefkowith, J. B., Gordon, J. 1., &
Li, E. (1991) J. Biol. Chem. 266, 24404-24412.

Cogan, U., Kopelman, M., Mokady, 8., Shinitzky, M. (1976) Eur. J. Biochem. 65, 71-78.

23

Cowan, S. W., Newcomer, M. E., & Jones, A. (1993) J. Mol. Biol. 230, 1225-1240.

Daly, A. K., & Redfem, C. P. F. (1988) Biochim. Biophys. Acta 985, 118-126.

de Bruijn, D. R., Oerlemans, F ., Hendriks, W., Baats, E., Ploemacher, R., Wieringa, B.,
& Geurts van Kessel, A. Differentiation 58, 141-148.

Durand, B., Saunders, M., Leroy, P., Leid, M., & Chambon, P. (1992) Cell 71, 73-85.

Eads, J., Sacchettini, J. C., Kromminga, A., Gordan, J. I. (1993) J. Biol. Chem. 268,
26375-26385.

Eller, M. S., Oleksiak, M. F ., McQuaid, T. J., McAfee, S. G., & Gilchrest, B. A. (1992)
Exp. Cell Res. 199, 328-336.

Evans, R. M. (1988) Science 240, 889-895.

Fiorella, P. D., Giguere, V., & Napoli, J. L. (1993).]. Biol. Chem. 268, 21545-21552.

Fiorella, P. D., & Napoli, J. L. (1991).]. Biol. Chem. 266, 16572-16579.

Fiorella, P. D., & Napoli, J. L. (1994) J. Biol. Chem. 269, 10538-10544.

Fogh, K., Voorhees, J. J ., & Astrom, A. (1993) Arch. Biochem. Biophys. 300, 751-755.

Giguere, V., Lyn, s., Yip, P., Siu, C-H., & Amin, s. (1990) Proc. Natl. Acad. Sci. 87,
6233-6237.

Giguere, V., Ong, E. S., Segui, P., & Evans, R. M. (1987) Nature 330, 624-629.

Herr, F. M., & Ong, D. E. (1992) Biochemistry 31, 6748-6755.

Heyrnan, R. A., Mangelsdorf, D. J ., Dyck, J. A., Stein, R. B., Eichele, G., Evans, R. M.
& Thaller, C. (1992) Cell 68, 397-406.

Hodson, M. E., & Cistola, D. P. (1997a) Biochemistry 36, 1450-1460.

Hodson, M. E., & Cistola, D. P. (1997b) Biochemistry 36, 2278-2290.

Hodson, M. E., Ponder, J. W., & Cistola, D. P. (1996) J. Mol. Biol. 264, 585-602.

24

Hodson, M. E., Toner, J. J., & Cistola, D. P. (1995) J. Biomol. NMR 6, 198-210.

Hofmannn, C., & Eichele, G. (1994) In The Retinoids: Biology, Chemistry, and Medicine
(Sporn, M. B., Roberts, A. B., & Goodman, D. S., eds), pp. 387-441, 2“d ed., Raven
Press, New York.

Hong, W. K., & Itri, L. M. (1994) In The Retinoids: Biology, Chemistry, and Medicine
(Spom, M. B., Roberts, A. B., & Goodman, D. S., eds.), pp. 597-630, 2“d ed., Raven
Press, New York.

Jamison, R. S., Newcomer, M. E., & Ong, D. E. (1994) Biochemistry 33, 2873-2879.

Jing, Y., Waxrnan, S., & Mira-y-Lopez, R. (1997) Cancer Res. 57, 1668-1672.

Jones, T. A., Bergfos, T., Sedzik, J., & Unge, T. (1989) EMBO J. 7, 1597-1604.

Kitomoto, T., Momoi, T., & Momoi, M. (1988) Biochem. Biophys. Res. Commun. 157,
1302-1308.

Kleywegt, G. J., Bergfors, T., Senn, H., 1e Motte, P., Gsell, B., Shuao, K., & Jones, T. A.
(1994) Structure 2, 1241-1258.

LaLonde, J. M., Bernlohr, D. A., & Banaszak, L. J. (1994) Biochemistry 33, 4885-4895.

Lassen, D., Lucke, C., Kveder M., Mesgarzadeh, A., Schrrridt, J. M., Specht, B., Lezius,
A., Spener, F., & Ruterjans, H. (1995) Eur. J. Biochem. 230, 266-280.

Levin, A. A., Sturzenbecker, L. J ., Kazrner, S., Bosakowski, T., Huselton, C., Allenby,
G., Speck, J ., Kratzeisen, C., Rosenberger, M., Lovey, A., & Grippo, J. F. (1992)
Nature 355, 359-361.

Lucke, C., Lassen, D., Kreienkamp, H.-J., Spener, F., & Ruterjans, H. (1992) Eur. J.
Biochem. 210, 901-910.

Mangelsdorf, D. J., & Evans, R. M. (1995) Cell 83, 841-850.

25

Mangelsdorf, D. J ., Ong, E. S., Dyck, J. A., & Evans, R. M. (1990) Nature 345, 224-229.

Mangelsdorf, D. J ., Umesono, K., & Evans, R. M. (1994) The retinoid receptors. In The
Retinoids: Biology, Chemistry, and Medicine (Spom, M. B., Roberts, A. B., &
Goodman, D. S., eds.), pp. 319-349, 2"d ed., Raven Press, New York.

Maden, M. (1991) Dev. Biol. 2, 161-170.

Maden, M., Ong, D. E., Surnmerbell, D., & Chytil, F. (1988) Nature 335, 733-735.

Melhus, H., Gobl, A., & Ljunghall, S. (1994) Biochem. Biophys. Res. Comm. 200, 1125-
1129.

Morriss-Kay, G. M., & Sokolova, N. (1996) FASEB. J 10, 961-968.

Muller-Fahmow, A., Egner, U., Jones, A., Ruedel, H., Spener, F., & Saenger, W. (1991)
Eur. J. Biochem. 199, 271-276.

Nilsson, M. H., Spurr, N. K., Saksena, P., Busch, C., Norlinder, H., Peterson, P. A.,
Rask, L., & Sundelin, J. (1988) Eur. J. Biochem. I 73, 45-51.

Norris, A. W., Cheng, L., Giguere V., Rosenberger, M., & Li, E. (1994) Biochim.
Biophys. Acta 1209, 10-18.

Norris, A. W., Rong, D., d’Avignon, D. A., Rosenberger, M., Tasaki, K., & Li. B. (1995)
Biochemistry 34, 15564-15573.

Ong, D. E., & Chytil, F. (1975) J. Biol. Chem. 250, 6113-6117.

Ong, D. E., & Chytil, F. (1978) J. Biol. Chem. 253, 4551-4554.

Ong, D. E., Newcomber, M. E., & Chytil, F. (1994) In The Retinoids: Biology,
Chemistry, and Medicine (Spom, M. B., Roberts, A. B., & Goodman, D. S., eds),
pp. 283-317, 2““ ed., Raven Press, New York.

Peck, G. L., & DiGiovanna, J. J. In The Retinoids: Biology, Chemistry, and Medicine.

26

(Spom, M. B., Roberts, A. B., & Goodman, D. S., eds.), pp. 631-658, 2"d ed., Raven
Press, New York.

Petkovich, M., Brand, N. J ., Krust, A., & Chambon, P. (1987) Nature 330, 444-450.

Rizo, J., Liu, Z. P., & Gierasch, L. M. (1994) J. Biomol. NMR 4, 741-760.

Ross, A. C., Adachi, N., & Goodman, D. S. (1980) J. Lipid Res. 21, 100-109.

Sacchettini, J. C., Gordan, J. 1., & Banaszak, L. J. (1988).]. Biol. Chem. 263, 5815-5819.

Sacchettini, J. C., Gordan, J. I., & Banaszak, L. J. (1989a) Proc. Natl. Acad. Sci. U. S. A.
86, 7736-7740.

Sacchettini, J. C., Gordan, J. I., & Banaszak, L. J. (1989b) J. Mol. Biol. 208, 327-339.

Sacchettini, J. C., Scapin, G., Gopaul, D., & Gordan, J. I. (1992) J. Biol. Chem. 267,
23534-23545.

Sani, B. P., & Hill, D. L. (1974) Biochem. Biophys. Res. Commun. 61, 1276-1282.

Sarri, J. C. ( 1982) Methods Enzymol. 81, 819-826.

Scapin, G., Spadon, P., Marnmi, M., Zanotti, G., & Monaco, H. L. (1990) Mol. Cell.
Biochem. 98, 95-99.

Shubeita, H. E., Sambrook, J. F., & McCormick, A. M. (1987) Proc. Natl. Acad. Sci.
U. S. A. 84, 5645-5649.

Siegenthaler, G., Tomatis, I., Chatellard-Gruaz, D., Jaconi, S., Erikson, U., & Saurat, J.
H. (1992) Biochem. J. 287, 383-389.

Sporn, M. B., Roberts, A. B., & Goodman, D. S. (1994) The Retinoids: Biology,
Chemistry, and Medicine, 2“d ed., Raven Press, New York.

Stoner, C. M., & Gudas, L. J. (1989) Cancer Res. 49, 1497-1504.

Stump, D. G., Lloyd, R. S., & Chytil, F. (1991) J. Biol. Chem. 266, 4622-4630.

27

Szuts, E. Z., & Harosi, F. I. (1991) Arch. Biochem. Biophys. 287, 297-304.

Takase, S., Ong, D. E., & Chytil, F. (1986) Arch. Biochem. Biophys. 24 7, 328-334.

Thompson, J. R., Bratt, J. M., & Banaszak, L. J. (1995).]. Mol. Biol. 252, 433-446.

Torma, H., Stenstrom, E., Andersson, E., & Vahlquist, A. (1991) Skin Pharmacol. 4,
246-253.

Tsai, M- J ., & O’Malley, B. W. (1994) Ann. Rev. Biochem. 63, 451-486.

Wald, G. (1968) Science 162, 230-239.

Wang, L., Li, Y., & Yan, H. (1997) J. Biol. Chem. 272, 1541-1547.

Winter, N., Bratt, J., & Banaszak, L. J. (1993) J. Mol. Biol. 230, 1247-1259.

Xu, 2., Bernlohr, D. A., & Banaszak, L. J. (1992) Biochemistry 31, 3484-3492.

Xu, Z., Bernlohr, D. A., & Banaszak, L. J. (1993).]. Biol. Chem. 268, 7874-7884.

Young, A. C. M., Scapin, G., Kromminga, A., Patel, A. B., Veerkamp, J. H., &
Sacchettini, J. C. (1994) Structure 2, 525-534.

Yu, V. C., Delsert, 0, Andersen, B., Holloway, J. M., Devary, O. V., Naar, A. M., Kim,
S. Y., Boutin, J. M., Glass, C. K., & Rosenfeld, M. G. (1991) Cell 67, 1251-1266.

Zanotti, G., Scapin, G., Spadon, P., Veerkamp, J. H., & Sacchettini, J. C. (1992) J. Biol.
Chem. 267, 18541-18550.

Zhang, J., Liu, Z., Jones, T. A., Gierasch, L. M., & Sambrook, J. F. (1992) Proteins:

Structure, Function, and Genetics 13, 87-99.

Part 2. Physical Basis of Protein NMR

Section 1. Theory of Fourier Transform Pulse NMR

Nuclear magnetic resonance (NMR) spectroscopy, the detection of the Zeeman A
levels of nuclear spins in bulk condensed phase by resonant method through
electromagnetic means, was pioneered by two research groups headed by Felix Bloch
(Bloch et al., 1946a) and Edward M. Purcell (Purcell et al., 1946), respectively. It has
since become an important experimental tool for physicists, chemists and more recently
for structural biologists as well as radiologists (medical diagnosis). Its wide applications
are made possible by continuous improvement in experimental techniques and advances
in theory. The former is stimulated by technological breakthroughs in other fields such as
electronics, computer science (e. g. fast Fourier transformation algorithm and the
sophisticated controls of experimental parameters), and the manufacturing of stable high
field superconducting magnets. The latter takes advantage of the weakness of the
interactions involving nuclear spins. The weakness makes it relatively easy to manipulate
the interaction Hamiltonian through experimental methods such as radio frequency
pulses, and justifies the treatment of the evolution of the states of a spin system by
perturbation methods. The simplicity of the derived mathematical expressions facilitates

28

29

study biological macromolecules. In this review I will summarize the underlying physical
principles of NMR, and will show how they can be applied to practical problems
especially those in structural biology.

Basic equations. Nuclear spin, a quantum mechanical phenomenon without
classical analogs, can only be treated by quantum mechanics. However, for an isolated
system consisting of non-interacting spins, the equation for the expectation value of the
total magnetization <M> is the same as that for the classical magnetization under the
influences of a static field HO and a rotating external radio-frequency field Hl (r. f. field)

(Rabi etal., 1954).

d<M>/dt=y<M> me (l)

where Heff is the total magnetic field in a frame rotating with the frequency of the r. f.
field, y is the gyromagnetic ratio. Therefore, it is legitimate to treat such a system
classically. In reality the spin system is not isolated but interacts with the molecular
environment (bath) including all degrees of freedom other than the spins. The interaction
causes a decay of the observed magnetization. With the assumptions that the decay is
exponential, and the interactions with the external fields and with the bath are additive,
Bloch derived a phenomenological equation for a non-interacting spin system (Bloch,

1946b).

dM/dt = y M x H — MVT, — k (M,— M,)/ T, (2)

30

where H is the sum of a static field H0 = kl!O and a r. f. field H], k is the unit vector on
the z-axis, M0 = XoHo is the equilibrium magnetization with x0 being the static
susceptibility, M is the total magnetization, M is the transverse magnetization. T, and T 2
are, respectively, the longitudinal and transverse relaxation times representing the
interactions between the spin system and the bath. The nuclear spins in a real spin system
interact with each other. The descriptions of such a spin system must resort to quantum
mechanism. For convenience, the Hamiltonian H of the conserved whole system is
divided into three parts: the Hamiltonian H,(s, t) of the spin system including the
interactions with the external static and r. f. fields, the Hamiltonian F (f) of the bath and
the interaction Hamiltonian GO? 3, t) between them. The state of a spin system can be
represented by a reduced density operator o(t) (von-Neumann, 1927, 1955; Fano, 1957),
which could be defined as an incoherent superposition of pure states. The dynamical
evolution of o(t) under the influence of the spin Hamiltonian H,(t) follows the Liouville-

von Neumann equation.

50/5t = -i [H.(t), 60)] (3)

o(t) is simply related to the density operator p(t) of the whole system: o(t) = tr,{p(t)},
here tr, means the partial trace over the variables of the bath. The expectation value
<Q(t)> of any spin operator Q(t) can be calculated according to <Q(t)> = tr,{Q(t)o}, here
tr, means the partial trace over the variables of the spin system. The separation of the

whole system into two subsystems (the spin system and bath) is very helpful in the

31

development of NMR theory because of the weakness of the interaction between the spin
system and the bath. For example, in the operator product formalism discussed later, an
exponential term multiplied into a spin operator can be used to represent the effect of
relaxation on the spin system caused by the bath. In relaxation theory, irreversible
processes are assumed to occur exclusively in the spin system though only reversible
processes could happen in the conserved whole system.

Pulse NMR. For most pulse NMR experiments, the exact analytic solutions of the
Liouville-von Neumann equation (3) are very difficult to find because of the complexities
of the interactions among the spin system itself, and between the spin system and its
environment. To make the problem tractable and provide useful qualitative insights into
NMR experiments, various approximations have been proposed to simplify the
Harniltonians and/or the reduced density operator depending on the nature of the practical
system under investigation. One of the simplified versions of the density operator
formalism called operator product formalism (Banwell & Primas, 1962; Sorensen et al.,
1983; Van de Ven & Hibers, 1983) has been frequently employed to describe the
experiments performed on a weakly coupled spin system in the context of high resolution
NMR. Following a suggestion by Fano (F ano, 1957), the main idea behind this formalism
is to expand the o(t) in a tensor base constructed from the products of spin operators (E,
6,, 0y, 0', of the Pauli matrices) (Sorensen et al., 1983; Ernst et al., 1987).

o(t) = 2, b,(t) B, and B, = 2‘4-’>1_[(1,,)“s* (4)

k=l

32

where B, is a base operator in the Liouville space consisting of all spin operators, N = the

total number of I = '/2 nuclei in the spin system, k = index of nucleus, v = x, y, or z, q =

number of single-spin operators in the product, as, = 1 for q nuclei and as, = 0 for the N-q

remaining nuclei. The spin Hamiltonian H, of a weakly coupled spin system in the
context of high resolution NMR in liquid phase could be approximated by 0,1,2, MHZ] ,,
and B1,“, (here B is the flip angle of per unit time about v axis), respectively. The first two
are time-independent, representing the interaction with the static field (the chemical
shifls) and the interaction between nuclear spins (spin-spin coupling), respectively. The
other is a time-dependent term, representing the interaction with external r. f. fields
(pulses), which could be made time-independent for a finite time segment by selecting a
suitable rotating frame (Rabi et al., 1954). The solution of the equation (3) for a time-

independent propagator U is very simple

(3 (t) = exp(—iUt) 6(0) exp(+iUt) (5)

Most of the NMR experiments could be calculated by means of this equation and
commutation relations. With the simplified Harniltonians mentioned above the we could
derive all the results from operator product formalism. However, the physical picture was
more clear in the latter, whose advantage originates from the simplification of the spin
Hamiltonian H, such that each of the three terms is itself a base operator. Therefore, the

evolution of any product operator under the effect of any of the three terms could be

33

calculated according to the commutation rules among them. If there exists the following

commutation relation among operators A, B and C (and their cyclic permutations)

[A, B] = iC (6)

the operator A under the propagator U = H, = C will evolve as

exp(—i0C) A exp(+i0C) = A cost) + B sin0 (7)

Therefore, the space spanned by the three independent operators A, B and C is a subspace
of the Liouville space. The majority of pulse NMR experiments applied to the structural
biological problems, where the H, could be approximated by the three terms above, could
be easily treated in this manner. However, this formalism is not adequate for a complete
mathematical description of NMR relaxation in liquid.

Fourier Transform NMR. The method of recording NMR signals free of pulse
was widely used in the T , and T 2 measurements by the spin-echo method (Hahn, 1950;
Carr & Purcell, 1954; Meiboom & Gill, 1958). A non-echo method has also been
developed to record NMR signals free of pulse by means of Fourier transform. It has
been shown that after applying a strong external r. f. field H, for a very short time 1: (r. f.
pulse) the transverse magnetization M (t) = M, + i My evolves in the absence of a pulse

according to (Lowe & Norberg, 1957)

M (t) = M,, sin (1),“: exp(iroot) I f(a),, + a) ) exp(—i(ot) do) (8)

L 4

Git)

34

where M0 is the equilibrium magnetization, (0,, is the central Larmor frequency, a), is the
frequency of the applied r. f. field, f(a),, + (o) is the shape function related to the
imaginary part of the r. f. susceptibility x". The integral is from —oo to +00. M (t)
processes with a time-dependent amplitude G(t) called correlation function, which is the
Fourier transform of the shape function f((o(, + 03). Therefore, the recording of the decay
of a NMR signal (called free induction decay) represented by G(t) gives the same
information about the shape function as the observation of the resonance with a vanishing
small r. f. field (the continuous wave NMR). The superiority of the Fourier transform
NMR over the continuous wave NMR becomes obvious when it was applied to complex
systems (Ernst & Anderson, 1966). One of the advantages is the sensitivity improvement,
which becomes more apparent for complicated high resolution spectra in liquid.
Furthermore, Fourier transformation consists in the basis for extending the dimensionality
of NMR experiments.

Multidimensional NMR. Two-dimensional (2D) NMR experiments, first
proposed by Jeener (Jeener, 1971), are natural extensions of double resonance
experiments (Bloch, 1954, 1958; Royden, 1954; Bloom & Schoolery, 1955). Both
techniques are designed to observe the properties of a system, which are related to its
nonlinear responses to external perturbations such as the connectivity of transitions in
various energy levels. In contrast to the normal double resonance techniques, the
parameter t, (time in the evolution period) in 2D techniques is changed systematically as
a variable, and a Fourier transform is applied to both time variables t, and t2 (Aue et al.,

1976)

35

S ((01, (02) = I dtl exp(—iw,t,) l dtz [exp(-io)2t2)] 5+0], ’2) (9)
S+(tl' t2) = trs {PO-(’1' t2)}=zrs Zruexp{(“i(”nM)—A73“!))t2} er tu exp{(_iwm(C)—)\'tu(6))tll (10)

0 (to ’2) = €XP{-(i"7/§d’ + F,(d))t_,} WeXPI-(it‘y/J” + 11“”)0} '4?00 (11)

where S((o,, 0),.) represents the cross-peaks of the 2D spectrum, s+(t,, t2) is the observed
complex magnetization signal (FID) and F = Z l,+i1,, is the observable spin operator.

c’ 4“”, F ‘d’ and 6%,”, 1“,“) are the Hamiltonian and relaxation superoperators in the

3

detection and evolution periods, respectively. Z,, W = F, .% (.9f00)m, here .Jfoo is the

rs in
density operator after the preparation period, .44 is the preparation superoperator and <50 is
the density operator at t = 0. .A’is a rotation superoperator representing the r. f. pulses,
which couples the coherences in the evolution period with those in the detection period.
In other words, .%induces coherence transfers which are representative of the molecular
system under investigation, and is the characteristic feature distinguishing the various
forms of 2D experiments. In three- and four-dimensional experiments, there may be two
or three evolution periods, and two or three rotational superoperator .Ws transferring the
coherences between the different evolution periods and between the last evolution period
and the final detection period.

Selection of pathway. For the success of a 2D experiment, certain desired
coherence transfer pathways must be selected while others are suppressed by either
phase-cycling (Wokaun & Ernst, 1977) or pulsed field gradients (Maudsley et a1, 1978;

Bax et al., 1980). Both methods rely on the different transformational behaviors of the

36

coherences with different orders under phase-shifting or gradients. In NMR experiments
coherences are transferred between the different orders by a propagator U((p) (5), which

changes with its phase (p in the following manner.

U(<p) = eXP(-i<pF.) U(0) eXP(+i<pF.) (12)

where (p is defined as the angle of the displacement from the x-axis towards the y-axis,
U(0) is the propagator with phase zero and F, = Z 1,. A coherence (o p) with order p
transforms under a rotation about the z-axis in a manner characteristic of its order: the

change in phase is proportional to the order.

exm-ich. )0 ” exr>(+i<pF. ) = o ” exr>(-ip<p) (13)

Therefore, different coherence order can be separated from each other by means of
shifting the phase of the propagator U according to the equation (12) and (13). More
importantly, according to (13) a certain desired coherence order such as p can be selected
by means of a discrete Fourier transformation of the combined data acquired with distinct
phases (p.

The selection of the desired coherence transfer pathways by pulsed field gradients
is essentially based on the same mechanism: the proportionality between the phase angle
4), rotated by the application of a shaped gradient and the coherence order p (Maudsley et

a1, 1978; Bax et al., 1980; John et al, 1991; Davis et al., 1992)

37

(l). = Y(SGT)I‘P (14)

where s is the shape factor, G is the gradient strength, I is the gradient duration, s0: is
the area of the gradient pulse and r is the distance from the gradient isocenter.

Pure phase spectrum. In 2D NMR, a single coherence transfer pathway
invariably leads to mixed peaks with unwanted features, each of them is a superposition
of a pure absorptive peak and a pure dispersive peak. A 2D spectrum with pure phase can
be obtained, for example, by either adding or subtracting the signals from the two mirror-
image coherence pathways followed by a 2D Fourier transformation, which is real with
respect to t, and complex with respect to t, (other factors in equations (9) and (10) not

relevant to the discussion here are dropped unless stated otherwise).

S(tl’ t2) : 2008(wtrt(6)tl) exp (_iwrs(d)t2) (15)
so», 0),) = 21 dt, cos(w,t,) 1 dr, [exp(—io)2t2)] s(t,, t2)

= (anion) + am(_(DI)} {0..(002) - id..(03.)}= a...(wr)a..(w2) + additional terms (16)

where (1(0)) and (1(6)) are the Lorentzian absorption and dispersion component,
respectively. It is assumed that the two pathways have the same magnitude. Here, only
the added signal is retained and the subtracted one is disregarded. Therefore, without
regard to the noise, the sensitivity decreases two-fold in order to obtain a pure phase
spectrum.

Quadrature detection. In the direct detection dimension (t2), the sign of a

resonance frequency w,,‘“’ can be discriminated by recording complex signals s+(t,, t2) =

38

s,(t,, t) + i s,(t,, t2) (Redfield & Gupta, 1971). Following the same principle the sign of a
resonance frequency in the indirect (t,) dimension could be distinguished (States method)
by recording the two signals 5,, (t,, t.) and 38 (t,, t2) resulting, respectively, from the two
distinct phases (p, = 0 and (p3 = rt / 21p] of the pulse at the beginning of the evolution
period (t,) (States et al., 1982). Either of them may be understood as a superposition of
the two equally weighted mirror-image coherence transfer pathways with frequencies

com“) and -w,,,‘°’. After a complex Fourier transformation with respect to t, we have

SA(tI’ t2)=2COS((Dm(c)tl) exp(—imrs(d)t2) and SA(tl’ (02):2[ars((02)_ idrs(w2)] c()S(a)tri(¢)tl) (l7)
and

SB“ 1’ t2)=2 sin(w...‘°’t 1)exp(—iwrs(d)t2) and 5301. w2)=2[ars(w2)_idrs(w2)]Sin(('0tu(e)tl) (18)

With the assumption that the magnitudes of the two signals are the same irrespective of
the possible differences of their transfer pathways due to the relaxation, a pure phase
spectrum with sign discrimination can be obtained by first constructing a complex signal
s((t,, (02) = ReSA(t,, (1),) + i ReSA(t,, m2), and subsequently transforming it by a complex

Fourier transformation with respect to t, (keeping only the pure absorption term).

S(‘((Dl' (1)2) = 2 ars((02) atu(0)l) (19)

Please note that the signal from one of the two mirror-image coherence pathways is
discarded with two-fold decrease in sensitivity without regard to the noise. The sign

discrimination in the indirect detection dimension can also be achieved by increasing the

39

phase of the pulse at the beginning of the evolution period (t,) in proportional to the

evolution time t, (TPPI method) (Bodenhausen et al., 1980b; Marion & Wuthrich, 1983)

(p = rt/(2lplt,At,) = (0 ”PP” / t, with to ”PP” = 1t/(2lplAt,) (20)

where I/At, is the sampling rate in t, dimension. Therefore, the signals in t, dimension
appear to be shifted in frequency a), = comm + pm comp", here pm is the coherence order in
the evolution period. The signals from the coherences with opposite orders such as i] are
shifted in opposite directions. A pure phase spectrum could be obtained by superimposing
the different signals calculated by real Fourier transformations with respect to t,. The
third method (States-TPPI) to discriminate the sign is essentially a States method with the
exception that the axial peaks are moved to the edge of the spectrum by the TPPI method
(Marion et al, 1989a).

Sensitivity enhancement. As shown before, the signal from one of the two mirror-
image pathways is usually discarded to obtain a pure phase spectrum. However, under
certain circumstances, it can be retained to increase the sensitivity, for example, by
shifting the phase of a pulse in the mixing period (Cavanagh et al., 1991, Palmer et al.,
1991). This can be best illustrated with a simple pulse sequence on a system consisting of

non-interacting spin 1/2.

90°B —-t,— 90°,— 90°, —t2 (detection)

where d) = ix and B = y or —x. The two signals correspond to B = y and d) = ix are

40

Sy—x (tlt t2) = Cv—x exp(_imrs(d)t2) Re[1+ exp(-iwm(e)t 1)] and

s." (t,. 1,): C."exp(—iw..“”t2) Relf exp(iw..‘e’t,)l (21)

where C ’s represent other factors in equations (9) and (10). These two signals look like
they originate from two mirror-image coherence transfer pathways, and are related with
each other by equation (12) with (p = it. Therefore, amplitude-modulated signals (pure
phase spectrum) can be obtained by either adding (sf) or subtracting (s,‘) the two signals
assuming C," = C," = C}.

mm) 6X}? (-iw..‘d’t2) (22)

"I

s," = C, 2I,cos(a),,,‘°’t,) exp (—i(o,,‘“’t2) and s,‘ = C, 211,5in((r)

The two signals with B = —x can be obtained from (20) by equation (12).

s_, “ (t,, t) = C_, "‘ exp(—ia),,‘“’t2) Re[exp(—irt/2 F,)F exp(+itr/2 F, ) exp(—ia),,,“’t,)]
= C _, "‘ exp(—i(c,,‘“’t2) Re[exp(—ird2 )F exp(—i(o,,“’t,)]

5.30:. t2) = C_.’eXp(-iwr.‘d’tz) Relexm-iTE/Z )1+ eXP(i(D...‘°’tr)l (23)

The signals similar to (21) could be obtained by either adding (s“_,, ) or subtracting (s’_, )

the two signals assuming C_,"‘ = C_," = C_,.

S”-.= -C_. 21.81n(w...‘°’tr) eXp(-i00..“”tz) and if C. 211,.COS(C0...“’tr) eXp(-iw..‘d’t2) (24)

41

With the assumption that C_, = C ,.= C new complex time domain data can be constructed
by combining the two signals s’(t,, t2) and s’(t,, t2) with respect to t, as in the usual States

method prior to the complex Fourier transformation.

3’01: t2) = Syd (to ’2) " i SS—x : 4 CCOS(wm(C)t/) exP (“103”“le and

s‘(t,, t2) = s“_, (t,, t_,) + is‘y = —4 C sin(a),,,‘°’t,) exp (—ico,,‘“’t2) (25)

The sensitivity is enhanced by two-fold compared to the usual States method (19). It
originates from the preservation and combination of the two orthogonal transverse
magnetization components I, and I, with the condition that I, and 1,, must be equivalent in
the coherence transfer processes. However, the enhancement is only 2”2 with regard to the
noise, and in reality it is still lower because the four C’s above are not necessarily equal
to each other due to the different relaxation rates of different coherence transfer pathways.
This method has been successfully incorporated into many pulse sequences for studying
the structure and dynamics of proteins.

Composite sequence. Composite pulse sequences have been widely used to
suppress off-resonance effects, and to accomplish broadband decoupling and
homonuclear isotropic mixing. They are designed to have the same effect as a single
pulse but can tolerate the imperfections of r. f. pulses. In particular, cyclic composite
sequences designed to make their propagators as close as possible to the unit operators
have been employed most frequently. Of them, WALTZ-l6 (Shaka et al., 1983), GARP-l
(Shaka et al., 1985), DIPSI-l, 2 or 3 (Shaka et al., 1988), MELEV-l6 (Levitt et al., 1983) _

and SEDUCE-l (McCoy & Mueller, 1992a, b) have been used for broadband decoupling.

42

While WALTZ-l6, MELEV-17 (Bax & Davis, 1985) and DIPSI-2 or 3 have been
employed for homonuclear isotropic mixing.

Water suppression. The huge solvent signal of an H20 sample can be suppressed
by presaturation (Anil-Kumar et al., 1980b; Wider et al., 1983), 1-1 hard pulse combined
with phase alternated l-l hard pulse (Sklenar & Bax, 1987), spin-lock (Messerle et al.,
1989), WATERGATE (Piotto et al., 1992) and water flip-back techniques (Gr'zesiek &
Bax, 1993a). The last technique has been incorporated into many pulse sequences of
multidimensional heteronuclear NMR experiments to minimize the dephasing and
saturation of the solvent signal, and consequently to minimize the exchange of amide

protons with the solvent.

Section 2. NMR Experiments for Protein Studies

NMR experiments for sequential resonance assignments. Homonuclear 2D 'H-
1H through-bond correlation experiments (Wiithrich, 1986) such as COSY (Aue et al.,
1976), DQF-COSY (Piantini et al., 1982) and TOCSY (Braunschweiler & Ernst, 1983;
Bax & Davis, 1985; Griesinger et al., 1988) can be used to identify amino acid spin
systems. Homonuclear 2D lH-‘H NOESY, a through-space correlation experiment,
(Jeener et al., 1979; Macura & Ernst, 1980; Anil-Kumar et al., 1980a) can be used to
establish the sequential connectivities along the polypeptide chain. These homonuclear

experiments can only be used for resonance assignments of proteins with ~100 residues

43

because the through-bond experiments require that the coupling constant UH” be not
much smaller than the 'H resonance line width. The first sequential resonance assignment
of a protein called BPTI was obtained by this method (Wagner & Wiithrich, 1982).
Furthermore, it has been extended for larger proteins with M, less than 15 kDa by adding
another dimension (”N) to these 2D experiments to alleviate spectral overlap.

The assignments of proteins with M, larger than 15 kDa, but less than 25 kDa,
can be achieved by a series of triple resonance experiments (Ikura et al., 1990a; Kay et
al., 1990a) using a ”C and ”N doubly labeled protein. These 3D heteronuclear correlation
experiments are designed to take advantage of the relatively large heteronuclear coupling
constants to alleviate the problems associated with large line widths. INEPT (Morris &
Freeman, 1979), or refocused INEPT (Burum & Ernst, 1980), or DEPT (Doddrell et al.,
1983) sequences have been incorporated into these experiments to accomplish the
coherence transfers between the heteronuclei 1H and ”N and '3C. For relatively rigid
proteins with M, about 20 kDa, these transfer steps are quite efficient since the line widths
of 1H, ”N and '3C resonances are normally significantly smaller than the relevant one-
bond ('JHN, 'JHC, lJNCm, lJNCO, lJam, 'JCC) (Figure 1.4) and two-bond (ZJNCQ) coupling
constants. The sequential resonance assignment is obtained in a conforrnation-
independent or through-bond manner. These experiments have been further optimized by
techniques such as constant-time (Grzesiek & Bax, 1992a; Zhu & Bax, 1990), gradient
coherence selection (Kay et al., 1992a; Kay, 1995) and gradient artifact suppression (Bax
& Pochapsky, 1992; Wider et al., 1993), composite pulse, and sensitivity enhancement
(Kay et al., 1992a). The experiments most frequently performed for sequential

assignments are HNCO, HNCA, HCACO, HN(CO)CA and their extensions such as

 

 

 

 

 

 

 

H “C CB-——H H——-CB H
8
___(i{.}) '— ofia 320.131) ‘JNC. .. _
II I II
F H H H 0

Figure 1.4. The polypeptide backbone and side chains showing the one—bond coupling
constants through which the coherences are transferred in double and triple resonance

NMR experiments.

45

HNCACB (Wittekind & Mfieller, 1993), CBCA(CO)NH (Grzesiek & Bax, 1992b) and
HBHA(CO)NH (Grzesiek & Bax, 1993b). The side-chain resonances are mainly assigned
by HCCH-COSY (Kay et al., 1990b) and HCCH-TOCSY experiments (Bax et al., 1991;
Kay et al., 1993), and in favorable cases by C(CO)NH and H(CCO)NH experiments
(Logan et al., 1993; Grzesiek et al., 1993) (Figure 1.4). For proteins with M, larger than
25 kDa, the recently developed deuterium labeling is normally needed to simplify the
spectrum, to reduce the line width, and to enhance the efficiency of coherence transfer.
For example, the HCCH-COSY and HCCH-TOCSY experiments which depend on lJCC
(~30-35 Hz) become less efficient for proteins with M, larger than 25 kDa. With the help
of a series of experiments specifically designed for 2H, ”C and ”N triply labeled proteins,
the solution structure of a protein with 259 residues has been reported recently (Garret et
al,1997)

NMR experiments for extracting structural restraints. The distance restraints
used for structural calculation are derived from NOEs. The experiments most frequently
performed include 2D homonuclear NOESY, 3D ”N-edited NOESY-HMQC (Fesik &
Zuiderweg, 1988; Zuiderweg & Fesik, 1989; Marion et al., 1989b) or NOESY-HSQC
(Marion et al., 1989b), 3D ”C-edited NOESY-HMQC (Ikura et al., 1990b; Zuiderweg et
al., 1990) as well as 4D ”N, ”C-edited NOESY (Kay et al., 1990c; Muhandirarn et al.,
1993) and 4D ”C, ”C-edited NOESY (Zuiderweg et al., 1991). These experiments are
built on two highly sensitive 2D heteronuclear correlation experiments HMQC (Miieller,
1979; Bax et al., 1983) and HSQC (Bodenhausen & Ruben, 1980) ('JNH = ~92 Hz and

lJCH = ~120-160 Hz). The torsion angle restraints are obtained from coupling constants

46

according to the Karplus equation (Karplus, 1959). The experiments most frequently
performed to extract such coupling constants are homonuclear 2D COSY, E. COSY
(Griesinger et al., 1985), P. E. COSY (Mueller, 1987) and heteronuclear 2D lH-”N
HMQC-J (Kay & Bax, 1990) and 3D HNHA (Vuister & Bax, 1993). The last two
experiments need ”N-labeled samples and are used to measure 3J,,N,,,. Furthermore, ”C
secondary chemical shifts have also been included in the structural refinements
(Kuszewski et al., 1995). Stereospecific assignment at a chiral center can be obtained
from the three-bond scalar coupling, where appropriate, with intraresidue and sequential
interresidue NOE data (Clore & Gronenbom, 1991a, b). The first protein NMR solution
structure (Williamson et al., 1985) was determined by distance geometry (Havel &
Wiithrich, 1984). However, most of the recent NMR structures are calculated by the
hybrid distance geometry-simulated annealing (DGSA) protocol (N ilges et al., 1988).
NMR experiments for studying other structural and biochemical problems. In
addition to structure determination and dynamical studies, solution NMR techniques can
be used to study other structural and biochemical problems. Bound waters can be
identified by employing the different translational diffusive behaviors of water and
protein molecules (Wider et al., 1996) in terms of intermolecular translational Brownian
motions (Torrey, 1953, 1954; Resing & Torrey, 1963; Stejskal & Tanner, 1965). Binding
sites for ligands or substrates can be identified by measuring intermolecular NOEs
(Gronenbom & Clore, 1995), or quickly but approximately by monitoring the changes in
the chemical shifts of the residues upon ligand binding. A protocol for screening the
ligands and subsequently optimize them has been developed for drug discovery (Shukert

et al., 1996). The ionization states of the ionizable groups in proteins can also be

47

measured by NMR. For example, two aspartyl groups in the active site of HIV-1 protease
have different ionization states. The carboxyl of Asp 25 is protonated while that of
Asp125 is not over a pD range extending from 2.5 to 6.2 (Wang et al., 1996). This critical

information has been used in the design of its inhibitor.

Section 3. Theories of NMR Relaxation in Liquid

Internal motions are critical for many firnctions of proteins in solution such as
ligand binding and enzymatic catalysis (Karplus & McCammon, 1983; Karplus & Petsko,
1990), although their relevance in many systems has yet to be proven experimentally.
Solution NMR spectroscopy, especially NMR relaxation measurement, has become one
of the most powerful techniques to quantitatively characterize internal motions. For
example, the rotations of aromatic side chains were discovered by NMR (Snyder et al.,
1975; Wiithrich & Wagner, 197 5). More recently, NMR relaxation study has shown that
the long helix connecting the two globular domains of the protein calmodulin observed in
x-ray diffraction is not stable in solution (Ikura et al., 1992). Internal motions that occur
in partially folded proteins are important for understanding protein folding pathways. The
residual structures of unfolded proteins have been identified by heteronuclear lH-”N
NOE experiment (Farrow et al., 1997). Relaxation study can be beneficial to NMR
spectroscopy itself. The quantitative characterization of the relaxation behaviors helps to

optimize the design of NMR experiments. It is well known that the efficiency of

48

coherence transfer, no matter whether by means of spin-spin coupling or cross-relaxation,
depends on both the global and internal motions. The fast decay of the signals of large
proteins on the NMR time scale is the key factor limiting the applicability of NMR
techniques to such systems. The quantitative relaxation information may also be used to
improve the precision of NMR-derived structures because the distance restraints used for
structural calculation are derived from cross-relaxation rates, which in turn depend on
internal motions. In the following I will review two of the most frequently used methods
for extracting quantitative information about the internal motion from NMR relaxation
measurement. Although NMR relaxation measurements have been reported for quite a
few proteins, the interpretation of such data is itself far from an easy task.

Fourier series method. A general theory of relaxation, an irreversible dissipative
process by which a disturbed closed system returns to its equilibrium state in the absence
of external perturbation (see, c. g. Landau & Lifshitz, 1969), is a difficult quantum-
statistical problem starting from the time reversal Schrodinger equation (see, c. g.
Prigogine, 1962). It has not been completely solved. In contrast to the thermodynamic
fluctuations at equilibrium, irreversibility means a direction in time. To make the problem
tractable, it is usually assumed that relaxation is caused by the same underlying random
process (Brownian motion) as that giving rise to the thermodynamic fluctuations at
equilibrium. Irreversibility is introduced as an ad hoc assumption. The random process of
Brownian motion is assumed to be a stationary Markofl process to be described to the
accuracy of a joint probability or conditional probability (Wang & Uhlenbeck, 1945).
There exist at least two methods to describe Brownian motion, the Fourier series method

(spectrum method) and the diffusion equation method. The latter will be discussed in

49

later sections. The former analyzes the Brownian motion by Fourier series or integral
(Rice, 1944). A random function F,(t) of time can be expanded as a Fourier integral, the

spectrum of the random process.

F,(t) = L,” d0) Aa((o)e "w' (26)

where the random variable A,((r)) is the amplitude with frequency (1) of the Fourier
expansion. A random process with two random functions Fa(t) and F b(t) can be
characterized by the spectral density J,,,(co) defined as Lim H, (A,(c))A,,‘(w)) / T, a time
average. The same random process can also be described in terms of a time correlation

function Cab(t, t+t) defined as

C,,,(t, m) = Lim,_,,, ( l_,,+°° F,(t) F,‘(t+r) dt ) /T (27)
C,,,(t, t+r) depends only on the time interval 1. The equivalence of the two descriptions is
established by the Wiener-Khintchine theorem, which states that the time correlation
function of a random function is a Fourier transform of its spectral density (Wang &

Uhlenbeck, 1945).

C,,,(t, t+t) = 2 10“” do) J,,,(a))coswt (28)

50

Furthermore, the ergodic hypothesis is employed to justify the replacement of time

average with ensemble average

C,,,(z, m) = Limr_,,,,1/Tl_,,f"o F,(t) F,‘(z+r) dt = <F,,(t) F,‘(t+z)>,, (29)

where angular bracket represents ensemble average.

From the exact Liouville-von Neumann equation (3), several authors have derived
relaxation theories in terms of the correlation function or spectral density of the molecular
environment (bath) appropriate for a spin system with interactions such as scalar, dipolar
and quadrupolar couplings (Wangsness & Bloch, 1953; Fano, 1953, Bloch, 1956, 1957;
Redfield, 1958, 1965; Abragam, 1961; Hubbard, 1961; Argyres & Kelley, 1964). These
theories share features similar to the general Onsager’s thermodynamic theory on
irreversible processes (Zwanzig, 1961), and are constructed by means of various
simplifying assumptions on the bath and the magnitude of the coupling between the bath
and the spin system. Their final results are a set of equations relating the NMR-
measurable relaxation parameters such as relaxation time and NOE enhancement to the
microscopic properties characteristic of the bath. The key steps of the derivations are
recapitulated as follows.

Quantum relaxation theory in operator form. An isolated system with total

conserved Hamiltonian H is divided into two subsystems, a bath with time-independent

Hamiltonian hF(/) and a spin system with time-dependent Hamiltonian hH,(s, t) including

51

the interactions with external r. f. fields (Hubbard, 1961). These two subsystems interact

with each other and the interaction Hamiltonian 1560? s, t) is a random function of time.

H = 2H,(s, t) +2; F(f)+ hG(/f s, t) (30)

The separation is necessary since the two subsystems possess quite different properties
and play distinct roles in NMR relaxation theory.

Several simplifying assumptions are introduced for the bath at the different stages
of the derivation. The bath is assumed to possess quasi-continuous energy levels, and to

be in thermal equilibrium at all times and independent of the spin system.

p(fi s ,t) = o(s, t) pT(/) = G(S, t) exp(—/iF / k7) / tr,.{exp(—/iF / kT )} (31)

where p(fl s ,t), pT(f) and o(s, t) are density operators for the whole system, the bath in
thermal equilibrium and the spin system, respectively. T is the absolute temperature, k is
the Boltzmann constant and tr, means taking the trace over an ensemble of the bath
subsystems. These assumptions characterize the basic feature of a relaxation process:
irreversibility. With these assumptions, a syrnmetrized quantum-mechanical correlation

function C,,(r ) can be defined as

Cu“ ) = <F'(t)1“"(t+1)>.tv = '/2 [trfiprplt + I) F'(t)} + trle‘lt + 09710)] (32)

52

where F’s are the variables of the bath whose dissipative behavior is rather loosely
characterized by a temperature-dependent correlation time t, with the property that
|C,,,(t ) I z 0 if I >> rc >>/r‘/ kT. It is further assumed that the bath correlation time t, is
much shorter than the relaxation time of the spin system, that is, |R|‘I or [N]-1 >> r,, where
l R I"1 or | N l’I is a function of both the spin operators and the spectral densities of the
bath, the latter is defined as J“ (to) = l C,, (t) exp(iwt) dt {see equation (75) and (76) of
Hubbard, 1961}. The magnitude of lRl‘l or lM" represents a measure of the relaxation
times of the spin system.

Two assumptions are also made for the interaction Hamiltonian hG(s, f, t). It is

assumed to be so weak that the relaxation of a spin system caused by it can be treated by

time-dependent perturbation up to the second order.

ao/at = - tr, {i [60). pm] +1.;dr [G(t). [Ga—r). mom} (33)

Furthermore, G(t) is expanded as a sum of products of a bath operator F‘(t) and a spin

operator V(t) with the latter itself being expanded as a Fourier series or integral

G(t) = z , 15*(z) Va) and V0) = 2 , V,‘ exp (rm/‘1) (34)

where 0)," is the difference of the energy levels in frequency of the spin system. With

these assumptions the change in o(s, t) for a time interval At satisfying the condition lRl’l

or [N |‘1 >> At >> I, could be expressed as

53

o(t+At) = 6(1) + if“ {R (G(t), t') + i [G(t), N (t’)]} dt’ (35)

Finally, with the assumption that G(t) in (34) can be replaced by G(t’) (this is equal to the
introduction of the irreversibility into the problem according to Fano), the

integrodifferential equation (35) becomes a linear differential equation.

66(0/ 6: = i [G(t), N (t)]+ R (G(t), t) (36)

This approximation implies that the response of the spin system to an external
perturbation is simultaneous without any memory effect (Zwanzig, 1961). Hubbard
argued that this assumption was the same as the assumption |R|", WI“l >> At >> t,.

In order to take advantage of the symmetrical properties of the bilinear dipolar-
dipolar, quadrupolar and chemical shift anisotropy (CSA) interactions G(t) could be
expanded as a scalar contraction of second-rank spherical irreducible tensor operators 7",
of the spin system and F“,(t) of the molecular degrees of the bath (see, e. g. Brink &

Satchler, 1993; Goldman, 1984)

G(t) = 2M (—1)" I“, F“_q(t) (37)

where u specifies different interactions and the summation q runs from —1 to +1, here 1

represents the rank of the tensor, in this case, I = 2. The corresponding correlation

function involving only F",(t) can be defined similarly. This expansion makes it much

54

easier to treat the cross-correlation between different relaxation mechanisms but with the
same rank (Werbelow & Grant, 1977; Vold & Vold, 1978). The cross correlation between
different ranks vanishes irrespective of the value of q.

Semiclassical theory. In semiclassical approximation the bath is further simplified
to act as a randomly fluctuating magnetic field with infinite temperature. The classical
bath correlation function is identified with the symmetrized quantum-mechanical
correlation function (31) without regard to the commutation relation and the density

operator of the bath pT(/).

Cum = tr.{F*‘<t+r)1~"'(t)} = mute) m» = <F'(t) F*‘( t+r )>.. (38)

The semiclassical formalism widely used in the interpretation of protein NMR relaxation
data is due to Abragam (Abragam, 1961). In an interaction representation with o’(t) =

tr, {p'(t)} = tr, {exp(i H,+F) p exp(—iH,—F)} equation (32) and (34) can be simplified as

0’0) = exp(iH.+F) G(t) exp(-iH.-F) = Z). 1E*(t)A"(t) = Z]. F0) Vrkexpfiwrkt) (39)

6676i = —1/2 2,, ,J,(m,*) [V,"‘,[V,", o‘— 0T1] (40)

where spectral density J,,(oo,") = l <F*(t) F"( t+t )>exp(ico,"t) dt is defined as before, and
k"(t) is a spin operator. oTrepresents the reduced density operator in a thermal equilibrium
state. There is a difference between the H,(s, t) in Hubbard’s notation and the H,(s, t) in
Abragam’s notation, namely, the former includes the average over an ensemble of baths

in thermal equilibrium while the latter does not, so no 0’ appears in (36). The cross-

55

correlation between different dipolar-dipolar interactions, and between the dipolar and
chemical shift anisotropy (CSA) is neglected in (40). They are supposed to be suppressed
experimentally (Kay et al., 1992b).

Usually T, and T , relaxation and heteronuclear NOE enhancement o(H,—)N,) such
as that between 'H and ”N are dominated by the dipolar-dipolar interaction between ”N
and the attached proton, as well as the CSA interaction with an external magnetic field.
These three relaxation parameters can be related to spectral densities J((o)’s at four

fi'equencies (Abragam, 1961)

UT] = 72H Yzth /4r6NH {10)(wH-(DN) 1' 3fl)(mN) + 6J2)((DH+CDN) } 1" 1/3A2 OJZNJCCUDN)

...... (41)
1 / T 2 = 72 H 11214712 /8r6~H {4J°’(0)+J‘°’((DH-co~) + 3.1"(69~) + 6J‘ ”((011) +6J‘Z’(wn+w~)}
+ l/3A2w2N {2/3 rem) + 1/2 J“(o)N)} (42)
G(Hz-Wz) = 72 H 1(2th Mr’iw—r {612’(0)H+wN) —J(O)(O)H_wN)l (43)

where J‘“)(co)'s are the spectral densities for the different components of the dipolar-
dipolar interaction, and J “((0) is the spectral density for CSA interaction, respectively.
The CSA is assumed to have an axially symmetrical axis 2 with A = (5,, — l/2 (0,,+oyy). rm,
is the distance between the two nuclei. It is often further assumed that all the spectral

densities J (w)’s are the same

.110)((r)) .__ 10(0)) = J”)(CO) = J ‘c (0)) = J C“ ((0) / P2(COS9 ) (44)

56

where the J “’(m) is the spectral density for the cross-correlation between the dipolar-
dipolar and the chemical shift anisotropy interactions if it is not suppressed, P,(cosG) is
the second Legendre polynomial, and 0 is the angle between the internuclear vector and
the symmetrical axis of CSA. This assumption is valid for isotropic rotational diffusions,
and usually a good approximation for most practical systems in liquid. According to
equations (41)-(43) it is obvious that only certain frequencies of the bath are directly
responsible for NMR relaxation. Those are 0, a)”, (1),, (DH-(0N and (0,,+05N causing zero,
single and double quantum transitions of the spin system. Similar equations for the
relaxation of antiphase and two spin order coherences have also been derived (see, e. g.
Peng & Wagner, 1992), and the equations keeping the cross-correlation term between the
dipolar-dipolar and CSA interactions have been published (Goldman, 1984).

Relaxation theory in matrix form. The formalisms of the relaxation theories
above are expressed in terms of operators. The main result of the semiclassical formalism
equation (39) could also be cast in a matrix form. The result is a set of linear equations
for the evolution of the matrix elements owls in a representation of the eigenstates

or, or', B, B' etc. of the unperturbed spin Hamiltonian H,(s, t) (Redfield, 1965).

doa,./ dt = i (CU—or )6m, + 2139' R,,.BB. [CW — 5,39,01,33,] (45)

where BBB. is the Kronecker delta. R ,3. is a time-independent element of the relaxation

aa'B

matrix, and is related to the spectral density Jaw, of the bath.

57

R.......=J......(9'—a' > + J.,...(a —9> «3.4.2 , J.,... (y —a) — 6.. 2 . J... .. (or — y) (46)

The spectral density Jam, is the Fourier transform of the time correlation function

Cam, (t).

JaBa'B' ((1 ‘13) = 1 dt em 43): CaBa'B' (I) and CaBa'B’ (T) = (GaB (t) Grow—1)) (47)

The interaction Hamiltonian G(t) responsible for the relaxation of the spin system is a

random function of time, and Gm. (t) could be further expanded as before

Gaa' (t) = Z1: 11*“) Vaa' (48)

where V‘aa. ’s are the spin operators and H(t) ’s are the random functions representing
local fluctuating fields. A spectral density involving only bath operators could be defined

as

1.,, (a — 13) = 1 dr ,1. ' B" Wt) Ftp-r» (49)

The semiclassical theory in matrix form presented above can be easily generalized into a
quantum-mechanical theory. One result of such extensions is that the temperature appears
in the definition of the quantum spectral density j, p (or — B) = k,,, (or -— B) exp[/r‘ (B — or) /

2.4T], here / is the Boltzman constant. Therefore, we have

58

Jrr(a‘B) =lrk(B"a) exp[/i(B—0t) //Tl (50)

This equation represents mathematically the fact that the bath induced transition where

the bath gains the energy Ii (B—or) is more probable than the opposite one by a factor

exp[/i (B—a) / / T]. Therefore, in quantum-mechanical relaxation theory Raawi
Rama, and the disturbed spin system will approach its equilibrium state as it should be.
This is in contrast to the semiclassical theory where R0,,” = Rama, and the approach to an
equilibrium state is introduced in (40) and (45) as an ad hoc assumption.

BPP theory. The semiclassical theory becomes identical with BPP theory for an
isolated system consisting of non-interacting nuclear spins (Bloembergen et al., 1948) by
an additional assumption that there exists no phase correlation between the eigenstates of
the spin Hamiltonian. Consequently, all the off-diagonal elements of the reduced density
operator G(t) can be neglected. The underlying mechanism is clear in this theory. Random
fluctuations of the bath cause not only single quantum transitions between the nuclear
spin energy levels with energy transfer between the bath and the spin system, but also
zero and double quantum transitions with energy conservation within the spin system
alone. A master equation about the probability of a transition per unit time Wag (= Rm”)

could be derived

dpa/dtzzB WaB [(pB_pBO)—(pa—pa0)] (51)

59

where p,l and p, are, respectively, the populations of or and B energy states and p00, p,0
their corresponding equilibrium values. The macroscopic magnetization such as M,,
which commutes with the spin Hamiltonian of a system of nuclear spins I could be

calculated according to

M. (t) = 2'4 m19...(t) (52)

where pm(t) is the population of an energy level with magnetic number m.
Cross-Relaxation. The BPP theory was extended to a two spin system coupled by

dipolar-dipolar interaction with four eigenstates ll) = lord), [2) = laB), I3) = |B0t) and I4) =

|BB). The following equations for longitudinal components 1,. 3,, and similar ones for

transverse components could be derived (Solomon, 1955).

dIz / dt = —-p (1,—10) — o (S,— 0) and dS, / dt = -p’ (S,— 0) — o (I,— 0) (53)
p = U,,+2U,+U, and p’ = U0+2U,' +U, (54)
0' = U2— U,, (55)
and
U0 =W323W23, U, = W,,=W2,, U, = WI, and U,’ = W.,: W,, (56)
where W.,, i, j = 1, 2, 3, 4, is the transition probability per unit time between the

eigenstates li) and 1]) of the longitudinal component of the combined system of spin I and

S. p and p’ are auto relaxation rates of spin I and S, respectively. 0 is the cross-relaxation

60

rate between the two spins I and S. One manifestation of the cross-relaxation is nuclear
Overhauser effect (NOE) predicted by Overhauser (Overhauser, 1953) and extended to
the case of nuclear dipolar-dipolar interaction by Bloch (Bloch, 1954). The distance
restraint used in NMR structure determination is extracted from the measurement of the
cross-relaxation rate 6 between protons by a NOESY type experiment. For a system
consisting only of two protons 'HI and 'H2 their cross-relaxation rate 0’ depends on the

internuclear distance r, and the spectral densities J‘q’(ar) at two frequencies.

6 (lHlHIHZ) = Waltz/4r" {6 12120)”) -J‘°’(0)} (57)

In NMR structural determination it is assumed that 6 oc l/r“ for any pair of protons in a
protein. This essentially is equivalent to the assumption of an isotropic global rotation
without internal motion. However, in reality the situation is much more complicated. The
spectral densities are related to both the global and internal motion. Moreover, global
motion is anisotropic for non-spherical molecules. For any system with three protons
sharing a common partner there exists cross-correlation (Hubbard, 1970). Other
mechanisms such as random-field may also contribute to the relaxation (Werbelow &
Grant, 1977). The compromise of extracting only distance range rather than one accurate
number from the NOESY experiment may be the best choice. However, the structures
derived in such a manner may be biased, especially for proteins with high internal

flexibility. This offers an excellent example that a clear understanding of the underlying

61

physical principles is a prerequisite for the assessments of the results obtained through
NMR techniques.

Bloch phenomenological equation. The simplest classical relaxation theory (1)
was proposed by Felix Bloch in 1946 (Bloch, 1946b) based on phenomenological
arguments. It assumes that relaxation is solely determined by two distinct relaxation times
T, and T 2.

Spectral density mapping. The theories outlined above correspond to the
treatment of the Brownian motion by the spectrum method. Their final results are similar:
a set of equations such as (41)—(43) relating the measurable NMR relaxation parameters
to the spectral densities of the bath at several frequencies. The equations (41)—(43) and
two similar ones for two spin relaxation are the starting point of the so-called spectral
density mapping method. The original version (Peng & Wagner, 1992) tried to determine
the spectral density at the five frequencies 0, 0),, 0),”, to”, or,” (here X represents
heteronucleus) through the measurements of both the one- and two-spin relaxation.
Modified versions dealing with the spectral density at only three frequencies have been
developed to compensate for the experimental errors at high frequencies by assuming that
J,” (00,”) z «Am (0)”) 2: un (0),”) (Farrow et al., 1995; Peng & Wagner, 1995; Lefevre et
al., 1996), here ny (0)) is the spectral density of the X-H vector. It is true that spectral
densities at certain frequencies are the only parameters that can be directly determined by
NMR relaxation experiments at least nowadays. However, the information context about
internal motion obtained by the mapping method is very limited. Furthermore, without
resort to a certain model the result can not be correlated or compared with the

microscopic properties of the bath obtained by means of theoretical or other experimental

62

methods. In fact, employing a model to describe relaxation is related to the diffusion
equation method (Wang & Uhlenbeck, 1945) for analyzing the Brownian motion. The
two methods are not completely equivalent under all circumstances but they give
identical results for the processes much longer than the correlation times characteristic of
the Brownian motion of the bath (Chandrasekhar, 1943). In contrast to the mapping
method the macroscopic diffusion constant obtained from the diffusion equation method
can be related to other microscopic quantities of the bath.

Diffusion equation method. This method assumes that NMR relaxation in liquid
is caused by the discrete random walk processes through short distances or over small
angular orientations. Passing over to the continuous case the probability distribution
satisfies a particular partial differential equation of diffusional type with certain boundary
conditions (Chandrasekhar, 1943). For a random Markofl process there exists a general
solution POT t), and a Green function P, (f,, t; f, t+r), here P(fl t) df is a priori probability
(at equilibrium) of finding f in the range of (t: f + df) at a time t, and P,(f,, t; f}, t+t) df, is
the conditional probability of given f, at time t one finds f, in the range of (f,, f, + df2) a

time I later. These two probabilities are simply related

P(f,, t+r) = l P (f, t)P_.(f,, t; f, m) df, (58)

Furthermore, lim,_,,, P,(f,, t; f, t+t) is equal to P(f,' t) for the thermodynamic fluctuations
at equilibrium but for an irreversibility process it becomes P03, 00) of the newly

established equilibrium state. For NMR relaxation we have P03, 00) = P( f, t) according to

63

the assumptions about the bath (31). The equivalence of the spectrum and the diffusion

equation methods is established by the following basic relationship.

Cult. t+I) = <F..(t) 1""r,‘(t+1)>rv = 11 PM, 0PM. tif.» t+1)17.11?)Fz.'(f2)dfr dfz (59)

Motional models. As is well known both the overall and internal rotations
contribute to NMR relaxation in liquid. The overall rotation can be either isotropic or
anisotropic depending on the shape of the molecule under investigation. The internal
rotation can be either free or restricted, about one bond or multiple bonds, and the bond
can be either fixed or wobbling. A model is distinguished by the particular diffusion
equation, and by its boundary condition suggested by physical instinct and the nature of
the system. A general model appropriate for an anisotropic overall rotation, and a
restricted internal rotation about multiple wobbling bonds will be quite complicated.
Some simplifications are usually necessary for extracting the dynamical information from
NMR relaxation data.

Sphere without internal motions. The simplest model considers a protein in
solution as an isotropically rotating sphere without internal motions (Abragarrr, 1961). Its

correlation function C(t) is

C(T) = expf-ITl/T.) (60)

where the parameter r, is the correlation time characteristic of the bath. The irreversibility

is encoded in the absolute value of It]. The conditional probability P(Q,,, Q, r), where 0,,

64

and Q are the Euler angles specifying the orientation of the molecule in a laboratory
frame, is the solution of a diffusion equation for the free Brownian rotational motion with

the initial condition (MS), 0) = 5(0 — £20)

dry/6t = D, A,\V (61)

where D, is the rotational diffusional coefficient and A, the spherical Laplacian operator.
This diffusion equation is the special case of the more general Smoluchowski equation

with the potential V

But/6t = div[D gradw + vf" grad V] (62)

where D and f are the diffusion constant and the frictional coefficient of the rotational
motion, respectively (Chandrasekhar, 1943). The solution P620, (1, r ) can be found by

first expanding w in spherical harmonics Y,’, and then substituting the result into (61).

P610. 9. I ) = 21.... YmP(Qo) Y...’ (Q) “pl-110+1 )D.] (63)

With a priori probability P(Q,,) = l/(8Tt2) the corresponding correlation function can be
obtained from P(Q,,, Q, t ) by (59).
Arbitrary shape without internal motion. The distribution function P(Q,,, Q, t)

for the overall motion of an arbitrary shaped molecule is a solution of the following

65

diffusion equation for the free Brownian rotational motion with the initial condition w“),

0) = 8(Q — 00) (F avro, 1960),

fill/at = Zkak L k2‘l’ (64)

where k = x, y, z, and Lf, Lyz, L," are the components of angular momentum L. The
molecule-fixed frame with the axes x, y and z diagonalizes the diffusion tensor D, and D,,,
D, and D,z are the three principal diffusional coefficients. The solution could be obtained
similarly as before except that w is now expanded in terms of functions D’m,,(Q), which
are the elements of Wigner rotation matrix D’(Q) (Hubbard, 1970). They are the
eigenfunctions of the total angular momentum of a rigid body (Brink & Satchler, 1993).

These eigenfunctions form a complete set in the space of Euler angles.

P610. 0. l‘) = 2°10 (21 +1) /(873) tr{lexm-Q '0] D"(Qo) ”(0)1 (65)

The (21+!) x (21+1) real, symmetric matrix Q1 has matrix elements

QC... = <1ml D+ L2 + (D...— -D+)L.2 + D- (U - 142) |ln> (66)

where |lm> is the eigenket of L2 and L,2 with eigenvalues 1 (1+1) and m, respectively, and

D, a 1/2 (Dni- D,,,). If one introduces the quantities

66

b... sl<l+1>D.+m2(D..-D.) = b.-. (67)

then in the special case with axially symmetry (D_ = 0) the matrix Q1 becomes diagonal,

[Cxp(—Q l(t)]rnn : 5 mn exp(—blmt) (68)
and

P(Q(lr Q! t) = Elm," (21 +1) /(81'[2) [exp(—blmt)] Dlnm(00) Dlnm(Q) (69)

where summation 1 runs from 0 to +00 and m, n from —1 to +1. For relaxation dominated
by the dipolar-dipolar and CSA interactions 1 is equal to 2. Isotropic rotation of a sphere
corresponds to D”: D”: D,, and (69) becomes (63). With a priori probability P(Q,,) =
1/(8112) the corresponding correlation functions can be obtained from (59).

Model with internal motion. With internal motion the solutions to the diffusion
equation become much more complicated. As the first step of simplification it is usually
assumed that the internal motion is independent of the overall rotation. This is valid for
the isotropic overall rotations of spherical molecules, and normally a good approximation
for arbitrary shaped molecules. The correlation function in the laboratory frame can be
related to that in a molecule-fixed frame diagonalizing the diffusion tensor by Wigner

rotation matrices D2 '5, since the elements of the correlation functions of most practical

"Ill

NMR systems are second-rank spherical tensors (Brink & Satchler, 1993).

CM, (0 = 2 (17.... [0(0)] D7 [0(01) (02.0. [9(0), ¢(0)] Dim [G(t), ¢(t)]) (70)

67

The summation n, n’ runs from —2 to +2. Where Q is referred to the laboratory frame and
0, d) to the molecule-fixed frame, respectively. The overall rotation is characterized by the
first average while the internal motion by the second average over a bath ensemble.
Model-free approach. One of the simplest ways to include internal motion is to
assume that the correlation function C,(t) for the internal motion can be divided into two
parts (Lipari & Szabo, 1982a, b): the average value S2 of the components of C,(t)
representing the restrictions of the internal motion, and a weighted normal correlation
function Q(t) with the properties that lim,_,,, Q(t) = 0 and Q(O) = 1 characterizing the

randomness of the motion. In other words, the average value of Q(t) is zero.

C,(t) = S2 + (1— S2)Q(t) (71)

S2 is called order parameter and equal to lim,_,,o C,(t). It does not depend on time and is

not a random function. With the assumption that the overall rotation is isotropic, and is

independent of the internal motion the total correlation firnction C(t) becomes

C(t) = C0(1) C[(t) = [”5 CXPH /IM)] C10) (72)

where C,(t) is the correlation function for the overall motion. The corresponding spectral

density J(o)) is

J((r)) = 2S2tM /5 [1+(th)2] + 2/5 (1— S2) 1,,” dt cosmt Q(t) exp(—t /rM) (73)

68

If we assume that Q(t) = exp(— t /r,) we have

J(o3) = 2/5{S"t,,, / [1+(corM)2] +(1— SZ) 1: / [1 +(wr)2]} (74)

where t " = 1M" + t,", and t, is defined according to r, (1— S") = 10“” (C,(t) — S2) dt. If the
internal motion is much faster than the overall motion, and lies in the extreme narrowing
range, that is, the correlation time for internal motion r, satisfies the condition 1:, << 1,,

and t, << (1), then (74) becomes

J((r)) = 2/5 {Sth / [1 +(corM)2]+( 1 —S")r,} (75)

For internal motion much slower than overall motion we instead have
J((r)) = ZIM /5 [1+(wrM)2].

Order parameter. The physical meaning of the order parameter S’ is quite clear
according to the definition of the correlation function (59). With the modified spherical
harmonics C2,, (Brink & Satchler, 1993) substituting for the general random function in

(59) we have

52 =1im..... an, t; 0... 1+1) =1im E <C2,.(0,. t) €2.10.» t+r)>Av

T—‘NXJ ”I

= 2m <C2m(QI’ t) >Av <C2m.(02’ w)>Av: zm i<C2m(Q' t) >Avi2 (76)

69

where the summation runs from —2 to 2, and C2,,1(Q,, 00) is the spherical harmonics at the
final equilibrium state reached by relaxation. We have <C,,,,(Q,, t) >Av = <C,,,,(Q,, oo)>,,=
<C,,,,(Q,, t)>Av based on the assumptions about the bath introduced before (31). In terms

of the probability distribution function the order parameter S2 is

5" =1im..., 2,, ll P(o,, t)P,(o,, 1; 9,, 1+1) C,,,(o,, t) C,,,‘(o,, 1+1) do, do,

= 2,1113“), t)P,(Q,, :; o, oo) C,,,,(Q,, z) C,,,‘(rr,, 00) do, do,

= 2,, {lP(o,, t) C,,,(o,. t) (10,} {l C2,,‘(o,, 0102(1), t) 210,}
= 2.. l{lP(Q. t) €2.19. 9 amt = 2,. I<C...(o. t) >112 (77)
where the relations P,(Q,, t; (2,, oo) = P,(Q,, co) = P,(Q,, t) = P,(Q,, t) = P(Q, t) of the
equilibrium distribution have been used. Therefore, S2 is nothing more than the average
value over a bath ensemble of the spherical harmonics of the Euler angles specifying the
orientation of a vector relative to a molecule-fixed frame. It is also the first nontrivial
term in the expansion of P(Q, t) in a series of Legendre polynomials (Lipari & Szabo,
1980). However, how well it represents P(Q, t) depends on how fast this series
converges. Therefore, a model is needed to establish the relationship between S2 and the
distribution of the orientation. The information context of S2 corresponds to the
description of a random process by only the first set of the probability distribution P(Q, t)
(Wang & Uhlenbeck, 1945). The conformation with high energy or rare probability is not
being sampled properly by S2. The claim that S" is insensitive to motions slower than

nanosecond is misleading. The insensitivity is not due to the time scale but due to the

energies of these conformations. Though P(Q, t) can not be easily obtained for a complex

70

system such as a protein even a random Markofl process must be specified by a joint
probability or conditional probability P,(Q,, t; (2,, 1+1) together with the equilibrium
distribution P(Q, t).

Extension of model-free approach. For some proteins it was found that two
exponential terms instead of one were needed to fit their relaxation data by the Lipari &
Szabo approach (Clore et al., 1990). Moreover, an exchanging term R, has been added to
the transverse relaxation rate constant 1/ T 2 to account for the contribution to it from the
possible slow motions (Kay et al., 1989; Clore et al., 1990) other than the dipolar-dipolar
and CSA interaction. The term R“ is usually explained as the contribution fi'om
conformational exchanges. However, the physical meanings of these added terms are not
well defined. They are related to the diffusion tensor in some complicated ways.
Furthermore, for proteins with arbitrary shapes the total correlation function can not be
separated rigorously into an overall and an internal part. Therefore, one should always
bear in mind the range of validity of the original formalism or its extension when using it
to interpret NMR relaxation data. Recently the Lipari & Szabo approach has been
extended to the cross-correlation by several authors (Kay & Torchia, 1991; Zhu et al.,
1995; Daragan & Mayo, 1995; Daragan & Mayo; 1997). The cross-correlation is
responsible for the difference of relaxation behaviors in the two outer lines from the two
inner ones, and manifests itself through the nonexponential behavior of the relaxation. It
has been treated in detail in two reviews (Werbelow & Grant, 1977; Vold & Vold, 1978).

Restricted rotational diffusion model. Several more sophisticated models are in
vogue. The first class may be referred to as restricted rotational diffusion model

(Wittebort & Szabo, 1978, London & Avitabile, 1978). The conditional probability P(y', t

71

; y, 0) of the internal rotation is the solution of one dimensional diffusion equation dw/o‘t
= D EN! /827 with the initial condition w“), 0) = 5(0 — £20) and the boundary condition
(hwy/61y | in. = 0, where y is azimuthal angle. The boundary condition is the mathematical

statement that the internal rotational diffusion is restricted between iy,,.

P(r'. t: v. 0)=1/(210){1 + 2..=zcos[nn(r-ra)/2101 coslnn(r’-ro)/2nl exr)(-t/I..)} (78)

where t, = 4702 /(Dn"rt"). With a priori probability P620) = 1/(2y0) the corresponding
correlation function can be obtained from (59). With this model we can include excluded
volume effects accounting for the restricted space an internuclear vector could have to
diffuse. The Woessner model (Woessner, 1962) of free internal rotation is included as a
special case of this general model with 17,, = i180°. The model above is good for the
internal rotation about one bond. It can be easily generalized into a model for internal
rotations about N bonds (Wallach, 1967). The transformation from a frame F attached to

a nucleus whose relaxation is of interest to the laboratory frame L can be achieved by a
series of Wigner’s rotation matrix D2's with Euler angles (20,, (2,2. QM, N specifying the
orientations of the successive coordinate systems.

D2q()(QLF) = Za,b,---b"D2qa(QLD)D2 (QB/)D2b,b2(012)‘”Dzbn,,bn(QN,N—I) D2b"0(QNF) (79)

ab,

The internal rotations are usually assumed to be independent of each other in order to

simplify the problem.

72

Discrete-jump model. The discrete-jump model with M configurations (Wittebort
& Szabo, 1978) tries to give a more realistic description of concerted internal rotations
with mutual dependence. It is also an extension of the Woessner jump model with only
three configurations (Woessner, 1962). The underlying mathematics is the same: the
conditional probability P(k, t; l, 0) of the side chain of a macromolecule has configuration
k at time t if it has configuration I at time 0 is the solution of the mater equation (81) with

the initial conditions (80):

P(k, t; 1, 0) = 5,, 1 = 0 and lim,_,,,, P(k, t; 1, 0) = P,,(k) for all 1. (80)

dry/6t: ZMJ.=,R,j\V (81)

where R, is the rate constant for the transition from configuration j to i with M total
available configurations. P,q(k) is a priori probability at equilibrium satisfying the
equation )3 ”H R]. P,q(i) = 0. The solution can be obtained by matrix method, and is
expressed in terms of the eigenvectors X ,'= (X ,0, mX 1M) and eigenvalues 1.,/s of the matrix

Q with elements Q”: (R, R1,)”.

1’09 t: 1, 0) = X10 (1110)"l 2,. X1" X1" eXP(-1~,.t) (82)

where X ,0 = [P,q(k)]"'2.

Wobbling in a cone model. This model assumes that an internuclear vector is

wobbling uniformly in a cone (Kinoshita et al., 1977; Lipari & Szabo, 1980). The

73

conditional probability P(QO; Q, t) is the solution of a diffusion equation for the free
Brownian rotational motion with the initial condition w(Q, 0) = 5(0 — Q”), and the

boundary condition dul(0)/60 l80 = 0, here 0 is the polar angle.

(NI/61 =1), AN (83)

where D, is the diffusion coefficient of the wobbling motion. The boundary condition is
the mathematical statement that the internal rotational diffusion is restricted within a
polar angle 0,, but does not depend on the azimuthal angle. The solution could be obtained
as an expansion of associated Legendre functions. However, the resulting expressions for
P(Qo; Q, t), and consequently for the correlation function C,(t) are not closed analytic

firnctions.

CAI) = 21' Ai exp(—Dwt/Gi) (84)

where A , and o, are functions of 00. Instead an approximate expression for C,(t) is derived
which is exact at time t = 0 and t = 00 (order parameter S2), and has the property that the
area under the approximate correlation function is the same as that under the exact

correlation function.

C,(t) = .S‘2 + (l— SZ) exp (-D,t/<o> ) (85)

74

where S2 = A,o and <o> = Z,,.,, A,o, / (l- 82). The a priori probability P(00) =
[21t(1—c0300)]". The wobbling motion has been combined with a free rotation to give a
more realistic description of the internal motion (Richarz et al., 1980).

Intermediate and slow motions. The NMR relaxation measurement discussed
above can provide dynamical information about the fast motion on the time scale ranging
from 10"2 to 10'8 second. The motions in the intermediate time scale from microsecond
to millisecond are much more difficult to be studied by NMR. The measurement of the
exchange term R“, sensitive to motions on the microsecond to millisecond time scale,
will be discussed later. The slow motion in the time scale ranging approximately from
milliseconds to seconds gives cross-peaks between separate resonances which could be
observed in homonuclear NOESY, ROESY and in heteronuclear longitudinal
magnetization or two-spin order exchange experiments. Most of NMR techniques for
measuring intermediate or slow motions have been discussed in a recent review (Palmer
et al., 1996).

Pulse sequences for measuring NMR relaxation parameters. The 2D pulse
sequences for NMR relaxation experiments have been reviewed in a recent paper (Peng &
Wagner, 1994), and the versions with pulsed field gradients are also available (Farrow et
al., 1994). Frequently the inversion-recovery (Carr & Purcell, 1954; Vold et al., 1968)
and the Carr-Purcell-Meiboom-Gill (CPMG) (Carr & Purcell, 1954; Meiboom & Gill,

1958) techniques are used to measure T, and T2, respectively.

114.0) = M0 [(1-2 eX13(-t/ T 1) l (86)

M,, (211:) = M, (0) exp(—2nt/T,) exp(—2y2DG2nt3/3) (87)

75

where G is the gradient, D diffusion coefficient and n the number of echo. The
heteronuclear NOE enhancement is measured by steady-state NOE technique. The
experimental techniques for measuring two-spin order or antiphase coherence have also
been developed (Peng & Wagner, 1992).

These pulse sequences normally include pulses to suppress cross-correlation for
minimizing its contribution to the relaxation (Boyd et al., 1990; Kay et al., 1992; Palmer
et 1a., 1992). The interference from the chemical exchange or slow motions on the
microsecond time scale can be measured by T ,p measurement (Szypersky et al., 1993) or
spectral density mapping method. The exchange term Ra manifests itself by an
anomalous decrease in T 2 relative to that predicted for dipolar, CSA or quadrupolar
relaxation in a free-precession NMR spectrum (the line width), or in a CPMG (Allerhand
& Thiele, 1968), or T ,0 experiment. In the T ,p experiment R, can be identified by directly
measuring 1,, according to the equation (Davis et al., 1990, where a number of general

cases are also discussed)

R... = (5(0)2 PIPE T.,. / (1+1...2(D.2) (88)

by varying the effective spinlock field a), (Akke & Palmer, 1996). Where 801) is the
difference in chemical shift, and p A, p, are the time fractions spent in the conformation A
and B, respectively.

T he future of NMR relaxation study. A more realistic model needs to be derived

from a general diffusion equation with external forces (Chandrasekhar, 1943) with the

76

considerations of more relevant factors such as solvent effect and cross-correlation effect.
A unified relaxation theory is also very helpful for distinguishing and quantitating the
contributions to the total relaxation of the various internal motions on different time
scales. However, even with a quite realistic model the concerted movements of large
fragments, which may be the most important internal motions for protein functions, could
not be obtained because the dynamical information obtained from NMR experiments is
local. Certain calculation procedures need to be developed to identify the collective
motion. Most importantly, experimental techniques need to be improved or developed to
measure accurately the relaxation parameters, or to measure new relaxation parameters.
Finally, dynamical information should be combined with biochemical data to establish

the roles of internal motions in the biochemical functions of proteins.

77

References

Abragam, A. (1961) Principles of Nuclear Magnetism, Clarendon Press, Oxford.
Chapter VIII, XXI.

Akke, M., & Palmer A. G. (1996) J. Am. Chem. Soc. 118, 911-912.

Allerhand, A., & Thiele, E. (1968) J. Chem. Phys. 45, 902-916.

Anil-Kumar, Ernst, R. R., & Wfithrich, K. (1980a) Biochem. Biophys. Res. Commun.
95, 1-6.

Anil-Kumar, Wagner, G., Ernst, R. R., & Wiithrich, K. (1980b) Biochem. Biophys.
Res. Commun. 96, 1156-1163.

Argyres, P. N., & Kelley, P. L. (1964) Phys. Rev. 134A, 98-111.

Aue, W. P., Bartholdi, E., & Ernst, R. R. (1976) J. Chem. Phys. 64, 2229-2246.

Banwell, C. N., & Primas, H. (1962) Molec. Phys. 6, 225-256.

Basus, V. J. (1989) Methods Emzymol. 177, 132-149.

Bax, A., Clore, G. M., & Gronenbom, A. M. (1991).]. Magn. Reson. 88, 425-431.

Bax,A., & Davis, D. G. (1985) J. Magn. Reson. 65, 355-360.

Bax, A., DeJong, P. G., Mehlkopf, A. F. , & Smidt, J. (1980) Chem. Phys. Lett. 69, 567-
570.

Bax, A., Driscoll, P. C., Clore, G. M., Gronenbom, A. M., Ikura, M., Kay, L. E. (1990)
J. Magn. Reson. 87, 620-627.

Bax, A., Griffey, R. H., & Hawkins, B. L. (1983) J. Magn. Reson. 55, 301-315.

78

Bax, A., & Pochapsky, S. S. (1992) J. Magn. Reson. 99, 638-643.

Bloch, F. (1946b) Phys. Rev. 70, 460-474.

Bloch, F. (1954) Phys. Rev. 93, 944-944.

Bloch, F. (1956) Phys. Rev. 102, 104-135.

Bloch, F. (1957) Phys. Rev. 105, 1206-1222.

Bloch, F. (1958) Phys. Rev. 111, 841-853.

Bloch, F ., Hansen, W. W., & Packard, M. (1946a) Phys. Rev. 69, 127-127.

Bloembergen, N., Purcell, E. M., & Pound, R. V. (1948) Phys. Rev. 73, 679-712.

Bloom, A. L., & Schoolery, J., N. (1955) Phys. Rev. 97, 1261-1265.

Bodenhausen, G., & Ruben, D. J. (1980) Chem. Phys. Lett. 69, 185-188.

Bodenhausen, G., Vold, R. L., & Vold, R. R. (1980) J. Magn. Reson. 37, 93-106.

Boyd, J ., Hommel, U., Campbell, I. D. (1990) Chem. Phys. Lett. 175, 477-482.

Braunschweiler, L., & Ernst, R. R. (1983) J. Magn. Reson. 53, 521-528.

Brink, D. M., & Satchler, G. R. (1993) Angular Momentum, Oxford University Press,
New York.

Burum, D. R., & Ernst R. R. (1980) J. Magn. Reson. 39, 163-168.

Carr, H.Y., & Purcell, E. M. (1954) Phys. Rev. 94, 630-638.

Cavanagh, J., Palmer, A. G., Wright, P. E., & Rance, M. (1991) J. Magn. Reson. 91,
429-436.

Chandrasekhar, S. (1943) Rev. Mod. Phys. 15, 1-89.

Clore, G. M., Bax, A., Driscoll, P. C., Wingfield, P. T., & Gronenbom, A. M. (1990)
Biochemistry 29, 8172-8184.

Clore, G. M., Driscoll, P. C., Wingfield, P. T., Gronenbom, A. M. (1990) Biochemistry

79

29, 7387-7401.

Clore, G. M., & Gronenbom, A. M. (1991a) Science 252, 1390-1399.

Clore, G. M., & Gronenbom, A. M. (1991b) Annu. Rev. Biophys. Chem. 21, 29-63.

Clore, G. M., Szabo, A., Bax, a., Kay, L. E., Driscoll, P. C., & Gronenbom, A. M.

(1990).]. Am. Chem. Soc. 112, 4989-4991.

Daragan, V. A., & Mayo, K. H. (1995) J. Magn. Reson. B107, 274-278.

Daragan, V. A., & Mayo, K. H. (1997) Prog. NMR Spectrosc. 31, 63-105.

Davis, D. G., Perlman, M. E., London, R. E. (1990).]. Magn. Reson. B104, 266-275.

Davis, A. L., Boelens, R., & Kaptein, R. (1992).]. Biomol NMR 2, 395-400.

Dayie, K. T., Wager, G., & Lefevre, J. F. (1996) Annu. Rev. Phys. Chem. 47, 243-282.

Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J ., & Bax, A. (1995).].
Biomol.NMR 6, 277-293.

Doddrell, D. M., Pegg, D. T., & Bendall, M. R. (1982).]. Magn. Reson. 48, 323-327.

Ernst, R. R., & Anderson, W. A. (1966) Rev. Sci. Instrum. 37, 93-102.

Ernst, R. R., Bodenhausen, G., & Wokaun, A. (1987) Principles of Nuclear Magnetic
Resonance in One and Two Dimensions, Clarendon Press, Oxford.

Fano, U. (1953) Phys. Rev. 96, 869-873.

Fano, U. (1957) Rev. Mod. Phys. 29, 74-93.

Farrow, N. A., Muhandiram, R., Singer, A. U., Pascal, S. M., Kay, C. M., Gish, G.,
Shoelson, S. E., Pawson, T., Kay, J. D. F., & Kay, L. E. (1994) Biochemistry 33,
5984-6003.

Farrow, N. A., Zhang, 0., Forrnan-Kay, J. D., & Kay, L. E. (1997) Biochemistry 36,

2390-2402.

80

Farrow, N. A., Zhang, 0., Szabo, A., Torchia, D. & Kay, L. E. (1995) J. Biomol. NMR 6,
153-162.

Farvo, L. D. (1960) Phys. Rev. 119, 53-62.

Fesik, S. W., & Zuiderweg, F. R. P. (1988) J. Magn. Reson. 78, 588-593.

Garret, D. S., Seok, Y., Liao d., Peterkofsky, A., Gronenbom, A. M., & Clore, G. M.
(1997) Biochemistry 36, 2517-2530.

Goldman, M. (1984) J. Magn. Reson. 60, 437-452.

Griesinger, C., Otting, G., Wfithrich, K., & Ernst, R. P. (1988) J. Am. Chem. Soc. 110,
7870-7872.

Griesinger, C., Sorensen, O. W., & Ernst, R. P. (1985) J. Am. Chem. Soc. 107, 6394-
6396.

Gronenbom, A. M., & Clore, G. M. (1995) CRC Crit. Rev. Biochem. Mol. Biol. 30,
351-385.

Grzesiek, S., Anglister, J., & Bax, A. (1993) J. Magn. Reson. BI 01, 114-119.

Grzesiek, S., & Bax, A. (1992a) J. Magn. Reson. 96, 432-440.

Grzesiek, S., & Bax, A. (1992b) J. Am. Chem. Soc. 114, 6291-6293.

Grzesiek, S., & Bax, A. (1993a) J. Am. Chem. Soc. 115, 12593-12594.

Grzesiek, S., & Bax, A. (1993b) J. Biomol. NMR 3, 185-204.

Hahn, E. L. (1950) Phys. Rev. 80, 580-594.

Havel, T. F., & Wiithrich, K. (1984) Bull. Math. Biol. 46, 673-698.

Hubbard, P. S. (1961) Rev. Mod. Phys. 33, 249-164.

Hubbard, P. S. (1970) J. Chem. Phys. 52, 563-568.

Huntress, W. T. (1970) Adv. Magn. Reson. 4, 1-37.

81

Ikura, M., Clore, G. M., Gronenbom, A. M., Zhu, G., Klee, C. B., & Bax, A. (1992)
Science 256, 632-638.

Ikura, M., Kay, L. E., & Bax, A. (1990a) Biochemistry 29, 4659-4667.

Ikura, M., Kay, L. E., Tschudin, R., & Bax, A. (1990b) J. Magn. Reson. 86, 204-209.

Jeener, J. (1971) Ampere International Summer School, Basko Polje, Yugoslovia

Jeener, J., Meier, B. H., & Ernst, R. R. (1978).]. Chem. Phys. 71, 4546-4553.

John, B. K., Plant, D., Heald, S. H., & Hurd, R. D. (1991) .1. Magn. Reson. 94, 664-669.

Karplus, M. (1959) J. Phys. Chem. 30, 11-15.

Karplus, M., & McCammon, J. A. (1983) Ann. Rev. Biochem. 53, 253-300.

Karplus, M., & Petsko, G. A. (1990) Nature 347, 631-639.

Kay, L. E. (1995) Prog. Biophys. Molec. Biol. 63, 277-299.

Kay, L. E., & Bax, A. (1990) J. Magn. Reson. 86, 110-126.

Kay, L. E., Clore, G. M., Bax, A., & Gronenbom, A. M. (1990c) Science 249, 411-414.

Kay, L. E., Ikura, M., & Bax, A. (1990b) .1. Am. Chem. Soc. 112, 888-889.

Kay, L. E., Ikura, M., Tschudin, R., & Bax, A. (1990a) J. Magn. Reson. 89, 495-514.

Kay, L. E., Keiffer, P., & Saarinen, T. (1992a) J. Am. Chem. Soc. 114, 10663-10665.

Kay, L. E., Nicholson, L. D., Delaglio, F., Bax, A., & Torchia, D. (1992b) J. Magn.
Reson. 97, 359-375.

Kay, L. E., & Torchia, D. (1991).]. Magn. Reson. 95, 536-547.

Kay, L. E., Torchia, D., & Bax, A. (1989) Biochemistry 28, 8972-8979.

Kay, L. E., Xu, G. Y., Singer, A. U., Muhandiram, D. R., & Forman-Kay, J. D. (1993)
J. Magn. Reson. BIO], 333-337.

Kinoshita, K. Jr., Kawato, S., & Ikegami, A. (1977) Biophys. J. 20, 289-305.

82

Kuszewski, L., Qin, J., Clore, G. M., & Gronenbom, A. M. (1995).]. Magn. Reson.
3106, 92-96.

Landau, L. D., & Lifshitz, E. M. (1969) Statistical Physics, pp. 6, Addison-Wesley,
Reading, Massachusetts.

Lefevre, J. F., Dayie, K. T., Peng, J. W., & Wagner, G. (1996) Biochemistry 35, 2674-
2686.

Levitt, M. H., Freeman, R., & Frenkiel, T. (1983) Adv. Magn. Reson. 11, 47-1 10.

Lipari, G., & Szabo, A. (1980) Biophys. J. 30, 489-506.

Lipari, G., & Szabo, A. (1982a) J. Am. Chem. Soc. 104, 4546-4559.

Lipari, G., & Szabo, A. (1982b) J. Am. Chem. Soc. 104, 4559-4570.

Logan, T. M., Olejniczak, E. T., Xu, R. X., & Fesik, S. W. (1993) J. Biomol. NMR 3,
225-231.

London, R. E., & Avitabile, J. (1978) J. Am. Chem. Soc. 99, 7765-7776.

Lowe, I. J., & Norberg, R. E. (1957) Phys. Rev. 107, 46-61.

Macura, S., & Ernst, R. R. (1980) Mol. Phys. 41, 95-117.

Marion, D., Driscoll, P. C., Kay, L. E., Wingfeld, P. T., Bax, A., Gronenbom, A. M., &
Clore, G. M. (1989d) Biochemistry 28, 6150-6156.

Marion, D., Ikura, M., & Bax, A. (1989c) J. Magn. Reson. 84, 425-430.

Marion, D., Ikura, M., Tschudin, R., & Bax, A. (1989a) J. Magn. Reson. 85, 393-399.

Marion, D., Kay, L. E., Sparks, S. W., Torchia, D. A., Bax, A. (1989b) J. Am. Chem. Soc.

111,1515-1517.
Marion, D., & Wuthrich, K. (1983) Biochem. Biophys. Res. Commun. 113, 967-974.

Maudsley, A. A., Wokaun, A., & Ernst, R. R. (197 8) Chem. Phys. Lett. 55, 9-14.

83

McCoy, M. A., & Miieller, L. (1992a) J. Am. Chem. Soc. 114, 2108-2112.

McCoy, M. A., & Mueller, L. (1992b) J. Magn. Reson. B98, 674-679.

Meiboom, S., & Gill, G. (1958) Rev. Sci. Instrum. 29, 668-691.

Messerle, B. A., Wider, G., Otting, G., Weber, G, & Wiithrich, K. (1989) J. Magn.
Reson. 85, 608-613.

Morris, G. A., & Freeman, R. (1979) J. Am. Chem. Soc. 101, 760-762.

Miieller, L. (1979) J. Am. Chem. Soc. 101, 4481-4484.

Miieller, L. (1987) J. Magn. Reson. 72, 191-196.

Muhandiram, D. R., & Kay, L. E. (1994) J. Magn. Reson. B 103, 203-216.

Muhandiram, D. R., Xu, G., & Kay, L. E. (1993).]. Biomol. NMR 3, 463-470.

Nilges, M., Clore, G. M., & Gronenbom, A. M. (1988) FEBS Lett. 229, 129-136.

Overhauser, A. W. (1953) Phys. Rev. 89, 689-700; ibid, 92, 411-415.

Palmer, A. G., Cavanagh, J., Wright, P. E., & Rance, M. (1991) J. Magn. Reson. 93,
151-170.

Palmer, A. G., Skelton ,N. J ., Chazin, W. J ., Wright, P. E., & Rance, M. (1992) Mol.
Phys. 75, 699-711.

Palmer, A. G., Williams, J ., & McDermott, A. (1996) J. Phys. Chem. 100, 13293-13310.

Peng, J. W., & Wagner, G. (1992) J. Magn. Reson. 98, 308-332.

Peng, J. W., & Wagner, G. (1994) Methods in Enzymology 239, 563-596.

Peng, J. W., & Wagner, G. (1995) Biochemistry 34, 16733-16752.

Piantini, U., Sorensen, O. W., & Ernst, R. P. (1982).]. Am. Chem. Soc. 104, 6800-6801.

Piotto, M., Saudek, V., & Sklenar, V. (1992) J. Biomol. NMR 2, 661-665.

Prigogine, I. (1962) Non-Equilibrium Statistical Mechanics, John Wiley & Sons, New

84

York, New York.

Purcell, E. M., Torrey, H. C., & Pound, R. V. (1946) Phys. Rev. 69, 37-38.

Rabi, I. I., Schwinger, J., & Ramsey, N. F. (1954) Rev. Mod. Phys. 26, 167-171.

Redfield, A. G. (1958) IBM J. Res. Dev. 1, 19-31.

Redfield, A. G. (1965) Adv. Magn. Reson. 1, 1-32.

Redfield, A. G., & Gupta, R. K. (1971) Adv. Magn. Reson. 5, 81-115.

Resing, H. A., & Torrey, H. C. (1963) Phys. Rev. 131, 1102-1104.

Rice, S. O. (1944) in Selected Papers on Noise and Stochastic Processes edited by
Nelson Max pp. 133-294, Dover, New York, New York.

Richarz, R., Nagayama, K., & Wuthrich, K. (1980) Biochemistry 19, 5189-5196.

Royden, V. (1954) Phys. Rev. 96, 543-544.

Shaka, A. J., Barker, P. B., & Freeman, R. (1985) J. Magn. Reson. 64, 547-552.

Shaka, A. J ., Keeler, J., Frenkiel, T., & Freeman, R. (1983) J. Magn. Reson. 52, 335-338.

Shaka, A. J., Lee, C. J., & Pine, A. (1988).]. Magn. Reson. 77, 274-293.

Shukert, S. B., Hajduk, P. J., Meadows, R. P., & Fesik, S. W. (1996) Science 274, 1531-
1534.

Sklenar, V., & Bax, A. (1987) J. Magn. Reson. 74, 469-479.

Snyder, G. H., Rowan, R., Karplus, S., & Sykes, B. D. (1975) Biochemistry 14, 3765-
3777.

States, D. J ., Haberkon, R. A., & Ruben, D. J. (1982) J. Magn. Reson. 48, 286-292.

Stejskal, E. O., & Tanner, J. E. (1965) J. Chem. Phys. 42, 288-292.

Solomon, 1. (1955) Phys. Rev. 99, 559-565.

Sorensen, O. W., Eich, G. W., Levitt, M. H., Bodenhausen, G., & Ernst, R., R. (1983)

85

Prog. NMR Spectrosc. 16, 163-192.

Szypersky, T., Luginbuhl, P., Otting, G., Guntert, P., & Wiithrich, K. (1993) J. Biomol.
NMR 3,151-164.

Torrey, H. C. (1953) Phys. Rev. 92, 962-969.

Torrey, H. C. (1954) Phys. Rev. 96, 690-690.

Van de Ven, F. J. M., & Hibers, C. W. (1983).]. Magn. Reson. 54, 512-520.

Vold, R. L., & Vold, R. R. (1978) Prog. NMR Spectrosc. 12, 79-133.

Vold, R. L., Waugh, J. S., Klein, M. P., Phelps, D. E. (1968) J. Chem. Phys. 48, 3831-
3832.

Von-Neumann, J. (1927) J. Gottinger Nachr. 245 and 273.

Von-Neurnann, J. (1955) Mathematical Foundations of Quantum Mechanics, Princeton
University Press, Princeton. New Jersey.

Viiister, G. W., & Bax, A. (1993) J. Am. Chem. Soc. 115, 7772-7777.

Wagner, 6., & Wiithrich, K. (1982) J. Mol. Biol. 155, 347-366.

Wallach, D. (1967) J. Chem. Phys. 4 7, 5258-5268.

Wang, Y-X., Freedberg, I. D., Yamazaki, T., Wingfield, P. T., Stahl, S. J ., Kaufman, J.
D., Kiso, Y., & Torchia, D. A. (1996) Biochemistry 35, 9945-9950.

Wang, M., & Uhlenbeck, G. E. (1945) Rev. Mod. Phys. 17, 323-342.

Wangsness, R. K., & Bloch, F. (1953) Phys. Rev. 89, 728-739.

Werbelow, L. G., & Grant, D. M. (1977) Adv. Magn. Reson. 9, 189-299.

Wider, G., Hosur, R. V., & Wiithrich, K. (1983) J. Magn. Reson. 52, 130-135.

Wider, G., Riek, R., & Wiithrich, K. (1996) J. Am. Chem. Soc. 118, 1 1629-11634.

Wider, G., & Wiithrich, K. (1993) J. Magn. Reson. A102, 239-241.

86

Williamson, M. P., Havel, T. F., & Wiithrich, K. (1985) J. Mol. Biol. 182, 295-315.

Wittebort, R. J., & Szabo, A. (1978) J. Chem. Phys. 69, 1722-1737.

Wittekind, M., & Mfieller, L. (1993) J. Magn. Reson. 101, 201-205.

Woessner, D. E. (1962).]. Chem. Phys. 36, 1-4.

Wokaun, A., & Ernst, R. R. (1977) Chem. Phys. Lett. 52, 407-412.

Wiithrich, K. (1986) NMR of Proteins and Nuclear Acids, Wiley, New York.

Wiithrich, K., & Wagner, G. (1975) FEBS Lett. 50, 265-268.

Zhang, 0., Kay, L. E., Olivier, J. P., & Forman-Kay, J. D. (1994) J. Biomol. NMR 4,
845-858.

Zhu, G., & Bax, A. (1990) J. Magn. Reson. 90, 405-410.

Zhu, L., Kemple, M. D., Landy, S. B., & Buckley, P. (1995) J. Magn. Reson. 3109, 19-
30.

Zuiderweg, E. R. P., & Fesik, S. W. (1989) Biochemistry 28, 2387-2391.

Zuiderweg, E. R. P., McIntosh, L. P., Dahlquist, F. W., & Fesik, S. W. (1990) J. Magn.
Reson. 86, 210-216.

Zuiderweg, E. P. R., Petros, A. M., Fesik, S. W., Olejniczak, E. T. (1991).]. Am. Chem.
Soc. 113, 370-372.

Zwanzig, R. (1961) Phys. Rev. 124, 983-992.

CHAPTER 2

Structure-Function Relationships of

Cellular Retinoic Acid Binding Proteins

(Reprinted with the permission of the American Society of
Biochemistry and Molecular Biology)

87

88

1115 Juuaxru. or 8101.04.21ch Crtsunmrv
O 1997 by The American Society for Biochemistry and Molecular Biology. Inc.

Vol. 272. No. 3. Issue ofJanuary 17, pp. 1541-1547. 1997
Printed in USA.

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”

(Received for publication, August 1, 1996, and in revised form, September 27, 1996)

Lincong Wang, Yue Li, and Honggao Yanil:

From the Department of Biochemistry, Michigan State University, East Lansing, Michigan 48824

It has been suggested that electrostatic interactions
are critical for binding of retinoic acid by cellular reti-
noic acid-binding proteins (CRABP-I and CRABP-II).
However, the roles of two conserved arginine residues
(Art-111 and Arg-l3l in CRABP-I; Arg-lll and Arg-132 in
CRABP-II) that interact with the carboxyl group of ret-
inoic acid have not been evaluated. A novel competitive
binding assay has been developed for measuring the
relative dissociation constants of the site-directed mu-
tants of CRABPs. Arg-lll and Arg-l32 of CRABP-II were
replaced with methionine by site-directed mutagenesis.
The relative dissociation constants of R1 11M and R132M
(K4 (auras/Ks (CRABP-fl) ”“1 Kd (army/Kdtcmpanl were
determined to be 40-45 and 6-8, respectively. The ring
protons of the aromatic residues of the wild-type
CRABP-II and the two mutants were sequentially as-
signed by two-dimensional homonuclear NMR in con-
junction with three-dimensional heteronuclear NMR.
Detailed analysis of the nuclear Overhauser effect spec-
troscopy spectra of the proteins indicated that the con-
formations of the two mutants are highly similar to that
of the wild-type CRABP-II. These results taken together
showed that Arg-ll l and Arg-l32 are important for bind-
ing retinoic acid but contribute to the binding energy
only by ~2.2 and 1.2 kcal/mol, respectively. In addition,
the relative dissociation constant of CRABP-II and
CRABP'I (Kd (CRABP-TD’Kd (cmp_[)) was determined to be
2-3, in close agreement with that calculated using the
apparent K, values determined under the same condi-
tions by fluorometric titrations.

 

Retinoic acid (RA).1 a hormonally active metabolite of vita-
min A, has profound effects on cell growth, differentiation, and
morphogenesis. Two types of proteins have been found to bind
RA: nuclear retinoic acid receptors (RARs and RXRs) and cel-
lular retinoic acid-binding proteins (CRABPs). RARs and RXRs
are RA-activated transcriptional factors that regulate expres-
sion of target genes (1). Although the physiological roles of

‘ This work was supported by funds from the REF Center of Protein
Structure and Design and the Cancer Center at Michigan State Uni-
versity. The costs of publication of this article were defrayed in part by
the payment of page charges. This article must therefore be hereby
marked ‘advertisement' in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.

3 To whom correspondence should be addressed. Tel.: 617-353-8786;
Fax: 517-353-9334; E-mail: yan®nmr1.bch.msu.edu.

‘The abbreviations used are: RA, all-trans-retinoic acid; CRABP.
cellular retinoic acid-binding protein; DQF-COSY, double quantum-
filtered correlation spectroscopx NOE. nuclear Overhauser effect;
NOESY, nuclear Overhaqu effect spectroscopy; RAR, retinoic acid
receptor; RXR, retinoid X receptor; TOCSY, total correlation spectros-
coPY; WT. wild-type CRABP-II; MT, CRABP-Il mutant.

 

This paper is available on line at hitp://www-ibc.stanford.edulibcl

CRABPs are not clear at. present, they are thought to be in-
volved in cellular transport and metabolism of RA (2). Two
isoforms (CRABP-I and CRABP-II) have been characterized.
Both CRABP-I and CRABP-II bind specifically to all-trans-
retinoic acid, but they differ in several respects: (1) CRABP-I
has higher affinity for RA than CRABP-II (3—6); (ii) CRABP-I is
expressed in many adult. mouse and human tissues, but the
expression of CRABP-II is limited to skin (7, 8); and (iii) RA
stimulates expression of CRABP-II but not that of CRABP-I (8).
It appears that the two isoforms may have distinct functions.
The idea is supported by the fact that. the sequence identity of
human and mouse CRABP-I (99.3%) or human and mouse
CRABP-II (93.5%) is much higher than the sequence identity
(73.7%) between the two isoforms from the same source.

CRABPs are members of a family of intracellqu lipid-bind-
ing proteins that bind small hydrophobic molecules such as
retinoids and fatty acids (9). The family of proteins includes
CRABPs, cellular retinol-binding proteins, fatty acid-binding
proteins, P2 myelin protein, a mammary gland protein, and
gastropin. The structures of 11 different intracellular lipid-
binding proteins, including CRABP-I and CRABP-II, have been
determined by x-ray crystallography (9-11). Although the se-
quence identity among intracellular lipid-binding proteins is
rather low (~20%), these crystal structures are remarkably
similar with respect to backbone folding. They are composed of
two nearly orthogonally packed five-stranded B-sheets and two
short a-helices. The ligand binding pocket in each protein is
deep inside the interior of the B-barrel formed by the two
B-sheets. The helix-turn-helix motif is located at the ligand
entrance. These proteins apparently lack a true hydrophobic
core that is important for folding and stability of many other
proteins (12).

Electrostatic interactions are thought. to play major roles in
binding of RA by CRABPs because the two proteins bind only
retinoic acid but not. retinol and retinal. No arginine residues
(Arg-lll and Arg-131 in CRABP-l; Arg-lll and Arg-132 in
CRABP-II) have been identified by crystallography to interact
with the carboxyl group of the bound RA (10). However, previ-
ous site-directed mutagenesis studies suggest that both Arg-
111 and Arg-131 are critical for CRABP-I to bind RA (13) but
that only Arg-132 is important for CRABP-II to bind BA (14).
Because of the limitations of the current methods for meas-
uring RA binding, the dissociation constants of the critical
mutants have not been measured. Furthermore, the conforma-
tions of the CRABP-I mutants have been characterized only by
CD, and the conformations of the CRABP-II mutants have not
been characterized. Thus, the contributions of the electrostatic
interactions between the arginine residues and RA to binding
are still uncertain.

We are interested in studying the quantitative structure-
function relationships of CRABPs with respect to binding of

1541

89

1542

RA. In this paper, we report a novel competitive binding assay
developed for measuring the dissociation constants of the site-
directed mutants of CRABPs. We have used the novel method
to evaluate the contributions of both Arg-lll and Arg-132 of
CRABP-II to binding of RA in conjunction with sitedirected
mutagenesis and NMR. The results show that like CRABP-I,
both Arg-lll and Arg-132 in CRABP-II are important for bind-
ing of RA, contrary to the results of Chen et a1. (14). However,
Arg-lll and Arg-132 contribute to the overall binding energy
only by ~2.2 and 1.2 kcal/mol, respectively.

EXPERIMENTAL PROCEDURES

Materials—Nonredioactive RA was purchased from Sigma. [11,12-
’H]RA was purchased from DuPont NEN. A DNA sequencing kit was
obtained from US. Biochemical Corp. Enzymes for recombinant DNA
experiments were purchased from Life Technologies. Inc., or New Eng-
land Biolabs. Other chemicals were analytical or reagent grade from
commercial sources.

Cloning and Expression—The cDN As encoding human CRABP-I and
CRABP-II were kindly provided by Dr. Anders Astrom (8). 'Ihe genes
were subcloned into the expression vector pET-17b (Novagen) by po-
lymerase chain reaction. The primers for polymerase chain reaction
were: 5’-GGAA'!'1‘CCATATGCCCAAC’I'I‘CGCCGGC-3' (forward,
CRABP-I), 5‘-CGGGATCCTCA'I'I‘CCCGGACATAAA’I'I‘C.3' (reverse,
CRABP-I), 5‘-CGGGATCCATATGCCAAAC'ITCTCTGGCAACTG-3'
(forward, CRABP-II), and 5'-GGAA'I'I‘C’I‘CAC'I‘CTCGGACGTAGAC-3'
(reverse, CRABP-II). To ensure that there were no undesired mutations
in the amplified genes, they were sequenced by double-stranded DNA
sequencing from both orientations. The expression constructs were then
transformed into the bacterial strain BL21(DE3)pLysS (15). The trans-
formed bacterial cells were grown at 37 'C in a LB agar plate containing
both ampicillin (100 ng/ml) and chloramphenicol (25 pg/ml). Expression
of CRABPs was verified by growing the colonies in 5 ml of LB medium
followed by induction with isopropyl-l-thio-R-ego' ‘ r, " and
SDS-polyacrylamide gel electrophoresis of the harvested cells.

Site-directed Mutagenesis—The oligonucleotides for making human
CRABP-II mutants were 5'-TCGTGGACCATGGAACTGACCA-3'
(RlllM) and 5'-GTGTGCACCATGGTCTACGTC-3' (R132M). The mu-
tants were generated by the method of Kunkel (16) and screened by
DNA sequencing. In order to ensure that there were no unintended
mutations in the mutants, the entire sequences of the mutated genes
were determined.

Protein Purification—All proteins were purified by the same proce-
dure as described below. 100 ml of LB medium containing 100 ug/ml of
ampicillin and 25 pg/ml of chloramphenicol was inoculated by a small
piece of a frozen seed culture and incubated at 37 'C with vigorous
shaking (200 rpm) overnight. It was then used to inoculate 4 liters of LB
medium containing both antibiotics and grown at 37 'C. When Am of
the culture reached 0.6-0.8, isopropyl-l-thio-fi-D-galactopyranoside
was added to a final concentration of 0.4 mu. The culture was allowed
to grow further at 30 'C for 3-5 h. The bacterial cells were then har-
vested by centrifugation and suspended in 100 ml of buffer A (10 mil
Tris-HCl, 1 ma dithiothreitol, pH 8.2). The suspension was aonicated on
ice and centrifuged (27,000 x g) at 4 'C for 30 min. The supernatant
was applied to a DEAE-cellulose column equilibrated with the same
buffer. The column was washed with buffer A until Am of the eluent
was less than 0.05. Elution of the column was achieved by a linear N aCl
gradient (0—200 mil in buffer A) and monitored by A,” and 15%
SDS-polyacrylamide gel electrophoresis. The fractions containing
CRABP were pooled and concentrated by an Amicon ultrafiltration cell
using a YM 10 membrane. The concentrated protein solution was cen-
trifuged (27,000 x g) for 20 min. The supernatant was loaded onto a
Sephadex G-50 column equilibrated with phosphate-buffered saline (4
mt! NaH,PO,, 16 mil Na,HPO., 150 mM NaCl. pH 7.3) and eluted with
the same buffer. The fractions containing CRAB? were located byA,.o
and 15% SD8polyacrylamide gel electrophoresis. The fractions of
greater than 99% purity were pooled and concentrated. The protein
solution was dialyzed against doubledistilled water and lyophilized.

Fluorometric Titration—Fluorescence binding assays were carried
out by a procedure modified from that of Cogan et al. (17) using a
Hitachi 4500 fluorometer. Briefly, CRABPs were dissolved in phos-
phate-buffered saline. The concentrations of the protein stock solutions
were measured by Am. The absorption coefficients used were 19,868
is“ cm" for CRABP-I and 19,480 M" cm‘l for CRABP-II determined
by the method of Gill and Hippel (18). RA stock solutiOns were prepared
in absolute ethanol. The concentrations of the RA stock solutions were

 

Cellular Retinoic Acid-binding Proteins

determined by A,“ using the absorption coefficient of 45,000 It" cm".
The final ethanol content for each titration was kept less than 2%. The
samples were excited at wavelength of 283 nm with a slit width of 2.5
nm. The emission wavelengths were 330 nm for CRABP-I and 338 nm
for CRABP-ll with a slit width of 20 nm. The excitation shutter was
closed between measurements. The inner filter effects were not cor-
rected because they were negligible at the protein and RA concentra-
tions we used in the experiments. The data were analysed by nonlinear
least square fit to Equation 1.

F 1 (F. l)x(P,+R,+K,,—JP,+R,+K,,)’-4P,R,
r,‘ i F,‘ 2?.

 

 

) (Eq. 1)

where F is the observed fluorescence, I", is the fluorescence of the bound
CRABP. F,is the fluorescence of the free CRABP. P, is the total protein
concentration, and R, is the total RA concentration.

Competitive Binding Assay—The assay was carried out in a Spectral
Por equilibrium dialyzer at room temperature. The two compartments
of each dialysis cell were separated by a semipermeable membrane with
a molecular mass cut-off of 6—8 kDa. One compartment was filled with
the wild-type CRABP-ll (1 ml), and the other was filled with a
CRABP-Il mutant or the wild-type CRABP-I. The buffer used was
phosphate-buffered saline plus 5 mil dithiothreitol. An equal amount of
[’HIRA (100 nu) was added to each compartment. The protein concen-
trations in both compartments were much greater (at least 20—fold) than
the RA concentration. 100 pl of samples was taken from the two com-
partments after various times of incubation at room temperature,
mixed with 5 ml of scintillation fluids, and counted by a liquid scintil-
lation counter. The equilibria in the two compartments that contained
the wild-type CRABP-Il and mutant proteins can be described by Equa-
tions 2 and 3 (similar equations can be written for CRABP-I and
CRABP-II).

[WTIRA]

Km " W “5“. 2’
IM'I‘ RA

Klilfl‘l . [MT F RAll (Eq. 3)

where WT. MT, WT-RA, and MT-RA represent the wild-type CRABP-II,
a CRABP-II mutant, and their RA complexes, respectively. Therefore.

’31“ .. L______M I I w I'm
Km—n [WTIMT ' RA]
Since the concentrations of the proteins were much greater than their
respective dissociation constants and the concentration of RA, [WT] ~

lw'rlm... [MT] ~ (mm... [WT'R-Al » [RA]. and [MT'RM >9 [M-
Then the relative dissociation constant can be calculated by Equation 6,

Km .. immen- (Eq 5)

ann [WTlianCirr '
where Cw... and Cm. are the measured radioactivities of the two com-
partments containing the wild-type CRABP-II and the mutant, respec-
tively. It turned out that the system could not reach equilibrium in 2
days, presumably because of few free RAs in solution to diffuse across
the membrane. Since RA is not stable even in the dark, the assay was
redesigned to match the equilibrium conditions by varying the ratio of
the protein concentrations of the wild-type and the mutant ([MTIWJ
lWT],,,.,). Thus, the concentration of the mutant was varied while
keeping the concentration of the wild type at ~2 use. Initially, the
concentration of the mutant was increased in an exponential manner
(e.g. 2. 20, 200 nM). Then it was varied in a small range. Since an equal
amount of RA was added to the two compartments of the dialysis cell,
the two compartments should have the same RA concentration and
radioactivity at the beginning of the assay. If [MT]«,uuillii‘i'l‘lh,“| e
Kama/K.,, «am: there would have a net transfer of RA across the semi-
permeable membrane separating the two compartments. Thus, the
radioactivity counts of the two compartments (CW,- and Cm) would
differ after incubation for a certain period. When [M'I‘lm/Ml‘lw <
Karim/Knurn- then er ‘ Cm > 0- When maul/masts! > Kenn-l
KltW'hv then CWT ‘ Ctr-r < 0- When mun/mural ' Kaiser/Karena.
than Cm " Cm - 0.

NMR Spectroscopy—All NMR measurements were performed at
32 'C on a Varian VXR-SOO spectrometer operating at a proton fre-
quency of 500 MHz. The proteins were dissolved in 20 mil sodium
phosphate. pH 7.5 (direct pH meter reading). 100 mu NaCl, 5 mu
dithiothreitol in D20. The protein concentration was ~2 mu. The spec-

(Eq. 4)

90

Cellular Retinoic Acid-binding Proteins

tral width was 7200 Hz. For wild-type CRABP-II, one phase-sensitive
DQF-COSY spectrum (19), two MLEV-l'l clean TOCSY spectra (20-22)
with mixing times of 20 and 40 ms and one NOESY spectrum (23. 24)
with a mixing time of 150 ms were acquired. For RlllM. one DQF-
COSY spectrum, three TOCSY spectra with mixing times of 20, 40. and
75 ms and one NOESY spectrum with a mixing time of 150 ms were
acquired. For R132M. one TOCSY spectrum with a mixing time of 40 ms
and one NOESY spectrum with a mixing time of 150 ms were acquired.
All of the spectra were acquired in the hypercomplex mode with stand-
ard phase cycling schemes. The data were usually acquired with 2048
complex points in the t2 dimension and 256 complex points in the t1
dimension. 96 transients were collected for each FID. Data processing
was performed on a Sun Sparc 10 station using VNMR software from
Varian. The time domain data were zero-filled once and multiplied by
shined sine bell or Gaussian functions before Fourier transformation in
both dimensions. Base lines were corrected in t2 dimensions using a
5-order polynomial. Chemical shifts were referenced to internal sodium
3-(tn'methyl silyl)-propionate-2.2.3.3-d..

RESULTS

Fluorometric Titration—Probably because of the simplicity
of the method. fluorometric titration has been frequently used
for measuring the affinities of CRABPs for RA. However, there
are large discrepancies in the K., values obtained by the
method. These could be due to variations in the assay method-
ology and/or conditions. Some problems with the method have
been recently discussed (2. 4. 6). In order to compare the ligand
binding properties of CRABPs under the same conditions, we
carried out fluorometric titration experiments. The results are
shown in Fig. 1. The apparent K, values obtained by nonlinear
fitting were 0.63 nu for CRABP-I and 1.9 nu for CRABP-II.
While our work was in progress, Norris et al. (6) reported the
apparent Kd values of mouse CRABPs measured by an im-
proved fluorescence titration method. The major changes were
the lowering of protein and RA concentrations and the use of
nonlinear least square fitting for data analysis. These modifi~
cations were in line with our approach. Our fluorometric titra-
tion data were analyzed by nonlinear least square fitting. The
protein concentrations we used in the assays (~50 nM) were the
lowest among those used in previous studies except those of
Norris et al. (6). We did not include the additive gelatin used by
Norris et al. (6) in the assays. The apparent Kd values deter-
mined by us are in close agreement with those (0.4 all for
CRABP-l and 2 nu for CRABP-II) reported by Norris et al. (6).

Competitive Binding Assay—In order to measure the affini-
ties of the mutants of CRABPs for RA. we developed a novel
competitive binding assay. We first used the method to meas-
ure the relative affinity of CRABP-I and CRABP-II for RA. The
result is shown in Fig. 2A. When the ratio of the concentrations
of the two proteins ([CRABP-IH/[CRABP—ll) was 52, there was
a not transfer of RA from the compartment containing
CRABP-ll to the compartment containing CRABP-I. When the
ratio of the concentrations of the two proteins was 23, there
was a net transfer of RA from the compartment containing
CRABP-I to the compartment containing CRABP-ll. As de-
scribed under “Experimental Procedures,‘ the relative K., of the
two proteins lies between the points with opposite net trans-
fers. Thus the Kd of CRABP-II relative to that of CRABP—I
(K.,(CRABP-IIVKJCRABP-D) is 2—3. It is in close agreement
with the results of the fluorometric titration studies. We then
used the competitive binding assay to measure the K., values of
RlllM and R132M relative to that of the wild-type CRABP-II.
The results are shown in Fig. 2 (B and C). In the case of R111M
(Fig. 28), there was a net transfer of BA from the compartment
containing RlllM to the compartment containing the wild-
type CRABP-II when [R111Ml/[WT] s 40. When [R111MIIIWTl
z 45. there was a net transfer of BA from the compartment
containing the wild-type protein to the compartment contain-
ing R111M. Therefore. the relative affinity of R111M for RA

 

 

 

 

 

 

 

 

1543
1.0- o.._ A CRABP-l
a: ‘
8 he
m 03-
m o.
5
2 0.6-
E .
1.3 0.44 -.
a are-
a: ~xe-.._-..______.
0.2-
6 2'0 ' 4'0 ' e’o ' 33 T 100
Retinoic Acid (nM)
1.0-I 0
\. B CRABP-Il
5 03- "
0
m
9 ‘ 'b.
O
g os~
o ‘ in“
.. 0.4-4 “N...
é ‘ k“.“~— .--—._.D
02 1 V I " I V I ' I ' I
O 20 40 60 80 100
Retinolc Acid (nM)

Fla. 1. Fluorometric titration of wild-type CRABP-l (A) and
CRABP-II (B). The concentrations of CRABP-l and CRABP-II were 40
and 60 ml. respectively.

(Kd‘RI‘lu/de is 40-45. As shown in Fig. 20, the relative
affinity of R132M for RA (Kama/Kay") is 6—8.
Sequential Resonance Assignment of the Ring Protons of the
Aromatic Residues of the Wild-type CRABP-II and Site-directed
Mutants—Since perturbations in the ligand binding property of
a mutant may be due to conformational changes caused by the
mutation (25), we characterized the conformations of the mu-
tants by NMR. A prerequisite for NMR structural analysis is
sequential resonance assignment. Currently, we are in the
process of making total sequential resonance assignments of
the wild-type CRABP-Il and R111M by multidimensional
multinuclear NMR. We have sequentially assigned the ring
protons of aromatic residues of the wild-type and mutants
based on the homonuclear two-dimensional NMR (DQF-COSY.
TOCSY, and N OESY) and heteronuclear three-dimensional
NMR.3 The chemical shifts of the ring protons are listed in
Table l. The sequential resonance assignment is labeled in Fig.
3 for R132M. In comparison with the chemical shifts of the
wild-type ring protons, there are quite a few chemical shift
changes in the ring protons of the mutants. Some chemical
shift changes can be easily rationalized, while the causes for
other chemical shift changes are not obvious. For example, the
guanidino group of Arg-132 stacks on the aromatic ring of

 

' L. Wan. Y. Li, and H. Yan. unpublished results.

91

Cellular Retinoic Acid-binding Proteins

 

 

 

200+

. 4004

40m

 

 

 

 

 

 

 

 

 

 

 

 

[CRABP-lMCRABP-l]

[mural/[camera

[R132M1/[CRABP-ll]

Fla. 2. Competitive binding assays for measuring the relative dissociation constants of CRABP; The relative radioactivity inA is the
radioactivity count of the compartment containing CRABP-l minus that of the compartment containing CRABP-II. The relative radioactivity in B
is the radioactivity count of the compartment containing CRABP-ll minus that of the compartment containing R111M. 'flie relative radioactivity
in C is the radioactivity count of the compartment containing CRABP-Il minus that of the compartment containing R132M.

Tan
Chemical shifts of the ring protons of the aromatic residues

as I
of wild-type CRAB? II. RIIIM mutant. and R132M mutant

The underlined values are the mutant resonances that differ by >0.02 ppm from the corresponding resonances of the wild-type CRABP-II.

 

 

 

Residue mm”. ems-n sum 3132M

Pin-3 7.60‘ 7.34‘ 7.152‘ ‘L§_1‘ g 7.46! 7.60- my 7.52‘

Pile-15 7.42- 7.02‘ 7.11‘ 7 39- LB" 110‘ 7.40- 1._1_1‘ 7.11c

Pine-50 6.66' 7.26‘ 6.99‘ 6.68' 7.24‘ 6.99‘ 6.64- 7.240 6.99'

Phe-65 6.42- 6.70‘ 6.48‘ §-3_8‘ 6.70‘ 6.44‘ 6.42‘ 6.69“ w

Phe-‘ll 7.48‘ 7.00‘ 6.70‘ 7 44- 6.99‘ m 7.48‘ 7.00‘ 6.69‘

I‘m-.51 7.12‘ 6.79' 7.124 6._74_r 7.134 6.78‘

Tyr-134 6.98“ 6.65‘ 7.00!' §.8_6‘ 7.034 w-

Trp-‘l 7.25! 7.00! 7.41‘ 7.30‘ 7_._2_2_’ 6.99! 7.39‘ 7.29" 7.25! 6.99! 743* 730‘
W7 8.16’ 7.13! 7.39‘ 7.80‘ s 16’ M 7.39‘ 116‘ 8.17’ 19;! 7.39“ 7.80‘
Trp-B'l" 8.07’ 7.05! 7.33“ 7.76‘
Trp-109 7.0V 6.63! 6.61“ 6.87‘ 'L_l’ gm! 6.59" 6.89‘ 7.09’ 6.62! 6.61“ 6.87‘

 

'" 2.68. 3.514. and 4B of phenylalanine. respectively.
"‘ 2.6!! and 3.5}! of tyrosine. respectively.
"‘ 4H. 5H. 6H. and 7B of tryptophan. respectively.

’ Tip-87 has two sets of resonances because of conformational heterogeneity.

Phe-15 and is hydrogen-bonded to the hydroxyl group of 'lyr-
134 in the crystal structure of the wild-type CRABP-II (10).
Removal of the guanidino group is likely to be the cause for the
chemical shift changes in Phevl5 and Tyr-134 of R132M. The
crystal structure also reveals that the guanidino group of Arg-
111 interacts with Trp-109 through an amide/aromatic-ring
hydrogen bond (10). Thus. removal of the guanidino group of
Arg-lll may cause the chemical shift changes in M109 of
R111M. Some small chemical shift differences may result from
changes in pH. temperature. and other conditions. They are not
necessarily indicative of conformational changes. Conforma-
tional changes are best characterized by analysis of NOESY
spectra as described below. Interestingly. 4H and 7H of the
Tip-87 ring protons in R132M have two sets of NMR signals
(Fig. 3). indicating that there is localized conformational het-
erogeneity and that the 'hp ring is in two distinct conforma-
tional states. The lifetimes of the two conformational states are
at least 50 ms. The ratio of the pepulations in the two states
(major and minor forms) is about 4:1. 1H of the Tip-87 ring
protons in the wild-type protein and RlllM also show two
distinct resonance signals (data not shown). indicating that the
two proteins also have localized conformational heterogene-
ities. The lifetimes of the two conformational states are at least
40 ms in the wild-type protein and at least 33 ms in R111M.
The ratio of the major and minor forms is about 3:1 in the
wild-type protein and about 4:1 in RlllM. Conformational
dynamics may be important for RA to get into the deep binding
pocket of the protein. However. Tip-87 is not at the RA-binding

pocket and is far away from the ligand entrance. The functional
significance of the conformational heterogeneity is not clear.

Conformational Comparison of the Wild-type CRABP-II.
RIIIM. and R132M—The aromatic-aromatic NOEs are shown
in the lower panels of Fig. 4 (wild-type CRABP-II). Fig. 5
(R111M). and Fig. 6 (R132M). Some aromatic-aliphatic NOEs
are shown in the upper panels of the figures. Interresidue
NOEs are labeled. because they are indicative of the conforma-
tion of a protein. The interresidue NOE peaks of the three
proteins are the same except for some minor variations in peak
intensity. The observed NOEs are also consistent with the
atomic distances measured from the coordinate of the crystal
structure of CRABP-II complexed with RA. The identities of the
aliphatic protons shown in the upper panels of these figures are
not known at present. However. qualitatively. the three pro-
teins have very similar aromatic-aliphatic NOE patterns. The
results of the N OESY analysis suggest that the conformations
of R111M and R132M are highly similar to that of the wild-type
protein. Thus. the decreases in affinity for RA of the two muo
tants are most likely due to the disruption of the interactions of
the arginine residues with RA by mutagenesis. Arg-lll and
Arg-132 contribute to the binding energy by ~2.2 and 1.2
kcal/mol. respectively.

DISCUSSION

Competitive Binding Assay—Two types of methods have
been in general use for measuring binding of RA to CRABPs:
fluorometry and radiometry. The radiometric method involves

92

Cellular Retinoic Acid-binding Proteins

 

 

 

 

 

, "I . . .. .
‘I . 4 'l . 0
n (on)
FIG. 3. TOCSY spectrum of the aromatic protons of R132M at
32 '0 showing two sets of resonances for e mixing time

was 40 ms. All of the cross-peaks have been sequentially assigned. The
chemical shifts of the resonances are listed in Table 1.

separation of bound from free RA by dextran—coated charcoal.
gel filtration, and other means. Substantial loss of bound li-
gand during the separation process makes the method unsuit-
able for measuring the dissociation constants of site-directed
mutants with greatly decreased affinity for RA. The very lim-
ited solubility of RA in water (~200 nM) (26) also makes the
fluorometric method inapplicable for determining the dissocia-
tion constants of these mutants. Studies of the quantitative
structure-function relationships of CRABPs have been ham-
pered by the lack of methods for measuring the affinities of
site-directed mutants for RA (13, 14). We have developed a
novel competitive binding assay for measuring the affinities of
site-directed mutants for RA relative to that of the wild-type
CRABP. The essence of the method is to monitor the competi-
tion between a mutant and the wild-type protein for binding of
limited RA. The method uses an equilibrium dialyzer. The two
compartments of each dialysis cell are filled with the wild-type
and mutant proteins. respectively. The absolute concentration
of RA is not important as long as the concentration of free RA
is much smaller than that of bound RA. There is no need to
separate bound from free RA by dextran-coated charcoal, gel
filtration. and other means. The transfer of RA fi'om one com-
partment to the other is determined by measuring the radio-
activities of the samples taken from the two compartments.
The direction of the net transfer is dependent on the relative
affinity of the proteins and the ratio of the protein concentra-
tions of the two compartments. If the solubility of a mutant is
the same as that of the wild-type CRABP-II (~2 mM). then the
ratio of the concentrations of the wild-type and mutant proteins
can be varied by more than 1.000-fold. and a relative K.,, of as
large as 1000 (Kd(m/Kd (w-n = 1000) can still be measured by
the method. More importantly. determination of the relative
dissociation constant of a point mutant is sufficient for estimat-
ing the energetic contribution of the amino acid residue to
ligand binding (AAG= RTln(Kd (MT/KtHWT)

In addition to vitamin A, many other hydrophobic molecules
such as vitamin D and steroid hormones play vital roles in a
variety of cellular processes. Studies of the stmcture-function
relationships of the receptor proteins for these bioactive hydro—
phobic molecules are of fundamental interest and biomedical

1545

 

 

 

 

0.: 0.0 7.0 7.6 7.4 7.2 7.0 6.6 6.6 6.6 6.2
'2 (vs-I)

 

I rout

 

 

 

 

I I I I I I I I l I
s.z 6.6 7.s 7.6 7.: 7.: 7.0 6.0 6.6 6.6 6.:
n (vs-I)

FIG. 4. Parts of the two-MHz NOESY spectrum of the wild-type
CRABP-II at 32 ‘C. The mixing time was 150 ms. Only the interresi-
due NOE: are labeled. The identities of the NOEs are as follows. A,
Phe-65 2. 6H to Phe-71 2, 6H; 8 Phe-65 2. 6H to Phe-71 3. 5H; C. Rho-65
4H to’l‘rp-1096;H D. 4Phat-35 3.6H99to'l‘I-p~106;H E. Phe-BO 2.6Ht0

87 7H;F,Phe-71to-’I‘rp 109 7H; G. Phe-65 4H to Tip-109 7H;
H, Phe-65 2. 6H to 'l‘rp-109 7H; 1. Phe-60 3. 5H to Phe-65 4H; J. 'l‘rp-8'I
5H to Phe-3. ~
Trp-;877HM.Phe-504Hto'l‘rp— 877H ;.N
Phe-50 2. 6H to 'I‘rpv87 6H; P. Phe-50 2. 6H to [I‘ll-2.611; Q. 'h'p-S‘ISH
Phe-3 2. 6H; R. Phe-S 4H to ’I‘rp-87

significance. The method can also be used for measuring the
relative dissociation constants of the site-directed mutants of
these important proteins and may facilitate these studies.
Relative Afl'inity of CRABP-I and CRABP-II for RA—
CRABP-I and CRABP-II differ in affinity for RA, expression
pattern, and regulation. Although it is generally accepted that
CRABP-I has a higher affinity for RA than CRABP-II. it has
been difficult to determine precisely the relative affinity of the
two proteins. A wide range of apparent Kd values have been
reported. The apparent K., values for purified CRABP-I are in
the range of 0.4-39 nM (4—6, 13. 27-29). The apparent K.,
values for purified CRABP-Il are in the range of 2-64 nu (3-6,
14) The best estimation of the relative affinity of the two
proteins for RA was made by Fiorella et al. (5) In their study,
RA was first incubated with CRABP- II and then mixed with
CRABP-I I. The protein RA complexes (CRABP-I- -RA and
CRABP-II~RA) were separated by chromatofocusing. The result
indicated that CRABP-II has about 3-fold lower affinity for RA
than CRABP-I. More recently. Norris et al. (6) compared the
affinities of CRABP-I and CRABP-II for RA by fluorometric
titration under the same conditions. The results showed that

93

 

 

   
 
    

1546
2.1 ° 3"? l '
£23 0 fit as no
5232 a: $9.63? .. ~
3:? v o 0 so?» 0 ° .

 

0.: 0.0 7.0 7.6 7.0 7.: 7.0 6.0 s.s 0.0 6.:
n (0000)

 

 

 

' uuuuu ' I ' i""‘ I |

0:: I 0.0 1.0'1it '74 '7.0 1.0 6.0 I0.‘
I: (00-)
FIG. 5. Parts of the GOO-MHz NOESY spectrum of RlllM at

32 ‘C. The mixing time was 160 m0. Only the interresidue N020 are
labeled. The identities of the N030 ars given in the legend to Fig. 4.

0.0 0.3

CRABP-I has 4~fold higher affinity for RA than CRABP-II
(although a 10-fold stronger affinity of CRABP-I for RA. com-
pared with CRABP-II. was stated in the paper. apparently due
to an error in calculation). The apparent dissociation constant
or CRABP-II relative to that Of CRABP-I (K4 (cmp.u/
Kamp.n) is 3, based on our fluorometric measurements. The
results of our competitive binding assays indicated that the
relative apparent dissociation constant of CRABP—II
(K4 (CRABPoll/Kd(cmP-D) is “.1118 mats Of our fluorometric
measurements and competitive binding studies are in close
agreement. They are also in line with the results of Fiorella et
al. (5). The relative affinity of CRABP-I and CRABP-II meas-
ured by Fiorella et al. (5) may be the upper limit. because some
RA bound to CRABP-Il could be preferentially lost during
separation of the protein'RA complexes by chromatofocusing.
The difference in affinity for RA between CRABP-I and
CRABP-II is rather small (1—2-fold).

Both Ami-111 and Amt-132 of CRABP-II Are Important for
Binding of RA—Arg-lll. Arg-132. and 'l‘yr-134 of CRABP-II
have been predicted to be involved in binding of RA on the basis
of homologous modeling (13, 30, 31). The crystal structure of
CRABP-II in complex with RA reveals that one of the carbox-
ylate oxygens of RA is hydrogen-bonded to the guanidino group
Arg-132 and the hydroxyl group of 'l‘yr-134. and the other
carboxylate oxygen of RA interacts with Arg-lll via a water
molecule (10). We have substituted Arg-lll and Arg-l32 with
methionine. Our biochemical and structural characterizations
show that guanidino groups ot'Arg-lll and Arg-132 contribute
to the binding energy by ~2.2 and 1.2 kcal/mol, respectively.
The results indicate that both Arg-lll and Arg-132 are indeed

Cellular Retinoic Acid-binding Proteins

 

 

 

 

0.3 0.0

'2 (son)

 

 

 

 

     

- L——.-___..-.-- ..._. - _.--.————- _-

0.0 0.0 1.0 1.0 1.0 1.3 1.0 0.0 0.0 0.0 0.3

'2 (0’)

FIG. 6. Parts of the mun: NOESY spectrum of R132M at
32 'C. The mixing time was 150 m0. Only the interresidue NOEs are
labeled. The identities of the NOEs are given in the legend to Fig. 4.

involved in binding of RA and that Arg-lll contributes more to
the binding energy than Arg-132. Fluorometric titration also
suggests that R132M has higher afiinity for RA than RlllM
(data not shown). Zhang et al. (13) have replaced the corre-
sponding residues in CRABP-I with glutamine (RlllQ and
R131Q). Quantitative comparison of our results with those of
Zhang et al. (13) is not possible, because the dissociation con-
stants of their mutants have not been determined. Qualita-
tively, the ligand binding properties of our mutants are similar
to those of the corresponding CRABP-I mutants. Zhang et al.
(13) observed that both the wild-type CRABP-I and R131Q can
induce CD of RA in the wavelength range between 320 and 360
am. but R111Q cannot induce the CD signal. The results of the
CD experiments with CRABP-I are consistent with our meas-
urements of the energetic contributions of Arg-lll and Arg-132
to the binding of RA. Chen et al. (14) have mutated Arg-lll of
CRABP-II to alanine (R111A) and Arg-132 of CRABP—II to
alanine and glutamine (R132A and R132Q). In contrast to our
R11 1M mutant, the apparent dissociation constant of their
R111A mutant is very similar to that of the wild-type
CRABP-II (only 50% higher than that of the wild type). The
causes of the discrepancy are not clear, but several possibilities
can be ruled out. (i) It is unlikely that there are mutations other
than substitution of Arg-lll with a methionine in our RlllM
mutant. We have sequenced the entire gene after mutagenesis
and subcloning into the expression vector. (ii) Steric hindrance
is unlikely to be the cause. because the side chain of methionine
is smaller than that of arginine. (iii) Although small conforma-
tional changes cannot be ruled out. NMR characterizations
suggest that RlllM is properly folded, and its conformation is

94

Cellular Retinoic Acid-binding Proteins

very similar to that of the wild-type protein. Furthermore,
RlllM has been crystallized. The preliminary crystal struc-
ture at 2.2-A resolution shows that the structure of RlllM is
indeed very similar to that of the wild-type protein.3 The ligand
binding property of our R132M mutant is qualitatively similar
to those of the R132A and R132Q generated by Chen et al. (14).
However, quantitative comparison is again not possible, be-
cause the apparent dissociation constants of the two mutants
have not been determined.

The results of our mutagenesis studies appear to be partly
inconsistent with the crystallographic data. The crystal struc-
ture of holo~CRABP-II suggests that Arg-132 interacts directly
with the carboxyl group of RA. whereas the interaction between
Arg-lll and the carboxyl group of RA is mediated by a water
molecule (10). Our results agree with the crystallographic ob-
servation that both Arg-lll and Arg-132 are involved in bind-
ing of RA. However, in contrast to the crystallographic data,
our mutagenesis results indicate that Arg-lll is more impor-
tant than Arg-132 for RA binding. But this is not necessarily
inconsistent with the crystallographic studies. Since Arg-132
and Tyr-134 are hydrogen-bonded to the same carboxylate ox-
ygen of RA, the loss of the hydrogen bond between Arg-132 and
RA in R132M may be partially compensated by the interaction
between Tyr-l34 and RA. On the other hand, the other carbox-
ylate oxygen of RA only interacts with Arg-lll (albeit via a
water molecule), the loss of the interaction in RlllM may not
be compensated. Alternatively, in solution, the position of the
carboxyl group of the bound RA may be slightly different from
that observed in the crystals. It is noted that the RA binding
cavity is much larger than necessary to accommodate the li-
gand, and a large portion of the deep binding cavity is not
occupied by RA. It appears that the fi-ionone ring of the bound
RA is well fixed, but the isoprene tail and carboxyl group have
room to move. The guanidino group of Arg-lll could be closer
to the carboxyl group of RA than to that of Arg-132 in solution.
Solution NMR study of holo-CRABP—II is in progress.

CRABPs have very stringent retinoid specificity (2). The
proteins bind only RA and reject both retinol and retinal. On
the other hand, cellular retinol-binding proteins bind both ret.
inol and retinal and reject RA. Amino acid sequence analysis
reveals that the two conserved arginine residues in CRABPs
are replaced with glutamine in cellular retinal-binding pro-
teins. It has long been thought that the electrostatic interac-
tions between the guanidino groups of the two conserved argi-
nine residues and the carboxyl group of RA play major roles in
binding of RA (31). Our mutagenesis studies indicate that
Arg-lll and Arg-132 are indeed involved in RA binding, but

 

’ Ji, K., L. Wang, Y. Li, and H. Yan, unpublished results.

1547

the electrostatic interactions between the guanidino groups of
Arg-lll and Arg-132 and the carboxyl group of RA contribute
to the overall binding energy only by ~2.2 and 1.2 kcal/mol.
Thus, amino acid residues that interact with the hydrophobic
moiety of RA may be more important for binding of RA.

Acknowledgment—We are indebted to Dr. Anders Astrom for provid-
ing the cDNA clones of human CRABP-I and CRABP-II. We thank Drs.
David G. McConnell and Yanling Zhang for critical reading of the
manuscript and Dr. Xinhua J i for useful discussion.

REFERENCES

. Mangelsdorf, D. J., Umesono, K, and Evans. R. M. (1994) in The Retinoids:
Biology, Chemistry, and Medicine (Sporn, M. 8., Rorberts, A. 3., and Good-
man, D. S., eds) pp 319-349, Raven Press, New York

2. Ong, D. E., Newcomer, M. E.. and Chytil, F. (1994) in The Retinoids: Biology,
Chemistry, and Medicine (Sporn, M. 8., Rorberts. A. 3.. and Goodman. D.
S., eds) pp 283-317, Raven Press. New York

. Bailey, J. S., and Siu, C.-H. (1988) J. Biol. Chem. 203. 9326-9332

. Fogh, K. Voorhees, J. J., and Astroin (1993) Arch. Biochem. Biophys. soo,
751-755

5. Fiorella, P. D.. Giguere, V., and Napoli, J. L (1993) J. Biol. Chem. 298.
21545-21552

. Norris, A. W., Cheng, L. Gigubre, V., Rosenberger, M., and Li. 3. (1994)
Biochim. Biophys. Acta 1209. 10-18

. Giguere, V., Lyn, S., Yip, P., Siu, C.-H., and Amin, S. (1990) Proc. Natl. Acad.

Sci. U. S. A. 87. 6233- 6237
Astrom, A., Tavakkol, A., Pettersson, U., Cromie, M., Elder, J. T., and Voo-
rhees, J. J. (1991) J. Biol. Chem. 206, 17662-17666
. Banaszak, L, Winter, N., Xu, 2.. Bernlohr, D. A., Cowan, S. W., and Jones, T.
A. (1994)Adv. Protein Chem. 45. 89-151
10. Kleywegt, G. J., Bergfors, T., Senn, H., be Motto. P., Gsell, 8.. Shudo. K., and
Jones, T. A. (1994) Structure 2. 1241-1258

11. Thompson, J. R, Bratt, J. M., and Banesuk. J. (1995) J. Mol. Biol. 852.
433—446

12. Scapin, G., Gordon, J. 1., and Sacchettini, J. C. (1992) J. Biol. Chem. 987.
4253-4269

13. Zhang, J., Liu, Z.-P., Jones, T. A., Gierasch, 1.. M., and Sambrook, J. I". (1992)
Proteins Slrucl. Funcl. Genet. 13. 87-89

14. Chen, L. X., Zhang, Z.-P., Scafonas, A., Cavalli, R. C., Gabriel, J. L, Soprano,
K. J., and Soprano, D. R. (1995) J. Biol. Chem. 270, 4518-6525

15. Studier. P. W., Rosenberg, A. K., Dunn. J. J., and Dubendorfl'. J. W. (1987)
Methods Enzymol. 1985. 60-89

16. Kunkel, T. A. (1985) Proc. Natl. Acad. Sci. U. S. A. 92. 488-492

17. Cogan, U., Kopelman, M., Mokady, S., and Shinitzky, M. (1976) Eur. J. Bio-
chem. “. 71-78

18. Gill, 8. C., and Rippel, P. T. ( 1989) Anal. Biochem. 193, 319-326

19. Piantini, 0., Sorensen, O. W., and Ernst, R. P. (1982) J. Am. Chem. Soc. 10‘.
6800-6801

20. Braunschweiler, L, and Ernst. R. P. (1983) J. Magn. Reson. 53. 5521-5528

21. Bax, A., and Davis, D. V. (1985) J. Magn. Reson. 86, 355—360

Griesinger, C.. Otting, G., Wuthrich, K., and Ernst, R. P. (1988) J. Am. Chem.

Soc. 110, 7870-7872

Jenneer, J., and Ernst. R. P. (1979) J. Chem. Phys. 71. 4548-9653

Macura, S., and Ernst. R. P. (1980) Mol. Phys. ‘1. 95—117

Tsai., M.-D., and Yan. H. (1991) Biochemistry 30. 6806-6818

Szuts, E. Z., and Harosi, F. l. (1991) Arch. Biochem. Biophys. 287. 297-304

Daly, A. K., and Redfem, C. P. P. (1988) Biochem. Biophys. Acta 995. 118-128

Fiorella, P. D., and Napoli, J. L. (1991) J. Biol. Chem. 906. 16572-16579

Ong, D. 8., and Chytil, P. (1978) J. Biol. Chem. 253, 4551-‘554

Cowan, S. W., Newcomer, M. S., and Jones, T. A. (1993) J. Mol. Biol. 8”.

1225-1246

. Jones, T. A., Bergfors, T., Sedzik, J., and Unge, T. (1988) EMBO J. 7.

1597-1604

b“ D‘

Q“

g

8883383?

a
one

CHAPTER 3
Solution Structure of Type-ll Human Cellular

Retinoic Acid Binding Protein by NMR

Introduction

One of the key questions concerning the family of iLBPs including CRABPs is
how the ligands get in and out of their respective binding pockets, since they are
encapsulated by the B-sheets and helix-turn-helix motif in most cases. The crystal
structures of holo-CRABPI and holo-CRABPII (both in complex with all-trans-RA) have
been determined (Kleywegt et al., 1994). In each structure, RA is buried deeply in the
RA-binding pocket, with its carboityl group interacting with the side-chains of two
arginines and one tyrosine at the bottom of the pocket. The B-ionone ring which is fixed
snugly at the entrance of the RA-binding pocket with only one edge of the ring accessible
to the solvent, is forced to adopt an unusual cis—like conformation. It appears that RA
cannot enter or exit the deep binding pocket in the absence of major conformational
changes in the protein.

More recently, the crystal structure of apo-CRABPI has been solved (Thompson

95

96

More recently, the crystal structure of apo-CRABPI has been solved (Thompson
et al., 1995). The ligand entrance of apo-CRABPI is slightly more open than that of holo-
CRABPI and thus a bit more accessible to RA. Surprisingly, apo-CRABPI is a dimer in
the crystalline state, held together by an intermolecular B-sheet. Dimerization has been
suggested to be the mechanism that allows RA to enter or exit the RA-binding pocket and
by which CRABPI interacts with RA-metabolizing enzymes such as cytochrome P4503.
However, it is unclear whether apo-CRABPI is dimeric in solution. CRABPI has been
studied by NMR (Rizo et al., 1994), but no solution NMR structure has been reported.

In this chapter, the solution structure of human apo-CRABPII determined by
NMR spectroscopy is described. The sequential assignments of the 1H, 13C and ”N
resonances of apo-CRABPII were established by multinuclear multidimensional NMR
experiments. The solution structure of apo-CRABPII was derived from 2,382
experimental NMR restraints. The solution structure of apo-CRABPII is similar to the
crystal structure of holo-CRABPII, but significant conformational differences were
observed, especially in the ligand entrance region. In comparison to holo-CRABPII, apo-
CRABPII exhibits a concerted movement of the second helix, the BC-BD loop and the
BE-BF loop, so that the ligand entrance of apo-CRABPII is greatly enlarged and readily
accessible to RA. Furthermore, the ligand-binding pocket of apo-CRABPII is rather
disordered. The results suggest that binding of RA induces significant changes in the

conformation and dynamics of CRABPII.

97

Experimental Procedures

Sample preparation. ['3C] Db-glucose, ISNH4C1, ['SN]-leucine and ['SN]-va1ine
were purchased from Cambridge Isotopic Laboratories. Unlabeled amino acids and other
compounds were purchased from Sigma (St. Louis, Missouri).

Human CRABPII gene was over-expressed and the unlabeled protein was purified
as detailed in Chapter 2. The same procedure was used for isotopic labeling of CRABPII
except for some minor modifications as described below. M9 medium was used for the
uniform isotopic labeling of CRABPII with 15NH4C1 and ['3C] Dé-glucose as the sole
nitrogen and carbon sources for 15N- and 13C~-labeling, respectively. For 15N--labeling of
leucine and valine residues, a rich medium containing [”NJ-valine or ['SN]-leucine was
used (Muchmore et al., 1989). The E. coli strain DL49PS pLysS (kindly provided by Dr.
David M. LeMaster) was used for the selective isotopic labeling. D20 samples were
prepared by dissolving lyophilized unlabeled CRABPII in PBS buffer (20 mM sodium
phosphate, 150 mM sodium chloride in 99.6% D20, pD 7.5, uncorrected). H20 samples
were prepared by directly concentrating CRABPII fractions eluted with the PBS buffer
(pH 7.3) from a gel-filtration column at the final step of the purification procedure
(Chapter 2). About 10% D20 was added to the concentrated protein solutions. The protein
concentrations of the NMR samples were ~2 mM.

NMR spectroscopy. All NMR spectra were acquired at 25°C. Typical carrier
frequencies for various experiments were as follows: lH, 4.70 ppm; l5N, 119.7 ppm;

”CO, 175 ppm; 13C“, 54 ppm and 13CW", 43 ppm. Quadrature detection in the

98

nonacquisition dimensions was achieved by the hypercomplex method (States et al.,
1982) for homonuclear experiments and by the States-TPPI method (Marion et al., 1989a)
for heteronuclear experiments.

Homonuclear 2D spectra were recorded on a Varian IN OVA-600 spectrometer.
The acquisition times and numbers of complex points were as follows: DQF-COSY
(Piantini et al., 1982) F 1 61.0 ms, 512 and F2 243.8 ms, 2048 (128 transients); NOESY
(Jeener et al., 1978; Macura & Ernst, 1980; Kumar et al., 1980) F1 38.1 ms, 320 and F2
243.8 ms, 2048 (64 transients); SSNOESY (Smallcombe, 1993) and a WATERGATE
(Piotto et al., 1992) NOESY, Fl 61.0 ms, 512; F2 243.8 ms, 2048 (64 transients and 150
ms mixing time); clean-TOCSY (Griesinger et al., 1988) F1 38.1 ms, 320 and F2 243.8
ms, 2048 (64 transients). Two TOCSY spectra were acquired, one with a mixing time of
30 ms and one of 50 ms. Two NOESY spectra were recorded, one with a mixing time of
100 ms and one of 150 ms. The data were processed with the program VNMR v. 5.2F
(V arian Associates). A gaussian and minus LB combination was applied in F2 dimension
and a ~75°~shified sine-bell in F1 dimension in processing NOESY and TOCSY spectra.
A ~30°-shificd sine-bell was applied in both F1 and F2 dimensions in processing of
DQF-COSY data. The data sets were zero-filled to 4096 x 2048 real points for TOCSY
and NOESY and 8192 x 4096 real points for DQF-COSY. In general, a five-order
polynomial was applied for baseline correction in F2 dimension after Fourier
transformation.

Heteronuclear double resonance NMR spectra of uniformly ”N labeled protein
were recorded with the following acquisition times and complex data points: 2D lH-‘sN

HSQC (Bodenhausen & Ruben, 1980; Kay et al., 1992; Zhang etal., 1994), 'H (F2) 121.9

99

ms, 1024, ”N (F1) 53.3 ms, 128 (32 transients); 3D lH-”N NOESY-HSQC (Zhang et al.,
1994), 1H (F3) 141.5 ms, 1024, ”N (F2) 34.0 ms, 68, 1H (F1) 33.2 ms, 200 (16 transients,
150 ms mixing time). 2D lH-”N HMQC spectra (Miieller, 1979; Bax et al., 1983) of the
selectively labeled proteins ([”N]-valine or [”N]-leucine) were recorded with the
following acquisition times and complex data points: lH (F2) 141.5 ms, 1024 and (F1)
”N 32.0 ms, 64 (32 transients). Heteronuclear triple resonance NMR spectra (Kay et al.,
1990; Grzesiek & Bax, 1992a; Muhandiram & Kay, 1994; Kay, 1995) of ”N/”C-doubly
labeled protein were recorded with the following acquisition times and complex data
points: CT-HNCO (Grzesiek & Bax, 1992a), ”N (F1) 16.0 ms, 32, ”CO (F 2) 34.0 ms, 64,
lH (F3) 127.80 ms, 1024 (16 transients); CT-HNCA and CT-HN(CO)CA (Grzesiek &
Bax, 1992a), ”N (Fl) 16.0 ms, 32, l3CCl (F2) 10.9 ms, 64, 1H (F3) 127.80 ms, 1024 (16
transients); HNCACB (Wittekind & Miieller, 1993; Muhandiram & Kay, 1994), ”0""
(Fl) 7.1 ms, 64, 15N (F2) 15.2 ms, 32, 1H (F3) 128.0 ms, 1024 (32 transients);
CBCA(CO)NH (Grzesiek & Bax, 1992b; Muhandiram & Kay, 1994), ”CM3 (Fl) 6.4 ms,
64, ”N (F2) 16.0 ms, 32, lH (F 3) 127.80 ms, 1024 (32 transients); C(CO)NH(G1'zesiek et
al., 1993; Muhandiram & Kay, 1994), 13c (Fl) 9.7 ms, 96, ”N (F2) 15.2 ms, 32, ‘H (F3)
128.0 ms, 1024 (32 transients); H(CCO)NH (Grzesiek et al., 1993; Muhandiram & Kay,
1994), IH (Fl) 15.9 ms, 128, l5N (F2) 25.0 ms, 50, 1H (F3) 127.80 ms, 1024 (32
transients); gradient HCCH-TOCSY (Bax etal., 1991; Kay et al., 1993), 1H (F1) 35.8 ms,
128, 13C (F2), 10.5 ms, 75, 1H (F3) 127.80 ms, 1024 (16 transients, 16 ms mixing time).
A low power GARP-I sequence (Shaka et al., 1985) was applied for ”N broad-band
decoupling during the data acquisition. 'H broadband decoupling in the CT-HNCO, CT-

HNCA, CT-HN(CO)CA, C(CO)NH and H(CCO)NH experiments were achieved by a

100

DIPSI-2 sequence (Shaka et al., 1988), and in the CBCA(CO)NH experiment by a
WALTZ-l6 sequence (Shaka et al., 1983). The CO decoupling during the CM3 chemical
shift and Cam-”N scalar coupling evolution periods in the HNCACB experiment, and Ccl
decoupling during the constant time t2 (”N) in the CBCA(CO)NH experiment were
accomplished by applying a SEDUCE-l sequence (McCoy & Mueller, 1992a, b). ”C
isotropic mixings in the 3D HCCH-TOCSY (Bax et al., 1991; Kay et al., 1993),
H(CCO)NH and C(CO)NH experiments were accomplished by a DIPSI-3 sequence
(Shaka etal., 1988). The HSQC, HMQC, HNCACB and C(CO)NH spectra were acquired
on the INOVA-600 spectrometer and processed with the program VNMR v. 5.3 (Varian
Associates). All other spectra were recorded on a Briiker DMX-SOO spectrometer at the
National Magnetic Resonance Facility at Madison and processed by Felix 95 (Biosym
Technologies inc.). A ~45°-shified sine-bell was applied in both dimensions of the 2D
HSQC and HMQC data. The HSQC data was zero-filled to 2048 x 512 real points and
the HMQC data to 2048 x 256 real points. A ~75°-shified sine-bell was applied in both
F1 and F3 dimensions, and a cosine-bell in the F2 dimension of the 3D NOESY-HSQC
and HCCH-TOCSY data. For the other 3D data, a 75°-shified sine-bell was applied in the
F3 dimension and a cosine-bell in the other two dimensions. The first points of both F1
and F2 dimensions were calculated by linear predictions. The data size in indirect
detection dimension was extended by backward-forward linear predication (Zhu & Bax,
1992). Solvent suppression was improved by convolution of time domain data (Marion et
al., 1989b). The data size was doubled in each dimension by zero-filling once. The empty
regions of the 3D data (the aliphatic parts in the F3 dimension of the ”N-edited data and

the amide part of the HCCH-TOCSY data) were removed to reduce the data size.

101

Derivation of structural restraints. Two types of structural restraints were
derived from experimental NMR data: interproton distance restraints and hydrogen bond
restraints. Approximate interproton distance restraints were derived from the NOE data.
NOE cross peaks between aliphatic protons were picked from the homonuclear 2D
NOESY spectrum with a mixing time of 100 ms using the program VNMR v. 5.2. NOE
cross peaks involving amide protons were picked from the 3D 'H-”N NOESY-HSQC
with a mixing time of 150 ms using the program Felix 95, except for those of Ile9, IlelO,
Ile63, Arg132, Va1133 and Tyrl34, which were picked from the 2D homonuclear
NOESY data. The integrated peak volumes were converted into approximate interproton
distances by normalizing them against the calibrated volumes of NOE cross-peaks
between backbone protons within the identified B-sheet regions. The upper limits of the
interproton distances were calibrated according to the equation Va = V, (rb/ra)6, here Va, V,
were the volumes and rd, rb the internuclear distances. The distance bounds were set to
1.8—2.7 A (1.8—2.9 A for NOE cross peaks involving amide protons), 1.8—3.3 A (1.8—3.5
A for NOE cross peaks involving amide protons) and 1.8—5.0 A corresponding to strong,
medium and weak NOEs, respectively. Pseudoatom corrections were made for non-
stereospecifically assigned methylene and methyl resonances (Wfithrich et al., 1983). An
additional 0.5 A was added to the upper bounds for methyl protons.

Protein backbone hydrogen bonds were derived from patterns of cross peaks
remaining in the fingerprint regions of the homonuclear 2D TOCSY and NOESY spectra
recorded in D20. Hydrogen bond restraints were set to 2.0 i 0.5 A and 3.0 i 0.3 A for

HNmO and N---O, respectively.

102

Structure calculation. The structures were calculated by the hybrid distance
geometry-simulated annealing (DGSA) protocol (Nilges et al., 1988) in the program X-
PLOR v. 3.1 (Briinger, 1992) on an SGI IndigoII workstation. A square-well potential
function with a force constant of 50 kca1.mol“.A’2 was applied for the distance restraints.

The X-PLOR f function was used to simulate van der Waals interaction with atomic

repel
radii set to 0.80 times their CHARMM values (Brooks et al., 1983). Hydrogen bond
restraints within the regions of regular secondary structures were introduced at a later
stage of structural refinement. Seventy structures were generated using this protocol at
the beginning. The structures were inspected with the programs QUANTA96 (Molecular
Simulations. inc.) and InsightII (Biosym Technologies inc.), and analyzed by
PROCHECK—NMR (version 3.4.4) (Laskowski et al., 1993; 1997). An iterative strategy
was used for the structural refinement. The first round calculation employed mainly
NOEs between backbone protons. Then the newly computed NMR structures were
employed to assign more NOE restraints, to correct wrong assignments and to loosen
NOE distance bounds if spectral overlapping was deduced. Then another round of
structural refinement was carried out with the modified NMR restraints. Afier several

rounds of such refinement, an ensemble of 25 structures was selected according to their

best-fits to the experimental NMR restraints and the low values of their total energies.

103

Results

Sequential resonance assignment. HNCACB and CBCA(CO)NH spectra were
first analyzed to obtain sequential connectivities. Representative results are shown in
Figure 3.1. About one hundred residues with relatively strong peaks in the 2D lH-”N
HSQC (Figure 3.2) spectrum could be easily linked on the basis of these two sets of data.
These residues were then sequentially assigned by the combined analyses of the 3D
C(CO)NH, H(CCO)NH, HCCH-TOCSY, and homonuclear 2D TOCSY spectra.

The majority of the remaining residues had weak or no detectable peaks in the 2D
'H-”N HSQC spectrum, even with minimum water perturbation using the water flip-back
technique (Grzesiek & Bax, 1993). Because these residues also had weak or no detectable
peaks in the triple resonance experiments, their sequential assignment posed a big
challenge. The following residues had weak cross-peaks in the 2D lH-”N HSQC
spectrum: LysS—Glu17, Leu28, Lys30, Ala32—A1a35, Ala40, Va141, Glu62, Ile63,
Leu119, Leu121 and Thrl31—Va1135. They were assigned by spin system identifications
combined with sequential and long range (inter-strand) NOEs. Because of the excellent
chemical shift dispersion of the protein, 2D homonuclear TOCSY and NOESY data were
especially useful for assigning these residues. In particular, most of the inter-strand NOE
cross-peaks between Ha-H“ protons of the adjacent B-strands were well resolved in the
2D NOESY spectra. The sequential assignments of many of these residues were also
facilitated by the comparisons with the sequential assignments of the wild type holo-

CRABPII in complex with all-trans-RA (Chapter 5) and the apo-form of the site-directed

104

Figure 3.1. Strips of the HNCACB (a) and CBCA(CO)NH (b) spectra showing sequential
connectivities for residues Gly68—Gln74. Both inter-residue and intra-residue peaks in the

HNCACB strips (a) are labeled. The dotted lines indicate negative peaks.

105

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

—25.o
.3 ' A
B l3__ P 4 —3o.o
é}?— [xi-1)
-—35.0
B —4o.o
9 ..,,_._o
$__. ““5” —45.o
—so.o
D a 0 $
., a(i-1)
3:3 .. a D . 4 _
a a @__o éd 55.0
5.
b o ”G —60.0
so ' © C018
3 b a a a b a a (ppm)
G68 E69 E70 F71 E72 E73 Q74

Residue

106

Figure 3.2. Gradient- and sensitivity-enhanced 2D lH—”N HSQC spectrum of unifomly
”N-enriched apo-CRABPII. Sequential assignments are indicated with one-letter amino
acid code and residue number. The side-chain amides of Asn and Gln are indicated by #.

The unlabeled peaks are most probably from the side-chains of arginines.

F1

(ppm)?

110

112

114

116

118

120

122

124 V

126

128

130

132

134

136’

1.. .A.__L_L_...._._._._..s_... - _

11.0

. I . .- l . .

10.5

0

107

 

678°.623 0'
30451: 0451
9
G117
6102‘)
Gs G104 o
T110 0 N91o_l:l4#N14# N141! 1.125410?“
0 D77 N251:
° ° S10132
V135 T75 0.0034 N2#I=N91#
8891.7”154; ~091#N115#N64#
N64#°337
09711
a T124 011.6 0 037,,
812'?
T114 90V33 0W1099 .°K30
’ 357%”
1 K101 4 L22
6470 °r1o7' E16E103 (85389
11}, 0mg F15 87K?” 0'23
V94 $572 ~64" M27
T1220 K92 L34 131
°°V26 Na R11
K660 R111W87oo127 £69 “79 $01134
a M1230 OF71 R136 @7135 V21 097
F3 D126QV76 E623V133°°Y513L100L18N24
W7 I43 F50 15112:3 128/13826
0
W7N€ {Aw/E13 “T8231
9 265%“ A125 sV41
”20 19 o V129
M93 L99 9 0 flE137
p 0 348 L113 0
Y134 95'& O
a 523%.”: 045 E96
R132 K44 o
E42 E46
0130 fives
E70 0
K53
61.121
‘1 """ 1i? ’l' 'l’l'FWT'll
0 9.5 9.0 8.5 8.0 7.5 7.0 6.5

108

mutant R1 1 1M (Chapter 4). Most weak peaks in the 2D lH-”N HSQC spectrum of wild
type apo-CRABPII became much stronger upon the binding of RA or with the point
mutation, and thus could be assigned unambiguously for holo-CRABPII and apo-R11 1M.
Furthermore, neither the binding of RA nor the mutation changed the chemical shifts of
the majority of residues. Such comparisons were especially helpful for the assignments of
the selectively lSN-labeled residues such as Leu119, Va133, Val41, Val76, Va1133 and

Leu121. Such comparisons also expedited greatly the assignments of the forty-six
residues with amide protons remaining in the fingerprint regions of the homonuclear 2D
spectra in D20 (Figure 3.3). Almost exactly the same residues remained in the 2D
homonuclear D20 spectra of apo-CRABPII and apo-R1 11M (Figure 4.8 of Chapter 4). In
particular, such comparisons helped the assignments of residues LysS, Ile9, IlelO, Ser12,
Ile63, Thr131, Arg132 and Tyrl34. Nine residues, namely Glu13, Asn14, PhelS, Lys30,
Ile3l, A1340, Asp77, Gly78 and Cysl30, were assigned based on NOEs and with the help
of the comparisons. The HN and ”N resonance assignments of Va133 were based on
selective labeling, spin system identification and several NOEs between the ring H53 of
Phe15 and HS of Va133 after the first round of structural calculation. The HN and ”N
resonances of Ala32, Ala34 and Ala35 were assigned based on the comparisons and a few
weak, sequential NOEs. The HN or ”N resonance of Ser37 was assigned tentatively by
spin system identification and the comparisons of the three sets of 2D lH-”N HSQC
spectra. The peaks corresponding to the two adjacent residues, Ala36 and Lys38, could
not be observed in the 2D lH-”N HSQC spectrum of apo-CRABPII. The aliphatic
protons of the residues with weak peaks in the 2D lH-”N HSQC spectrum were assigned

by the inter-strand NOEs, by spin system identifications obtained from the homonuclear

109

2D spectra and 3D HCCH-TOCSY spectrum, and by the comparisons with the
assignments of holo-CRABPII and apo-R1 11M. For examples, the spin systems of Ser12,
A1a40, Val4l, Thr54, Glu62 and Val76 could be easily identified. The H“ of Arg132 was
assigned by its inter-strand NOE to Ser12 observed in the homonuclear 2D NOESY
spectrum.

The following eleven residues were not observed in the 2D lH-”N HSQC
spectrum: Asn2, Arg29, Ala36, Lys38, Thr56—Thr61 and Asn115. They were first
grouped into several spin systems based on the homonuclear 2D DQF-COSY, TOCSY
and 3D HCCH-TOCSY experiments. The aliphatic protons of Asn2, Thr56, Thr57,
Va158, Thr60, Thr61 and Asn115 were assigned by spin system identifications, the inter-
strand NOEs, and by the comparisons with the assignments of holo-CRABPII and apo-
Rl 11M. The assignments of the aliphatic protons of Thr56 and Thr57 were further helped
by the similar but unique chemical shifts of their H“ and HB resonances in the spectra of
apo- and holo-CRABPII and apo-R1 11M: H“ 4.83 ppm and HB 4.61 ppm for Thr56, and
H“ 4.00 ppm and H“ 4.23 ppm for Thr57 (in apo-R1 1 1M). Two possible alanine peaks
from Ala32 and Ala36 could be identified in the 2D homonuclear and 3D HCCH-TOCSY
spectra, but their sequential assignments could not be obtained due to the lack of strong,
well-resolved sequential NOEs. None of the 1H, ”C and ”N resonances of Arg29, Lys38,
Pro39 and Arg59 were assigned (Table 3.1).

The ring protons of aromatic residues, except for PhelS and Tyrl34, could be
easily assigned by analysis of the 2D DQFCOSY and TOCSY data, and by the
comparisons with the assignments of holo-CRABPII and apo-R11 1M (Chapters 4, 5).

Their identities were confirmed later by sequential assignment and the NOEs between H”-

110

Figure 3.3. The fingerprint region of the homonuclear 2D TOCSY spectrum with 40 ms
mixing time. The TOCSY experiment was initiated more than 24 h after dissolving a

lyophilized protein sample in PBS, pD 7 .5, in D20.

lll

__.._ _. _-____._ _..___ _.___ 7
0
F1 '; T122(HB) '
(pm);
4.4 V94 K86 l
s 635 °
4-6‘3 s 0 F50 ”2‘0ch
K44 "0 .0573
4'8 3 0 . A125
; I120 s .63
5.0‘j . K5 M123 0.8 V133
T: R132 W7 .
5.2; 7110 , F71 L119 Y51
K08 C130 '43 1.113
5.4 :f V135/E 2
L121 4 , K.
5.6"}:9 ° 1 O 92,
Y134 0 111 0
5'8: T122 51°12 W109

 

 

"7‘71'l'TT'iNUTT'T‘TTTT‘ITE' " l! TIT!) ll lljlinllquTqfi WITH W11 TITO T1 TT'lTiTTTl'lTTTlUTTTPlTTTlTTTITTTTTTqu

10.2 9.8 9.4 9.0 8.6 8.2 7.8

F2 (PM)

112

Table 3.1. 'H, ”N and ”C assignments of apo-CRABPII in PBS at 25°C.

 

Residue
P 1
N2
F3
54
GS
N6
W7

K8
19
110
R11
512
E13
N14
F15
E16
E17
L18
L19
K20
V21
L22
023
V24
N25
V26
M27

P39

T49
F50
Y51
152

K53
T54
S55
T56
T57
V58

ISN

121.9 (9.82)
112.5 (8.49)
110.6 (9.22)
118.6 (8.15)
123.3 (9.35)

123.2 (l0.l7)

125.6 (9.05)
118.8 (9.24)
120.8 (7.57)
115.8 (8.33)
124.0 (8.71)
117.8 (9.25)
119.3 (8.72)
117.4 (8.36)
118.5 (8.36)
121.4 (7.59)
114.5 (7.36)
118.9 (8.18)
121.3 (7.83)
117.5 (7.24)
106.5 (7.69)
121.4 (7.62)
127.9 (8.68)
120.3 (8.47)
119.7 (7.95)
118.6 (7.82)

116.5 (7.66)
119.2 (7.16)
122.6 (8.54)
116.7 (8.66)
121.0 (7.31)
121.6 (8.44)

113.9 (7.43)

123.9 (8.71)
125.2 (7.93)
130.0 (9.30)
123.1 (9.30)
128.9 (9.56)
128.0 (8.58)
128.9 (8.53)
1 17.9 (9.10)
126.6 (8.54)
116.7 (8.08)
124.2 (9.08)
122.5 (8.35)
123.3 (8.41)
133.1 (9.30)
124.4 (8.85)
122.7 (9.43)
124.0 (8.71 )

CO

171.9
173.5
171.8
167.9
171.5
172.5

172.4
173.3
173.3

171.0

172.1
175.9
174.9
176.0
178.3
174.9
173.6
172.0
173.6
172.8
173.4

177.7

174.8

174.4
171.5
172.1
173.9
170.8
172.2
173.5
170.6
172.5
170.1
173.1
172.8
171.6
172.0

173.5

C0.
63.0 (4.32)
52.7 (5.03)
60.4 (4.32)
60.7 (4.70)
44.9 (3.96, 3.76)
53.0 (5.52)
56.7 (5.14)

54.1 (5.28)
55.5 (3.92)
61.3 (4.61)

62.8 (5.12)
61.7 (4.80)
55.7 (4.95)
55.3 (4.59)
63.2 (3.30)
59.6 (3.68)
58.5 (3.73)
57.6 (3.62)
60.5 (3.83)
66.0 (3.78)
55.4 (3.91)
45.9 (3.96, 3.58)
63.3 (3.72)
54.2 (4.38)
66.7 (3.46)
58.5 (4.14)
(4.08)
(4.32)
59.6 (3.87)
64.3 (3.72)

66.6 (3.62)
(4.08)
(4.21)

52.2

60.4 (4.70)

51.9 (4.96)

61.2 (5.13)

54.6 (5.36)

60.0 (5.34)

56.4 (4.60)

53.9 (4.45)

55.1 (4.37)

48.0 (3.94, 3.58)

54.5 (5.18)

63.0 (4.70)

57.6 (4.56)

56.3 (5.24)

61.0 (4.60)

54.5 (5.08)

62.1 (4.88)

58.0 (5.30)
(4.85)

64.9 (4.00)

62.2 (4.16)

(:13

33.2 (2.37, 1.91)
39.3 (2.95, 2.89)
32.3 (3.28, 2.70)
64.5 (4.30, 3.99)

40.2 (2.69, 2.52)
32.2 (3.06, 2.66)

36.1 (1.90, 1.85)
33.6 (1.76)
41 .8 (1.83)

(3.61, 3.43)
40.1 (2.19, 2.08)
41.4 (3.31, 3.22)
33.7 (2.98, 2.85)
32.6 (1.96)
28.1 (1.84, 1.73)
42.4 (2.17, 1.34)
41.1 (1.62, 1.56)
32.6 (1.93, 1.61)
31.7 (2.15)
41.5 (1.39)

31.7 (1.45)
39.3 (3.09, 2.74)
32.1 (2.03)

32.0 (2.07, 2.01)
(1.63, 1.32)
(2.20, 1.97)

32.2(189, 1.81)

39.2 (1.80)

(2.07)
(1.38)
(1.38)

19.1
(4.23, 4.02)

20.9 (1.45)
35.3 (1.73)
33.4 (2.06, 1.99)
40.8 (2.27)
35.3 (1.87, 1.76)
30.6 (1.82, 1.71)
31.0 (1.62, 1.26)

42.1 (3.15, 2.68)

69.6 (4.14)

44.6 (1.65)

41.0 (3.01, 2.70)

40.4 (1.84)

34.9 (1.76, 1.58)

70.0 (3.86)

65.0 (3.83, 3.74)
(4.61)

66.7 (4.23)
(1.96)

Others
H" 2.10, 2.08; H“ 3.32, 3.25

2, 6H 7.42; 3, 5H 7.16; 4H 7.34

2H 6.78; 4H 7.07; 5H 6.82; 6H 7.23; 7H 7.12
H“ 9.20; N‘ 124.6

H" 1.63, 1.52; H“ 1.37, 1.33; H‘2.85

H" 1.65, 1.56; H“ 0.48; H""' 1.39

H""‘0.87

H’l.93

H“ 7.35, 7.00; N“ 109.8

2, 6H 7.14; 3, 5H 6.84; 4H 6.93
H" 2.16

H" 2.38, 2.19

H" 0.81; H“ 0.74, 0.70

H" 1.12; H“ 0.36, 0.32

H" 1.35, 1.14; H“ 2.91

H" 1.10, 0.95

H“ 0.49

H" 0.83

H“ 7.73, 6.91; N“ 112.3
H" 0.96, 0.91

H" 2.59, 2.46

H" 1.09; H“ 0.82, 0.76
H" 1.74; H“ 2.70

H" 1.68, 1.59

H""‘ 0.88; H“ 0.80

H" 0.94

H" 0.63, 0.46

H" 2.25

H1 1.66; H"‘“ 0.85; H“ 0.51

H" 1.54, 1.41; H“ 1.33, 1.24

H" 2.03; H“ 7.83, 7.11; Nc 106.8
H" 2.02

H" 0.99

2, 6H 6.48; 3, 5H 7.08; 4H 6.81
2, 6H 6.94; 3, 5H 6.61

H’ 1.26; H""' 0.80; H“ 0.31

H" 1.17; H“ 0.44, 0.29; H“ 2.84
H" 0.94

H" 1.07
H" 1.27
H" 0.79, 0.74

 

113

 

Residue
R59
T60
T61
E62
163
N64
F65
K66
V67
G68
E69
E70
F7 1
E72
E73
Q74
T75
V76
D77
G78
R79
P80
C8 1
K82
583
L84
V85
K86
W87

E88
S89
E90
N91
K92
M93
V94
C95
E96
Q97
K98
L99
L100
K101
6102
E103
G104
P105
K106
T107
8108
W109

T110
R111
5112
L113
T114
N115
D116
Gll7
E118

ISN
123.5 (8.58)
122.9 (8.73)

120.4 (8.10)
117.2 (8.34)
120.8 (9.32)
127.5 (9.01)
114.9 (8.91)
120.2 (7.94)
130.3 (9.21)
121.4 (9.06)
119.6 (8.67)
122.4 (8.44)
117.9 (7.59)
111.8 (9.18)
122.4 (9.18)
112.0 (8.54)
107.2 (7.93)
121.0 (7.86)

118.8 (8.87)
120.0 (8.87)
124.9 (9.06)
120.7 (7.36)
130.1 (9.20)
123.8 (8.47)
120.2 (8.65)

127.7 (9.21)
112.5 (8.59)
116.7 (8.71)
111.2 (8.05)
120.1 (7.84)
127.0 (9.35)
119.3 (8.86)
127.3 (8.85)
127.7 (7.70)
121.0 (7.43)
123.1 (8.94)
126.1 (8.78)
121.9 (7.77)
117.4 (8.05)
108.6 (8.18)
117.2 (8.10)
109.1 (8.14)

123.5 (8.58)
118.7 (8.79)
112.3 (7.80)
116.3 (8.21)

111.2 (9.56)
120.1 (8.79)
123.3 (8.87)
126.4 (8.25)
116.9 (9.35)

115.7 (7.75)
107.6 (7.96)
117.5 (7.43)

CO

170.5
169.7
169.1
175.2
174.1
170.7
171.5
172.6
169.5
173.1
171.6
174.3
175.4

171.4
173.6
168.5
173.0
168.7
173.7
171.5
173.3
174.9

173.3
171.2
173.1
171.3
171.0
171.0
170.0
171.4
172.6
173.1
172.9
175.4
174.8
172.7
169.7
173.1

172.8
175.1
170.8
171.2
171.8

171.7
170.9
170.5
175.1

173.1
173.5
171.7
172.0

CG

60.3 (5.20)
60.1 (4.61)
(4.98)
59.5 (4.88)
52.3 (5.56)
55.9 (4.65)
53.9 (4.65)
67.1 (3.23)
45.7 (4.38, 3.61)
54.9 (4.77)
57.6 (4.88)
55.8 (5.11)
55.3 (5.05)
55.3 (4.73)
54.2 (5.30)
62.4 (4.55)
65.6 (3.82)

45.9 (4,10, 3.55)
54.0

63.8 (4.66)

55.9 (5.05)

54.5 (4.73)

58.7 (4.86)

54.9 (4.23)

60.6 (4.35)

54.2 (4.48)

57.0 (5.1 1)

58.0 (4.35)
57.0 (4.55)
59.1 (3.66)
52.9 (4.96)
56.7 (5.46)
53.4 (5.18)
60.3 (4.30)
55.6 (4.48)
55.2 (4.31)
54.7 (4.91)
54.7 (4.60)
56.0 (4.00)
57.0 (4.07)
55.4 (4.37)
45.1 (4.02, 3.71)
54.9 (4.60)
45.5 (4.10, 3.90)
62.7 (4.48)
56.6 (4.38)
60.1 (5.36)
57.3 (4.63)
56.1 (5.76)

61.0 (5.24)

56.1 (5.57)

53.9 (5.78)

52.8 (5.28)
(4.76)

55.5 (4.44)

54.3 (4.74)

46.1 (4.17, 3.69)
(4.98)

(:13

71.7 (3.90)

71.7 (3.97)
(2.02, 1.91)

42.2 (1.85 )

42.8 (2.54, 2.50)

39.8 (2.15)

35.4 (1.63, 1.49)

31.9 (1.85)

35.3 (2.16, 1.96)
31.4(1.99,1.91)
41.4 (3.53, 3.20)
31.9 (202,191)
29.7 (202,190)
33.1 (1.73, 1.70)
71.2 (4.16)

32.1(200)

29. 8

33.0 (2.38, 1.86)
31.6 (2.58)

34.3 (1.54, 1.47)
68.0 (2.78, 1.54)
45.2 (2.00, 1.58)
32.6 (1.35)
36.9(1.6l, 1.43)
29.5 (3.22, 2.99)

32.3 (1.80, 1.66)
65.0 (4.1 1, 3.98)
29.5 (1.84, 1.78)
40.4 (3.15, 2.85)
37.2 (1.98, 1.47)
35.1 (1.99)

34.9 (1.77)

27.9 (0.36, -0.28)

31.6(1.81,1.73)
30.8 (1.75)
35.0 (2.10)
42.0 (1.98, 1.63)
42.5 (1.64, 1.47)
34.9 (1.72, 1.61)

32.9 (1.96, 1.78)

32.7 (2.03, 1.33)
32.6(1.81, 1.51)
72.5 (4.46)

66.8 (3.86, 3.78)
33.1 (3.48, 3.29)

72.2 (4.16)

35.6(185. 1.50)

34.9 (1.94)

43.9 (1.25)
(4.70)

38.1 (2.79, 2.76)

41.2 (2.79, 2.39)

(2.02, 1.98)

Others

H" 1.06
H" 1.06

1171.16, 101;va 1.09;H“0.80
H“ 7.62, 6.92; N“ 113.1

2, 6H 6.24; 3, 5H 6.52; 4H 6.32
117 1.45, 1.28; 115 2.92

H, 0.64, 0.43

H" 2.19

H" 2.20, 2.13

2, 6H 7.32; 3, 5H 6.82; 4H 6.52

H" 2.30, 2.13

H" 2.14, 2.14

H" 2.17, 1.99; H‘ 7.49, 6.70; N" 110.3
H" 1.03

H" 1.01, 0.81

H" 2.15, 1.99
H" 1.26, 1.22; H“ 1.03

H" 1.17; H“ 0.68, 0.51

H" 0.23, 0.02

H" 1.26, 1.12

2H 7.30; 4H 7.98; 5H 6.95; 6H 7.21; 7H 7.62
H‘ 10.66; N“ 129.6

H" 1.99

H" 2.08

H“ 7.50, 6.83; N“ 112.6
H" 1.22; H“ 1.10; H8 2.68
H" 2.19, 2.08; H“ 0.41

H" 0.73, 0.70

H" 2.14, 1.95

H" 2.43, 2.02; H“ 7.38, 6.71; Nc 114.4
H" 1.96, 1.78; H“ 1.24; H‘ 3.16, 2.84
H" 1.36; H“ 0.86, 0.65

H" 1.35; H“ 0.80, 0.76

H" 1.37, 1.27

H"2.10

H" 1.84, 1.76; H“ 3.60, 3.29
H" 1.35, 1.30;
H" 1.13

2H 6.89; 4H 6.90; 5H 6.35; 6H 6.43; 7H 6.59
H“ 11.37; N‘ 137.8
H" 1.09
H" 1.35, 1.23; H“ 3.00
H" 2.03
H" 0.94; H“ 0.51, 0.44
1.25
H 7.67, 6.95; N“ 112.6

 

114

 

Residue
L119
1120
L121
T122
M123
T124
A125
D126
D127
V128
V129
C130
T131
R132
V133
Y134
V135
R136
E137

ISN

124.6 (8.73)
125.2 (9.47)
135.5 (9.88)
120.1 (9.13)
121.7 (9.29)
115.2 (8.26)
125.3 (8.39)
122.4 (9.28)
121.1 (8.48)
123.4 (8.41)
125.7 (8.38)
129.4 (9.56)
124.6(8.11)
128.3 (9.77)
122.2 (8.50)
127.2 (9.79)
112.1 (9.55)
123.2 (8.55)
125.6 (8.08)

CO
172.6

172.8
171.4
171.5
172.1
172.7
172.7
172.8
171.2
172.9

171.2
172.5
172.8
174.5
178.7

CG
53.4 (5.08)
60.8 (4.83)
53.3 (5.56)
60.4 (5.76)
55.1 (5.14)
60.7 (5.46)
51.6 (4.84)
57.3 (4.14)
54.7 (4.50)
63.6 (3.95)
61.4 (4.50)
56.4 (5.16)
(5.13)
(5.01)
62.2 (4.90)
56.5 (5.59)
59.7 (5.33)
57.3 (3.89)
58.9 (3.86)

(:13

43.3 (0.89)

39.4 ( 1.77)

45.3 (2.15)

71.0 (4.03)

38.3 (1.84, 1.81)

71.2 (3.93)

21.3 (1.03)

39.6 (2.72)

41.4 (2.66)

32.9 (2.14)

34.4 (1.83)

35.4 (3.00, 2.66)
(3.74)
(1.78, 1.67)
(1.66)

43.6 (3.23, 3.01)

36.2 (2.36)

30.3(l.29,1.11)

31.6 (1.80, 1.68)

Others
H" 0.44; H“ 0.18, —0.36
H" 1.35; H"“’0.75; H“ 0.99
H" 1.22; H“ 0.85, 0.74
H" 1.06
H" 2.55, 2.28
H" 0.96

H" 0.92, 0.69
H" 0.77, 0.69

H" 0.98

H" 1.50

H" 0.85, 0.80

2, 6H 6.82; 3, 5H 6.37

H" 1.04, 0.85

H" 0.59, 0.06; H“ 2.70, 2.79
H" 2.05, 1.90

 

115

H“ and between Ha-H“ for tyrosines and phenylalanines, between HB-H“l and between H“-
H“3 for tryptophans. The ring protons of PhelS and Tyrl34, which were overlapped with
other peaks in the aromatic regions, were assigned based on the NOEs between HB-H“ and
between Ha-H“ and the comparisons. In the corresponding spectra of holo-CRABPII and
apo-R1 1 1M they were separated from the other peaks.

Stereospecific resonance assignment. In order to better determine the solution
structure of CRABPII, stereospecific assignments were made whenever possible. The
methylene H‘3 protons were stereospecifically assigned on the basis of the qualitative
estimations of 33613 coupling constants from DQF-COSY spectrum in conjunction with the
relative volumes of the NOE cross-peaks between H"‘—Hl3 picked from the 2D
homonuclear NOESY with 100 ms mixing time and the NOE cross-peaks between HN—H“
from 3D lH-‘SN NOESY-HSQC (Basus, 1984). The methyl protons of valine residues
were stereospecifically assigned in a similar manner except that the relative volumes of
the NOE cross-peaks of H“—H" and HN—H" were used. Methyl protons of leucine residues
and the ring protons of Phe and Tyr residues were stereospecifically assigned after two
rounds of structure refinement. Stereospecific assignments were obtained for H'3
methylene protons of 43 residues, for methyl protons of six valine residues and four
leucine residues.

Secondary structures. Secondary structural elements were obtained from NOE
data. The NOEs used to characterize secondary structures were d,m and dag from the
homonuclear 2D NOESY and dam d- and dim from the 3D 'H-‘SN NOESY-HSQC and/or
homonuclear 2D NOESY data. Regular a-helix was characterized by strong short-range

NOEs dNN, do“, (i, i+3) and de (i, i+3) and weak dew, and regular B-sheet by strong short-

116

range NOEs dew and de and strong or medium long-range NOEs do“, and do,” between
adjacent strands. The following ten B-strands forming two anti-parallel B-sheets were
identified (Figure 3.4): Ser4—SerlZ (BA), Ala40—Glu46 (BB), Thr49—Ser55 (BC), Thr61—
Lys66 (BD), G1u69—Thr75 (BE), ProSO—G1u88 (BF), Lys92—Lys98 (BG), K106—L113
(BH), L119—A125 (BI) and V128—R136 (BJ). Long-range NOEs, a",m and daN, were
observed between adjacent BA, BB, BC and BD, between adjacent BE, BF, BG, BH, BI and
BJ, and also between the N-terminal half of BA and the C-terminal half of BJ. No such
NOEs were observed between the adjacent BD and BE. LysS-Ser12 of BA, LysS3 of BC,
Ile63 of BD and Thr131—Vall35 of BJ had weak peaks in the 2D lH-‘SN HSQC spectrum.
However, the amide protons of Lys8, Ile10, LysS3, Ile63 and Arg132—Va1135 were
observed in the homonuclear 2D spectra recorded in D20 (Figure 3.3). In particular, most
of their inter-strand NOE distances due, and do,” was observed in the 2D NOESY spectra in
D20. Therefore, Lys8—Ser12 and Thrl31—Va1135 were considered to adopt B-sheeted
structures. They formed part of the so-called B-bulge structure. A1a40—Val4l of BB and
Thr60—Glu62 of BD showed weak or no observable peaks in the 2D lH-‘SN HSQC and
3D lH-‘SN NOESY-HSQC spectra. However, strong NOEs (dad) were observed for these
residues. The first two B-strands were linked by a long segment (Glul3-Pr039). Along
this segment a short a-helix 01A, Glul7—Gly23, could be recognized. Glu17 was weak in
the 2D lH-‘SN HSQC spectrum. It was assigned as the N-terminus of the helix according

to the NOE (110,,3 (i, 1' +3) observed in the D20 spectra.

117

Figure 3.4. Summary of the sequential and medium-range NOEs involving backbone HN
and H“ atoms, slow-exchange backbone amide protons and the deduced secondary
structures of apo-CRABPII. Line thickness for (11,,N and a"NN sequential NOE distances
reflects the intensities of the cross peaks. Asterisks indicate the residues whose amide
protons remained after 24 h of exchange with D20 in PBS buffer at room temperature and

pD 7.5 as determined by 2D homonuclear NMR experiments.

118

 

 

 

 

 

 

 

 

 

10 20 30
| 1 I
PNFSGNWKIIRSENFEELLKVLGVNVMLRKIAVAAA
Slow NH exchange ' " ‘ " "‘ “ ‘ ‘ “
daN(i,i+1) I — _-_ _ H .—
dNN(i,i+1) — _ I - _- .—
daN(i,i+2) — —"=="——-— '—
dNN(i,i+2) —‘—=—— —
36140.16) —— —‘===—-
NN('.°+3)
Secondary/“structures Ct, ,j PA ,f I > . OLA 1
4'0 5'0 60 70
l l
SKPAVEIKQEGDTFYIKTSTTVRTTEINFKVGEEFEE
Slow NH exchange ‘ “ ’ ‘ ‘ ‘ ‘ ‘ “ " ‘
daN(i,i+1) — _ I - - __
dotN(i,i+2)
dNN(',1+3) _
Secondar; structures CDZEIBZII» Snag” cur-£12m) 1:231:53:
1° 1° 11°
QTVDGRPCKSLVKWESENKMVCEQKLLKGEGPKT
Slow NH exchange ‘ "' ‘ " “ “ ‘
d01N(i,i+l) - — _——-—- -
dNN (1,1+1) — — _ - - - fl — -
daN(i,i+2) —-=-—
dNN(i,i+2) — —
dNN(i,1+3) BF ' " BO
Secondarystructuresuzb “11.11.,fitjp 111111114> ccn
1'10 120 130
SWTRELTNDGELILTMTADDVVCTRVYVRE
Slow NH exchange . t I t I O t t t I 3 t t t 3 t t i .
daN(i,i+l) _ __ __-
dNN (Li-F1) — - i — - —
daN(1,1+2) —
dNN(i,i+2) “:—

G10: (1 1+3)
1 H
Secondarystructures 12:21:23) Egg” mxrxlgjnxxb

 

119

Glul3—Glu16 had very weak peaks in the 2D 'H-‘SN HSQC, homonuclear 2D SSNOESY
and WATERGATE NOESY spectra.

NOE distance restraints. A total of 2290 structurally useful NOE distance
restraints were obtained from the analyses of the homonuclear 2D NOESY and 3D IH-‘SN
NOESY-HSQC spectra. Of the NOE restraints, 647 were intra-residue, 560 were
sequential, 248 were medium-range and 835 were long-range NOEs (Table 3.2). On
average, each residue had ~17 NOE restrains, but the numbers of NOE restraints per
residue varied greatly along the amino acid sequence (Figure 3.5). The residues situated
in several turns or loops and the entrance region had NOE restraints well below the
average. A total of 92 hydrogen bond restraints obtained fiom 46 hydrogen bonds were
included in the structural refinement.

Quality of the structure. Figure 3. 6A shows the C“ traces of the superimposed 25
computed structures with no NOE violations exceeding 0.3 A. The statistics of the
structures are summarized in Table 3.2. Except for the entrance region, the structures
were well defined. Excluding residues Ala32-Pro39 and Thr57-Glu62, the RMSD of the
25 coordinates was 0.54 A for the backbone (N, C“, C’, O) and 0.92 A for all heavy
atoms. The stereochemical qualities of the structures were checked by the program
PROCHECK—NMR (Table 3. 2). The restrained minimized mean structure (Clore &
Gronenbom, 1989) had 55.6% residues in the most favorable regions of the
Ramachandran plot, and 37.1% in the additional allowed regions. Two residues, Phe3 and
Asn115, and eleven others were at the disallowed and generously allowed regions,

respectively. The relative low percentage of the residues in the most favorable regions

120

Figure 3.5. The distribution of NOEs along the amino acid sequence of CRABPII. For

intra-residue NOEs only those that are structurally useful are included.

No. of NOEs

 

No. of NOEs

121

 

 

  

 

 

  

Long

Medium

Sequential

lntraresidue

1 .113 ' ‘1
gt}? 1" ’ .1 I 1
“1:33.35: 11: ' ,1. 1;
PNFSGNWKIIRSENFEELLKVLGVNVMLRK1AVAAASKPAVEIKOEGDTFVIKTSTTVRTTEINFKVGE
5 10 15 20 25 30 35 40 45 50 55 so es

Residue Number

 

 

 

 

so
40
20 .l.Ԥ '
1]
3311.3‘ig : "113'; 5" l' 1.; ' 1‘;
o s...€.~.«.24:~391..<.§.u.~ 1. 1 it
EFEEQTVOGRPCKSLvxweseuxuvceoxtLKGEGPKTSWtRELTNDOEL1Lruerovvcravvvns
70 75 80 85 90 95 100 105 110 115 120 125 130 135

Residue Number

122

Figure 3.6. (A) Stereoview of the C“ traces of the superimposed 25 refined structures of
apo-CRABPII. (B) Stereoview of the C“ traces of the restrained minimized mean
structure of apo-CRABPII (thin line) superimposed with the C“ traces of the crystal
structure of holo—CRABPII (thick line). (C) Stereoview of the C“ traces of the restrained
minimized mean structure of apo-CRABPII (thin line) superimposed with the C“ traces of

the crystal structure of apo-CRABPI (molecule A) (thick line).

123

 

124

Table 3.2. Restraint and structural statistics of apo-CRABPII

 

Restraint Statistics

Numbers of experimental NOE restraints

Intra-residue 647
Sequential 560
Medium 248
Long 835
Total 2290
Number of hydrogen bonds 92

Structural Statistics

<SA>3 <SA>,b
RMSD from experimental distance restraints (A) (2290) 0.036 t 0.001 0.028
Deviation from idealized covalent geometry
Bonds (A) 0.004 i 0.001 0.003
Angles (deg) 0.742 t 0.028 0.583
Impropers (deg) 0.604 3: 0.006 0.482

Measures of Structural Quality (By Procheck)

% residues in most favorable regions of the Ramachandran plot 55.6 i 2.4 54.8
% residues in additional allowed regions of the Ramachandran 37.1 :1: 1.6 37.1
plot
No. of bad contacts 8 :1: 2 8
H-bond energy 0.40 i 0.10 0.50
Overall G-factor -0.30 :1: 0.02 -0.30

Coordinate Precision

RMSD for Ca trace (A) 0.84 i 0.21 0.47 :t 0.15 °
RMSD for backbone atoms (A) 0.89 3: 0.25 0.54 i 0.18 °
RMSD for all heavy atoms (A) 1.12 :t 0.28 0.92 :1: 0.20c

 

“ <SA> are the final 25 simulated annealing structures.

“ <SA>r is the restrained minimized mean structure obtained by restrained regularization
of the mean structure, which is obtained by averaging the coordinates of the individual
SA structures best fitted to each other.

° Residues from Ala32-Pro39 and Thr57-Glu62 were excluded from the RMSD calculations.

125

was due to the looseness or absence of the backbone NOE restraints for more than one-
fourth of the total 137 residues.

Description of the solution structure of apo-CRABPII. The solution structure of
apo-CRABPII mainly consists of two nearly orthogonal anti-parallel B-sheets twisted to
form a flattened barrel, a structural feature shared by all the published structures of iLBPs
(Banaszak et al., 1994). As in other iLBPs, owing to a large gap, there are no main-chain
hydrogen bonds between BD and BE. The gap gradually widens fi'om the N-terminus to
the C-terminus of BD (Figure 3. 6A). The C-terminal half of BC and the N-terminal half
of BD together with the loop connecting them move away from the main protein body
and extend into the solvent. The C-tenninal half of BB together with the loop connecting
BE and BF also moves away the main protein body. The entrance to the binding pocket is
formed by the hydrophobic side-chains of Leu19, Va124, Leu28—Pro39 and the residues
in the BC—BD loop and BE—BF loop. The distance from the C“ atom of Va158 in the BC—
BD loop to the C“ atom of Val76 in the BE—BF loop is about 20.7 A. The distance
between the C“ atom of Va124 of the loop connecting 01A to 018 and the C“ atom of Val76
is 17.0 A. The ligand binding pocket is wide open and the side chains of the trio (Argl 1 l,

Arg132 and Tyrl34) are readily accessible to the external ligand.

126

Discussion

Assignment and secondary structures. The amide resonances of eleven residues
were not observed, and about twice the number of residues had weak peaks in the 2D 1H-
15N HSQC spectrum. This posed a challenge for the sequential assignment method based
on a series of ISN-edited heteronuclear NMR experiments because these experiments
depend on magnetization transfer of amide resonances. Their assignments could not be
accomplished without the help of the comparisons with the sequential assignments of
both the wild type holo-CRABPII and the site-directed mutant apo-R11 1M. However, the
intensities of the amide proton signals from Arg29, Ala36, Lys38 and Arg59 were so low
in both the 2D lH-‘SN HSQC and the homonuclear 2D SSNOESY and WATERGATE
NOESY spectra that none of their 1H and 15N resonances has been assigned. Secondary
structure elements were identified according to the NOE patterns. For residues with weak
or no observable amide peaks in NOESY spectra only the a",m and daB(i, i +3) NOEs were
used. Several residues had strong peaks in the fingerprint regions of the 2D homonuclear
spectra recorded in DZO but had only very weak peaks in the 15N-edited heteronuclear
spectra. Their secondary structural identities were established by the backbone hydrogen
bond, and the NOEs observed in the homonuclear spectra. All the available data, in
particular, the homonuclear 2D SSNOESY, WATERGATE NOESY spectra were
analyzed carefully to assign the resonances of Met27-Ala36, which assume an or-helical

structure (018) according to the crystal structure of holo-CRABPII. Some weak sequential

127

NOEs typical of the regular helical structure were observed for Met27-Ile31 but their
intensities and numbers were well below the average (Figure 3.5). No structurally useful
sequential NOEs could be identified for residues Ala32-Ala35. However, the chemical
shift indices suggested that Lys30-Ala35 are helical. No NOEs could be identified
unambiguously for Ser37 except its intra-residue NOEs. These experimental data
suggested that the residues Met27-Ala35 adopt an unstable or-helical structure. Residues
Ala32-Pro39 are disordered in the NMR structures because no chemical shift indices
were included in the structure refinement (Figure 3.6A).

Comparisons with the crystal structures of halo-CRABPII and apo-CRABPI.
Figure 3.6B shows the overlay of the C“ traces of the restrained minimized mean structure
of the apo-CRABPII and the crystal structure of holo-CRABPII. The structure can be
superimposed with C“ deviations of 3.52, 2.04, and 1.65 A for all residues, the regular
secondary structures, and all residues excluding Va124-Ser37, Glu74-Pro80 and Leu100-
Gly104, respectively. The main differences between the two structures are located in the
regions around the ligand entrance. The second helix is well defined in the holo crystal
structure but appears flexible in solution, especially the C-terminal half. The BC-BD loop
moves away fiom the main body of the protein by ~5 A in the apo solution structure. The
C-terminal half of BE and the N-terminal half of BF in the apo solution structure do not
twist toward the center of the protein to constrict the ligand entrance as they do in the
holo crystal structure. Consequently, the long loop between BG and BH also moves away
from the main body of the apo protein. The distance between the C“ atom of Val76 in the
BE—BF loop and the C“ atom of V3158 in the BC—BD loop is ~8.0 A longer in the apo

solution structure than in the holo crystal structure. The distance between the C“ atom of

128

Val76 and the C“ atom of Va124 in the orA-orB loop is ~6.7 A longer in the apo solution
structure than in the holo crystal structure. As a result, the ligand entrance in the apo
solution structure is greatly enlarged with a much more exposed binding pocket. In
particular, residues Va124, Leu28, Ile31, Va158, Arg59, and Val76 move away from each
other so that the side-chains of Argl 1 l, Arg132 and Tyrl34 that interact with the
carboxyl group of RA are easily accessible to the ligand. However, the relative positions
of Argl l 1, Arg132, and Tyrl34 are quite similar to those observed in the holo crystal
structure: their backbone atoms can be superimposed rather well.

Compared with other parts of the protein, the ligand entrance region of apo-
CRABPII is not well defined in the solution structure. Most residues in the region have
weak or no observable peaks in all 'SN-edited spectra. The number of the experimental
distance restraints per residue for the region is well below the average (Figure 3.5).
However, except for Ala32—Pro39, the differences between the apo solution structure and
the holo crystal structure (Figure 3.6B) are larger than between any two conformations
from the ensemble of 25 refined NMR solution conformations (Figure 3.6A). Thus, the
differences between the apo solution structure and the holo crystal structure are real.

Most of the differences between the apo solution structure and the holo crystal
structure of CRABPII are also observed between the solution structure of apo-CRABPII
and the crystal structure of apo-CRABPI (Thompson et al., 1995) (Figure 3.6C).

Dynamical properties. As detailed earlier, about a quarter of the residues showed
weak or no signals in the 2D 'H-‘sN HSQC spectrum (Figure 3.2). Some of them were
situated in the turns or loops between elements of regular secondary structure.

Surprisingly, many others assumed regular a-helical or B-sheet structures. Only the

129

residues in the BC-BD loop and the BE-BF loop and Asn115 in the BH-BI loop showed
weak or no cross-peaks in the lH-‘SN HSQC spectrum. The residues in the longest loop
between BG and BH showed the highest intensities in the lH-‘SN HSQC spectrum. By
contrast, residues in the long loop of holo-CRABPII had high temperature factors in the
X-ray structure. Interestingly, most of these residues that showed weak or no cross-peaks
in the lH-‘SN HSQC spectrum are located in the RA-binding pocket (Figure 3.7). Among
the 21 residues that constitute the RA-binding pocket (Kleywegt etal., 1994), 16 residues
exhibited weak or no detectable cross-peaks in the 'H-‘sN HSQC spectrum.

The cluster of the weak or missed lH-‘SN correlations suggested that the RA-
binding pocket of CRABPII is rather dynamic in the absence of RA. It is unlikely that
these peaks were affected by water saturation, because the lH-‘SN HSQC spectrum was
recorded with a “water flip-back” pulse sequence that minimized water saturation and
dephasing (Grzesiek & Bax, 1993). Furthermore, for several weak or missing IH-‘SN
correlations, the amide protons remained two days after the lyophilized protein was
dissolved in D20 and showed cross-peaks in the fingerprint regions of the homonuclear
2D spectra recorded in D20 (Figure 3.3), indicating that these amide protons exchange
slowly with water and should not be perturbed by water saturation. Rather, the lH-‘SN
correlations were most likely weak or missing as the result of broadening of the amide
proton and/or nitrogen resonances due to conformational exchange at rates intermediate
on the NMR time scale. For Lys8, Ile10, Leu19, Leu121 and Argl32-Vall35, clearly,
line broadening of their nitrogen resonances was responsible for the weak peaks in the
lH-‘SN HSQC spectrum because their amide protons exhibited sharp signals and showed

cross-peaks with a-protons in 2D homonulcear spectra recorded in D20 (Figure 3.3). A

130

Figure 3.7. The relative peak intensity of the 2D lH—‘SN HSQC spectrum of apo-
CRABPII color-coded along the C“ trace of the solution structure. The strong peaks

are in white, the weak peaks in pink and the missing peaks in cyan. RA (green) is
positioned on the basis of the superposition of the solution structure of apo-CRABPII and

the crystal structure of holo-CRABPII.

l3l

 

132

slow conformational exchange with a lifetime of at least 40 ms was observed as described
in Chapter 2, resulting in two sets of cross-peaks for Trp87. It was noted that Trp87 is
located at the N-terminus of BF far away from the RA-binding pocket. The residues in the
RA-binding pocket likely undergo intermediate conformational exchange, probably on
the microsecond to millisecond time scale, because only one set of NMR signals was
observed for these residues. The clustering of the mobile residues and the relatively low
frequency of conformational exchange suggest that the RA-binding pocket may undergo
collective motions in the absence of the RA ligand.

Apart from the RA-binding pocket, the structure of apo-CRABPII appears well
ordered and less dynamic. Most residues, including those at the turns and loops, showed
intense cross-peaks in the 2D lH-‘SN HSQC spectrum. Many of them also showed cross-
peaks in the fingerprint regions of the 2D homonuclear spectra recorded in D20 (Figure 3.
3). Remarkably, the hydroxyl protons of Ser83 and Thr107 were observed in 2D
homonuclear spectra recorded in H20, and the indole e'-proton of Trp7 was detected in
2D homonuclear spectra recorded in D20, suggesting that these side-chains are not only
buried but also immobile.

The dynamic nature of the RA-binding pocket is consistent with biochemical and
NMR studies of CRABPI. Limited proteolysis showed that helix 012 of CRABPI is
significantly more susceptible to proteolysis in the apo-form than in the holo-form
(Jamison et al., 1994), suggesting that RA binding induces a conformational change or
that helix 012 is more mobile in the apo-form. Hydrogen exchange measurements revealed
that helix a2 and the BC-BD loop of CRABPI have much higher exchange rates in the

apo-form than in the holo-fonn (Rizo, et al., 1994), indicating that these parts of the

133

molecule are more dynamic in the apo-form. Other iLBPs also have been found to
exhibit similar dynamic properties. For example, it was shown by NMR that the second
helix of rat intestinal F ABP is disordered in solution (Hodsdon and Cristola, 1997a, b).

Is CRABPII dimeric in solution? Unlike other iLBPs, apo-CRABPI was
crystallized in a dimeric form (Thompson et al., 1995). The crystalline dimer is held
together by an intermolecular B-sheet formed by the BD strands of two CRABPI
molecules, resulting in a 20-stranded double B-barrel with slightly more open RA-binding
pockets. It was suggested that dimerization may be the mechanism by which CRABPI
opens the ligand entrance so that RA can enter or exit the binding pocket without steric
hindrance. In light of the potential significance of the dimerization to the ligand entrance
problem, we carefully examined the possibility of the formation of the intermolecular B-
sheet. We concluded that apo-CRABPII is predominately monomeric in solution for the
following reasons. (i) On a Sephadex G-50 gel filtration column, apo-CRABPII eluted
with a volume characteristic of monomeric CRABPII. (ii) For the most part, the line
widths of the NMR signals of CRABPII were consistent with a monomeric structure. (iii)
None of the predicted NOEs indicative of formation of the intermolecular B-sheet was
observed. For example, according to the dimeric crystal structure of apo-CRABPI, the
distance between H“ of Thr61 of molecule A and H“ of Thr60 of molecule B is 2.5 A but
the distance between H“ of Thr61 and H“ of Thr60 of the same molecule is 5.9 A.
Therefore, if there is a stable dimer in solution, we should see a NOE cross-peak between
the two protons. On the other hand, if there is no stable dimer, there would be no NOE
between the two protons. The two protons were sequentially assigned, but we did not see

any NOE cross-peak between them. Although the formation of transient dimers could not

134

be ruled out, the transient dimers, if they exist, can only be a minor population under the
conditions of the NMR or gel filtration experiments. What is the likelihood of the
dimerization in viva? Since the cellular concentration of CRABPII is much lower than the
concentrations of the NMR samples, the concentration of dimeric species would be even
lower in vivo.

Implications for RA binding. The crystal structures of holo-CRABPI and
CRABPII suggested that RA cannot enter or exit the ligand binding pocket of the proteins
in the absence of significant conformational changes. On the basis of limited proteolysis
of several intracellular lipid binding proteins, it was proposed that a rigid body movement
of the helix-turn-helix motif may serve as the mechanism by which the family of proteins
opens up their ligand binding pocket (Jamison et al., 1994). In light of the solution
structure of apo-CRABPII, it is unlikely that the two helices move as rigid rods as
previously proposed. Rather, the second helix of the helix-tum-helix motif is flexible and
undergoes conformational exchange. The dynamical nature of the second helix may be
the cause of its high susceptibility to proteolysis.

Based on the comparisons of the crystal structures of apo- and holo-CRABPI, it
was suggested that movement of the BC—BD loop is responsible for the opening of the
ligand entrance (Thompson et al., 1995). Furthermore, it was also proposed that the
movement of the BC—BD loop is dependent on the formation of an intermolecular B-sheet
as mentioned above. Comparison of the solution structure of the apo-CRABPII and the
crystal structure of holo-CRABPII confirms the hypothesis that BC—BD loop moves when

RA is bound. However, the movement of the BC—BD loop that opens the binding pocket

135

is shown by NMR not to involve dimerization of the protein because CRABPII is
predominately monomeric in solution.

The solution structure of apo-CRABPII suggests that the ligand entrance is widely
open and readily accessible to RA. The enlargement of the ligand entrance of apo-
CRABPII compared with holo-CRABPII is mainly due to a concerted conformational
change in three structural elements, namely the second helix, the BC—BD loop and the
BE—BF loop. A widely opened entrance may be essential for binding of RA, because RA
is a long, relatively rigid, negatively charged molecule. A narrow entrance lined with the
hydrophobic side-chains would result in unfavorable interactions with RA. The entry of
RA into the deep binding pocket may be further facilitated by the positively charged
potentials generated by Arg29, Arg59, Argl 11 and Arg132 (Chen et al., 1998). Binding
of RA induces significant conformational changes in CRABPII. Furthermore, the
interactions between RA and CRABPII may stabilize the structural elements constituting

the RA-binding pocket, especially the second helix, the BC—BD loop and the BE—BF loop.

136

Conclusion

Using multidimensional NMR spectroscopy, we have determined the first solution
structure of a CRABP. Comparison of the solution structure of apo-CRABPII with the
crystal structure of holo-CRABPII indicates that the largest conformational differences
between the two structures are localized at the ligand entrance. The ligand entrance of
apo-CRABPII is greatly enlarged and readily accessible to RA, mainly due to a concerted
movement of the second helix and the BC-BD and BE-BD loops. Furthermore, the ligand
binding pocket of apo-CRABPII is rather dynamic as indicated by analysis of the cross-
peaks intensities of the lH-‘SN HSQC spectrum. CRABPII is predominately monomeric
in solution as determined from NMR and biochemical evidence. Although the formation
of transient dimers could not be ruled out, dimerization apparently is not a prerequisite

for entry of RA into the ligand binding pocket of CRABPII.

137

References

Anil-Kumar, Ernst, R. R., & Wiithrich, K. (1981) Biochem. Biophys. Res. Commun. 95,
1-6.

Banaszak, L., Winter, N., Xu, Z., Bernlohr, D. A., Cowan, S. W., & Jones, T. A. (1994)
Adv. Protein Chem. 45, 89-151.

Basus, V. J. (1989) Methods Enzymol. 1 7 7, 132-149.

Bax, A., Clore, G. M., & Gronenbom, A. G. (1991) J. Magn. Reson. 88, 425-431.

Bax, A., Griffey, R. H., & Hawkins, B. L. (1983) J. Magn. Reson. 55, 301-315.

Bodenhausen, G., & Ruben, D. J. (1980) Chem. Phys. Lett. 69, 185-188.

Brooks, B. R., Bruccoleri, R. E., Olafson, B. D., States, D. J ., & Karplus, M. (1983).].
Comput. Chem. 4, 187-217.

Briinger, A. T. (1992) X-PIOR: A System for Crystallography and NMR. Yale University,
New Haven, CT.

Chambon, P., Olson, J. A., & Ross, A. C. (coordinators) (1996) The retinoid revolution.
FASEB. J 10, 938-1048.

Chen, X., Tordova, M., Gilliland, G. L., Wang, L., Li, Y., Yan, H., and 11, X. (1998) J.
Mol. Biol, in press.

Clore, G. M., & Gronenbom, A. G. (1989) CRC Crit. Rev. Biochem. Mol. Biol. 24, 479-
564.

Griesinger, C., Otting, G., Wiithrich, K., & Ernst, R. P. (1988).]. Am. Chem. Soc. 110,

138

7870-7872.

Grzesiek, S., & Bax, A. (19923) J. Magn. Reson. 96, 432-440.

Grzesiek, S., & Bax, A. (1992b) J. Am. Chem. Soc. 114, 6291-6293.

Grzesiek, S., Anglister, J., & Bax, A. (1993) J. Magn. Reson. Ser. B 101, 114-119.

Grzesiek, S., & Bax, A. (1993).]. Am. Chem. Soc. 115, 12593-12594.

Hodson, M. E., & Cistola, D. P. (1997a) Biochemistry 36, 1450-1460.

Hodson, M. E., & Cistola, D. P. (1997b) Biochemistry 36, 2278-2290.

Jamison, R. S., Newcomer, M. E., & Ong, D. E. (1994) Biochemistry 33, 2873-2879.

Jeener, J ., Meier, B. H., & Ernst, R. R. (1978) J. Chem. Phys. 71, 4546-4553.

Kay, L. E. (1995) Prog. Biophys. Molec. Biol. 63, 277-299.

Kay, L. E., & Bax, A. (1990) J. Magn. Reson. 86, 120-126.

Kay, L. E., Ikura, M., Tschudin, R., & Bax, A. (1990) J. Magn. Reson. 89, 495-514.

Kay, L. E., Keiffer, P., & Saarinen, T. (1992) J. Am. Chem. Soc. 114, 10663-10665.

Kay, L. E., Xu. G. Y., Singer, A. U., Muhandiram, D. R., & Forman-Kay, J. D. (1993) J.
Magn. Reson. BIO], 333-337.

Kleywegt, G. J ., Bergfors, T., Senn, H., 1e Motte, P., Gsell, B., Shuao, K., & Jones, T. A.
(1994) Structure 2, 1241-1258.

Laskowski, R., MacArthur, M. W., Moss, D. S., & Thornton, J. M. (1993) J. Appl.
Crystallogr. 26, 283-291.

Laskowski, R., Rullmannn, J. A., MacArthur, M. W., Kaptein, R., & Thornton, J. M.
(1997) J. Biomol. NMR 8, 477-486.

Logan, T. M., Olejniczak, E. T., Xu, R. X., & Fesik, S. W. (1993) J. Biomol. NMR 3,

225-231.

139

Mangelsdorf, D. J ., Umesono, K., & Evans, R. M. (1994) In The Retinoids: Biology,
Chemistry, and Medicine (Sporn, M. B., Roberts, A. B., & Goodman, D. S., eds.), pp.
319-349, 2nd ed., Raven Press, New York.

Marion, D., Ikura, M., Tschudin, R., & Bax, A. (1989a) J. Magn. Reson. 85, 393-399.

Marion, D., Ikura, M., & Bax, A. (1989b) J. Magn. Reson. 84, 425-430.

McCoy, M. A., & Mfieller, L (1992a) J. Am. Chem. Soc. 114, 2108-2112.

McCoy, M. A., & Miieller, L (1992b) J. Magn. Reson. B98, 674-679.

Muchmore, D. C., McIntosh, L. P., Russell, C. B., Anderson, D. E., & Dahluist, F. W.
(1989) Methods Enzymol. I 7 7, 44-73.

Muhandiram, D. R., & Kay, L. E. (1994) J. Magn. Reson. 3103, 203-216.

Nilges, M., Clore, G. M., & Gronenbom, A. M. (1988) FEBS Lett. 229, 129-136.

Ong, D. 13., Newcomber, M. E., & Chytil, F. (1994) In The Retinoids: Biology,
Chemistry, and Medicine (Spom, M. B., Roberts, A. B., & Goodman, D. S., eds.), pp.
283-317, 2“d ed., Raven Press, New York.

Piantini, U., Sorensen, O. W., & Ernst, R. P. (1982) J. Am. Chem. Soc. 104, 6800-6801.

Rizo, J ., Liu, Z. P., & Gierasch, L. M. (1994) J. Biomol. NMR 4, 741-760.

Sacchettini, J. C., Gordan, J. 1., & Banaszak, L. J. (1988) J. Biol. Chem. 263, 5815-5819.

Shaka, A. J ., Keeler, J., Frenkiel, T., & Freeman, R. (1983).]. Magn. Reson. 52, 335-338.

Shaka, A. J., Lee, C. J., & Pine, A. (1988) J. Magn. Reson. 77, 274-293.

Shaka, A. J ., Barker, P. B., & Freeman, R. (1985) J. Magn. Reson. 64, 547-552.

Smallcombe, S. H. (1993) J. Am. Chem. Soc. 115, 4776-4785.

Sporn, M. B., Roberts, A. B. & Goodman, D. S. (1994) The Retinoids: Biology,

Chemistry, and Medicine, 2"d ed., Raven Press, New York.

140

States, D. J ., Haberkon, R. A., & Ruben, D. J. (1982) J. Magn. Reson. 48, 286-292.

Thompson, J. R., Bratt, J. M., & Banaszak, L. J. (1995) J. Mol. Biol. 252, 433-446.

Wittekind, M., & Miieller, L. (1993) J. Magn. Reson. 101, 201-205.

Wiithrich, K., Billletter, M., & Braiin, W. (1983) J. Mol. Biol. 169, 949-961.

Zhang, J ., Liu, Z., Jones, T. A., Gierasch, L. M., & Sambrook, J. F. (1992) Proteins:
Structure, Function, and Genetics 13, 87-99.

Zhang, 0., Kay, L. E., Olivier, J. P., & Forman-Kay, J. D. (1994) J. Biomol. NMR 4,
845-858.

Zhu, G., & Bax, A. (1992).]. Magn. Reson. 90, 405-410.

Zuiderweg, F. R. P., & F esik, S. W. (1989) Biochemistry 28, 2387-2391.

CHAPTER 4

Solution Structure of a Site-Directed

Mutant (R1 11 M) of Human CRABPII

Introduction

The crystal structures of holo-CRABPs reveal that RA is buried in the binding
pocket with its carboxyl group interacting with Argl l 1, Arg132 and Tyrl34 (CRABPII
numbering) at the bottom of the pocket (Kleywegt et al., 1994). Arg132 and Tyrl34
interact directly with the carboxyl group of RA, whereas the interaction between Argl 11
and the carboxyl group of RA is mediated by a water molecule. Electrostatic calculations
based on a modeled structure of CRABPI have suggested that the electrostatic repulsion
between the guanidinium groups of the above two conserved arginine residues provides
the conformational flexrbility to CRABPs for RA to enter the binding pockets (Zhang et
al., 1992).

Site-directed mutagenesis and competitive binding studies have shown that both
Argl 11 and Argl32 are important for CRABPII to bind RA but Argl ll contributes more
to the binding energy than Argl32 (Chapter 2). This chapter describes the structure

determination of the CRABPII mutant protein R1 1 1M in solution by NMR spectroscopy.

141

142

The results suggest that Argl 11 play a critical role in determining the structure and

dynamical properties of CRABPII.

143

Experimental Procedures

Sample preparation. Unlabeled amino acids, nucleotides, vitamins and D-
glucose were purchased from Sigma (St. Louis, Missouri). ['5N]-labeled amino acids and
'SNH4C1 were purchased from Cambridge Isotopic Laboratories.

Site-directed mutagenesis and expression, and the purification of the recombinant
mutant protein R11 1M have been described in Chapter 2. Uniform 15N-labeling and ”N-
labeling of leucine were performed as detailed in Chapter 3. The procedure for 15N-
labeling of isoleucine and phenylalanine was the same as that for isoleucine and leucine.
For 15N-labeling of lysine, the expression strain BL21(DE3)/pLysS containing the
plasmid pET-l7b/R111M was used. The strain was grown in M9 medium supplemented
with 110 mg/L [‘SN]-lysine added ten minutes before the induction with IPTG. D20
sample was prepared by dissolving the lyophilized protein in PBS buffer (20 mM sodium
phosphate, 100 mM sodium chloride in 99.9% DZO, pD 7.5, uncorrected). The protein
solution was incubated for six hours at 37°C and lyophilized again. The lyophilized
protein was then dissolved in 99.96% D20. HZO samples were prepared by directly
concentrating the protein fractions eluted with the PBS buffer (pH 7.3) from a gel-
filtration column (G-50) at the final step of the purification procedure (Chapter 2). About
10% D20 was added into the concentrated protein solutions. The protein concentrations

of the NMR samples were ~ 2 mM.

144

NMR spectroscopy. All NMR spectra were recorded at 27°C. Homonuclear 2D
spectra of unlabeled R11 1M in D20 and heteronuclear 2D lH-‘SN HMQC (Miieller, L.
1979; Bax et al., 1983) spectra of selectively l5N-labeled proteins in H20 were acquired
on a Varian VXR-SOOS spectrometer. All 2D experiments were recorded in the phase-
sensitive mode with F1 quadrature detection achieved by the hypercomplex method
(States et al., 1982). The relaxation delay was 1.5 second, and the residual water
resonance was suppressed by low-power presaturation. The spectral width was 7238 Hz
for 'H and 2000 Hz for l5N. The following 2D homonuclear spectra were recorded: one
DQF-COSY (Piantini et al., 1982) spectrum, one E.COSY (Griesinger et al., 1985)
spectrum, three NOESY (Jeener et al., 1978; Kumar et al., 1981) spectra with mixing
times of 50, 100 and 150 ms, and three clean-TOCSY (Griesinger et al., 1988) spectra
with mixing times of 20, 45 and 75 ms. The time domain data were composed of 2048 X
320 complex points (128 transients) for the E.COSY and DQF-COSY experiments and
2048 X 256 complex points (80 transients) for all other homonuclear 2D experiments.
The 2D lH-‘SN HMQC spectra were recorded with 1024 (‘H) X 64 (”N) complex points
(32 transients). The data were processed with the program VNMR v. 5.1 (Varian
Associates). A ~30°-shified sine-bell was applied in both dimensions for processing the
DQF-COSY and E.COSY data, and a ~45°-shified sine-bell for the lH-‘SN HMQC data.
A gaussian and minus LB combination was applied in F2 dimension and a ~75°-shifted
sine-bell in F1 dimension for processing the NOESY and TOCSY data. The time domain
data were then zero-filled to 2048 X 512, 4096 X 2048, 4096 X 2048, and 8192 X 4096

real points for the 2D HMQC, TOCSY, NOESY, DQFCOSY and E.COSY, respectively.

145

A five-order polynomial was applied for baseline correction in the F2 dimension after
Fourier transformation.

Heteronuclear NMR experiments of the uniformly ”N labeled protein were carried
out on a Briiker DMX-600 spectrometer. The carriers for 1H and ”N were set at 4.70 ppm
and at 115.7 ppm, respectively. A 2D lH-”N HSQC spectrum (Bodenhausen & Ruben
1980; Cavanagh et al., 1991; Palmer et al., 1991; Kay et al., 1992) was acquired with
coherent selection by pulsed field gradients and sensitivity enhancement. The acquisition
times and complex data points were IH (F2) 113.9 ms, 1024 and ”N (F1) 56.9 ms, 256
(32 transients). Both 3D lH-”N TOCSY-HMQC (Marion et al., 1989b) and NOESY-
HMQC (Zuiderweg and Fesik, 1989; Marion et al., 1989c) spectra were recorded with the
following acquisition times and complex data points: lH (F3) 113.9 ms, 1024; ”N (F2)
11.9 ms, 32; 'H (Fl) 14.3 ms, 128 (8 transients and 45 ms mixing time for TOCSY-
HMQC, and 16 transients and 150 ms mixing time for NOESY-HMQC). In both 3D
experiments the solvent resonance was suppressed by low-power presaturation.
Quadrature detection in indirectly detected dimensions were accomplished by the States-
TPPI method (Marion et al., 1989a). Homonuclear isotropic mixing in TOCSY-HMQC
experiment was achieved by a DIPSI-3 sequence (Shaka et al., 1987). Two dimensional
IH-”N HMQC-J (Key and Bax, 1990) data was recorded on a Briiker DMX-750
spectrometer at the National Magnetic Resonance Facility at Madison with the following
acquisition times and complex data points: lH 136.4 ms, 2048 and ”N 232.7 ms, 512 (16
transients). A low power GARP-I sequence (Shaka et al., 1985) was applied for ”N
broad-band decoupling during the data acquisition periods of these experiments. The

spectra were processed with the program NMRPipe (Delaglio et al., 1995). A ~45°-

146

shifted sine-bell was applied in both F2 ('H) and F l (”N) dimensions for both 2D lH-”N
HSQC and HMQC-J spectra. The lH-”N HSQC and HMQC-J data were zero-filled to
2048 (F2) x 512 (F1) and 2048 (F 2) x 1024 (F1) real points, respectively. A ~75°-shifted
sine-bell was applied in both Fl ('H) and F3 ('H) dimensions and a cosine-bell in F2
(”N) dimension for processing of the 3D data. The first points of F l and F2 dimensions
were calculated by linear prediction. Both 3D data were zero-filled to 2048 (F3) x 128
(F2) x 512 (F1) real points. The aliphatic part in the F3 dimension was removed after
Fourier transformation.

Resonance assignment. The sequential assignments were carried out following
the standard procedure developed by Wiithrich and others (Wiithrich, 1986; Clore &
Gronenbom, 1989 and 1991). Briefly, the spin systems were identified by the analysis of
the homonuclear 2D TOCSY and DQF-COSY spectra and the 3D lH-”N TOCSY-
HMQC spectrum or by selective labeling. Sequential linking was obtained from NOE
connectivities observed in 3D lH-”N NOESY-HMQC experiment. The stereospecific
assignments of B-mythylene protons were obtained by the joint analyses of the qualitative
“JG, coupling constants derived from the E.COSY and DQF-COSY spectra and the
relative volumes of the NOE cross peaks between H“—H“ and HN—H“ from the 2D
homonuclear NOESY and 3D lH-”N NOESY-HMQC spectra (Basus, 1984). The methyl
protons of valine and leucine residues and the ring protons of aromatic residues were
stereospecifically assigned as detailed in Chapter 3. Secondary structural elements were
identified according to their characteristic NOE patterns.

Derivation of structural restraints. Three types of structural restraints were

derived from the experimental NMR data: interproton distance restraints, torsion angle

147

restraints and hydrogen bond restraints. Approximate interproton distance restraints were
derived from the homonuclear 2D NOESY with 100 and 150 ms mixing times, and the
3D ”N-edited NOESY-HMQC with 150 ms mixing time. The same procedure for
establishing the upper limits of interproton distances was followed as described in
Chapter 3. The only difference was that the NOE cross peaks from the 3D lH-”N
NOESY-HMQC were picked by the programs CAPP and PIPP (Garret et al., 1991). The
NOEs were classified as strong, medium, weak and very weak corresponding to
interproton distance 1.8—2.7 A (1.8—2.9 A for NOEs involving NH protons), 1.8—3.3 A
(1.8—3.5 A for NOEs involving NH protons), 1.8—5.0 A and 2.5—5.0 A, respectively.
Pseudoatom corrections were made for non-stereospecifically assigned methylene protons
and methyl protons (Wiithrich et al., 1983). Additional 0.5 A was added to the upper
bounds for methyl protons. The backbone torsion angle 11) was restrained according to the
value of “JNHQ obtained from 2D HMQC-J experiment. It was restrained —120° i 40° for
UN”, 2 7.5 Hz, and —60° i 20° for 3.1M,CL .<_ 5.0 Hz and when NOE data indicated the
presence of a helix. Backbone hydrogen bond restraints were derived from the cross
peaks remaining in the fingerprint regions of the homonuclear 2D TOCSY and NOESY
spectra recorded in D20. The HN---O and N---O were restrained to be 2.0 i 0.5 A and 3.0
i 0.3 A, respectively.

Structure calculation. The structures were calculated as detailed in Chapter 3
except that an additional square-well potential fimction with force constant of 200

kcal.mol".rad'2 was applied for the dihedral angle restraints.

148

Results

Isotopic enrichment. The level of ”N enrichment in the uniform l“N-labeling was
high as judged by the signal to noise ratio of the 2D 'H-”N HSQC spectrum (Figure 4.1).
No l5N isotopic scrambling to other residues was observed for selective labeling of
isoleucine, leucine, lysine and phenylalanine. For example, the number of the cross peaks
in the 2D 'H-‘SN HMQC spectrum of the [”N]-leucine-labeled protein was exactly equal
to the total number of leucine residues in R11 1M (Figure 4.2).

Spin system identification. Aliphatic spin systems were first identified by the
analysis of the homonuclear 2D DQF-COSY and TOCSY spectra in D20 and then
extended to amide resonances by the 3D lH-”N NOESY-HMQC experiment. Most of the
threonine, alanine, valine and glycine residues could be identified easily by their unique
cross-peak patterns. The majority of the residues belonging to either AMX or AM(PT)X
spin system could also be identified. However, except for a few, the residues with long
side-chains (lysine, leucine, isoleucine, arginine and proline) could not be distinguished
from each other because of severe spectral overlapping in the upfield region (1.00 ppm-
2.00 ppm) and the inefficiency of the TOCSY experiments for long-relayed coherence
transfer. Their spin system identifications were achieved or assisted by selective l“N-
labeling. All the leucine, isoleucine and lysine residues could be identified by selective
”N-labeling. Furthermore, proline and arginine could also be distinguished from each

other. In general, the H“ protons of prolines have larger down field chemical shifts than

149

Figure 4.1. Gradient- and sensitivity-enhanced 2D lH—‘SN HSQC spectrum of uniformly
”N enriched of apo-R1 1 1M. Sequential assignments are indicated with one-letter amino

acid code and residue number. The side-chain amides of Asn and Gln are indicated by #.

 

150

 

 

 

105
623
I 0
6370454 04514
610200 G78
G104 ‘ - —110
ON,“ N14» 0
T110 T g 074* N115“ ‘ Q
N 1
wean N “337 '\' 0N“
.889 3.4 8108. N9 0356;; N28!“ :4#
N9”! N1
077 N115#IN25# N25“
698 T‘s 90'9" 69.771 ’115
812 o L 9
4
N14 0 7‘2 oovvo109 T49 0116
T122 V339 0g
T114 °.E16° v 0 F65 513°01‘10" K330913118
C8129163 E1476 2
110. T61 T17ROE72A126°5N°M027L38K9R59031h120
”"5 M111 F15
K66 .K 2 419R110L
55 W87 E7370012,,“26 “1569“" L1
15
7 M13 :3; .133 V1. 14%,: 6197
V 8. T60E627Y15 100 434 (ppm)
. 0126 00208?
1
F3 “3' T:E1 74136 1228 492
“’7’ F50 . . —125
9 W7N‘0 ‘513 L119 "29 T131
K11 1120 583 9
.0 0 0 A1 5 L113
Y134 "31321193 v 7 L99 '9 .04“ t,
A40/N25 5137 596
K34 1538 (:95 45
Q 0 0542 E46 —130
warm C13“ V85”
E70
0
K53
-'135
0
L121
1 1 1 1 1 1 1
10 5 10 0 9.5 .5 8 0 7 5 7 0

o 1 a
H(ppm)

151

Figure 4.2. 2D 'H—‘SN HMQC spectrum of the [‘SN]-leucine selectively labeled apo-
R1 1 1M. Sequential assignments are indicated with one-letter amino acid code and residue

number.

152

 

 

0
L19
0
L28
L100
0 0
L18
1.119
a
1.99

0
L113

L121

[:30

—

126.0 120.0 115.0
DZ (ppm)

130.0

135.0

 

 

9.6 910 8.4 7.8
D1 (ppm)

153

those of arginines. For example, Prol was identified in this way. The cross peaks
corresponding to H“ protons of four arginines could be identified in the 2D lH-‘SN HSQC
spectra. Consequently, the H‘ was linked to the HY and H‘3 protons, and eventually to H“
protons by the 3D lH-‘SN TOCSY-HMQC experiment. The residues belonging to AMX
spin system were further distinguished from each other based on distinct chemical shifis
and NOEs. Four of the seven serines could be identified by the down field positions of
their HB protons. All four asparagine residues could be identified by the NOE cross-peaks
between Hfl and H6 protons observed in the 3D 'H-‘SN NOESY-HMQC spectrum. All five
phenyalanines were identified by selective labeling, and confirmed by the homonuclear
2D NOESY experiments. One tyrosine and two tryptophan residues could be identified
after the assignments of their aromatic ring protons. Two glutamines could also be
differentiated from the other residues belonging to the same AM(PT)X spin system by the
NOE cross-peaks between H7 and H‘ protons. All remaining residues were classified into
several groups, and identified later along with sequential assignment. The positional lH
resonance assignments of the side-chains were obtained mainly through the ZD DQF-
COSY spectrum, and three TOCSY spectra with different mixing times. Remarkably, the
hydroxyl protons of both Ser83 and Thr107 were observed in both homonuclear and
heteronuclear NMR experiments performed in H20 (data not shown), and were assigned
by the homonuclear TOCSY and NOESY experiments. The indole 8' proton of Trp7 was
observed in the NMR spectra recorded on D20 sample.

The ring protons of aromatic residues. The assignments of aromatic ring protons
were derived from the 2D DQF-COSY spectrum, and three TOCSY spectra with different

mixing times. The ring protons of phenylalanines and tyrosines were linked to their HB,

154

HCl and HN protons by NOE cross peaks between H‘3 and H5 protons and between H"1 and
H5 protons observed in the 2D homonuclear NOESY. The H" and H51 protons of
tryptophans could be assigned by the 3D lH-‘SN NOESY-HMQC experiment, and H83
protons by NOEs between H“1 and H‘13 protons and between HB and H£3 protons observed
in the homonuclear 2D NOESY experiments.

Sequential assignment. The sequential assignment (Table 4.1) was primarily based
on sequential NOE connectivities such as daN, dNN and day The assignment was initiated
after over sixty percent of the total 137 residues had been identified. Most residues from
the ten B-strands and the first a-helix could be easily assigned (Figure 4.3). The
sequential assignments of ValZ6—Lys38 posed a challenge. Val26, Arg29, Ile31 and
Lys38 were weak or very weak in the 2D 1H-‘sN HSQC spectrum. No correlations
between their amide and aliphatic protons could be observed in the 3D 'H-‘SN TOCSY-
HMQC spectrum. Only a few NOEs useful for sequential assignments were observed in
the 3D lH-‘SN NOESY-HMQC spectrum. The 1H chemical shifts of the four alanines
(Ala32, Ala34, Ala35 and Ala36) were quite similar. The 'H and 15N resonances of both
Va126 and Arg29 were assigned tentatively by comparison with the assignments of wild-
type holo-CRABPII (Chapter 5), and a few weak sequential NOE peaks. The amide
resonances of both residues became stronger upon RA binding. The aliphatic protons of
Va126 were assigned by spin system identification. The assignments of Ile3l and Lys38
were helped greatly by selective labeling, and by comparison with the assignments of
wild-type holo-CRABPII. The spin system of Asnl 15 could be easily identified by the
homonuclear NMR experiments but its sequential assignment was achieved only at the

later stage because of the weakness of its peaks in all l5N-edited spectra. The H"l and HB

155

Table 4.1. 'H and '5N resonance assignments of apo-R1 1 1M in PBS at 27°C

 

Residue

P 1
N2
F3
S4
G5
N6
W7

K8
I9
110
R1 1
S 12
E13
N14
F15
E16
E17
L18
L19
K20
V21
L22
623
V24
N25
V26
M27
L28
R29
K30
131
A32
V33
A34
A35
A36
S37
K38
P39
A40
V41
E42
143
K44

15N

123.2
112.6
110.7
119.0
124.4

125.2
126.1
118.9
120.6
115.6
124.1
117.5
120.5
117.8
118.5
121.8
115.3
118.7
121.4
118.0
106.7
121.8
127.4
120.7
119.6
119.1
120.9
118.1
119.3
123.5
117.1
122.0
119.0
120.2
112.2
120.1

127.4
120.8
129.4
123.7
129.1

HN

9.76
8.39
9.36
8.17
9.31

10.06
8.62
9.23
7.43
8.32
8.83
9.10
8.54
8.80
8.20
7.62
7.48
8.07
7.81
7.27
7.65
7.62
8.59
8.20
7.91
7.74
8.66
7.66
7.56
8.01
8.41
7.58
8.19
8.20
7.48
7.68

8.60
7.78
9.19
9.20
9.48

HO.

4.14
5.11
4.38
4.31
4.13, 3.97
5.24
5.08

5.27
3.67
4.59
4.60
5.28
4.83
4.96
4.60
3.32
3.74
3.75
3.64
3.89
3.79
3.97
4.00, 3.62
4.60
4.46
3.51
4.15
4.11

3.92
4.07
4.10
3.72
4.20
4.35
4.54
4.60

4.90
4.90
4.98
5.40
5.30
4.63

1.13

2.21, 1.75
2.99, 2.83
3.40, 3.22
3.99, 3.99

2.74, 2.56
3.08, 2.91

1.85, 1.85
1.39

1.87

1.72, 1.72
3.70, 3.61
1.93, 1.93
3.38, 3.27
3.14, 2.97
1.88, 1.88
1.97, 1.78
1.56, 1.56
1.61, 0.81
1.96, 1.69
2.19

1.70, 1.59

1.88
3.15, 2.80
2.05
2.09, 2.03
1.82, 1.66

1.93, 1.93
2.02
1.37
2.10
1.47
1.40
1.40
3.92, 3.82

1.82, 1.40
1.59
1.29
2.06, 2.10
2.25
1.84, 1.84

Others

1171.92, 1.94;H‘5 3.13, 3.11
H’ 7.64, 6.89; N7 112.5
2, 6H 7.43; 3, 5H 7.20; 4H 7.27

2H 6.80; 4H 7.04; 5H 6.81; 6H 7.21; 7H 7.11
H‘ 9.28; Nc 125.2

I-P 1.65, 1.41; H6 0.86, 0.86; Hc 2.86, 2.86

H1 0.92, 0.92; H15 0.60; 1-17'“ 0.69

1171.21 1.21;H51.00;HV“‘ 0.81

H" 1.59; H‘5 3.07; H‘ 7.29; N‘ 83.8

H7 2.14, 2.21

H1 7.29, 6.91; N1 109.8

2, 6H 7.21; 3,5H 6.94; 4H 7.02

H” 1.98, 2.18

H" 2.42, 2.24

H" 0.84; 1-15 0.79, 0.79

H" 1.09; H‘5 0.38, 0.22

HY 1.41, 1.41;H51.20, 0.95; H‘ 3.22, 2.96
H1 1.10, 0.94

H" 0.96; H‘5 0.86, 0.69

H" 0.92, 0.82
H" 0.95, 0.90
H1 2.60, 2.45
H7 1.35; H’5 0.90, 0.79

H7 1.74, 1.55; H‘5 1.47, 1.47; HE 2.92, 2.92
1171.85, 1.85;H‘"“ 1.01

1170.95, 0.95

H7 1.07, 0.93; Hz5 3.47, 3.47

H" 0.43, 0.41

H" 2.30, 2.30

W 1.66, 1.28; H‘5 0.69; H1m 0.91
1171.54, 1.35; H‘ 2.86, 2.78

156

 

Residue

Q45
E46
G47
D48
T49
F50
Y5 l
152
K53
T54
SSS
T56
T57
V58
R59
T60
T6 1
E62
163
N64
F 65
K66
V67
G68
E69
E70
F71
E72
E73
Q74
T75
V76
D77
G78
R79
P80
C8 1
K82
S83
L84
V85
K86
W87

E88
889
E90
N91
K92
M93

lSN

127.4
128.9
117.9
126.7
116.9
124.7
122.6
123.8
131.9
124.6
121.2
115.5
110.9
113.0
119.3
122.9
120.0
122.9
117.9
119.9
117.1
120.7
127.5
114.8
120.4
130.5
121.5
119.7
121.8
117.9
111.8
122.4
113.1
108.4
121.8

118.7
120.0
124.9
120.9
130.3
123.7
120.5

127.7
113.2
116.6
111.1
120.2
126.5

8.59
8.55
9.04
8.51
8.06
9.00
8.32
8.42
9.26
8.96
9.16
8.36
8.34
7.43
7.38
8.76
8.95
8.62
8.50
8.00
8.29
9.20
8.83
8.77
7.90
9.18
8.99
8.61
8.42
7.78
9.18
9.26
8.75
7.93
6.86

8.78
8.83
9.02
7.33
9.21
8.39
8.54

9.25
8.63
8.69
7.98
7.74
9.27

HO.

4.50
4.40
3.88, 3.62
5.19
4.75
4.60
5.20
4.72
5.17
5.10
5.34
4.85
4.05
4.19
4.60
5.23
4.65
5.13
4.85
5.59
4.67
5.31
3.05
4.40, 3.63
4.81
4.91
5.15
5.09
4.63
5.35
4.62
3.79
4.40
4.23, 3.58
4.63
4.70
5.08
4.79
4.93
4.24
4.40
4.52
5.02

4.38
4.61
3.77
4.01
5.71
5.16

HB

1.86, 1.86
1.84, 1.68

3.16, 2.74
4.18
1.74, 1.74
3.03, 2.77
1.59
1.77, 1.69
3.89
3.75, 3.75
4.72
4.27
2.02
1.72, 1.72
3.94
4.01
1.93, 1.93
1.86
2.51, 2.51
2.28, 2.28
1.67, 1.54
1.67

2.00, 2.00
2.02, 1.94
3.51, 3.22
2.06, 1.97
1.98, 1.90
1.79, 1.79
4.18

1.98

2.74, 1.98

1.82, 1.58
2.42, 1.90
2.64, 2.13
1.58, 1.51
2.80, 1.55
1.61, 1.29
1.32

1.64, 1.51
3.28, 2.99

1.95, 1.76
4.18, 4.01
1.87, 1.87
3.12, 2.92
1.96, 1.96
1.22, 1.22

Others

11* 1.03, 1.03; H5 7.71, 6.98; N5 106.9
H1 2.03, 2.03

HY 1.02

2, 6H 6.50; 3, 5H 7.06; 4H 6.81
2, 6H 6.94; 3, 5H 6.56

H1 1.38, 1.28; H8 0.59; H“ 0.43
H" 1.24, 1.16; H‘ 2.86, 2.86

H" 0.96

H" 1.16

11’ 1.31

Hy 0.82, 0.82

H" 1.40, 1.40

H1 1.11

H1 1.07

H" 2.10, 2.04

H" 1.36, 0.79; H5 0.91; I-P'"n 1.11
H1 7.51, 6.81;N7113.1

2, 6H 6.20; 3, 5H 6.52; 4H 6.26
H1 1.32, 1.32; H‘ 2.96, 2.96

H" 0.46, 0.25

117 2.22, 2.22

H" 2.24, 2.15

2, 6H 7.26; 3, 5H 7.81; 4H 6.55

H" 2.34, 2.16

H" 2.14, 2.14

H” 2.20, 2.03; H‘5 7.35, 6.58; N‘5 110.3
H" 1.06

H" 0.86, 0.86

H7 1.36, 0.69; H5 3.21, 3.12; H‘ 8.59; N‘ 87.9
H1 2.05, 2.17; H5 4.03, 3.66;

H1 1.28; H8 2.82;
m 6.26

H1 1.19; 1150.71, 0.71

HY 0.22, —0.03

H7 1.31, 1.22

2H 7.38; 4H 7.98; 5H 6.86; 6H 7.21; 7H 7.58
H‘ 10.50 (1057):; Na 129.6

H7 2.22, 2.12

117 2.12, 2.12

H" 7.37, 6.70; N7 112.6

H" 1.50, 1.50; Hz5 1.23, 0.93; H‘ 2.71, 2.56
H‘1 1.75, 1.75; H“ 0.31

 

Residue

V94
C95
E96
Q97
K98
L99
L100
K101
6102
E103
G104
P105
K106
T107
S108
W109

T110
M111
E112
L113
T114
N115
D116
G117
E118
L119
1120
L121
T122
M123
T124
A125
D126
D127
V128
V129
C130
T131
R132
V133
Y134
V135
R136
E137

15N

118.5
127.6
127.4
121.0
123.6
126.1
121.7
117.6
109.0
117.1
109.1

123.6
118.9
112.9
117.1

110.4
119.7
124.3
126.7
117.2
119.6
115.8
107.7
117.8
124.8
125.8
135.4
117.8
121.2
116.0
125.6
122.5
121.0
123.7
125.7
129.5
124.8
126.3
122.0
126.9
111.8
123.8
127.4

HN

8.64
8.91
7.70
7.31
8.97
8.75
7.70
7.97
8.18
8.08
8.10

8.54
8.75
7.84
8.29

9.44
8.62
8.85
8.17
9.29
9.14
7.75
7.94
7.40
8.64
9.38
9.69
8.85
9.20
8.42
8.44
9.23
8.44
8.39
8.35
9.57
8.06
9.43
8.42
9.73
9.19
8.56
8.15

Ha

4.24
4.46
4.32
4.92
4.60
4.00
4.10
4.40
4.06, 3.75
4.63
4.12, 3.95
4.54
4.42
5.38
4.75
5.72

5.15
5.46
5.78
5.29
4.62
4.46
4.77
4.21, 3.73
5.03
5.04
4.85
5.54
5.75
5.20
5.50
4.88
4.17
4.54
3.99
4.55
5.25
5.00
5.04
4.92
5.54
5.34
3.89
3.93

157

1.13

1.77

0.51,—0.10

1.86, 1.80
2.03, 1.85
1.57, 1.39
1.67, 1.51
1.47, 1.47
1.82, 1.82

1.79, 1.79

2.08, 2.08
1.85, 1.85
4.50

3.83, 3.83
3.58, 3.33

4.19
1.86, 1.83
1.93, 1.93
1.31, 0.77
4.52
2.82
2.82, 2.46

2.04, 1.92
0.95, 0.84
1.84
1.92, 1.36
4.08
1.87, 1.84
3.97
1.06
2.77, 2.73
2.70, 2.70
2.18
1.87
3.03, 2.74
3.77
1.87, 1.84
1.70
3.30, 3.17
2.38
1.31, 1.09
1.85, 1.75

Others

H" 0.75, 0.69

H" 2.17, 1.97

H" 2.50, 2.15; Hz5 7.51, 6.68; N15 114.8
H" 1.29; H‘ 2.85

H" 1.39; H5 0.69, 0.69

H" 0.78

H" 1.66, 1.66; H5 1.29, 1.29

H’2.12, 1.99

H" 1.87, 1.80; H5 3.63, 3.35
H" 1.56, 1.40; H5 1.81, 1.81; Hc 2.98, 2.98
H" 1.17, 1-1"I 6.04

2H 6.87; 4H 6.83; 5H 6.41; 6H 6.41; 7H 6.71
H‘ 11.17; N‘c 136.4

H" 1.07

H“ 2.41, 2.25

H" 2.09, 2.09

H" 1.14; H5 0.51, 0.48

H" 1.08

H" 7.54, 6.83; N" 112.7

H" 2.35, 2.20

H" 0.30; H5 —0.3 1, —0.38
H"1.10,1.10;H51.04;H"'" 0.79
H" 1.60; H6 0.77, 0.71

H" 1.08

H" 2.45, 2.37

H" 0.99

H" 0.96, 0.74
H" 0.81, 0.74

H" 1.02

H" 1.75, 1.75; H13 3.02, 1.96; Hc 7.17; N“ 84.4
H" 0.89, 0.84

2, 6H 6.82; 3, 5H 6.68

H" 1.08, 0.91

H" 0.62, 0.23; Hz5 2.82, 2.70; Hc 6.86; N‘ 85.2
H" 2.09, 2.09

 

"' Two sets of resonances.

158

Figure 4.3. Strip plots extracted from the 3D 15N-edited NOESY-HMQC spectrum of
apo-R1 11M showing the sequential and medium-range NOE connectivies of a-helix

(01A) (A) and B-strand ([1]) (B).

159

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

7 7 7 7 7 7777 7 7 7 7

”cream 0°90 $0 9 0 "9 “9‘: ° 1' "5‘3
e 000097690 3 =15 '6 2:, '00 E éi-‘i '25;
j 6 “9‘90 3 ,OW 5.?3‘ 60 ‘ o 6.2—04-?
9.. ea E99 17 0 0 EL 0 "é
'Qoewo 00:11 A; Q «0'00 '§
0 997.15 77.? ,No .e.. -2
9'0 r1560 0 co 0 :1 9%: 1 ° L?
as: o - o 0 g? =10 '° ° 0 —§
m 0 0 E? I“ Q _§
mon 00mm 0 0 Joe 1 10° 7:5
w a o 3 3"0‘ '9
.0“ o" - gpk ._‘z :7. 13
09110990 °° EH» 9° ”717?: 9 ‘7

H GEDGQGO ,, ,gfia 751$ o..- _
no ea 0090 $74 0'1: '3
1 °°®°° “10% O: 1.31-'07- '3
u 9 @009” 70° 0 .2 21¢-- ’2 -E
0% 9° “07 0° ' .2" 7b‘ ° hi
< . .0 “mo 0 9.0 on o cmflk'o LE

 

 

 

Residue

Residue

160

protons of Thr114, whose chemical shifts were close to the water signal, were not
observed in the two sets of the 3D spectra due to water presaturation. They were assigned
by the analysis of the 2D homonuclear spectra recorded in D20 helped by the comparison
with their assignments in wild-type holo-CRABPII. Aan and Thr57 were the only two
residues not observed in the 2D 1H—‘SN HSQC spectrum. Their aliphatic protons were
assigned by spin system identification after the sequential assignments. The side-chain
assignments were complete except for Argl 1, Leu22, Arg29, Ile31, Arg59 and eight
lysine residues. Stereospecific assignments were made for the B-methylene protons of 51
residues and the methyl protons of ten valine and six leucine residues.

Secondary structure determination. Regular secondary structure elements were
characterized by their distinct NOE patterns. The following ten strands forming two anti-
parallel B-sheets could be identified (Kleywegt et al., 1994) (Figure 4. 4): Ser4—Glu13
(BA), Pro39—Glu46 (BB), Thr49-Thr56 (BC), Thr60—Lys66 (BD), Glu69—Thr75 (BE),
ProSO—Glu88 (BF), Lys92—Lys98 (BG), K106—L113 (BH), E118—A125 (BI) and V128—
R136 (BJ). The second strand BB began at Pro39, one residue shorter than the
corresponding BB in the crystal structure of holo-CRABPII, where it began at Lys38.
Lys38 was very weak in the 2D lH-‘SN HSQC spectrum. Long-range NOEs were
observed between the backbone protons of all adjacent B-strands except between BD and
BE. There were also NOEs between the backbone protons of the N-terrninal half of BA
and the C-terminal half of BJ. The first two strands BA and BB were connected by two
anti-parallel (it-helices. The first (01A), PhelS—Leu22, showed NOE connectivity and

hydrogen bonding patterns typical of a regular a-helical structure. The N-terminal half

161

Figure 4.4. Summary of the sequential and medium-range NOEs involving backbone HN
and H“ atoms, slow-exchange backbone amide protons and the deduced secondary
structures of apo-R1 1 1M. Line thickness for daN and dNN sequential NOE distances
reflects the intensities of the cross peaks. Asterisks indicate the residues whose amide
protons remained afier 24 h of exchange with D20 in PBS buffer at room temperature and

pD 7.5 as determined by 2D homonuclear NMR experiments.

162

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

10 20 30
l | 1
PNFSGNWKIIFlSENFEELLKVLGVNVMLRKIAVAAA
Slow NH exchange " * " " ‘ " "
daN(i,i+1) -—_-— — ‘—
dNN(i,i+l) I _ --_-—
daN(i,i+2) — _ L———_— __ — .—
dNN(i,i+2) fi=——=__ ”‘3— '—
daN(i,i+3) a 1
dNN(i,i+3)
Secondary structures 1 1 . r L133: Ljfi’ . (1A “3 fl
4'0 5'0 60 7
I
SKPAVEIKQEGDTFYIKTSTTVRTTEINFKVGEEFEE
SlowNH exchange ‘ * " ‘ ' "I t t t t :- 9 t e
daN(i,i+1) “ .—
dNN(1,1+1) - - 4 — — _ —
daN(i,i+2)
dNN(',i+3) _
7 7 "1°
QTVDG RPCKSLVKWESENKMVCEQKLLKGEGPKT
Slow NH exchange “ ‘ " ‘ ' ‘ ' * ' *
d01N(i,i+1) - -— u -
dNN(iri+l) - * g i L -
daN(i,i+2) — —
dNN(i,i+2) —— 1— __
dNN (1.1+3) ‘— BF "_ [3G
Secondarystrucnnescrp ............ p ”1.1.1Lr, a:
1:0 120 130
SWTMELTN DGELILTMTADDVVCTHVYVRE
510w NH exchange a e e e t e e e e e e e e t e e e
d111N(i,i+1) - -_
dNN(1,1+1) — — — — — _
daN(i,i+2) —
dNN(i,i+2) ——- ——
dNN (1.1+3) pH pl — 3.1

 

 

Secondary structures cm 1:1:r::r==::> u xxxxxxx ¢

163

(Leu28—Ile31) of the second helix appeared to be partially unwound. The strength and
number of the NOEs observed for residues Leu28-Ile31were well below what is expected
for a regular a-helix. However, the chemical shifts of these residues indicated that they
are predominately in a helical conformation. No slow exchanging amide protons were
observed for the residues in the second helix. The loop (Gly23—Va126) between the two
helices could be detected but the link (Ser37—Pro39) connecting dB and BB appeared to
undergo conformational exchange in solution.

Structural restraints. A total of 2302 structurally useful distance restraints were
obtained by the analysis of the homonuclear 2D NOESY and heteronuclear 3D lH-‘sN
NOESY-HMQC data. Of the 2302 distance restraints, 704 were from long-range NOEs
(Table 4.2). The distributions of the distance restraints along the amino acid sequence
were quite uniform except for the two N-terminal residues, the residues in turns or loops
and in the second a-helix (Figure 4.5). The number of NOE restraints for each of them
was well below the average. A total of 77 dihedral angle restraints were obtained from the
2D 'H-‘SN HMQC-J data. A total of 98 hydrogen bond restraints from 49 hydrogen bonds
were included in the structural refinement.

Quality of the structures. Figure 4.6A shows the superposition of 28 refined
structures with no NOE violation exceeding 0.25 A and no dihedral angle violation larger
than 5.0° The statistics of the NMR structures is summarized in Table 4.2. The
stereochemical qualities of the structures were checked by the program PROCHECK—
NMR as described in Chapter 3. The Ramachandran plots and overall G-factors showed
that the stereochemical qualities of the structures were comparable to those X-ray

structures with 2.5 A resolution and R-factors less than 20 % (Table 4.2).

164

Table 4.2. Restraint and structural statistics of apo-R1 1 1M

 

Restraint Statistics

Numbers of experimental NOE restraints
lntra
Sequential
Medium
Long
Total
Number of experimental ¢ dihedral restraints
Number of hydrogen bonds

Structural Statistics

RMSD from experimental distance restraints (A) (2302)
RMSD from experimental dihedral restraints (deg) (77)
Deviation fiom idealized covalent geometry

Bonds (A)

Angles (deg)

Impropers (deg)

711

Measures of Structural Quality (By Procheck)

% residues in most favorable regions of the Ramachandran plot
% residues in additional allowed regions of the Ramachandran

plot
No. of bad contacts
H-bond energy
Overall G-factor
Coordinate Precision
RMSD for Ca trace (A)

RMSD for backbone atoms (A)
RMSD for all heavy atoms (A)

635
252
704
2302
77
98
<SA>al <SA>)"
0.032 i 0.001 0.024
0.926 t 0.102 0.836
0.004 i 0.001 0.003
0.621 d: 0.028 0.520
0.495 i 0.004 0.411
71.8 i 2.4 71.0
25.8 i 0.8 25.8
6 i 2 6
0.40 i 0.10 0.50
-0. 19 :h 0.01 -0.20
0.43 d: 0.14
0.54 :t 0.26
0.98 :t 0.23

 

“ <SA> are the final 28 simulated annealing structures.

b <SA>r is the restrained minimized mean structure obtained by restrained regularization
of the mean structure, which is obtained by averaging the coordinates of the individual

SA structures best fitted to each other.

165

Figure 4.5. The distribution of NOEs along the amino acid sequence of CRABPII. For

intra-residue NOEs only those that are structurally useful are included.

No. of NOEs

No. of NOEs

166

 

 

 

8

a
0

Long
Medium
Sequential
lntraresidue

 

1 * 1
1111 1111111 1

PN‘SGNWKIIRSENFEELLKVLGVNVMLRKIAVAAASKPAVEIKOEGDTFVIKYSTYVRYTEINFKVOE

5 10 15 2o 25 30 35 4o 45 so 55 so 65
Residue Number

 

.1 " i a
I 2 a. a ,1: .. 11; -

u
‘ I 3;!

mar

oilinixsluil 1u1111.11111511111..-.1h 1.1115.--1131.h!-.1111271111

EFEEDTVDGRPCKSLVKWESENKMVCEOKLLKGEGPKTSWTMELYNDGELILTMYADDVVCYRVVVRE

70 75 80 85 9O 95 100 105 110 115 120 125 130 135

Residue Number

167

Figure 4.6. (A) Stereoview of the C“ traces of the superimposed 28 refined solution
structures of apo-R11 1M. (B) Stereoview of the superimposed C“ traces of the restrained
minimized mean structures of apo-R1 1 1M (thin line) and apo-CRABPII (thick line). (C)
Stereoview of the C“ traces of the restrained minimized mean structure of apo-R1 11M
(thin line) superimposed with the C“ traces of the crystal structure of holo-CRABPII

(thick line).

168

 

 

 

 

 

 

 

 

169

Description of R] I 1M solution structure. The tertiary structure of apo-R1 1 1M
was very similar to the crystal structure of CRABPs (Figure 4.6C) (Kleywegt et al., 1994;
Thompson et al., 1995). The protein formed a compact, single domain structure
consisting of two antiparallel B-sheets and two short a-helices. Both B-sheets consisted of
five B-strands and were twisted into a nearly orthogonal flattened barrel. Inside the
protein a binding pocket was formed by the side-chains from eight of the ten B-strands
and lined by both hydrophobic and hydrophilic residues. It was closed off from the
exteriors except for a small tunnel-like opening. The entrance was formed mainly by the
hydrophobic side-chains from one side of helix 01B, the BC-BD loop and the BE-BF loop,
as well as Leu19 and Va124. The loop connecting (1A and (18 extended into the solvent.
In solution the ligand entrance was slightly more flexible than the other parts of the

protein, but it was still relatively well defined in the solution structure.

170

Discussion

Assignment. The sequential assignment of the apo-R1 1 1M posed a challenge for
amide-directed sequential assignment strategy based on a series of homonuclear 2D and
heteronuclear 2D and 3D 'SN-edited NMR experiments. The problem was caused by the
presence of three prolines in the middle of the amino acid sequence, the low percentage
of the residues whose spin systems could be identified unambiguously by homonuclear
2D NMR experiments, and the relatively low intensities of a dozen cross peaks in the 2D
lH-‘SN HSQC spectrum. The intensities of these weak peaks were even lower in the 3D
lH-‘SN NOESY-HMQC spectrum and the lowest in the 3D lH-‘sN TOCSY-HMQC
spectrum. Moreover, eight residues had H“ protons signals and one residue had an HB
proton signal quite close to the solvent resonance. These protons were not observed in the
two sets of 3D lSN-edited data. Consequently, their sequential assignments could not be
achieved by the 3D 'H-‘SN NOESY-HMQC experiment. Most of these difficulties were
overcome by selective 15N-labeling. Lysine, leucine, isoleucine and phenylalanine were
chosen because of the abundance of these residues in the protein, and the difficulty in
identifying them by the standard procedure (Wiithrich, 1986). After their identifications
by selective 'SN-labeling, the spin systems of over sixty percent of the total residues were
known, which made it possible to achieve the total sequential assignment.

Comparison of the solution structures of apo-R11 1M and wild-type apo-

CRABPII. The backbone fold of apo-R11 1M is quite similar to that of apo-CRABPII.

171

The restrained minimized mean NMR structures of the two proteins can be superimposed
with a RMSD of 3.2 A for all backbone C“ atoms (Figure 4.6B and Figure 4.7C), a
RMSD of 2.3 A for the backbone C“ atoms of the ten B-strands and the first or-helix.
However, there are significant conformational differences between the two structures,
mainly localized to three segments (Leu19-Ala36, Glu73-Cys81 and Leu99-Pr0105)
clustered around the ligand entrance. Surprisingly, the conformational differences occur
in the region far away from the site of the point mutation. If these segments are excluded,
the C“ RMSD between the two structures is reduced to 1.6 A. The first segment (Leu19-
Ala36) encompasses the middle of the first helix to the end of the second helix, the
second segment (Glu73-Cys81) the end of BE and the beginning of the BF, and the third
segment (Leu99-Pr0105) the long BG—-—BH loop. In apo-R1 1 1M, all three segments move
towards the center of the ligand entrance so that the opening of the ligand binding pocket
is much smaller than that in wild-type apo-CRABPII. The binding pocket in apo-R1 11M
is much less exposed to the external solvent than in the wild-type structure. For example,
the distance between the C“ atoms of V3158 in the BC—BD loop and Va124 in the turn
connecting two helices is 7.4 A shorter in apo-R1 11M than in wild-type apo-CRABPII.
The distance between the C“ atoms of Va158 and Asp77 in the BE—BF loop is 10.8 A
shorter in apo-R1 11M than in wild-type apo-CRABPII.

Comparison of the solution structures of apo-RI I 1M and the crystal structure of
wild-type halo-CRABPII. The restrained minimized mean NMR structure of apo-R1 11M
and the crystal structure of wild-type holo-CRABPII could be superimposed very well
(Figure 4.6C). The RMSD between the two superimposed structures is 0.96 A for the

Cu atoms of all regular secondary structures and 1.38 A for all C,l atoms. The

172

Figure 4.7. C“ atom deviations between apo-R11 1M and wild-type holo-CRABPII (A)

and between apo-R1 11M and wild-type apo-CRABPII (B).

C, Deviations (A)

C,ll Deviations (A)

173

 

>

14-

j

.3
N
l

A
O
14

A

030
.I.J

A
J

A

' . 1 . 1.1 1' ..
...111 11. 1.1.1111 11111111 1111 11111.111‘1. 1
10 20 30 40 50 60 70 80 90 100 110 120 130

 

 

N

o
0—

 

14. 3

12-

A
O
l n

A

1

1111 .111

1
1 11 1
80 90 100 110 120 130

 

 

 

2
011 1111 .111.111 .
50 60 70

0 10 20 30 40

Residue Number

174

superposition also shows good alignments of the side-chains that are involved in binding
of RA such as those of Arg132 and Tyrl34.

It is obvious from examining Figure 4.7 that the solution structure of apo-R1 11M
is more similar to the crystal structure of holo-CRABPII than to the solution structure of
apo-CRABPII. The ligand entrance of apo-R1 1 1M is much smaller than that of the apo-
CRABPII, but is still larger than that of holo-CRABPII. The distance between the
C“ atoms of Va158 in the BC—BD loop and Va124 in the turn connecting two helices is
15.0, 21.3 and 28.8 A in holo-CRABPII, apo-RlllM and apo-CRABPII, respectively.
The distance between the C“ atoms of Va158 and Asp77 in the BE—BF loop is 12.4, 13.4
and 24.4 A in holo-CRABPII, apo-R1 1 1M and apo-CRABPII, respectively. It appears
that the point mutation and the binding of RA cause similar conformational changes,
although the magnitudes of the changes induced by the mutation are not as large as those
caused by the binding of RA.

Dynamical properties of apo-R11 1M. Apo-Rl 1 1M was quite stable in solution.
The NMR sample lasted for several months with almost no changes in the spectral
properties and quality. More than 50 % backbone amide protons in the first a-helix and
two B-sheets remained after a day of exchange with D20 (Figure 4.8). The hydroxyl
protons of Thr107 and Ser83 were observed in the homonuclear 2D spectra recorded in
H20, and the indole 8‘ proton of Trp7 was observed in the homonuclear 2D spectra
recorded in DZO, indicating that these protons had very low solvent accessibility. All
these observation suggested that R1 11M is for the most part fairly stable and rigid.

The peak intensity of a 2D lH-‘SN HSQC spectrum is, in general, modulated by

both amide proton exchange and conformational exchange. Amide proton exchange

175

Figure 4.8. The fingerprint regions of the homonuclear 2D TOCSY spectra of apo-
CRABPII (A) and apo-R1 11M (B) recorded with 40 ms mixing time. The TOCSY
experiment was initiated more than 24 h after dissolving a lyophilized protein sample in

PBS, pD 7.5, in 13,0.

176

.23... .75...

 

 

 

3 N... . 3. ca . to _ 2. 2: 3 S 3. 2. ed a... 2:

15.3.. _...... 1.. :..::..:;..:_::r_:_..:_::_:_35:37.:_.. . _..f-.._1.::1r-P.r.E TREE _ _ _ . :FELE. ._. :3. .. __ ._ FL. .. __E._:._ H. BEFFPIF. .59:

_ 2:3 SE .1 z

.33. O 0255; a”? O 29; “New

. 3:32.. :3. 32>

. Es. o o _ c m .

. ° New N3. . a...— men
a n1. .

. 93> 3:2 .0 o 23 n: 28 3. m I

. S. O 82.5. :5. 2:. . _ 2%.”6 98.295 2:. .r 2

_ _ _ E m

_ 85 OR? 02: _ Sm 3. 22¢ on

H . mo...— °m~m . m2 mm; C cu: m.

U a on“. o: 3.x :2 . ..,. 3.

. o 0 . 2m 0. ..

_ 03.... “we a; 2.3 an“. o: .3. .

. N2 93 o o T .

. X> . 0 “8 Q m f.

. £52; 0 __

m 0 . 2.x 3> . 3.

m £5 0 o w

m o aim; 352:. W13.

. r1.

" o .W

1 1 -1- 1- w -3.

 

 

 

 

 

 

m
<

177

appeared to have little effects on the peak intensities of the HSQC spectra of apo-R1 11M
and apo- CRABPII (Chapter 2), partly owing to the inclusion of a “water flip-back” pulse
(Grzesiek & Bax, 1993) in the HSQC experiments that minimized water saturation and
dephasing. In fact, the average peak intensity of the slow exchanging amide protons was
the same as that of all amide protons. Thus, variation in peak intensity is primarily due to
conformational exchange. As expected, the residues in the loops such as Gly47 and
Asnl 15 had peak intensities lower than the average (Figure 4.1). In general, the turns or
loops were more mobile than the regular secondary structures. Of the all secondary
structural elements only helix aB, especially its N-terminal half Leu28—Ile31, were more
flexible than the others. All the amide protons in this helix exchanged completely with
the solvent in less than 12 hours. By contrast, most of the amide protons of the first helix
remained after more than two days of exchange with D20. The peaks of Leu28, Arg29,
Lys30 and Ile3l were much weaker than the average (Figure 4.1 and 4.7A). Va126 and
Lys38 had very weak peaks, and Thr57 was not observed in 2D lH-‘sN HSQC spectrum.
Furthermore, the number of NOEs observed for these residues was very low. Taken
together, these results suggest that conformational exchanges are occurring in the aA-aB
turn, the aB-BB link and the BC-BD loop. In other words, the ligand entrance was slightly
more mobile than most of the other parts of the molecule.

Did mutation change the dynamical properties of CRABPII? For the most part,
apo-R1 11M and wild-type apo-CRABPII had similar dynamic properties. Thus, both
proteins had about the same number of slow-exchanging backbone amide proteins.
Furthermore, the identities of most of the amide protons were the same for the two

proteins (Figure 4.7). However, apo-CRABPII appeared to be more mobile than apo-

178

R1 11M. The spectral qualities of the apo-CRABPII deteriorated in about two weeks as
opposed to several months for apo-R11 1M. For most peaks, the wild-type protein had
slightly larger line width than the mutant so that the DQF-COSY and TOCSY
experiments did not work as efficiently for the wild-type as for the mutant.

To compare the dynamic properties of the two proteins in more detail, we
examined the intensities of the peaks in their 2D lH-‘SN HSQC spectra. The intensities of
the 2D 'H-‘SN HSQC spectrum of the mutant were quite uniform (Figure 4.7A). As
mentioned earlier, of all secondary structure elements, only three residues (Arg29-Ile3l)
at the flexible N-terminal half of the second helix showed weak peaks. The rest of the
weak peaks invariably belonged to the residues in the loops or turns. However, the wild-
type protein displayed large variations in peak intensities (Chapter 3). Most residues in
the RA-binding pocket and near the ligand entrance showed weak or no peaks in the 2D
'H-‘SN HSQC spectrum, indicating that the RA-binding pocket, especially the ligand
entrance, is rather dynamic. Most of these peaks became much stronger afler the mutation
(Figure 4.9C). In fact, most of the residues with significant intensity differences between
the two spectra were clustered in these regions. Of the 21 residues that constitute the RA-
binding pocket, all but three (Ile3l, Thr54 and Met123) showed significant increases in
peak intensities after the point mutation. In particular, the mutation stabilized the B-bulge
structure composed of residues Ile9-Arg11 and Thr13l-V133. This may be due to the
attenuation of the flexibility of its neighboring residues Leu28-Ala36. The same
phenomenon was observed for the residues Ser12-Glul7, which may be caused by the

attenuation of the fluctuations of the two helices (01A and 013). The residues Lys8, Ala40,

179

Figure 4.9. Relative peak intensities of the 2D IH—‘sN HSQC spectra of apo-R1 1 IM (A)
and wild-type apo-CRABPII (B) as a function of the residue number. (C) was obtained by

subtracting (B) from (A).

180

 

 

 

 

 

80- C:
60.
40¢

20-
0
20-

40.

60.

80-

14“

 

v

120-

1 . . .
o o 0 0 o
8 6 4 2

100.

£22... 252$.

 

140

120- l\

100-

80-

60-

40-

20-

O-

70 80 90 100 110 120 130

60

20 30 40

10

Residue Number

181

LysS3, Glu62, Ile63, Leull9 and Vall35 also became much stronger after the point
mutation. Their implications will be discussed later.

Binding of RA substantially stabilized CRABPI as measured by thermal
denaturation. In particular, it stabilized the flexible parts of the ligand binding pocket in
CRABPI as revealed by NMR (Rizo et al., 1994). Our NMR studies showed that the
changes in dynamical properties caused by the point mutation were rather similar to those
induced by RA binding (Chapter 5). Thus, the dynamic properties of apo-R1 11M were
more similar to those of holo-CRABPI and holo-CRABPII than to apo-CRABPI and apo-
CRABPII. Binding of RA and the mutation caused similar changes in the dynamical
properties of CRABPII.

Roles of Arg] 1 1 . Argl 11 interacts with the carboxyl group of RA via a hydrogen
bond mediated by a water molecule. In Chapter 2 I have shown that Argl 11 contributes
to the binding of RA by ~2.2 kcal/mol, indicating that Argl 11 is indeed important for the
binding of RA. X-ray crystallographic and molecular modeling studies suggest that
Argl 11 is one of the four arginine residues that generate a positive potential that may
facilitate the binding of RA (Zhang et al., 1992). The NMR results presented in this
chapter suggest an additional role for Argl 11 related to the structure and dynamics of
CRABPII.

As mentioned earlier, RA binding induces significant but localized
conformational and dynamical changes in both CRABPI and CRABPII. The ligand
entrance, in particular, is much more open and flexible in the apo-forms than in the holo-
forms. The large opening of the ligand entrance and its high flexibility are probably

essential for the unhindered entry of RA into the deep binding pocket and the subsequent .

182

closure of the ligand entrance. Our NMR results show that the point mutation also causes
significant but localized changes in the conformation and dynamics of CRABPII.
Strikingly, apo-RlllM is more similar to wild-type holo-CRABPII than to wild-type
apo-CRABPII in both structure and dynamics, suggesting that Argl 11 may play a major
role in maintaining the ligand entrance in an open and dynamic state for the entry of RA.

The next question is how Argl ll affects the internal motions of CRABPII. At
first thought, the interactions leading to the high flexibility of apo-CRABPII was the
electrostatic repulsion between the positively charged guanidinium groups of two
conserved arginines, Arglll and Arg132 in CRABPII and Arglll and Argl31 in
CRABPI (Zhang et al., 1992). However, the two guanidinium groups in apo-CRABPII
were ~7.6 A apart. Furthermore, if the electrostatic repulsion played a critical role
substitution of Argl32 with methionine would most probably change the dynamical
properties to the same extent. However, this was not supported by our NMR
characterization of apo-R132M. The qualities of the homonuclear 2D spectra of apo-
R132M were slightly better than those of the wild-type but worse than those of the apo-
R1 1 1M, suggesting that apo-R132M is less flexible than apo-CRABPII but not as rigid as
apo-R1 1 1M. For most peaks apo-R1 11M had the smallest line widths among all mutants
that we generated (V24A, R59G, L121A, M123A and R132M). Thus the electrostatic
repulsion between the two arginines is, at most, only partially responsible for the
conformational flexibility of apo-CRABP.

Other interactions contributing to the dynamical properties of apo-CRABPII were
suggested by the solution structures of apo-R11 1M and apo-CRABPII (Chapter 3). In

both structures, the guanidinium group of the Arg132 is partially exposed and surrounded

183

by the side-chains of Ser12, Pro39, Thr54 (y-methyl group), Cysl30 and Tyrl34. The
hydroxyl groups of Ser12 and Tyrl34 and the thiol group of Cys130 provide the
favorable polar interactions with the guanidinium group of the Argl32. In particular, the
hydroxyl group of the phenolic ring of Tyrl34 is within the hydrogen bond distance to
the Ne nitrogen of the guanidinium group of Arg132. By contrast, in the solution
structure of apo-CRABPII, the guanidinium group of Argl 11 is almost completely buried
and in an environment much more hydrophobic than that of Arg132. It is surrounded by
the side-chains of Val4l, lle43, Ile52, Thr54, Ile63, Trp109, Leul l9 and Leu121. The
hydrophobic side-chains of Val4l, Ile63, Leul l9 and Leu121 had close contacts with the
guanidinium group of Argl l l. The only favorable interactions may be between the
guanidinium group of Arglll and the aromatic ring of Trp109 and between the
guanidinium group of Argl 11 and the hydroxyl group of Thr54. As mentioned earlier, the
peaks of the residues Val41, LysS3, Glu62, Ile63, Leull9 and Leu121 became much
stronger in the 2D lH-‘SN HSQC spectrum and/or 2D homonuclear spectra after the point
mutation. It is tempting to speculate that the charged guanidinium group of Argl 11 is
mobile because of its hydrophobic environment and possibly the repulsive interaction
with Argl32. The mobility of the guanidinium group of Arglll drives the
conformational flexibility of apo-CRABPII and results in a more open conformation for
the ligand entrance. The hydrophobic side-chain of Metl 11 is preferred by the
hydrophobic environment. Binding of RA neutralizes, at least partially, the positive
charge of Argl 11 and therefore reduces the mobility of its guanidinium group. CRABPs
are most stable in the holo-forms because there are additional hydrophobic interactions

between the proteins and RA. It is noted that the guanidinium group of Argl 11 is more

184

than 17 A away from the ligand entrance, the most flexible region of the apo-CRABPII. It

is not clear how the mobility of the guanidinium group propagates to the ligand entrance.

185

Conclusion

The solution structure of apo-R1 1 1M, a site-directed variant of human CRABPII,
has been determined by multidimensional heteronuclear NMR. The solution structure of
apo-R1 1 1M was more similar to the crystal structure of wild-type holo-CRABPII than to
the solution structure of wild-type apo-CRABPII. The size of the ligand entrance, the
mobility of the RA-binding pocket and the ligand entrance were reduced greatly after the
point mutation Argl 11 to Metl 11. The results suggest that Argl 11 plays a major role in

maintaining the ligand entrance in an open and dynamic state for the entry of RA.

186

References

Anil-Kumar, Ernst, R. R., & Wiithrich, K. (1980) Biochem. Biophys. Res. Commun. 95,
1-6.

Basus, V. J. (1989) Methods Emzymol. 1 7 7, 132-149.

Bax, A., Griffey, R. H., & Hawkins, B. L. (1983) J. Magn. Reson. 55, 301-315.

Bodenhausen, G., & Ruben, D. J. (1980) Chem. Phys. Lett. 69, 185-188.

Cavanagh, J., Palmer A. G., Wright, P. E., & Rance, M. (1991).]. Magn. Reson. 91, 429-
436.

Charnbon, P., Olson, J. A., & Ross, A. C. (coordinators) (1996) The retinoid revolution,
FASEB J. 10, 938-1048.

Delaglio, F ., Grzesiek, S., Viiister, G. W., Zhu, G., Pfeifer, J ., & Bax, A. (1995) J.
Biomol. NMR 6, 277-293.

Garret, D. S., Powers, R., Gronenbom, A. M., & Clore, G. M. (1991) J. Magn. Reson.
95, 214-220.

Clore, G. M., & Gronenbom, A. G. (1989) CRC Crit. Rev. Biochem. Mol. Biol. 24, 479-
564.

Clore, G. M., & Gronenbom, A. G. (1991) Annu. Rev. Biophys. Chem. 21 , 29-63.

Griesinger, C., Sorensen, O. W., & Ernst, R. P. (1985) J. Am. Chem. Soc. 107, 6394-
6396.

Griesinger, C., Otting, G., Wiithrich, K., & Ernst, R. P. (1988).]. Am. Chem. Soc. 110,

7870-7872.

187

Grzesiek, S., & Bax, A. (1993) J. Am. Chem. Soc. 115, 12593-12594.

Hodson, M. E., & Cistola, D. P. (1997) Biochemistry 36, 2278-2290.

Jeener, J., Meier, B. H., & Ernst, R. R. (1978).]. Chem. Phys. 71, 4546-4553.

Kay, L. E., & Bax, A. (1990) J. Magn. Reson. 86, 120-126.

Kay, L. E., Keiffer, P., & Saarinen, T. (1992).]. Am. Chem. Soc. 114, 10663-10665.

Kleywegt, G. J., Bergfors, T., Senn, H., 1e Motte, P., Gsell, B., Shuao, K., & Jones, T. A.
(1994) Structure 2, 1241-1258.

Marion, D., Ikura, M., Tschudin, R., & Bax, A. (1989a) J. Magn. Reson. 85, 393-399.

Marion, D., Driscoll, P. C., Kay, L. E., Wingfeld, P. T., Bax, A., Gronenbom, A. M., &
Clore, G. M. (1989b) Biochemistry 28, 6150-6156.

Marion, D., Kay, L. E., Sparks, S. W., Torchia, D. A., Bax, A. (1989c) J. Am. Chem.
Soc.111,1515-1517.

Mueller, L. (1979) J. Am. Chem. Soc. 101, 4481-4484.

Palmer, A. G., Cavanagh, J ., Wright, P. E., & Rance, M. (1991) J. Magn. Reson. 93, 151-
170.

Piantini, U., Sorensen, O. W., & Ernst, R. P. (1982) J. Am. Chem. Soc. 104, 6800-6801.

Rizo, J ., Liu, Z. P., & Gierasch, L. M. (1994) J. Biomol. NMR 4, 741-760.

Shaka, A. J ., Lee, C. J ., & Pine, A. (1988) J. Magn. Reson. 77, 274—293.

Shaka, A. J ., Barker, P. B., & Freeman, R. (1985) J. Magn. Reson. 64, 547-552.

Spom, M. B., Roberts, A. B. & Goodman, D. S. (1994) The Retinoids: Biology,
Chemistry, and Medicine, 2"“ ed. Raven Press, New York.

States, D. J ., Haberkon, R. A., & Ruben, D. J. (1982).]. Magn. Reson. 48, 286-292.

Wiithrich, K. (1986) NMR of Proteins and Nuclear Acids, Wiley, New York.

188

Wiithrich, K., Billletter, M., & Braiin, W. (1983).]. Mol. Biol. 169, 949-961.
Zhang, J ., Liu, Z., Jones, T. A., Gierasch, L. M., & Sambrook, J. F. (1992) Proteins:

Structure, Function, and Genetics 13, 87-99.

CHAPTER 5

NMR Studies of Human CRABPII

in Complex with All-trans-Retinoic Acid

Introduction

The crystal structures of holo-CRABPs suggest that RA cannot enter or exit the
deep binding pocket in the absence of major conformational changes in the protein
(Kleywegt et al., 1994). Indeed, significant conformational differences between the
crystal structure of holo-CRABPII and the solution structure of apo-CRABPII have been
observed, especially in the ligand entrance region (Chapter 3). In comparison to the holo
crystal structure, apo-CRABPII exhibits a concerted movement of the second helix, the
BC-BD loop and the BE-BF loop, so that the ligand entrance of apo-CRABPII is greatly
enlarged and readily accessible to RA. Furthermore, the ligand-binding pocket of apo-
CRABPII is rather disordered. The results suggest that binding of RA induce significant
changes in both conformation and dynamics of CRABPII.

This chapter describes the sequential resonance assignments and preliminary

189

190

results showed that RA binds to CRABPII in solution in the same manner as determined
by crystallography and the ligand entrance of holo-CRABPII is much less flexible than
that of apo-CRABPII. Relative motions between the bound RA and human CRABPII in
solution, particularly at the ligand entrance, were detected. Such motions may provide a

transient pathway for the exit of RA.

191

Experimental Procedures

Sample preparation. All-trans-RA and D20 were purchased from Sigma (St.
Louis, Missouri), dimethyl sulfoxide-d, (DMSO-d6 ) were purchased from Aldrich.

Expression of CRABPII gene and purification of the recombinant protein have
been described in Chapter 2. The NMR samples of holo-CRABPII were prepared as
follows. CRABPII was first dissolved in ~50 mi PBS buffer (20 mM sodium phosphate,
150 mM sodium chloride with 99.6% D20, pD 7 .5, uncorrected). The protein
concentration was ~40 uM. All-trans-RA was dissolved in DMSO-d6. The concentration
of the RA solution was ~3 mM. The RA solution was then added to the protein solution
by aliquots of ~5 11.1. The solution was quickly mixed by stirring after each addition. The
entire procedure was carried out in a dark room. The final DMSO concentration was less
than 1.5%. The sample was concentrated to ~06 ml. H20 sample of uniformly l“N-labeled
holo-CRABPII was prepared similarly except that the D20 in PBS buffer was replaced by
HZO. The final protein concentrations of the NMR samples were ~2 mM.

NMR spectroscopy. Homonuclear 2D spectra of the unlabeled DZO sample and
heteronuclear 2D and 3D spectra of the uniformly '“N-labeled HZO sample were acquired
and processed as detailed in Chapter 3 and 4. The temperature was set at 25 “C.
Homonuclear 2D spectra were acquired with a spectral width of 8000 Hz in both
dimensions. The carrier frequency was 4.70 ppm (solvent resonance). The relaxation

delay was 1.2 s. The residual water resonance was suppressed by low-power

192

presaturation. One DQF-COSY spectrum, three clean-TOCSY spectra with mixing times
of 20, 33 and 50 ms, and two NOESY spectra with mixing times of 100 and 150 ms were
acquired on the D20 sample. The time domain data were composed of 2048 X 320
complex points for the DQF-COSY (128 transients) and NOESY (64 transients)
experiments and 2048 X 256 complex points for the TOCSY experiments (64 transients).
The carrier frequencies were 4.70 ppm for 1H and 121.0 ppm for 15N for the
heteronuclear experiments. A 2D lH-‘SN HSQC spectrum was acquired with the
following acquisition times and complex data points: lH (F2) 121.9 ms, 1024 and 15N
(F 1) 53.3 ms, 128 (32 transients). A 3D 1H-‘SN NOESY-HSQC spectrum (16 transients,
150 ms mixing time) and a 3D lH-‘SN TOCSY-HSQC spectrum (8 transients, 35 ms
mixing time) were recorded with the same acquisition times and complex data points: IH

(F3) 121.9 ms, 1024, lsN(132) 15.2 ms, 32, 1H (Fl) 15.2 ms, 128.

193

Results and Discussion.

Protein ’ H and ”N resonance assignments and secondary structures. The
protein 'H and 15N resonances (Table 5.1) were assigned by the same strategy as detailed
previously (Chapter 4). Briefly, spin system identification was achieved by the
homonuclear 2D and 3D l5N-edited TOCSY experiments. The sequential linking was
accomplished by the analyses of the 2D and 3D NOESY spectra. The assignment was
helped greatly by comparison with the sequential assignments of the wild type apo-
CRABPII and the mutant apo-R1 1 1M (Chapters 3 and 4). Secondary structural elements
were identified as described in Chapter 4. They were essentially the same as those
observed in the crystal structure except that the first helix 01A appeared to be mobile in
solution (Figure 5.1 and 5.2).

Conformational and dynamical changes in CRABPII induced by the binding
of RA. The majority of the NOEs observed between the bound RA and the protein were
in agreement, at least qualitatively, with those predicated from the crystal structure of
holo-CRABPII, suggesting that the tertiary structure of holo-CRABPII in solution is most
likely very similar to that in the crystalline state. In particular, the contacts between the
bound RA and the residues at the ligand entrance observed by crystallography were also
detected by NOEs, indicating that the conformation of the ligand entrance of holo-
CRABPII in solution is similar to that in the crystalline state. These NMR results further

strengthened the conclusion based on the comparison of the holo crystal structure and the

194

Table 5.1. 1H and 15N chemical shifis of human CRABPII in complex with

all-trans-RA in PBS at 25°C.

 

No

P1
N2
F3
S4
GS
N6
W7

K8
19
110
R1 1
812
E13
N 14
F15
E16
E17
L18
L19
K20
V21
L22
G23
V24
N25
V26
M27
L28
R29
K30
131
A32
V33
A34
A35
A36
S37
K38
P39
A40
V41
E42
I43
K44

Q45

lSN

121.9
112.4
110.5
118.5
122.9

123.3
125.8
118.0
120.8

120.5

121.2
113.8
118.9
121.3
117.5
106.6
121.4
128.2
120.5
119.8
117.8
118.8
116.2
118.9
120.5
116.5
120.4
121.2
116.6
113.5
119.0

123.0
124.5
130.0
122.9
128.8
127.9

HN

9.86
8.48
9.22
8.19
9.31

10.17
8.60
9.24
7.49

8.52

7.59
7.25
8.24
7.84
7.15
7.68
7.60
8.80
8.57
7.94
7.80
8.81
7.63
7.00
8.56
8.85
7.19
8.18
8.69
7.38
7.05

8.77
7.95
9.29
9.29
9.57
8.60

HQ

4.35
5.08
4.35
4.70
4.04, 3.82
5.69
5.21

5.26
3.99
4.62
4.58
5.07
4.83
4.99
4.46
3.30
3.74
3.74
3.60
3.83
3.77
3.96
3.96, 3.61
3.74
4.42
3.49
4.13
4.09
3.58
3.90
3.75
3.88
3.50
4.15
4.24
3.96
4.22
4.80
5.09
5.03
5.22
5.35
5.38
4.65
4.51

H“

2.39, 1.92
2.98, 2.90
3.34, 3.28
4.26, 3.99

2.72, 2.55
3.23, 2.99

1.98, 1.88
1.91

1.70, 1.70
3.59, 3.33
1.94, 1.94
3.35, 3.16
2.99, 2.72
1.85, 1.85
1.87, 2.16
1.86, 1.69
1.66, 1.60
1.93, 1.63
2.15

1.84, 1.58

1.47
3.17, 2.64
2.07
2.13, 1.98
1.80, 1.64
1.95
1.91, 1.91
1.91
1.18
2.07
1.47
1.51
0.86
3.98
1.90, 1.61
1.92, 1.87
1.37
1.88
2.08, 2.02
2.26
1.87, 1.76
1.84, 1.79

Others

1172.11, 2.11; H“ 3.35, 3.28
yN 112.5;yN11, 7.75, 7.02
2, 6H 7.43; 3, 5H 7.38; 4H 7.17

2H 6.80; 4H 7.07; 5H 7.26; 6H 6.81;7H 7.14
N“ 124.8; H“ 9.17

H" 1.68, 1.63; H“ 1.37, 1.33; H“ 2.87, 2.83
H""' 0.83

H" 2.09, 2.19

2, 6H 7.06; 3, 5H 6.64; 4H 6.59
H" 1.98, 2.17

H" 2.27, 2.27

H“ 0.81, 0.81

H" 1.17; H6 0.32, 0.22

H" 1.56, 1.35; H“ 2.82, 2.93

H" 1.10, 0.94

H“ 0.92, 0.70

H" 0.82, 0.82

yN 112.4;er12 7.71, 6.92
H" 0.97

H" 2.60, 2.45

H" 1.33;}1‘5 0.88, 0.77

H" 1.72 1.53; H“ 1.45, 1.45; H“ 2.90, 2.90
H"'“ 0.83

H" 1.03, 0.96

H" 1.42, 1.31; H" 2.95, 2.95
H“ 3.65, 3.55

H" 0.76, 0.44
H"2.27, 2.27
H" 1.67, 1.34; H“ 0.48; H""' 0.85
H" 1.49, 1.36
H"2.04, 2.04

 

195

 

No

E46
G47
D48
T49
F50
Y51
152
K53
r54
SSS
T56
T57
vss
R59
T60
T61
E62
163
N64
F65
K66
V67
G68
E69
E70
F71
E72
E73
Q74
r75
v76
D77
G78
R79
P80
C81
K82
S83
L84
vss
K86
W87

E88
SS9
E90
N91
K92
M93

iSN

128.7
117.5
126.5
116.6
124.1
122.5
123.3
133.4
124.7
123.1
113.5

109.5
119.0
122.7
117.2
122.9
116.8
120.2
116.8
120.8
127.3
114.8
120.1
130.3
121.3
119.4
120.3
118.1
111.7
122.4
112.3
107.0
120.7

118.8
119.8
124.7
120.5
130.0
123.6
120.1

127.7
112.4
117.1
111.0
119.9
127.0

HN

8.53
9.30
8.56
8.08
9.10
8.35
8.35
9.31
8.84
9.56
7.74

6.71
7.09
8.73
8.72
8.77
8.29
8.10
8.36
9.35
8.98
8.93
7.96
9.25
9.06
8.72
8.44
7.68
9.25
9.25
8.75
7.96
7.91

8.85
8.89
9.05
7.34
9.24
8.47
8.69

9.25
8.60
8.72
8.05
7.83
9.40

HG.

4.37
3.96, 3.61
5.21
4.66
4.63
5.33
4.62
5.14
4.79
5.25
4.93
3.98
4.48
4.76
5.32
4.64
4.85
4.93
5.59
4.70
5.30
3.24
4.42, 3.62
4.76
4.80
5.14
5.05
4.76
5.32
4.61
3.95
4.46
4.06, 3.59
4.70
4.65
5.09
4.75
4.86
4.25
4.38
4.51
5.14

4.37
4.57
3.68
4.97
5.52
5.21

H“
1.80, 1.66

3.16, 2.72
4.15
2.57, 2.57
3.03, 2.69
1.89
1.71, 1.67
3.83
3.82, 3.77
4.33
4.18
2.18
2.02, 1.91
3.93
3.98
1.94, 1.94
1.90
2.58, 2.53
2.13, 2.13
1.65, 1.60
1.88

2.04, 1.92
2.00, 1.91
3.57, 3.18
1.96, 1.92
2.04, 1.92
1.69, 1.73
4.15

2.12

2.82, 2.13

180,156

2.62, 2.17
1.58, 1.50
2.78, 1.54
1.60, 1.30
1.37

1.64, 1.45
3.26, 3.00

1.98, 1.69
4.14, 3.99
1.87, 1.80
3.18, 2.88
2.01, 1.95
1.80, 1.80

Others

H" 2.01, 2.01

H" 1.01

2, 6H 6.46; 3, 5H 6.81; 4H 7.09

2, 6H 6.93; 3, 5H 6.61

H" 1.35, 1.35; H""1 0.28; H“ 0.37

H" 1.60, 1.48; H" 1.17; H“ 2.81, 2.81
H" 0.85

H" 1.18
H" 1.32

H" 0.81, 0.81

H" 1.65, 1.49

H" 1.06

H" 0.97

H" 2.21, 2.10

H" 1.33, 0.75; H“ 0.84; H“ 1.06
yN 113.0; mi, 7.62, 6.93

2, 6H 6.20; 3, 5H 6.51; 4H 6.28
H" 1.53, 1.47; H“ 2.94, 2.94
H"0.66, 0.45

H" 2.16, 2.16

H" 2.22, 2.17

2, 6H 7.24; 3, 5H 6.83; 4H 6.47
H"2.33, 2.14

H" 2.16, 2.16

H"2.24, 1.99

H" 1.07

H" 1.18, 0.86

H" 1.27, 1.27; H“ 2.80, 2.80

H" 1.17; H“ 0.73, 0.69

H" 0.24, 0.00

H" 1.52, 1.27; H“ 1.15

2H 7.32; 4H 7.98; 5H 6.85; 6H 7.23; 7H 7.63
N“ 129.5; H“ 10.68

H" 2.16, 2.02

H" 2.09, 2.09

yN 112.6; yNH, 7.49, 6.84
H" 1.49, 1.49

H"2.07, 199; He 0.43

 

No

V94
C95
E96
Q97
K98
L99
L100
K101
(3102
13103
6104
P105
K106
T107
8108
W109

T110
R111
E112
L113
T114
N115
D116
G117
E118
L119
1120
L121
T122
M123
T124
A125
D126
D127
V128
V129
C130
T131
R132
V133
Y134
V135
R136
E137

ISN

118.8
127.1
127.5
120.7
122.8
125.9
121.5
117.5
108.7
116.8
108.8

123.6
119.0
112.1
116.2

110.8
119.9
123.1
126.4
116.9

115.6
107.6
117.4
124.5
125.1
135.4
119.3
121.2
115.6
125.4
122.2
121.1
123.3
125.6
129.7
124.5
128.6
122.7
126.7
112.0
122.3
126.5

HN

8.80
8.87
7.67
7.41
8.96
8.84
7.74
8.03
8.27
8.17
8.16

8.58
8.80
7.80
8.26

9.58
8.81
8.91
8.23
9.38

7.74
7.98
7.42
8.92
9.51
9.98
9.15
9.28
8.27
8.40
9.26
8.49
8.41
8.42
9.59
8.45
9.85
8.56
9.79
9.52
8.44
8.10

HQ

4.30
4.51
4.35
4.93
4.59
4.02
4.10
4.42
4.06, 3.75
4.61
4.14, 3.93
4.51
4.47
5.39
4.70
5.79

5.24
5.62
5.80
5.31
4.60
4.46
4.76
4.19, 3.70
4.98
5.13
4.98
5.61
5.85
5.16
5.48
4.88
4.16
4.52
3.95
4.42
5.20
5.08
5.09
4.98
5.61
5.35
3.88
3.87

HB

1.80

0.41 —0.27

1.90, 1.80
1.77, 1.77
1.53, 1.53
1.64, 1.47
1.65, 1.48
1.80, 1.80

1.80, 1.80

2.03, 2.03
1.80, 1.80
4.49

3.93, 3.77
3.53, 3.33

3.96

1.80, 1.50
1.95, 1.95
1.28, 0. 75
4.761
2.82
2.81, 2.42

1.83, 1.72
0.87
1.82
1.90, 1.34
4.05
1.84, 1.80
3.96
1.04
2.75, 2.75
2.68, 2.68
2.14
1.84
2.73
3.76
1.75, 1.61
1.71
3.03, 3.03
2.39
1.32, 1.09
1.81 1.69

196

Others
H" 0.86, 0.75

H" 2.14, 2.14

H" 2.43, 2.04

H" 1.26, 1.26; H‘ 2.79, 2.79
H" 1.38; H‘5 0.67, 0.67

H" 1.28; H’5 0.78

H" 1.53, 1.37

H" 2.13, 1.98

H" 1.85, 1.73; H‘5 3.61, 3.25
H" 1.53, 1.42
H" 1.15

2H 6.92; 4H 6.90; 5H 6.35; 6H 6.44; 7H 6.67
N‘ 138.1; H‘ 11.38

H" 1.00

H" 1.32

H"2.04, 2.04

H" 1.12, H‘5 0.46, 0.46

H" 1.26

yN 112.6; yNH2 7.65, 6.95

H" 2.04, 1.91
H" 0.46; H5 0.15, —0.25

H6 0.78, 0.71

H" 1.09

H"2.64, 2.33 H‘ 1.63
H" 1.00

H" 0.92, 0.69
H" 0.78, 0.70

11" 1.00

H" 0.86, 0.86

2, 6H 6.83; 3, 5H 6.46

H" 1.06, 0.86

H" 0.58, 0.04; H6 2.81, 2.73
H" 2.07 2.07

 

197

Figure 5.1. Strip plots extracted from the 150 ms mixing time 15N-edited NOESY-
HSQC spectrum of the holo-form CRABPII showing the NOE connectivies for residues

Met27-Lys38.

198

 

 

 

 

 

 
 

>16»);

3)

(«555

  

1&1
1;

[All

LE2 M‘ .‘ ERR».

‘4,

1

3 ~ ~ _

é"

 

5* =1— N‘

., 1.11,? gt
\‘1'51
m0,

- saw
v .

 

 

 

 

 

 

 

 

 

 

 

i+1

i+1

i-2

f

i-l

 

 

 

 

 

 

1+1

 

 

0.0

 
  

1.0

 
 
 
  

”6.0

 

 

 

I I ‘T 1 T I 1 1 I I 1 1
M27 L28 R29 K30 131 A32 V33 A34 A35 A36 S37 K38

199

Figure 5.2. Summary of the sequential and medium-range NOEs involving backbone HN
and Hcl atoms and the deduced secondary structures of holo-CRABPII. Line thickness for

dGIN and dNN sequential NOE distances reflects the intensities of the cross peaks.

200

10 20 3o
PNFSGNWKIIRSENFEELLKVLGVNVMLRKIAVAAA

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SbWNngo t t :1 10 o .3. t ‘

(W1) _— _ ‘—_——-
daNth) __ _ -—-—__ _
(111110.02)

Kai-P3) —__— ___-*1—__

654-3) ' . — fl

Wm rIIIEA1111> $ k _
40 50 60 70
SKPAVEIKQEGDTFY‘IKTSTTVRTTEINFKVGEEFEE
WWW OOO OOOO O O
daN(i,i+1) - —-— ——- —1_-_-—.
dNN(i,i+l) I __ I. _ ‘ -
(161116.92) __ ___ __ —__

(W2) _. _

dNNG.’r*3) __
WW 0223;?» anpézzb 11111 PLOW”, czgcn
80 90 100
QTVDGRPCKSLVKWESENKMVCEQKLLKGEGPK
SWNHW t 1t t a o o t .
duN(i,'r1-1) — —-———-—-— -
dNNCLi'tl) —_ _— —* —
duN(i,i+2) -'_'-— ——-1_._
dNN(1,1+2) ___ —__ _
911196.113) -.__ __
mm c’ ”HEP-Fm, IIIIIBGIIIII) E2:
110 '1 120 g 130
SWTHELTNDGELILTMTA‘DDVVCTRVYVRE
”mm O O O O OO OOOO O OO O O O OO O

daNGJH) h.- nun—— —_-
daNam) —
“1010:1121 __ _
dNN(i.i+3) ——

. BH 31 N
mm m m fiT, _ _ _ rTLC.

 

201

apo solution structure that RA binding induces significant conformational changes at the
ligand entrance (Chapter 3).

Dynamical changes upon binding of RA were evidenced by the differences in
peak intensity between the 2D lH-‘SN HSQC spectra of apo- and holo-CRABPII (Figure
5.3 and 5.4) and the differences in the number and intensity of the observed NOEs.
Remarkable differences were observed in the second helix, which is part of the ligand
entrance. It had a typical a-helical NOE pattern in the holo-form (Figure 5.1 and 5.2) but
no such pattern was observed in the apo-form. Furthermore, the residues in the second
helix had much more intense cross peaks in the 2D lH-‘SN HSQC spectrum of the holo
form. The second helix is apparently disordered in the apo form (Chapter 3). Other
residues at the ligand binding pocket also showed more intense cross peaks in the 2D 'H-
15N HSQC spectrum upon binding of RA, including Thr56-Ile63, Val76-Gly78, Leu121,
Arg132 and Argl34. Thr56-Ile63 and Val76-Gly78 constitute the [SC-[3D loop and the
BE-BF loop, respectively, at the ligand entrance. Leu121, Arg132 and Arg134 are located
at the bottom of the binding pocket. Their side-chains interact with the carboxyl group of
RA. The results suggest that while the ligand binding pocket of apo-CRABP undergoes
conformational exchange, the conformational exchange is greatly reduced upon binding
of RA. Similar ligand-induced dynamical changes have been observed in the NMR
studies of CRABPI (Rizo et al., 1994) and intestinal fatty acid binding protein (I-FABP)
(Hodsdon et al., 1995; Hodsdon et al., 1996; Hodsdon & Cistola, 1997a, b).

The stability of the ligand entrance in holo-CRABPs may be brought about by the
hydrophobic interactions between the isoprene tail of RA and the protein. However, not

all HSQC cross peaks in CRABPII became stronger after the binding of RA. The cross

202

Figure 5.3. Gradient- and sensitivity-enhanced 2D lH—‘sN HSQC spectrum of uniformly
l5N enriched of the wild type holo-CRABPII in complex with all-trans-retinoic acid.

Sequential assignments are indicated with one-letter amino acid code and residue number.

120 ~1

"190112110 ‘

125——

1—<——--—- _... 11 ._.-.._..1_..-___._ .—

O
W87Ne

. W 109Ne

 

203

 

W., -—.-——.. .—

G780 &
o
0117 ‘
0102”
6104

19181
8108}. .

369331.19

T124
.
E90 W109. T49

966' 1141?”.

0116
O

V33 9

Tax-30.1538 m

F15

0126

o
w 043 El 2,140
555 9 1.1191(1

0
M93 '-

N25 0

.
c130 QV85
E70

Y134

0
L121

K1331 9&4

M2
6400192091214M4

G
87 °D 27EG99°~ V24
V263 M5523 3“

0
E96

rose-.9131
.9.

 

 

11.5

ITIIIIHITTllilll

11.0 10.5

10.0

Tj11111111111111111111711111[11111]

9.5 8.5

'11 (99m)

8.0 7.5

6.5

204

Figure 5.4. Relative peak intensities of the 2D lH—-‘5N HSQC spectra of apo-CRABPII
(A) and holo-CRABPII (B) as a fimction of the residue number. (C) was obtained by

subtracting (B) from (A).

Relative lntenslty

205

 

 

 

 

 

 

 

 

 

 

 

 

 

111 1 11 1 1

10 20 30 40 50 60 70 80 90 100 110 120 130

Resldue Number

206

peaks of Ser12, Glu13, Asn14, Glu16 and Glul7 could not be observed in the 2D lH-‘SN
HSQC spectra. The cross peaks of IlelO, Argl l, PhelS, Leu19 were very weak. Leu18
probably was also weak. These residues became more flexible afier the binding of RA.
Their high mobility compensates, at least partially, the decrease in entropy caused by
ligand binding.

’H resonance assignments of CRABPII-bound RA. The 'H resonances of the
bound RA were assigned by analysis of the homonuclear 2D DQF-COSY, TOCSY and
NOESY spectra of holo-CRABPII recorded in D20. The olefinic protons were well
resolved from the protein resonances. Through-bond correlations could be clearly
identified for CH(7), CH(8), CH( 10), CH(l 1) and CH(12) protons (Figure 5.5). CH(l 1)
was assigned by its distinct through-bond correlation pattern. The remaining olefinic
protons were assigned based on their NOE patterns (Figure 5.6). The methyl protons of
CH3(18), CH3(19) and CH3(20) were assigned based on their NOEs to olefinic protons.
The assignments of the methyl protons of CH3(16) and CH3(17) were tentative. None of
the methylene protons of CH2(2), CH2(3) and CH2(4) of B-ionone ring was assigned at
present. The 1H chemical shifts of the RA bound to human CRABPII (Table 5.2) were
almost identical to those reported for the RA bound to mouse CRABPII (Norris, et al.,
1994) except for CH(12) which differed by ~0.2 ppm. The chemical shifts of the
CRABPII-bound RA protons were also similar to those of free RA in CC], except for
CH(8), CH(l l) and CH3(20) protons, which were 6.06 ppm, 6.96 ppm and 2.36 ppm in
CC14, respectively (Waterhous & Muccio, 1990).

Conformation and dynamics of bound RA. Most of the NOEs observed for the

bound RA (Table 5.3, Figure 5.6 and 5.7) were consistent with the distances measured

207

Table 5.2. 'H resonance assignment of retinoic acid bound to human CRABPII in

PBS at 25°C.

 

Olefinic proton (CH) CH(7) CH(8) CH(10) CH(ll) CH(12) CH(14)

Chemical shift (ppm) 6.08 6.36 6.22 6.47 6.14 5.62
Methyl protons (CH3) CH3(16)’ CH3(17)' CH3(18) CH3(19) CH3(20)
Chemical shifi (ppm) 0.90 0.99 1.66 1.92 1.97

 

' the individual assignments of CH3(16) and CH3(17) were assigned tentatively.

208

Table 5.3. The relative intramolecular NOE volumes 3 of RA bound to human CRABPII

in PBS at 25°C (the mixing time was 100 ms).

 

Proton pair
Relative NOE volume
Proton pair
Relative NOE volume
Proton pair
Relative NOE volume
Proton pair
Relative NOE volume
Proton pair
Relative NOE volume

7-8
4.8
8-10
40.0
10-11
38.3
8-19
33.6
12-20
18.5

7-10
1.9
8-12

16.1
10-12
43.1
11-14
30.0
12-14
100.0

7-16
34.0*
8-14
10.7
10-14
36.7
11-12
9.1
14-20

strong"

7-17
40.1*
8-16
31.8*
10-18
33.2
11-19

strong“

7-18
26.9
8-17
31.6*
10-19
30.5
11-20

strong"

7-19
45.0
8-18
45.6
10-20
6.0
12-19
12.3

 

a The NOEs were picked by the program NMRCOMPASS 2.5 (Molecular Simulations)

from the homonuclear 2D NOESY spectrum with a mixing time of 100 ms. The relative

volumes were obtained by integration using NMRCOMPASS.

'These intramolecular NOE peaks may be overlapped with intermolecular ones.

" Strong peaks but no accurate values could be obtained because of overlapping.

209

Figure 5.5. Part of the 2D TOCSY spectrum with 33 ms mixing time of the RA bound to

human CRABPII at 25°C. The CH(7), CH(8), CH(lO), CH(l 1), CH(12) and CH(14)

protons are labeled.

210

 

   
 

’1
p

CH(14)

V 1
1111 111:1

1?

CH(10)-CH(11)

Ct at 01
O O 0
fi N 0

0’1
O
as

1111111111111111111111111111111111111111111111111:1

 

 

 

 

IIIIIIIIWTIIIIll[[11llllllllllllll[11'11111111111111IIIIIIIIPTIIIIIIIIITIIT

7.0 6.8 6.6 6.4 6.2 6.0 5.8 5.6

F2 (PM)

211

Figure 5.6. Part of the NOESY spectrum with 100 ms mixing time of the holo-CRABPII
in complex all-trans-RA at 25°C showing its intramolecular NOEs and the intermolecular
NOEs between RA and Phe15 of the protein. The identities of the NOEs are as follows.
A, CH(14)-CH(11); B, CH(14)-CH(8); C, CH(14)-CH(10); D, CH(14)-CH(12); E,
CH(10)-CH(12); F, CH(lO)-CH(7); G, CH(8)-CH(10); H, CH(8)-CH(12);

I, CH(8)-CH(7); J, CH(1 l)-CH(12); K, CH(11)-CH(10); a, 2, 6H(F15)-CH(14);

b, 4H(F15)-CH(14); c, 3, 5H(F15)-CH(14); d, 4H(F15)-CH(12); e, 3, 5H(F15)-CH(12); f,
4H(F15)-CH(10); g, 3, 5H(F15)-CH(10); h, 4H(FlS)—CH(8); i, 3, 5H(F15)-CH(8);

j, 4H(F15)-CH(11); k, 3, 5H(F15)-CH(14).

'1
H

l-fi
N

212

 

  
   

v
a

m.

.
U'
o

6.6
1.3

v '
L14111111111111111111111111111111111111111411111111111111111

m m
1.» E:
'9

at
o
h

01
O
at

 

 

m
e
m

 

7.0 ‘
4"" ‘3'; ‘3l:"

‘.
lllllllllllllllllllll[1111]lllllllllllllrlllllllll[ITTIlllllll

7.0 6.8 6.6 6.4 6.2 6.0 5.8 5.6

 

F2 (ppm) in

212

 

08
0A
am.

.O

Q.

 

 

 

. a
TIAJFCJ4211143J
m 8

O
1 5
r w

0

6

2

6

H
1

4

6

_~—ddd«q.—«ddfidqqud—d—qq—qq—q—qqfiuj

6

6

 

5.8 5.6

6.0

6.2

6.8 6.6 6.4

7.0

13

F2 (m)

213

Figure 5.7. The relative volumes (V "/6) of some of the NOEs observed for the bound
RA vs. the internuclear distance (r) measured from the crystal structure of the holo-
CRABPII in complex with all-trans-RA. The straight line represents a linear relationship

between the relative volume and the distance.

214

Relative NOE Volume (V"’6)

2v eon—~35 520352:—

QN

 

   

 

 

o6 mé o4. m.m o.m mN

CF.O . _ . p _ . _ . _ .

. Iiimr.
N; 1 -

. 13-12 1.75: INF-”2

- 213+: F .. 12in
3.0 1
$2.5

.. I

o F .o 1 622$: 3 5221-1...
I
. 5921-12 INTI:
. 1 I ..

3 o $1-13 :m-E
omd ..

. 12+:

I

mud 1

 

 

 

215

from the crystal structure of holo-CRABPII (Kleywegt et al., 1994), suggesting that the
conformation of the bound RA in solution is similar to that in the crystalline state. The
few discrepancies between the observed NOEs and the distances measured from the holo
crystal structure were probably caused by the internal motions of the bound RA in
solution rather than a different conformation as discussed below.

It is well known that NOE between an isolated pair of protons H,-H2 is
proportional to cross-relaxation rate 0, which is determined by the internuclear distance r
between the two protons and the spectral densities J‘q’(00) at two frequencies (Chapter 1).

o (HlHHZ) = 74Hh2/4r" {6.10" (20)") - J ‘0’ (0)}

where y” is the proton gyromagnetic ratio. However, for a pair of protons belonging to a
protein or a ligand bound to a protein the situation is usually much more complicated. In
particular, internal motions may have significant effects on the spectral density.
Furthermore, cross-correlation with additional spins may also change the apparent cross-
relaxation rate (Werbelow & Grant, 1977). While these factors may complicate the
accurate measurement of inter-proton distances, they are also part of the basis for
studying internal motions of biomacromolecules. If two pairs of protons in the same
molecule have the same inter-nuclear distance but different NOE intensities, it suggests
that their motions relative to other parts of the molecule are different.

Analysis of the intramolecular NOEs of the bound RA indicated that all NOEs
involving CH(7) are weaker than expected. For example, the distances between CH(7)
and CH(8), between CH(lO) and CH(1 l) and between CH(1 l) and CH(12) protons all are

~3.0 A. Medium NOE cross-peaks are expected for all three pairs of protons if their

 

216

relaxation times (T, and T 2) are much longer than the mixing time (100 ms) and their
positions in the binding pocket are fixed. However, the volume of the NOE cross-peak
between CH(7) and CH(8) was significantly lower that those of the other two pairs of
protons (Table 5.3 and Figure 5.6 and 5.7). By contrast, the corresponding peaks in both
2D DQF-COSY and TOCSY spectra were all relative strong (Figure 5.5), indicating that
the relaxation times of these protons were similar to the others, and were longer than the
100 ms mixing time. The weakness of the NOE cross-peaks involving CH(7) was most
likely due to the relative motions of the proton.

The nature of such relative motions could be either intramolecular or
intermolecular. For RA, if disregarding the methyl groups, the possible intramolecular
motion is about the 6-s bond (Honig etal., 1971), which affects the NOEs of the protons,
among the others, CH(7) and CH(8). The intermolecular motions could be either the
whole RA molecule moves as a unit relative to the protein or local motions restricted to
certain parts of the RA molecule. The motions of the whole RA molecule would attenuate
the intensities of all the NOEs involving the RA protons to similar extent. Thus, the
attenuation of the peak intensity is most likely due to the local motions relative to a
portion of RA. The weakness of the CH(7)-CH(8) cross-peak can be caused by either
intramolecular motions about 6-s bond or local intermolecular motions restricted to their
locations. In either case it suggested that the ionone ring of the bound RA is not fixed as
snugly in the protein as suggested by the crystal structure of holo-CRABPII.

The NOEs involving CH(14), on the other hand, were all stronger than the
average. In fact, the NOE peaks between CH(8) and CH(14) and between CH( 10) and

CH(14) protons were stronger than that between CH(7) and CH(8) protons (Table 5.3). In

217

the crystal structure of holo-CRABPII, the distances between the CH(10) and CH(14)
protons, between the CH(8) and CH(14) protons and between the CH(7) and CH(8)
protons were ~4.8 A, ~6.8 A and ~3.0 A, respectively. Obviously, spin diffusion
contributed to the appearance of the NOE cross-peak between CH(8) and CH(14) protons
and to the strength of the NOE cross-peak between CH(lO) and CH(14) protons. The spin
diffusions could be mediated by protons on the protein and/or RA. Since only the methyl
group of Met123 is within 3 A distance of CH(14) and the methyl group is ~7.5 A away
from CH(8), the spin diffusions are most likely through CH(12). Since only a structure
rigid on the time scale relevant to the NOESY experiment can provide efficient pathways
for spin diffusion, it suggests that the carboxylate end of the isoprene tail was essentially
immobile relative to the protein.

Very interestingly, the rigidity of the carboxylate end of the isoprene tail and the
flexibility of the B-ionone of the bound RA molecule correlate well with the binding
affinities and specificity of CRABPs for the retinoids. It has been shown that the
carboxylic group is critical for RA to bind to CRABPS, while the modifications at the B-
ionone ring and the C-7 and C-8 positions can be tolerated.

Our conclusion of the flexibility of the bound RA relative to the protein appears to
be inconsistent with what was reported for mouse holo-CRABPII (Norris, et al., 1994).
Their conclusion of a single, static conformation of the bound RA at the 6-s bond was
based solely on intramolecular NOEs between the olefinic protons and the methyl
protons. It was derived under the assumptions that both the protein and RA are rigid, and
the NOE volume is proportional to 1". Furthermore, only three NOEs picked from an

NOESY spectrum with 30 ms mixing time had values much higher than the noise. All

218

others have intensities similar to the noise. In fact, most NOEs involving CH(7) were also
lower than expected by their model, suggesting that there are also motions in the RA
bound to mouse CRABPII. Interestingly, similar intermolecular motions of the bound RA
relative to the protein was observed in mouse holo-CRABPI.

A mechanism for the exit of RA. The intensities of most of the intermolecular
NOEs between the bound RA and the protein agreed, at least qualitatively, with those
predicted from the crystal structure (Table 5 .3). However, differences between the
solution and crystalline state were also observed. The predicted NOEs between the H‘5
protons of Arg59 and CH(7) of the bound RA, between the H5 protons of Arg59 and
CH(8) and between the H5 protons of Leul9 and CH(8) were not observed, and the
intensity of the NOE between the H‘3 protons of Ala32 and CH(7) was much lower than
predicted. The intensities of these NOEs must be attenuated by intermolecular motions.
In the holo crystal structure, the side-chain of Arg59 extended from the BC-BD loop to
the Glu74 at the C-terminus of BB. The motion of this side-chain relative the bound RA
could temporarily open up the ligand entrance. Such flexibility may be necessary for the

exit of the bound RA.

219

Conclusion.

The 'H and [5N resonance assignments of holo-CRABPII have been completed.
Most of the observed intermolecular NOEs between the protein and bound RA and the
intramolecular NOEs of the bound RA are, at least qualitatively, consistent with those
predicted from the crystal structure of holo-CRABPII. The results suggested that RA
binds to CRABPII in solution in the same manner as determined by crystallography.
Comparison of the intensities of backbone amide resonances between apo- and holo-
CRABPII indicated that the ligand entrance of holo—CRABPII is much less flexible than
that of apo-CRABPII. Relative motions between the bound RA and human CRABPII in
solution, particularly at the ligand entrance, were observed. Such motions may provide a

transient pathway for the exit of RA.

220

References

Chambon, P., Olson, J. A., & Ross, A. C. (coordinators) (1996) The retinoid revolution,
FASEB J. 10, 938-1048.

Hodson, M. E., Toner, .1. J ., & Cistola, D. P. (1995).]. Biomol. NMR 6, 198-210.

Hodson, M. E., Ponder, J. W., & Cistola, D. P. (1996) J. Mol. Biol. 264, 585-602.

Hodson, M. E., & Cistola, D. P. (1997a) Biochemistry 36, 1450-1460.

Hodson, M. E., & Cistola, D. P. (1997b) Biochemistry 36, 2278-2290.

Honig, B., Hudson, 3., Sykes, B. D., & Karplus, M. (1971) Proc. Natl. Acad. Sci.
U. S. A. 68, 1289-1293.

Jamison, R. S., Newcomer, M. E., & Ong, D. E. (1994) Biochemistry 33, 2873-2879.

Kleywegt, G. J., Bergfors, T., Senn, H., 1e Motte, P., Gsell, B., Shuao, K., & Jones, T. A.
(1994) Structure 2, 1241-1258.

Norris, A. W., Rong, D., d’Avignon, D. A., Rosenberger, M., Tasaki, K., & Li. B. (1995)
Biochemistry 34, 15564-15573.

Rizo, J ., Liu, Z. P., & Gierasch, L. M. (1994) J. Biomol. NMR 4, 741-760.
Spom, M. B., Roberts, A. B. & Goodman, D. S. (1994) The Retinoids: Biology,
Chemistry, and Medicine, 2"d ed. Raven Press, New York.

Sporn, M. B., Roberts, A. B. & Goodman, D. S. (1994) The Retinoids: Biology,
Chemistry, and Medicine, 2"d ed. Raven Press, New York.

Waterhous, D. V., & Muccio, D. D. (1990) Magn. Reson. Chem. 28, 223-226.

Werbelow, L. G., & Grant, D. M. (1977) Adv. Magn. Reso. 9, 189-299.

111011an STATE UNIV. LIBRARIES
111111111111111111111111111111111111111111111111111111111111111
31293016900684