:9

932

E

I: N ‘7: vnjztptfrnwowpnmgufi ‘3, ‘ V , .~ _ ‘

 

N If! IDWFtt'HI 1

- yawn”.

 

:. p.22“ , i ‘ “mg...

.1 :21 . V an»
.f. . ix}..- 8
‘ . , «4.73.:- .
w E: t
... . , A z.

 

a,
n.» $12..

‘ .51... a...” «G. a...
$553.93. a. . is

3:» a!» 511+! s.

a £3315;

 

 

” A). l . .
g3 5-1,:
x (a;

 

2.
a: V
I .1. «3' 'al 7«
3.3:
v; .12 .i‘ :4... a!
12:31 11.15,
. A.

 

mu.

mu

 

This is to certify that the
dissertation entitled

STRUCTURE FUNCTION RELATIONSHIP STUDIES OF
ESCHERICHIA COLI
ADP-GLUCOSE PYROPHOSPHORYLASE

presented by

CLARISA MARIA BEJAR

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Biochemistry and
Molecular Biolgqy

 

 

 

Major Professor’s Signature

laud-06?

 

Date

MSU is an Affirmative ActiorVEquaI Opportunity Institution

 

 

LIBRARY
Michigan State
University

 

 

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/CIRC/DateDue.indd-p.1

 

 

 

STRUCTURE FUNCTION RELATIONSHIP STUDIES OF
ESCHERICHIA COLI
ADP-GLUCOSE PYROPHOSPHORYLASE

By

Clarisa Maria Bejar

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry and Molecular Biology

2006

ABSTRACT

STRUCTURE FUNCTION RELATIONSHIP STUDIES OF
ESCHERICHIA C OLI
ADP-GLUCOSE PYROPHOSPHORYLASE

By

Clarisa Maria Bejar

Alpha-1,4-polysaccharides are one of the main energy and carbon storage
components present in most living cells. Bacteria, higher plants and algae utilize ADP-
glucose as the activated form of glucose for polyglucan synthesis. This step is catalyzed
by ADP-glucose pyrophosphorylase, a tightly regulated enzyme responsible for
determining the rate of the entire pathway. The mechanism of allosteric regulation is
linked to the carbon assimilation pathway in the cell, suggesting the evolutionary
adaptation of this enzyme to respond to the specific carbon and energy necessities within

the respective cell.

The allosteric properties of the enzyme depend not only on specific residues that
have been identified by several genetic and biochemical studies, but also on more
extended regions of the protein. In the present work, the domain organization of the
Escherichia coli ADP-glucose pyrophosphorylase monomer and the functional and
structural role of the N- and C-terminus in regulation were studied. The biochemical data
presented here in combination with structural information allow a mechanism to be
proposed for how the putative allosteric sites communicate with the active site to regulate

enzyme activity.

Structural information on ADP-glucose pyrophosphorylases at the three-
dimensional level also led to the investigation of other regions of the enzyme with
predicted functional roles. Structure-guided site-directed mutagenesis was used to study
the molecular basis of substrate binding and affinity. The detailed characterization of the
E. coli enzyme’s glucose l-phosphate site architecture may have relevance not only for
the members of the ADP-glucose pyrophosphorylase family, but also for other

nucleotide-diphospho-sugar pyrophosphorylases. .

ACKNOWLEDGMENTS

My time in the Department of Biochemistry and Molecular Biology at Michigan
State University has been an incredible opportunity to grow both academically and
personally. I not only learnt about enzymes, carbohydrate metabolism (and other
scientific topics) and new experimental skills but I was also able to develop a critical
scientific thinking when addressing a biological problem. The classes (how to forget
BMB80l II), the seminars, the troubleshooting in the lab, the exposure to excellent
scientists and all the resources offered by this department have been essential for shaping
the scientist I am today. Outside the academic world I had the chance to be exposed to
different cultures, languages and traditions from around the world and I am very thankful
for the amazing people I have met and the wonderful experiences we have shared

together.

Thanks to my PhD mentor, Dr. Jack Preiss, who gave me the possibility to join
his lab and the “Pyro” team. I am very lucky and grateful for having the chance to work

next to and learn from someone so experienced and knowledgeable in the field as he is.

Thanks to the professors in my PhD Advisory Committee: Dr. Christoph Benning,
Dr. Shelagh Ferguson-Miller, Dr. Charles Hoogstraten, all of them from the Department
of Biochemistry and Molecular Biology, and Dr. Mark Worden, from the Department of
Chemical Engineering. Special thanks to Christoph for being open, for listening, for his

continuous support at every moment of this PhD and for considering me as another

iv

member of his lab. Thanks to Shelagh for her advice and her support to perform the
computational modeling which has been such an enjoyable approach for me to learn.
Thanks to Charlie for his positive point of View about enzymology and life, and for his
English grammar lessons in the correction of this thesis. Thanks to Dr. Worden who
introduced me to the Technologies for a Bio-based economy program which allowed me
to get better exposure to the biotechnological industry as well as to get fimding from the

Department of Energy of the US that supported me during three years of my PhD.

Invaluable thanks to the former members of the Preiss lab: Drs. Miguel Ballicora,
Alberto Iglesias and Alejandra Yep for their time, their advice, their support, their
patience, their never-ending jokes, the mates, the Sparty’s coffees and the talks about
Argentinean history, literature and old TV shows. I really felt like at home with them.
Special thanks to Miguel for teaching me to be like the water, to Alberto for the
Almafuerte’s poems and to Ale for all the time we spent working, talking, singing,
talking, crying, talking, traveling, talking, cooking and talking together. To Claire
Devillers for the great time during my first years in East Lansing. To Ana Bertolo, Pedro
Crevillen and Tiziana Ventriglia, who visited the Preiss’ lab in different summers giving
us the chance to learn from their research and to laugh with them so much! And to Dr.
Mirta Sivak who has always made sure that, although we are in the US, we have an

Argentinean mom around taking good care of all of us.

Thanks to the Department of BMB for their friendly and supportive environment.

Thanks to my dear friends in the department for their constant support, advice and fun

times: Hiroshi Maeda, Harini Krishnamurthy, Setsuko Wakao, Francisco Herrera, Dean
Shooltz, Clarie Viellie, Karla Ziegelmann-Field, Jake McKinley, Soledad Quiroz, Cora
Fong and the whole Benning lab. Thanks to Pappan and the administrative stuff (Julie,
Chris, Leslie, Melinda, Jessica and Teresa) for being always so helpful and cooperative.
To Joe Leykam for his advice and the interesting stories from California and to Dr. Leslie

Kuhn for her support during my postdoctoral job search.

Thanks to my great friends outside the lab. Very special thanks to Gisela and Nate,
for encouraging me to pursue this PhD, for their constant support, for the great times
together and for Sophie. To Ale Yep (again), Javier, Valentina, Rafael, Pako, Phil, Gaby
Ballicora and Ale Manzan for the many Argentinean laughs. To all the friends I met
during these years, the Spanish, the Italian, the French, the Latin Americans, the
Japanese... for the fun parties, the opera performances, the many trips, the wine-tasting

nights and the company. They have all became my family here.

Thanks to what is most important to me: my family in Argentina. To my mom
Dorita, my dad Hector, my sister Analia, my cousin Valeria, my dearest uncle Rubenacho,
my aunts, uncles and cousins for their constant communication, joy and support. Thanks
Lili and Gordi, Tati, Ceci, Vangi, Leonora, Irene, Gaby, Vale Palanza for being always,

always, always there for me. Thanks to Internet for bringing all these people much closer.

Thanks to Brad for showing up in my life, for his support, friendship and love

during this last year of my PhD and for all the exciting times to come.

vi

TABLE OF CONTENTS

List of Tables ...................................................................................................
List of Figures ..................................................................................................
List of Abbreviations ......................................................................................
Chapter 1:

Literature review .............................................................................................

1.1
1.2
1.3

Introduction to polysaccharides ................................................................
Storage a-1,4-polysaccharides: structure and function ............................
Elucidation of the a-l,4-polyglucan metabolic routes: historical

background ...............................................................................................

1.4 Metabolic pathways for a-l,4-polysaccharides synthesis in bacteria and
plants .........................................................................................................
1.4.1 ADP-glucose pyrophosphorylase ..............................................
1.4.2 Bacterial glycogen / starch synthase ..........................................
1.4.3 Branching enzyme .....................................................................
1.5 Alpha-1,4-polysaccharides in non-photosynthetic eukaryotes .................
1.6 ADP-glucose pyrophosphorylase: the regulatory enzyme in bacterial
and plant a-1,4-polysaccharides biosynthesis ..........................................
1.6.1 Quaternary structure ..................................................................
l .62 Regulation ..................................................................................
1.6.3 Kinetic mechanism of action .....................................................
1.6.4 Structure-function relationship: background and motivation of
this thesis work ..........................................................................
1.6.4.1 Important residues and domains ................................
1.6.4.2 Three-dimensional structure ......................................
Chapter 2:
Domain organization of the ADP-glucose pyrophosphorylase from
Escherichia coli ................................................................................................
2.1 Abstract .....................................................................................................
2.2 Introduction ..............................................................................................
2.3 Materials and methods ..............................................................................
2.3. 1 Materials ....................................................................................
2.3.2 Plasmids and expression vectors ...............................................
2.3.3 Linker-scanning mutagenesis ....................................................
2.3.4 Truncated protein constructs ......................................................
2.3.5 Construction of a plasmid for expression of the C-terminus
alone (EC323431) ..........................................................................

2.3.6 Bacterial strains and expression of the recombinant enzymes ..
2.3.7 Protein methods .........................................................................

2.3.8 Enzymatic assays .......................................................................
2.3.9 Purification of Ec-insl l7 ...........................................................
2.3.10 Purification of E01423 .................................................................
2.3.11 Purification of co-expressed Ec1-323 + Bo 323.43. .........................
2.3.12 Molecular mass determination ...................................................

vii

xi
xiii

1-39

ll
l4
l6
I9
21

22
22
24
29

32
35
37

40-64
41
42
43
43
43
43
44

44
45
45
47
47
48
so
so

2.3.13 Kinetic characterization ............................................................. 51

2.4 Results and discussion .............................................................................. 52
2.4.1 Purification and characterization of Ec-ins 117 ......................... 52
2.4.2 Purification and characterization of C-terminal truncated

forms .......................................................................................... 58
2.4.3 Characterization of EC1.323 + EC328-43| ........................................ 59

2.5 Conclusions .............................................................................................. 62

Chapter 3:

The structural role of the Escherichia coli ADP-glucose

pyrophosphorylase’s N-terminus in allosteric regulation ........................... 65-85

3.1 Abstract ..................................................................................................... 66

3.2 Introduction .............................................................................................. 67

3.3 Materials and methods .............................................................................. 69
3.3.1 Construction of N-terminal truncated enzymes ......................... 69
3.3.2 Expression of the recombinant enzymes ................................... 70
3.3.3 Purification of N-truncated ADP-Glc PPase mutants ................ 72
3.3.4 Purification of co-expressed EcNAl 5— CAlO8+ Ec32843, ,,,,,,,,, 72
3.3.5 Protein concentration assays ...................................................... 73
3.3.6 Protein electrophoresis and immunoblotting ............................. 73
3.3.7 Enzymatic activity assays .......................................................... 73
3.3.8 Kinetic characterization ............................................................. 74

3.4 Results and discussion .............................................................................. 75
3.4.1 Purification and characterization of E. coli ADP-Glc PPase N-

terminal truncated forms ............................................................ 75
3.4.2 Characterization of EcNAlS— CA108+ Ec32843, ,,,,,,,,,,,,,,,,,,,,,,, 31

Chapter 4

Molecular architecture of the glucose l-phosphate site in ADP-glucose

pyrophosphorylases ........................................................................................ 86-129

4.1 Abstract ..................................................................................................... 87

4.2 Introduction .............................................................................................. 88

4.3 Materials and methods .............................................................................. 90
4.3.1 Materials .................................................................................... 90
4.3.2 Homology modeling .................................................................. 90
4.3.3 Multiple sequence alignment ..................................................... 91
4.3.4 Site directed mutagenesis .......................................................... 91
4.3.5 Bacterial strains and expression of the recombinant ADPGlc

PPases ........................................................................................ 92
4.3.6 Purification of pMAB3-single mutants ..................................... 92
4.3.7 Protein methods ......................................................................... 93
4.3.8 Enzymatic assays ....................................................................... 95
4.3.9 Kinetic characterization ............................................................. 95
4.3. 10 Thermal stability ........................................................................ 96

4.4 Results ...................................................................................................... 97
4.4.1 Homology modeling .................................................................. 97

viii

4.4.2 Selection of residues for analysis .............................................. 101

4.4.3 Expression and purification of pMAB3-single mutants ............ 105

4.4.4 Kinetic characterization ............................................................. 105

4.4.5 Thermal stability ........................................................................ 1 13
4.5 Discussion ................................................................................................. 115
Chapter 5
Conclusions ...................................................................................................... 130-1 56
5.1 Structural and fimctional role of the N- and C-termini in allosteric

regulation .................................................................................................. 1 3 3
5.2 Structural basis of substrate binding ......................................................... 141
5.3 Structural basis of the catalytic mechanism ............................................. 150
5.4 Future directions ....................................................................................... 154
References ........................................................................................................ 1 57-1 82

ix

LIST OF TABLES

Table 1.1. Carbon metabolism and structural and regulatory properties of ADP-
Glc PPases from different ........................................................................................ 7-5

Table 2.1. Specific activities of mutant and wild-type E. coli ADP-Glc PPases 49

Table 2.2. Apparent affinity for the different effectors of the mutant Ec-insl 17
and wild-type ADP-Glc PPase ................................................................................. 57

Table 2.3. Kinetic parameters in the synthesis direction of E. coli ADP-Glc
PPase WIId type and E01323 '1’ 130323.431 form ............................................................ 61

Table 3.1. Specific activities of wild-type and N-terminal truncated E. coli ADP-
Glc PPases ................................................................................................................ 76

Table 3.2. Specific activities of wild-type and mutants E. coli ADP-Glc PPases 82
Table 3.3. Kinetic parameters of wild-type and mutants E. coli ADP-Glc PPases . 83
Table 4.1. Complementary oligonucleotides used to introduce single mutations 94

Table 4.2. Comparison of specific activities and apparent affinity for the

substrate GlclP of the E. coli wild type and mutant enzymes ................................. 106
Table 4.3. Kinetic parameters of the E. coli wild type and mutant ADP-Glc

PPases ...................................................................................................................... 110
Table 4.4. Thermal stability of the wild type and single mutants ........................... 1 l4

LIST OF FIGURES
Figure 1.1. Schematic organization of a-1,4-polyglucans structure .....................

Figure 1.2. Metabolic routes involved in the biosynthesis of (1-1,4-
polysaccharides in different systems .....................................................................

Figure 1.3. Schemes of the sequential ordered Bi Bi kinetic mechanism proposed
for ADP-Glc PPases ..............................................................................

Figure 1.4. Three-dimensional structures of NDP-sugar PPases monomers .......
Figure 2.1. E. coli ADP-Glc PPase constructs .....................................................

Figure 2.2. SDS-PAGE of E. coli wild-type ADP-Glc PPase and Ec-insl l7
mutant ....................................................................................................................

Figure 2.3. N-terrninal sequence analysis of the Ec-insl 17 mutant .....................
Figure 2.4. Sequence of the insertion 117 in the E. coli ADP-Glc PPase ............
Figure 2.5. Domain organization of the ADP-Glc PPase from E. coli .................

Figure 3.1. Summary of the E. coli ADP-Glc PPase constructs studied and their
activation properties ..............................................................................................

Figure 3.2. N-terrninal E. coli ADP-Glc PPase sequence and predicted secondary
structures ..............................................................................................

Figure 3.3. FBP-activation of wild-type and the N-terminal truncated E. coli
ADP-Glc PPases ....................................................................................................

Figure 3.4. AMP inhibition kinetics of wild-type and mutants E. coli
ADP-Glc PPases ....................................................................................................

Figure 3.5. F BP activation kinetics of wild-type and mutants E. coli
ADP-Glc PPases ....................................................................................................

Figure 4.1. Structural model of E. coli ADP-Glc PPase ......................................
Figure 4.2. E. coli ADP-Glc PPase- substrate interaction ....................................

Figure 4.3. Sequence alignment of the E. coli ADP-Glc PPase and its
homologues ...........................................................................................................

xi

12

30

34

46

53

55

56

64

71

77

79

80

84

98

102

103

Figure 4.4. Steady-state kinetic measurement for GlclP dependence for
wild type and E194A, E194D and E194Q mutant enzymes .................................

Figure 4.5. Superposition of the amino acids in the GlclP site from three
NDP-Glc PPases ....................................................................................................

Figure 4.6. Hydrogen bond network involving Ser2'2 in the E. coli ADP-Glc
PPase GlclP site ....................................................................................................

Figure 4.7. Stick representation of the SerZ'2 mutants’ modeled active sites .......
Figure 5.1. Three-dimensional structure of the potato tuber ADP-Glc PPase .....
Figure 5.2. Three-dimensional model of E. coli ADP-Glc PPase monomer ........

Figure 5.3. Overlap of the three E. coli ADP-Glc PPase models: non-ligand-,
ATP- and ADP-Glc-bound forms .........................................................................

Figure 5.4. Conformational changes in ADP—Glc PPase active site upon
ligand binding ........................................................................................................

Figure 5.5. Conformational changes experienced by Glu'94 and Lys'95 upon

ligand binding ........................................................................................................
Figure 5.6. Chemical reaction catalyzed by the ADP-Glc PPase .........................

Images in this dissertation are presented in color.

xii

107

118

121

122

135

139

I42

145

148

151

LIST OF ABBREVIATIONS

3-PGA: 3-phosphoglyceric acid

ADP-Glc PPase: adenosine 5’-diphosphoglucose pyrophosphorylase
ADP-Glc: adenosine 5’-diphosphoglucose

BSA: bovine serum albumin

CDP-Glc PPase: cytidine diphosphoglucose pyrophosphorylase
CrATP: chromium-ATP

DHAP: dihydroxyacetone phosphate

dTDP-Glc PPase: deoxythymidine diphosphoglucose pyrophosphorylase
EDTA: ethylenediaminetetraacetic acid

F6P: fructose 6-phosphate

FBP: fructose 1,6-bisphosphate

GallP: galactose l-phosphate

GlclP: glucose 1-phosphate

GIc6P: glucose 6-phosphate

IPTG: isopropyl-B-D-thiogalactoside

NDP—sugar PPase: nucleotide-diphospho-sugar pyrophosphorylases
PEP: phosphoenolpyruvate

Pi: inorganic phosphate

PLP: pyridoxal 5’-phosphate

PPiz inorganic pyrophosphate

SDS: sodium dodecyl sulfate

UDP-Glc PPase: urididine 5’-diphosphoglucose

TCA: tricarboxylic acid

xiii

CHAPTER 1

Literature review

1.1. INTRODUCTION TO POLYSACCHARIDES

Carbohydrates, proteins, lipids and nucleic acids are the four major classes of
biomolecules in nature. Carbohydrates or saccharides (Greek: sakcharon, sugar) account
for the most abundant group of compounds on earth and they play different roles, serving
as energy storage, metabolic fuels, main intermediates in the various carbon metabolism
routes, as well as structural components of the cells. These biomolecules are present in all
living organisms as mono-, di-, oligo- and polysaccharides, and their structural units have

multiple hydroxyl groups and are linked by O-glycosidic bonds (1, 2).

Monosaccharides are monomeric units of polyhydroxyaldehydes or
polyhydroxyketones. The hexoses D-glucose and D-fructose are the most important of
these simple sugars, whereas phosphorylated sugars like glucose 1-phosphate(GlclP),
glucose 6-phosphate (Glc6P), 3-phosphoglyceric acid (3-PGA) and dihydroxyacetone
phosphate (DHAP), are another class of monosaccharide essential for the cellular
metabolism. Reactive forms of monosaccharides, such as ADP-glucose (ADP-Glc) and
UDP-glucose (UDP-Glc), known as sugar nucleotides, are also needed for the formation
of 0- and N-glycosidic linkages. Disaccharides consist of two monomeric sugars joined
by an O-glycosidic bond. Sucrose (glucose-u-l ,Z-B-fructose), maltose (glucose- (1-1,4-
glucose) and lactose (galactose-B-l,4-glucose) are the three most abundant examples.
Oligosaccharides are relatively short polymers containing a small number of
monosaccharides (typically three to six). They are generally found either 0- or N-linked

to amino acid side chains and lipids and participate in molecular targeting and cell-cell

recognition, among other roles. Polysaccharides, which are also known as glycans, are
large linear or branched molecules consisting of a variable number and type of sugar
monomers linked in specific manners. Based on their chemical composition they can be
classified in homopolysaccharides, with one type of component monomeric unit, and
heteropolysaccharides, which are formed by more than one sort of monosaccharide. In
addition, polysaccharides can have either a structural role or serve as an energy source or
as carbon storage compounds. Examples of homopolysaccharides are glycogen and starch
(the most abundant cellular storage forms) and cellulose and chitin (the principal
structural components of the plant cell wall and invertebrate exoskeleton, respectively).
On the other hand, among heteropolysaccharides, hyaluronic acid, which is present in
connective tissues, is composed by units of D-glucuronic acid and N—acetyl-D-
glucosamine, whereas the bacterial cell wall component peptidoglycan (or murein)
consists of repetitive molecules of N—acetylmuramic acid and N-acetyl-D-glucosamine (1,

2).

1.2. STORAGE u-l,4-POLYSACCHARIDES: STRUCTURE AND FUNCTION

Alpha-1,4-polyglucans are one of the main strategies developed by living
organisms to store carbon and energy in a readily available manner to cope with
starvation conditions temporarily present in the environment (3 -9). The biosynthetic
process consists of the polymerization of large amounts of D-glucose surplus to form
glycans of high molecular weight. One of the principal advantages of using

polysaccharides as reserve compounds is that, due to their physiochemical properties they

produce negligible osmotic pressure, thus allowing for the accumulation of large amounts
of carbon without disturbing the water relations within the cell (3, 5, 7-10). Another
advantage is that the large amount of glycosidic linkages, present in amylopectin and
glycogen as a(1—*4) and (1(1 —>6) bonds, generates a particular structural organization
that allows for the access of degrading enzymes at the non-reducing ends and at internal
linkages (Fig. 1.1). Enzymes such as amylases and phosphorylases can act
simultaneously at different branches, thus speeding the conversion of large polymers into
single units of D-glucose derivatives (7, 11). The use of 01-1 ,4-polyglucans as reserves is
broadly seen in nature, including the starch in photosynthetic eukaryotes and the
glycogen present in bacteria, (including cyanobacteria), archea and non-photosynthetic

eukaryotes.

Glycogen is an essentially homogeneous water-soluble polymer of high molecular
weight (107-109) containing an average of 10-14 a(1-—>4)-Iinked D-glucose units,
interlinked by (1(1 —->6)-glycosidic linkages, to form a highly branched compound (Fig.
1.1.A) (5, 12, 13). The molecule has a spherical shape, with the chains organized in
concentric tiers (Fig. 118) (14). The structure of the bacterial glycogen has been less
studied than that of the mammalian one, but the available data point out the general

similarity between them (13).

Figure 1.1. Schematic organization of a-l,4-polyglucans structure. Drawing
of glycogen showing: (A) the homogeneous structure and (B) its spherical shape
formed by concentric tiers (marked by dashed lines) (13). The branches in the
most external tiers are the most readily available for hydrolysis. Starch
accumulates in granules (C) as a result of the growing helices of amylopectin (D)
which determine regions of crystalline and amorphous starch (E), viewed as
concentric rings (M. F. Chaplin, Starch, [http://www.Isbu.ac.uk/water/hysta.html]
adapted from T. P Coultate, Food: The Chemistry of its Components, 4th Ed.
Accessed on 13 Nov 2006).

0:1 ——>6
branchpoints

Reducing
end

,1».—

  
     
 
 
 

2. E3 51
6 glucose

units
K \ ' I .

6 1"“ \‘i. airs-2

   

‘\
‘\ 0-1.6 ,
- branchpoints ; _.-*

‘\. /

- Hilium

K

/

 

\ Amorphous
regions

Crystalline
regions

In mammals, the major deposits of glycogen are in liver and skeletal muscle but
many cell types are capable of glycogen synthesis and limited accumulation, like cardiac
and smooth muscle, the kidney, the brain and even adipose tissue (1 1). Its accumulation
and utilization are hormonally linked to the nutritional status. Glycogen synthesis in the
liver occurs after food intake as a consequence of the increased glucose level in blood
which stimulates the secretion of the pancreatic hormone insulin, whereas decreased
levels of glucose in the blood stream stimulate secretion of another pancreatic hormone,
glucagon, which promotes the breakdown of the hepatic glycogen and its conversion to
glucose by glucogenolysis (I, 2). Therefore, the synthesis and degradation of this
polysaccharide in liver determine the blood glucose homeostasis which is important for

providing the different tissues with energy supply.

In bacteria, the precise role that glycogen may play is still unclear; however its
presence is hypothesized to preserve cellular components susceptible to turnover in
stationary phase for production of energy during starvation (15). These microorganisms
usually accumulate glycogen as a result of limited growth conditions in the presence of
an excess source of carbon, generally when they reach the stationary phase of growth
(16). Exceptions to this are Streptococcus mitts (1 7), Mycobacterium smegmatis (I8) and
Rhodobacter capsulata (19) that have been reported to do it during the exponential phase
of growth, and Bacillus subtilis (20) and Streptomyces coelicolor (21) that accumulate it
prior to the onset or early in sporulation to supply the necessary resources for
differentiation. In Salmonella enteriditis, glycogen storage was described as related to

biofilm formation and virulence of this strain (22), whereas in Streptococcus mutans

accumulation of the polysaccharide was shown to be associated with cariogenic capacity

of the bacterium (23).

Higher plants and algae accumulate starch (9). It is composed of two
a-1,4-polyglucans: amylose and amylopectin. Amylose is referred to as a smaller (about
1500 D-glucose residues), mainly linear a(1—->4)-linked polysaccharide. Amylopectin is
the major component of starch and, like glycogen, is an a(l—+6)-branched a(1—*4)-
polyglucan (Fig. 1.1.D), although the degree of branching is only 5-6% (compared to the
glycogen 8-12%) (7, 24). Starch is organized in concentric growing rings around a central
hilum to form large granules that range from 0.1 um to 150 um in diameter, depending
on the species (Fig. 1.1.C) (7, [2). Starch granules are microcrystalline structures of
complex organization, with different polymorphisms depending on the source (7, 24-26).
Regions of amorphous and crystalline starch that are formed by the alignment of helical
external chains of the amylopectin molecule alternate in the granule (Fig. 1.1.E). Starch
granules are found in almost all green plants in various types of plant tissues and organs
(leaves, roots, shoots, fruits, grains, and stems) (7). Accumulation of starch in leaves
occurs in the chloroplast due to carbon fixation during photosynthesis, a phenomenon
that was demonstrated by Sachs in the XIX century (27). Exposure to low light or to
extended periods of dark (24-48 h) causes disappearance of starch from photosynthetic
tissues, as demonstrated by iodine staining (28) and light or electron microscopy (29).
Starch is degraded to products used mainly for sucrose synthesis, which serves as carbon
supply for sink tissues. In stomatal guard cells, however, starch is synthesized in dark

periods and degraded during the day while stomata are open (3 0-32). In cereals, starch is

mainly synthesized and accumulated within the amyloplasts present in the endosperm (7).
Deposits of this polysaccharide in potato tuber, maize and cassava, yarn and sweet potato
roots can range between 65% and 90% of their dry matter. In storage organs, such as
fruits or seeds, accumulation of starch occurs during development or maturation of the
tissue (7); whereas degradation takes place at the time of fruit ripening and seed or tuber

germination, generating metabolites used as sources of carbon and energy.

1.3. ELUCIDATION OF THE a-l,4-POLYGLUCAN METABOLIC ROUTES:

HISTORICAL BACKGROUND

By the late 19305, it was assumed that the synthesis and degradation of the
mammalian glycogen proceed by a common metabolic route. This was based on
observations made by Carl and Gerti Cori who obtained the first in vitro synthesis of a
polyglucan from GlclP by the glycogen phosphorylase, the enzyme responsible for the
mammalian glycogen degradation (33). It was only one decade later that a different
biosynthetic pathway was postulated after observation of large deposits of glycogen in
the skeletal muscle of patients with the “Me Ardle’s disease”, caused by a glycogen
phosphorylase B deficiency (34). Those patients were able to synthesize glycogen but not
to degrade it, leading to the conclusion that in vivo the synthesis and degradation of the
polyglucan occur through separate metabolic routes. What the Coris had observed in vitro

was merely the reversible reaction of the glycogen phosphorylase.

The elucidation of the metabolic route for polysaccharide accumulation begun in
the late 19403, when L.F. Leloir and his collaborators in Buenos Aires, Argentina,
discovered that the UDP-Glc had another role besides being a cofactor for the
transformation of the galactose l-phoshate (GallP) into GlclP (35-3 7). Cellular
fractions from several prokaryotic and eukaryotic sources were reported as able to
catalyze glycosidic transferences from UDP-Glc, and other sugar nucleotides that work as
activated forms of glucose, to their respective acceptors. First, the incorporation of
UDP-Glc in starch from plant extracts was reported (38); however, later studies on
specificity for synthetic sugar-nucleotides showed that ADP-Glc was a better substrate
for the synthesis of starch in higher plants (39). These results led to the isolation of ADP-
Glc from maize extracts (40) and to the first identification of an enzyme involved in

ADP-Glc synthesis, the ADP-glucose pyrophosphorylase (41).

Studies performed with mammalian liver extracts clearly determined that the
glucosyl donor for glycogen synthesis was UDP-Glc (42). On the other hand, the
metabolic reactions involved in the synthesis of bacterial glycogen were elucidated after
Sigal and collaborators reported UDP-Glc synthesis deficient Escherichia coli strains that
accumulated normal glycogen levels (43). The finding of large amounts of ADP-Glo-
specific glycogen synthase and ADP-Glc pyrophosphorylase in those microorganisms
ruled out that UDP-Glc was an important precursor of glycogen synthesis in
enterobacteria and strongly suggested that ADP-Glc was the main glucosyl donor for

polysaccharide biosynthesis in bacteria and plants (7).

10

1.4. METABOLIC PATHWAYS FOR a-l,4-POLYSACCHARIDES SYNTHESIS

IN BACTERIA AND PLANTS

The synthesis of the (it-1,4 linked, (1-1,6 branched polysaccharides can be divided
in the general following steps: [1] the synthesis of the glucosyl donor molecule fiom the
sugar-phosphate, [2] the transference of a single glucosyl moiety from the sugar
nucleotide to the glycan synthesis primer, and [3] the polysaccharide structure
remodeling. The biosynthesis of bacterial glycogen and plant starch occurs by utilizing

ADP-Glc as the glucosyl donor. The reactions taking place are as follows:

[1] ATP + a-GlclP : ADP-Glc + PP:

[2] ADP-Glc + a-l,4-glucan : 01-1,4-glucosyl-a-l,4-glucan + ADP

[3] Elongated a-l,4-glucosyl-a-1,4-glucan \—_.-> a-1,4-o-1,6 branched glucan

The enzymes catalyzing the different steps, [1] ADP-glucose pyrophosphorylase,
[2] glycogen and starch synthase and [3] branching enzymes, are described in sections
1.4.1, 1.4.2 and 1.4.3 and a more complete analysis of the biochemical characterization of
the ADP-glucose pyrophosphorylases is presented in section 1.6. A comparison between
the biosynthesis of a-l,4-polysaccharides in bacteria, plants and mammals is depicted in

Fig. 1.2

11

Figure 1.2. Metabolic routes involved in the biosynthesis of a-l,4-
polysaccharides in different systems. ADP-Glc PPase and UDP—Glc PPase
catalyze reaction [1], glycogen and starch synthases catalyze reaction [2] and
branching enzyme catalyze reaction [3], described in section 1.4. Bacterial
glycogen and starch utilize ADP-Glc as the activated form of glucose. (A) Rn:
Reaction. (B) De novo reaction is: UDP-Glc + apo-glycogenin :
(glycosyl)glycogenin + UDP; further glucose-residues incorporation occurs as
depicted in reaction [2] in the diagram. (C) In amylopectin synthesis. Enzymes in
the red boxes catalyze the rate limiting steps in the respective pathways and are

tightly regulated.

12

Rn (A)

[1]

[2]

I31

 

 

 

 

 

 

 

 

 

 

 

 

 

Bacterial Starch Mammalian
glycogen glycogen
G|c1P GIc1P GIc1P
/ ATP / ATP / UTP
ADP-Glc / ADP-Glc ( UDP-Glc /
PPase PPase PPase
\ 13131 K 13131 K PP]
ADP-Glc ADP-Glc UDP-Glc
o—1 ,4— o-1 ,4- o-1 ,4-
glucan glucan (B) glucan
Glycogen Starch Glycogen
synthase \ synthase synthase
I ADP (\ADP UDF’

 

q-1 ,4-glucosyl-
o-1,4-glucan

Branchin
9 Enzyme

o-1,4-d-1 ,6
branched glucan

 

q-1 ,4-glucosyl-
o-1,4-g|ucan
(C)

Branching
Enzyme

q-1 ,4-o-1,6
branched glucan

o—1 ,4-glucosyl-
o-1,4-glucan

Branching
Enzyme

a-1,4-a-1,6 branched
glucan

1.4.1. ADP-Glc pyrophosphorylase

Bacterial glycogen and starch biosynthesis route share a common first reaction
(Reaction [1]) catalyzed by the ADP-Glc pyrophosphorylase (ATP: a-D-glucose 1-
phosphate adenylyltransferase, EC. 2.7.7.27; ADP-Glc PPase). This enzyme is coded by
a single gene in most bacteria except those from the genus Bacillus (glgC in E. coli). But
in plants, at least two genes are found (8, 9). The reaction was first reported by Espada in
1962 (41) and, subsequently, Preiss and collaborators described its requirement of a
divalent metal ion, Mg2+ or Mn”, as well as its in vitro reversibility, with equilibrium
close to 1 (44). In vivo, however, the presence of inorganic pyrophosphatases in the cell
and the utilization of ADP-Glc for polysaccharide synthesis drive the equilibrium
towards the synthesis of the sugar-nucleotide (3). In addition, the ADP-Glc PPases from
many systems were found to be highly regulated by allosteric metabolic intermediates (4,

7-9, 16, 45-47).

The physiological role of ADP-Glc PPase as the regulatory enzyme in the
bacterial glycogen and plant starch biosynthetic pathways is supported by solid
experimental evidence. Mutants of E. coli and Salmonella enterica serovar Typhimurium
with affected capacity of glycogen accumulation after chemical mutagenesis, are the
basis of the previous statement. These mutant strains expressed ADP-Glc PPases with
altered regulatory properties and the point mutations found through gene sequencing
were the subject of further biochemical characterization. For instance, the E. coli B

mutants that accumulated glycogen at a faster rate than their parent strain expressed a

14

mutated ADP-Glc PPase with higher apparent affinity for the activator fi'uctose 1,6-
bisphosphate (FBP) and lower affinity for the inhibitor, AMP (48-50). On the other hand,
E. coli-glycogen deficient mutants, that synthesized glycogen at 40% the rate of wild

type, had a lower affinity for the activator, FBP (51, 52).

A similar situation was observed in oxygenic photosynthetic organisms, where
the main activator and inhibitor are 3-PGA and inorganic orthophosphate (Pi). Starch
deficient mutants of the unicellular algae, C hlamydomonas reinhardtdii, expressed an
ADP-Glc PPase insensitive to 3-PGA activation and Pi inhibition (53). Work with
Arabidopsis (54-57), maize (58), wheat (59) and potato (60) mutants also supported the
key regulatory role of ADP-Glc PPase in starch biosynthesis. More recently, metabolic
control analysis in potato tuber showed that the enzyme catalyzes a near rate-limiting step
in the pathway of starch synthesis (61). Therefore, because of the demonstrated
regulatory role of ADP-Glc PPase in starch biosynthesis in vivo, several studies have
been aimed to increase the polysaccharide yield and overall plant productivity by genetic

manipulation of this enzyme’s activity (58-60, 62, 63).

Overall, the finding that regulation in bacteria and plants occurs at the ADP-Glc

synthesis step agrees with the concept that effective regulation of a biosynthetic pathway

occurs at its first unique step.

15

In section 1.6., the particular regulatory properties of the ADP-Glc PPases from
different sources and information on their functional and structural characterization

known to date are presented.

1.4.2. Bacterial glycogen / starch synthase

The second step in the synthesis of bacterial glycogen and plant starch (Reaction
[2]) consists in the transfer of the activated glucose to the non-reducing end of a growing
a-1,4-Iinked glucan. Bacterial glycogen synthases (ADP-Glcza-1,4-glucan, 4-
glucosyltransferase; EC. 2.4.1.21; GS) and plant starch synthases (ADP-Glc:u-l,4-
glucan, 4-glucosyltransferase; EC. 2.4.1.21; SS) catalyze this chemical reaction. In

bacteria (e.g., E. coli) only one GS and GS gene (glgA) have been found (64).

Both GS and SS are non-regulated enzymes that use exclusively ADP-Glc as a
glucosyl donor, have molecular masses of 48-55 kDa and organize in dimers in the native
form (65-68). The sequence identity shared by the bacterial GS and the plant and algae
SS is between 30 and 36%, and it has been suggested that they could have a common
three-dimensional fold and catalytic mechanism. Both are glycosyltransferases that
operate retaining the anomeric configuration of the glucose transferred from ADP-Glc to

the non-reducing end of glycogen and amylose or amylopectin, respectively.

Secondary structure predictions and threading techniques postulated that GS and

SS have a glycosyltransferase-type B (GT-B) fold consisting of two distinctive

l6

Rossmann-like-B-u-B domains separated by a big cleft (69, 70) and in more recent reports
the E. coli GS three-dimensional structure was modeled from other related GT-B
glycosyltransferases. The model obtained was in its active conformation (“closed” form)
and a structure-guided site directed mutagenesis of conserved residues present in the
active site, which is formed in the cleft with amino acids belonging to both domains, and
a complete biochemical characterization of the mutants validated it (71, 72). Later, the
reported crystal structures of the A. tumefaciens GS (73) and, more recently, that of the
archaeal GS from Pyrococcus abyssi (74), agreed with the previous results, although the
structures obtained by crystallography were in the “open” inactive state which
complicated the identification of the studied amino acids. For a number of enzymes of the
GT-B superfamily it has been proposed that the inactive state undergoes a substrate-
triggered closing of the inter-domain cleft, thereby bringing together the catalytic

residues that make up a competent active center.

In higher plant tissues, in contrast to the bacterial GS, more than one SS gene has
been reported (75- 78). Two groups of starch synthases have been described: one is that
bound to the starch granule (granule bound starch synthase: GBSS) which can only be
solubilized by a-amylase digestion of the granule; the second one is found in the soluble

fraction of the plant extracts and, therefore called, soluble starch synthases (SSS) (7).

Work on several plant systems has shown that multiple forms of SSS and GBSS
are present, and the subject has been extensively reviewed by Sivak and Preiss (7). For

example, two major types of SSS (SSS I and SSS II) have been described and

17

characterized in maize (79). These two forms seemed to be distinct on the basis of their
physical, kinetic and immunological properties, and thus proposed to be products of two
different genes with different roles in starch granule formation. To date, SSS mutant or
SSS-knockout plants have been reported to display structural modifications of
amylopectin associated with increases in amylose content, suggesting a role for SSS in
amylopectin biosynthesis (80-83). In addition, two types of GBSS have been described in
maize: GBSS I and GBSS II (77). RNA antisense experiments in potato led to the
suggestion that these enzymes would be related to the synthesis of amylose fraction of
starch (84), although the precise molecular mechanism is still under debate (85). On the
other hand, mutation in the gene coding for the GBSS (waxy) in C. reinhardtii had been

suggested to has a role in amylopectin synthesis (86).

Molecular characterization of maize SSS forms have identified important residues
for catalysis and ADP-Glc binding (87-89). More recent reports describe the
identification of critical residues determining the differential activity of SSS isoforms in
rice plants (90). The varied activity of these SSS isoforms induced structural changes in
amylopectin determining some of the gelatinization and pasting properties of starch
granules (90, 91). The identified residues are close to a conserved motif first recognized
as critical for enzymatic activity in the E. coli GS by Yep and collaborators (72) and later
observed in the A. tumefaciens GS crystalline structure (73). Despite these findings,
studies aimed to determine the relationship between structure and function of the multiple

isoforms of SS still lag behind the other enzymes involved in starch biosynthesis.

18

1.4.3. Branching enzyme

The last step in the biosynthesis of glycogen and amylopectin (Reaction [3]) is
catalyzed by the branching enzyme (EC. 2.4.1.18; a-1,4-a-D-glucan: a-1,4-a-D-glucan
6-glycosyl-transferase; BE), and is responsible for the synthesis of the 01-1 ,6 linkages
seen in these polysaccharides. BE and starch BE (SBE) catalyze two reactions in
synthesizing a-I,6 glucosidic linkages: one is the cleavage of an a-(1—4) glucosidic
linkage and the next one is the transfer of the cleaved oligosaccharide to form a new a-
(1—6) branch (92). As observed with the previous enzymes in the pathway, in E. coli and
other bacteria only one BE and only one gene (glgB in E. coli) are present (20, 93-99) but

more than two genes have been described in higher plants (7).

Bacterial glycogen and amylopectin present different patterns of branching: the
first has shorter branches (10-13 glucose residues) and the a-I,6 linkages represent 10%
of the total glucosidic linkages, whereas the second polysaccharide is formed by longer
branches (20-24 residues) with only 5% of 01-1 ,6 linkages (7). It is thus expected that the
corresponding enzymes present different specificity with respect to the branch transfer, or
that the type of interaction between the BE and the GS in bacteria is different from that

between the SS and the SBE.

While amylose synthesis requires only an active GBSS (introduced in the section
1.4.2.), amylopectin is synthesized by a complex group of enzymes involving, among

others, several isoforms of SS, SBE, and starch-debranching enzymes [reviewed in ( 12)].

19

Potato contains two isoforms of SBE, SBEI and SBEII, and suppression of both isoforms
is required to generate starch with amylose content of >50% and a complete lack of
normal amylopectin (100). In maize, one SBEI and two forms of SBEII, SBEIIa and
SBEIIb, are found. Suppression of SBEIIb results in the "amylose-extender" phenotype,
with amylose contents from 50% to 90% (101), whereas suppression of SBEI or SBEIIa

has no impact on endosperm amylose content (102, 103).

The relationship in amino acid sequences between that of BE and amylolytic
enzymes in the "or-amylase family" of enzymes have been compared (7, 104). This family
includes (it-amylase (EC. 3.2.1.1), pullulanase (EC. 3.2.1.41), isoamylase (EC. 3.2.1.68),
neopullulanase (E.C. 3.2.1.135), cyclodextrin glucanotransferase (EC. 2.4.1.19). Among
these enzymes four specific regions within putative catalytic sites are conserved, and they
are also recognized in the SBE and the BE known sequences. The role of the amino acids
located in these regions (such as Asp405, Glum, and Asp526 in the E. coli BE numbering)

has been probed by site-specific mutagenesis (98, 105).

In addition, a structural conservation in the tit-amylase family has been proposed
by the prediction the (B-a)8-barrel structural domain in the various groups of enzymes in
the family (106). The recently reported three-dimensional structure of the E. coli BB
(107) supports this view. However, despite the conservation of this central (B-a)8-barrel,
the C-terrninus and N-terminus portions of BE from various bacteria as well as from
higher plants are dissimilar in sequence and in size (7). Chimeric constructs made from

different isoforms of the maize endosperm SBE with different specificity and pattern of

20

branching of the substrate assigned a role to the N-terrninal region in determining the size
of oligosaccharide chain transferred (108). Indeed, a progressive shortening of the
N-terminus led to a gradual increase in the length of the transferred chains, suggesting
that this domain provides a support for the glucan substrate during the processes of

cleavage and transfer of the a-( 1—4) glucan chains (109-111).

To date, several types of SBE have been reported to be useful for applications
such as changing the solubility of starch (112-114) or producing large cyclic glucans (99,
115). Therefore, understanding how BE and SBE work through their molecular
identification and characterization has also been important for the starch-processing

industry.

1.5. Alpha-1,4-POLYSACCHARIDES IN NON-PHOTOSYNTHETIC

EUKARYOTES

Several studies performed in mammalian cells determined that the glucosyl donor
obtained in the first reaction of the pathway (Reaction [1]) is UDP-Glc (116) (whereas in
bacterial and photosynthetic eukaryotes is ADP-Glc). The reaction is catalyzed by the
UDP-Glc PPase (EC. 2.7.7.9). Also different is that de novo synthesis of glycogen is
initiated by glycogenin (E.C. 2.4.1.186) (11 7, 118). This protein is self-glycosylated in a

reaction as follows:

UDP-Glc + apo-glycogenin —. (glycosyl)glycogenin + UDP

21

Following this reaction, a covalently bound oligosaccharide is formed by addition of
glucose monomers linked by a-1,4-linkages that serves as primer for the elongation

catalyzed by the eukaryotic GS (EC. 2.4.1.11) (Reaction [2]).

In contrast to what occurs in bacteria, the synthesis of mammalian glycogen
utilizes UDP-Glc and the production of this sugar nucleotide is not a regulated step. In
the mammalian pathway, the GS reaction is rate limiting (1, 2, 116). The mammalian GS
is regulated by allosterism and by phosphorylation/ dephosphorylation events, catalyzed
by protein kinases and phosphatases. On the other hand, the glycogen phosphorylase
(EC. 2.4.4.1) is also regulated by opposite phosphorylation/ dephosphorylation cycles, to
guarantee that they would not be active at the same time as the glycogen synthases, thus
avoiding futile cycles (1, 2). The balance between glycogen synthesis and degradation is
intimately regulated by insulin and glucagon hormone levels. The hormone epinephrine
has the same effect as glucagon but acts in muscular tissue rather than in liver, as

glucagon does (1, 2).

1.6. ADP-GLUCOSE PYROPHOSPHORYLASE: THE REGULATORY ENZYME

IN BACTERIAL AND PLANT a—l,4-POLYSACCHARIDES BIOSYNTHESIS

1.6.1. Quaternary structure

ADP-Glc PPase catalyzes the reaction converting GlclP and ATP into ADP-Glc

and PPi (41) in presence of a divalent cation (physiologically, Mg”) (44). The native

22

enzyme consists of four subunits, being, depending on the source, a homotetrarner ((14) or

a heterotetramer ((12132).

Most of the bacterial enzymes described to date, except for those from the
Bacillus genus, are homotetramers of 200 kDa and products of the expression of a single

gene (Table 1.1) (3, 4, 7-9, 20, 47, 119-122).

In contrast, higher plants and green algae ADP-Glc PPases described to date are
heterotetramers, composed by two types of subunits that differ in their molecular masses
and amino acid sequence (3, 4, 6-9, 20, 123, 124). Although in some cases the difference
in molecular masses between them is not more than 1 kDa (123, 125), these subunits
were called “small” (a subunit, 50-54 kDa) and “large” (B subunit, 51-60 kDa) subunits
for convenience. The small subunit of higher plants is highly conserved (85-90%
identity) and has a catalytic role, whereas the large subunit seemed to have diverged more
(they present 50-60% identity) and have a modulatory role without catalytic function
(126-128). Moreover, the most frequent situation in higher plants is the existence of one
functional small subunit gene and several large subunit genes that are differentially
expressed (7, 128-136). As reported in Arabidopsis (136, 13 7), the several large subunits
are differentially expressed in the plant and their particular association with the one
catalytic small subunit creates four types of heterotetramers with differential kinetic and
regulatory properties according to the necessities for starch synthesis. The similarity
between small and large subunits (~50% identity) suggests a common ancestry (128), and

it was recently shown that both subunits have most likely derived from a common

23

catalytic ancestor, with the large subunit maintaining the architecture of the active site,
including the substrates binding sites but exhibiting inability to perform catalysis due to

mutations of few essential residues (127).

1.6.2. Regulation

Most of the studied ADP-Glc PPases are allosterically regulated enzymes (8, 9).
Allosteric modulators of the enzyme so far characterized differ from source to source but
they share the common feature of being the key intermediates of the major pathway for
carbon assimilation/utilization in that given cell. The activators are markers or signals of
carbon and energy excess and the inhibitors represent indicators of low metabolic energy
levels. Therefore, these regulatory characteristics, in addition to the fact that ATP is one
of the substrates, agree with ADP-Glc PPase having maximal activity for storage

polysaccharide synthesis when cellular energy and carbon are in excess.

Specificity for metabolites behaving as allosteric regulators served as the basis for
the classification of the ADP-Glc PPases in different groups or “classes”, which has been
updated as novel enzymes were being characterized (3, 6-9, I 6, 138). Table 1.1. shows
the most recent classification reported (8, 9) including the principal pathway of carbon
assimilation in the respective cells, allosteric activators and inhibitors, and the quaternary

structure of the enzyme.

24

 

 

 

 

 

 

 

 

 

 

 

 

 

8.98 Eon 0:35.330 0.5262 32.8 Ezztimonosz
ocoz ofi>E§ _> 05 can Eon 3.3332; 05 9:553 553.... Eattinonoam

“$345.53 SNoDohosk

533.63% .meoucehosx

Q3 .5; @3358 tecovson— .nEEsohcoNM xmcoeconosx
65580502 5 .m2< .43“— .8m>E.E > -bficm 05 new max—8.2m wENES 682.23% useoucohetx

.5253: EEeeEEemesk

.3359... 55338 xmeoeco‘eexx
45¢. £2< .3... 6:33.35 2 through—-555 05 @2553 .5385; 53356.55

ESSEBSSE uEmwARore MEESKBR

QB .8353 Sheen...» xmuoanoxfiw
BEESSEo: E in? 52¢. n3..— .2a>E.E >_ bococsoflioficm o5 m5~=53 wastage...» EESBREMV

Eutuxxzmenon SEEKEEU

$3 $6.28wa .8353 BBSE.‘ xmeuucououtm
BEnboaoEo: n=2< 252 E ~050on -coEEm 05 9.353 €886.63 custom

ESEMsEu SateeouQeoAE

36.28.33 .3353 5.2.: 38.58665.
n5< ..=2< .5 mm“. : c2832 -535 2: BEE: easesexsgsosw

nmcuwebe b3eoob=$

.EEEEEPC.

A43 €330.33 .3358 .3328 318.3 SESSEW

58858050: 32¢. Lmn— _ .350on -coBEm o5 wank—53 So... eEotssomm
9.5%..8332
88.08.05
Baa—Em .CmEBmso 9:952:— Amrougsoa‘ awn—O

 

 

282: B otowmo=<

 

 

 

Omani o_O-mD<

 

Exam—882: :onau

 

EmEmwO

 

 

G .3 mEmEamuo 265me 805 38.5 smudge. mo 35.833 5825?: use .8335 55 85882: 5980 .—.— 03:.

25

 

 

 

 

 

 

 

 

 

 

N N Soto {8535
A n .3 . ._ 35>“: tom 25 <Om-m 28:3 BB .223
SEmbBEBo: mm; mo< .m 582% 252 v: .oNEE .8 shame—Em
ANQNUV 833.: 50553923 :85
.oEmbSobuo: E <Om-m :5 52.595 9883 9.315822 EBB 88o;
2.8 35.95280: 8335 05055323982
out .oNEE KERQEQEV
ANQNS 293
5285822»: 5 <Om.m _:> Elmo 05 £965 ~00 mExE .823 dog—Em mo 8384
moan: 05055382;
353 5&5
...:Eui..$.. §:©E%AES&U
Awngv 20.3 £333 33339
5522229: a <92 :5 =26 2: $85 .8 9.2... 833.826
cam—a 520
mouobuxam
=25... Baa—.2533.
om: 00n— 333335..
Ave 293 mono DO...— m..§0§um§.m.
552.2020: 5 <93 E> 5:8 2: 5:85 Noo 9.3m 88 on: 388525.
£855~mol~o<m
9.5968 MEESXSK 2.5830530
Amwmdv 5525.53 war—=5 20.3 w=-..&mofixmfiexufin Samoan
SEEHBEBo: 252. 252 :> Box 238553;. 25 wag—u: 3.35 Samoan
958:5 .CMEfimzo $38535 flmro->co< $20
282: 2 0:03.22 EmzonSoE 5930 E3890

 

 
 

 

 

 

 

 

08.; BOA—ado.

 

 

 

 

855.80 .2 2.3.

26

Enzymes organized in class I are those from enteric bacteria, such as E. coli and
Salmonella enterica serovar Thyphimurium. These microorganisms assimilate glucose
via glycolysis (the Embden-Meyerhof pathway) which is regulated at the site of FBP (the
phosphofructokinase step) (1, 2), and this metabolic intermediate is the main activator of
the ADP-Glc PPase (44, 139). These enzymes are inhibited by AMP and the native forms
are ~200 kDa homotetramers, products of a single gene (5, 8, 9). ADP-Glc PPases from
other microorganisms performing glycolysis have been grouped in classes 11 or 111,
depending on whether they are activated by F BP and F 6P, or if they are insensitive to any
activator, respectively. Both groups of enzymes are inhibited by AMP and class II is also

down-regulated by ADP (Table 1.1).

Class I V includes those bacteria catabolizing glucose to pyruvate by the Entner-
Doudoroff pathway, such as A. tumefaciens. The ADP-Glc PPase in these organisms is
sensitive to allosteric activation by, mainly, pyruvate and also F 6P, and to inhibition by
AMP and ADP (Table 1.1.) (19, 122). Interestingly, ADP-Glc PPases from organisms
that can assimilate carbon by both the Embden-Meyerhoff and the Entner-Doudoroff
pathways are regulated by the three main activators: Pyruvate, F BP and F 6P, depending
on the physiologic conditions (Table 1.1). They are also inhibited by AMP and Pi, form a
homotetramer which is the product of the expression of a single gene and have been

grouped in class V (121, 140).

27

ADP-Glc PPases in class VI are from anaerobic bacteria, like Rhadospirillum
rubrum and R. tenue, which are not able to catabolize glucose but grow on pyruvate,
lactate or C02. Pyruvate is the product of C02 fixation and it is the sole activator of the
R. rubrum and R. tenue ADP-Glc PPases (141, I 42). Grouped as class VII are the
ADP-Glc PPases fi'om sporulating bacilli (Table 1.1). These microorganisms synthesize
glycogen during early stage of sporulation, a process for survival to hostile environments
(120). Under the latter conditions, the tricarboxylic acids cycle (TCA) is the main
pathway for carbon utilization that fully metabolizes by-products of glycolysis (I 43). The
different recombinant enzymes from B. stearothermaphillus (99) and B. subtilis (20)
were insensitive to regulation by different metabolites typically affecting the activity of
other bacterial ADPGlc PPases. It has been shown that in these organisms the native
ADP-Glc PPase exhibits a heterotetrameric structure of the type (1262 (Table 1.1) product

of the expression of two different genes.

Class VIII and IX group ADP-Glc PPases from cyanobacteria, alga and higher
plants (Table 1.1). These organisms photoassimilate atmospheric C02 utilizing the
reductive pentose phosphate pathway that gives 3-PGA as the first intermediate product.
Under light conditions, P; is used to regenerate ATP through photophosphorylation (7-9,
144). ADP-Glc PPases of class VIII (from cyanobacteria and photosynthetic tissues in
higher plants and algae) are allosterically regulated by the ratio of 3-PGA (activating) and
Pi (inhibiting) (3, 145, 146). In non-photosynthetic tissues there are two types of ADP-
Glc PPases. One is activated by 3-PGA and inhibited by P, (14 7, 148). The prokaryotic

enzymes in this group are coded by a single gene, giving rise to a homotetramer, whereas

28

those from algae and higher plants, for instance the potato tuber enzyme, (from classes
VIII and IX) are the expression products of two genes that originate a native
heterotetramer (Table 1.1). A second type of ADP-Glc PPase present in non-
photosynthetic tissues shows distinctive regulatory properties and for this reason they
have been grouped in a different class (IX , Table 1.1). These enzymes revealed lower
sensitivity to activators (133, 149—153). The activity of the class IX enzyme from wheat
cannot be increased by allosteric activators, but 3-PGA and F6P can reverse the

inhibition caused by P,, ADP or FBP (149).

In addition, the enzymes from potato tuber and Arabidapsis leaves were shown to
be regulated by a redox mechanism in vitro and in vivo (14 7, 154-156). It consists on the
formation of an inhibitory disulfide bond between the two small subunits, involving an
N-terrninal cysteine (154, 15 7), which is not conserved in bacterial enzymes or in other

plant ADP-Glc PPases.

1.6.3. Kinetic mechanism of action

The chemical reaction catalyzed by ADP-Glc PPase proceeds following an
ordered sequential kinetic mechanism. It was shown first with the enzyme from
R. rubrum enzyme by Paule and Preiss in the early 19703 (158). Later binding studies
with the E. coli enzyme ([59), supported the model showing that ATP and ADP-Glc bind
to free forms of the enzyme (Fig. 1.2.A). These studies also pointed out the cooperative

properties of this enzyme and the heterotropic interactions between substrates. For

29

.2535 2: mo match 00¢ 23 9 come. m use K E38..— .82v 823— bonus 5 a ”an: :8 .m E can Amen: 83.5.2
.Q E A< dome?— ofiYA—a .5. coax—PE Sana—.85 032.2 a 5 1986.5 335:?» 2: .3 852.5 .Q— «.5»:—

 

 

m In. vain—(um Ao_0-n_o<._n_m-m I Qwo_0-n_._.<.wv n_._.<.m m
o_0-n_n_< in. Q 370 a._.<
.m—
m. o_O-n_n_<.m Ao_0-n_o<._n_n_-w I Q..0_O-n_._.<.mv n_._.<.m m
o_O-n_n_< in. Q P26 n.._.<

3O

instance, they observed that ATP displayed half-sites occupancy in the absence of GlclP,
but full occupancy when this second substrate was present. Also, Glcl P was not able to
bind the enzyme unless ATP and Mg2+ were present (159). Later, initial velocity kinetic
studies performed on barley leaf ADP-Glc PPase (160) also agreed with the sequential
ordered mechanism, but with ATP and ADP-Glc binding to different forms of free
enzyme (Fig. 1.2.B). This isomerization phenomenon (F H E in Fig. 12.8) may reflect
different conformations and/or aggregation states of the heterotetrameric plant protein,
not observed in the R. rubrum ADP-Glc PPase probably due to the less complex single-
subunit-type structure of the bacterial enzyme when compared to the two-subunit-type

structure of a higher plant enzyme (158).

The proposed sequential ordered Bi Bi type of mechanism also agrees with the
one postulated for other nucleotidyl-diphosphosugar pyrophosphorylases (N DP-sugar
PPases), such as the UDP-Glc PPase (161, I 62), the UDP-N—acetylglucosamine
pyrophosphorylase (GlmU) (I 63), the dTDP-Glc PPase (leA) (164, 165) and the
CDP-Glc PPase (166). Very recent data obtained by saturation transfer difference
(STD)-NMR studies on the UDP-Glc PPase from Leishmania major provided the first
direct proof for the sequential ordered bi-bi mechanism suggested in earlier studies (16]).
All these reports point out to a common mechanism of action by which the reactions
catalyzed by NDP-sugar PPases proceed. Therefore, it is possible that these enzymes
share a common ancestor and, if so, it would be expected that they also share common

structural features, like key catalytic residues or even similar structural fold.

31

1.6.4. Structure-function relationship: Background and motivation of this thesis

work

Research on the molecular and biochemical characterization of the ADP-Glc
PPases has radically changed over the last five to ten years. Before, structure-function
relationship studies provided information in one dimension, that is, in the context of the
known primary structure of the enzyme. For instance, determination of the quaternary
organization of the ADP-Glc PPases from different sources has been based on the
sequence (nucleotide or amino acid) analysis of the expressed gene products. Functional
residues, important for catalysis and regulation, have been identified by a combination of
chemical modification experiments, analysis of amino acidic sequence conservation and
site directed mutagenesis approaches. But studies aimed to investigate the relationship
between structure and function reached a new level of complexity and understanding
when new approaches designed to elucidate the organization of a protein structure in two
and three dimensions became available. First, the local secondary structures possibly
adopted by the enzyme polypeptide could be predicted (8, 9, 167). Later, the evidence of
specific interactions between certain regions in the ADP-Glc PPase allowed for drawing
hypothesis on the structural organization of the protein but in another dimension (168).
That line of work set the basis and inspired the investigation of the domain organization
of this enzyme and the closer analysis of their functional roles which are presented in

Chapter 2 and 3 of this thesis.

32

Efforts directed to obtain the three-dimensional structure of the ADP-Glc PPase
from E. coli by X-ray crystallization have been unsuccessful. During the last five years
the three-dimensional structures of related NDP-sugar PPases (GlmU (163), leA (169)
and CDP-Glc PPase (165) have been solved (Fig. 1.2.A, B, and C). This was certainly
beneficial since new and more efficient computational methods to predict protein folds
(like threading) and three-dimensional structures (such as the various approaches in
modeling: by homology, ab initio, etc) were being developed in parallel. Therefore, the
availability of structures of related enzymes and those predictive computational methods
became valuable to draw new hypotheses and design biochemical experiments to test
them. Indeed, the very recent elucidation of the first crystal structure of a plant ADP-Glc
PPase small subunit (Fig. 1.3.D), reported in 2005, was of great advantage for the
research in this field. Interesting to note when comparing the solved crystalline structures
of the PPases, is that the ADP-Glc PPase is a larger polypeptide and presents a C-
terminus, not present in the CDP-Glc PPase and the leA (Fig. 1.3.A and C,
respectively). The first ADP-Glc PPase crystalline structure provided us with a very
useful tool for understanding the function of those regions or residues that had been
already reported as important for activity or regulation. The opposite, however, did not
necessarily apply since the ability to visualize the structural organization of a protein still
left several questions unanswered, especially those regarding the functionality of specific
residues, regions or even whole domains. The research presented in Chapter 4 of this
thesis was motivated by the need of assigning a functional role, if any, to specific

residues located in the active site of the enzyme, specifically, at the GlclP site.

33

 

Figure 1.4. Three-dimensional structures of NDP-sugar PPases monomers.
CDP-Glc PPase (169) (A), and GlmU (I 63) (B) form trimers in the native
enzyme. leA (164) (C), and ADP-Glc PPase (170) (D), form tetramers. The
pyrophosphorylase domain presents a Rossmann-like-B-a-B fold, which is

conserved in the NDP-sugar PPases family. The difference is mainly in the

C-terminus.

34

The work presented in this thesis has been performed on the E. coli ADP-Glc
PPase, which has proven to be a very valuable model system to give insight into the

understanding of such a key enzyme in the biosynthesis of a-l ,4-polysaceharides.

1.6.4.1. Important residues and domains

The role of key single amino acids in ADP-Glc PPases was studied by site
directed mutagenesis after being identified by chemical modification experiments or
suggested by the conservation throughout this enzyme's family. In the E. coli enzyme,
Tyrl '4 was modified by the photoaffinity analogues of ATP and ADP-Glc, 8-azido-ATP
and 8-azido-ADP-Glc, respectively (171, 172). Also, site directed mutagenesis showed
that this residue is involved in affinity for ATP, but might be also close to the GlclP and
the activator FBP site (173). Following the same approach, it was shown that LysI95 in
the E. coli ADP-Glc PPase was involved in GlclP binding (1 74-176). Site directed
mutagenesis of the homologous residues in the small and large subunit of the potato tuber
enzyme, Lys198 and Lysm, respectively, confirmed a similar role for the homologous
residue in the small subunit, but not in the large (177), which agreed with the fact that the
large subunit has a modulatory role and is not involved in catalysis (147). More recently,
AspI42 in the E. coli enzyme, which is absolutely conserved among the ADP-Glc PPase
family, was characterized by site directed mutagenesis (I 67). The specific 10,000-fold
decreased kcat displayed by the D142A and D142N mutants clearly proved its key role in

catalysis. Similar results were obtained with the homologous Asp145 in the potato tuber

35

small subunit ADP-Glc PPase, but not with AspI60 in the large subunit, as expected due

to the lack of catalytic role of this subunit (126).

Likewise, the role of several residues in allosteric regulation was determined
[reviewed extensively in (8, 9)]. Interestingly, lysines and arginines involved in allosteric
regulation are located mainly close to the C-terminus in the plant [potato tuber (15 7) and
spinach (1 78, 1 79)] and cyanobacterial [Anabaena (3, 180-183)] ADP-Glc PPases,
whereas those characterized in the bacterial enzymes [E coli (1 75, 1 76, 184) and A.
tumefaciens (185)] are located in the N-terminus. These observations led to the
suggestion that the regulatory domains in the enzymes from enteric bacterial on one side,

and cyanobacteria and plants on the other, were located at different sites.

By the early 2000, information on the structural organization of the ADP-Glc
PPases was available only as a model of predicted secondary structures that fitted all the
enzymes from the family sequenced to date (7-9). Residues ~40-300 in the
ADP-Glc PPases (of a total of ~430-450 amino acids) aligned well with the
pyrophosphorylase domain of two related enzymes already crystallized: the GlmU and
the leA, despite the low sequence similarity between these enzymes (~14%).
Therefore, it was possible to postulate a common fold for the pyrophosphorylase domain
but yet, no information was available on the ADP-Glc PPase’s C-terminus. In 2002,
Ballicora and collaborators (168) performed a study with chimeric ADP-Glc PPases from
E. coli and A. tumefaciens determining that, even though important individual residues

involved in activator binding had been identified in their respective N-terminus (Lys39 in

36

E. coli and Arg33 and Arg45 in A. tumefaciens) the C-terminus was critical for activator
affinity and specificity. This became the first evidence of a combined role of the N- and
C-terminus in allosteric regulation (168). Whether the C-terminus was an integral part of
a unique domain of the ADP-Glc PPase monomer or a separate domain in intimate
interaction with the N-terminus, and whether it had a mere role in regulation or was
critical for proper enzyme activity were the questions that motivated the work presented

in Chapter 2.

On the other hand, the N-terminal tail (~ 40 N-terminal amino acids) had also
been involved in regulation. Apart form the amino acids characterized as involved in
binding the activators in the bacterial enzymes discussed above, up-regulated ADP-Glc
PPase mutants presenting deletions in this short stretch of the enzyme were reported (14 7,
186-188). Important also to consider is the fact that the reductive activation mechanism
regulating plant enzymes involves the formation of a disulfide-bond between Cys'2 of the
two small subunits present in the plant heterotetramer (189). Reduction of the disulfide
bond increases the affinity of the enzyme for substrates and the activator 3-PGA (154).
Work presented in Chapter 3 was performed addressing questions related to the structural

and functional nature of the N-terminus role in regulation.

1.6.4.2. Three-dimensional structure

The very recent availability of the first ADP-Glc PPase crystal structure in 2005,

that from the potato tuber small subunit homotetrameric form (1 70), shed light on the

37

monomeric structural organization of this enzyme’s family and its hypothetical
quaternary arrangement. In agreement with some of the previously obtained results,
presented in this thesis in Chapter 2, the ADP-Glc PPase monomer is organized in two
distinctive domains in close interaction. The N-terminal ~12- 300 residues adopt a similar
fold as the pyrophosphorylase domain of the already reported crystal structures of other
NDP-sugar PPases: GlmU (163), the leA (169) and the CDP-Glc PPase (165).
Observation of the potato tuber small subunit crystal structure validated all the previous
predictive studies performed on the ADP-Glc PPases (8, 9, 167), and allowed for a
visualization of the position of those individual residues already characterized. In the
potato tuber ADP-Glc PPase crystallization report (1 70), a chemical mechanism of action
was also postulated based on the orientation and interaction of side chains in the active
site with the complexed ligands (ATP or ADP-Glc) and by comparison with the other
NDP-Glc PPases, whose mechanisms had been also proposed (163-165). Despite the
large amount of structural information provided by the reported structure, several
questions concerning the functionality of specific residues in the ADP-Glc PPase
remained unanswered. For example, only AspI45 in potato tuber small subunit (and its
homologous AspI42 in E. coli) has been characterized and its essential role in catalysis
has been biochemically probed (126, 167). However, other residues postulated to be
involved in the chemical mechanism, or with other roles in the active site, have not been
studied to date. For instance, Asp280 (Asp276 in E. coli) and GluI97 (Glum in E. coli) were
proposed to have a key role in chelating the Mg2+ required in the chemical reaction and in
binding GlclP, respectively. These residues have been assigned functional roles based on

proximity to the ligands and conservation among the NDP-sugar PPases, but no

38

biochemical characterization had been performed to date. The need for a detailed analysis
of the functional role of these and other residues located in the active site formed the
basis that inspired the studies on the E. coli ADP-Glc PPase reported in Chapter 4 of the

present thesis.

In summary, the results introduced in the present thesis offer an integrated point
of view including information on the biochemistry of ADP-Glc PPases, including
structural information at the three-dimensional level together with biochemical data
regarding ligand binding and functional roles of the protein domains. This work has
aimed to contribute to a better understanding of the mechanism of action and regulation
of this enzyme that is fundamental in the bacterial and plant reserve polysaccharide

metabolism.

39

CHAPTER 2

Domain organization of the ADP-glucose pyrophosphorylase from

Escherichia coli1

 

1These results have been published in Bejar, C.M., Ballicora, M.A., Gemez-Casati, D.F., Iglesias, A.A.,

Preiss, J. (2004) The ADP-glucose pyrophosphorylase from Escherichia coli comprises two tightly bound
distinct domains. F EBS lett. , 573, 99-104.

40

2.1. ABSTRACT

Computational analysis of ADP-glucose pyrophosphorylases predicts a fold with
two domains. Co-expression of two polypeptides comprising residues 1-323 and 328-431
from the Escherichia coli ADP-glucose pyrophosphorylase yielded an enzyme form as
active as the wild type. The only difference with the wild type was a slightly modified
affinity for allosteric effectors. The two polypeptides could not be separated by
chromatographic procedures. Separate expression of these polypeptides produced
inactive unstable forms. All these results indicated that the ADP-glucose
pyrophosphorylase comprises two domains with a strong interaction between them. That

interaction is important for allosteric properties and structural stability.

41

2.2. INTRODUCTION

The ADP-Glc PPases from enterobacteria are homotetramers (a4) composed by
subunits of ~50 KDa (~440 residues) encoded by the glgC gene (8). The three
dimensional structure of these enzymes is not available, but structural and computational
analyses predict that the folds of ADP-Glc PPases have two domains (3, 8, 9, 16, 167).
Despite the low homology (~20% identity), the predicted structure of the central part of
the ADP-Glc PPases (residues ~20-300) resembles other NDP-sugar PPases and is
postulated to be the catalytic core of the enzyme (3, 8, 9, 16, 167). On the other hand, the
C-terminus (~100 residues) was proposed to be important for regulation and specificity
for activators (168). A similar conclusion has been reached for plant ADP-Glc PPases
(9). Both domains are postulated to interact to regulate the activity because there are
residues near the N-terminus that participate in the regulatory site (175, 1 76, 184, 185). In
addition, the presence of an N-terminal tail (~10-20 residues) strongly determines the

allosteric properties of the enzyme (175, 176, 184, 185, 187, 188).

Partial proteolysis has been a classical procedure to probe domain boundaries
(190, 191). In this chapter, molecular biology techniques were used to test the 2-domain
hypothesis. The enzyme was divided into two polypeptides corresponding to the putative
domains, they were co-expressed, and the resulting enzyme form was purified and its

biochemical properties characterized.

42

2.3. MATERIALS AND METHODS

2.3.1. Materials

Synthesis and purification of oligonucleotides and DNA sequencing were
performed by the Genomics Technology Support Facility at Michigan State University.
Perfect Protein TM marker was obtained form Novagen, Inc. The Mono-Q HR 10/10, 5/5
and Phenyl-Superose columns were purchased from Amersham Pharmacia Biotech.
[32P]PPi was purchased from Perkin Elmer Life Sciences and ['4C]Glc1P from
Amersham Pharmacia Biotech. All other reagents were purchased at the highest quality

available.

2.3.2. Plasmids and expression vectors

The construction of pETEC, a pET24a derivative with the E. coli ADP-Glc PPase,
has been previously described (168). Plasmids pMABS and pMAB6 are modified

versions of the respective compatible expression vectors pMON17335 and pMON17336

(148). They have an Ndel rather than NcoI site for cloning.

2.3.3. Linker-scanning mutagenesis

Random introduction of a single 15-bp insertion per plasmid (pETEC) was

performed by the technique of Hallet et al. (192) with the commercial GPSTM linker-

43

scanning mutagenesis kit from New England Biolabs. From 130 insertions mutants
obtained, 20 colonies had a single 15 bp inserted in the ADP-Glc PPase coding region.
The plasmid with the insertion mutant 117 (pETEC-ins] 17) was selected for further
studies and subcloned in pMABS (pMABS-Ec-insl 17) after digestion with Ndel-Sacl for

expression in AC70R1-504 cells.

2.3.4. Truncated protein constructs

To obtain an enzyme with a C-terminal truncation of 108 residues (Ec,-323), DNA
from the 3' coding region of the E. coli ADP-Glc PPase gene was removed as follows. A
PCR fragment was amplified using pETEC-insl 17 as a template, T7 promoter primer
(5'-TAATACGACTCACTATAGGG-3') and the oligonucleotide
5'-ACCGGAGAGCTCTGTTTAAACACG-3', which introduced a Sad site 3 bp afier the
previously inserted stop codon. The PCR fragment was subcloned into an Ndel-Sacl

digested pMABS to obtain pMABS-Ec1-323.

2.3.5. Construction of a plasmid for expression of the C-terminus alone (Ec323431)

The DNA fragment encoding the C-terminal portion (Ec328431) was amplified

from pETEC with the T7 terminator primer (5'-GCTAGTTATTGCTCAGCGG-3') and

the oligonucleotide 5'-GGTAGCCACCATATGACCCTTAACT-3', which introduced an

Ndel site. The fragment was subcloned into pMAB6 using the sites Ndel and Sad to

44

obtain pMAB6-Begun. The coding regions of all final products were verified by DNA

sequencing and the various constructs used in this work are illustrated in Fig. 2.1.

2.3.6. Bacterial strains and expression of the recombinant enzymes

E. coli AC70R1-504 cells without endogenous ADP-Glc PPase activity were used
for expression of the enzymes subcloned in pMABS and pMAB6 (I 48). The expression
of pMABS-Ec-insl 17 and pMABS-Ec.-323 was performed as described previously for
pMLlO (147). Expression of pMAB6-Begun was performed as described previously for
pMON17336 (148). Co-expression of pMABS-Ec1-323 and pMAB6-Ec323431 was

performed as was described previously for plasmids pMLlO and pMON17336 (147).

2.3.7. Protein methods

Protein assay, electrophoresis (SDS-PAGE) and immunoblotting were performed
as described previously (167). N-terrninal peptide sequence determination of blotted
samples onto PVDF membranes (Bio-Rad) and soluble pure protein samples were
performed at the Macromolecular Structure, Sequencing and Synthesis Facility at
Michigan State University. Desalting was performed on Bio-Rad 10 DO chromatography

columns. Samples were concentrated with Centricon-30 devices (Amicon Inc.).

45

.mozomoaam caesium wEmz cocoon—Eco 295

.213QO 98 320m .m_mo=owS=E wficcmomioxcz .3 3588 was» 2 EEém 59:32.8 3a.: caving. as .m 4." ensure

zlmmo - a 2.20
A m. om<2 + m Ransom + 220m

 

 

 

 

 

 

 

 

 

-mm<2& cowgwESMmcmbéU may:
2 _m=_-mm<2a 5. ask... a... E 3.3.”.
.mszmomemSZQ w . ,_ - .mimmom
mmm-_om-mm<_2a 230m
85¢ a 25.2;

9e 8: 9a 8sz :2 82v
mEEmaE «U .5386 eta—3.3 255.5 a Z 5895

46

2.3.8. Enzymatic assays

Assay A: Pyrophosphorolysis direction -— Formation of [32P]ATP from [32P]PP,
was determined by the method of Morell et al. (123). The reaction was carried out for 10
min at 37 °C in a mixture that contained 50 mM Hepes (pH 8.0), 7 mM MgC12, 1.5 mM
[32mph (1500-2500 cpm/nmol), 2 mM ADP-Glc, 4 mM NaF, and 0.05 mg/ml bovine
serum albumin, plus enzyme in a total volume of 0.250 ml. Unless other activators were
assayed or specifically stated, 2 mM F BP was added in the reaction mixture.

Assay B: Synthesis direction — Formation of ADP-['4C]Glc from [14C]GlclP was
determined at by the method of Yep et al. (193). The reaction was carried out for 10
minutes at 37 °C in a mixture that contained 50 mM Hepes (pH 8.0), 7 mM MgC12, 0.5
mM [14C]GlclP (~1000 dpm/nmol), 1.5 mM ATP, 1.5 units/ml pyrophosphatase, and
0.2 mg/ml bovine serum albumin, plus enzyme in a total volume of 0.20 ml, unless
specifically stated. Unless other activators were tested, 2 mM FBP was added in the
reaction mixture. One unit of enzymatic activity is one umol of product, either [32P]ATP

or ADP-[14C]Glc, formed per min at 37 °C.

2.3.9. Purification of Ec-insll7

After induction, twenty liters of AC70R1-504 cells transformed with
pMABS-Ec-insl 17 were harvested and crude extracts obtained as described previously
(147). The crude extract was precipitated with a 30-60% ammonium sulfate cut. After

centrifugation, the precipitate was redissolved in buffer A (50 mM Hepes pH 8.0, 5 mM

47

MgClz, 0.1 mM EDTA, 10% sucrose) and desalted. The sample was applied onto a
DEAE-Fractogel column (EMD Chemicals), and eluted with a linear NaCl gradient
(0-0.5 M). The active fractions were pooled, desalted, applied to a Green A (Amicon
Corp., Lexington, MA) affinity chromatography column (1 ml bed volume) and eluted
with a 20-ml linear gradient of NaCl (0-2 M). Purest fractions were pooled, desalted,
concentrated, and applied onto a Mono Q HR 5/5 (FPLC, Pharmacia) column
equilibrated with bufler A and eluted with a linear NaCl gradient (005 M). The post-
Mono Q fractions were pooled, concentrated, resuspended in buffer B (buffer A plus 1.2
M ammonium sulfate), and applied on a Phenyl-Superose (F PLC, Pharmacia) column
equilibrated with buffer B. The sample was eluted with a decreasing linear gradient of
ammonium sulfate (1.2-0.001 M). The purest fractions were pooled, concentrated, and
applied to a Pharmacia Superdex 200 HR 10/30 column equilibrated with buffer A. The
gel filtration chromatography was run at a flow rate of 0.25 ml/min and fractions with
activity were pooled. After this step, Ec-ins117 was >95% pure, with a specific activity

of 180 U/mg in the pyrophosphorolysis direction assay (Table 2.1).

2.3.10. Purification of Ec1-323

The protein was monitored by immunoblot throughout the purification steps.
AC70Rl-504 cells harboring pMABS-Ecng, were grown and induced in 4 liters of LB
medium. Crude extracts were precipitated with 0-30% ammonium sulfate and the pellet
was resuspended in buffer A and desalted. Samples were further purified with a Mono Q

HR 10/10 and a Green A affinity column as indicated above (Table 2.1). The fractions

48

Table 2.1. Specific activities of mutant and wild-type E. coli ADP-Glc PPases

Concentration of FBP for the mutants was raised to 4 mM. For the sample

Ec1-323+EC323-341, concentration of ATP (synthesis direction) was raised to

 

 

 

5 mM.
Specific Activity
Purity a

Sample Pyrophosphorolysis Synthesis

(%)

(U/mg) (U/mg)

Wild-type 131 i 3 54 d: 1 9O
Ec-insl 17 180 i 9 b ND C >95
EC1-323 < 0.001 <0.0001 50-60
EC1-323 + EC323-34| 132 :1: 5 43 i 1 >95

 

a Purity of the samples was estimated from SDS-PAGE gels.

b Another independent purification produced a pure sample with an activity
of 140 U/mg.

c not determined

49

containing the enzyme, ~50-60% pure as detected by immunoblot analysis, were pooled,

concentrated, and stored at -80 °C.

2.3.11. Purification of co-expressed Ec1-323 + Ec323431

AC70R1-504 cells co-transformed with pMABS-Ec1-323 + pMAB6-EC323431 were
grown and induced in a 1-liter culture. The resulting crude extracts were applied onto a
DEAE-Fractogel column (EMD Chemicals Inc.), and eluted with a linear NaCl gradient
(0-0.5 M). The active fractions were pooled and precipitated with 30-60% ammonium
sulfate. After centrifugation, the pellet was resuspended in bufler A and desalted. The
sample was applied to a Mono Q HR 10/10 (FPLC, Pharmacia) column equilibrated with
bufler A and eluted with a linear NaCl gradient (005 M). Purest fractions, assessed by
SDS-PAGE, were pooled, concentrated and applied to a Green A affinity
chromatography column. The sample was eluted with a linear gradient of NaCl (0-2 M).
After this step both co-expression products, Ec|-323 + 130323.431 accounted for >95% of the

protein (Table 2.1).

2.3.12. Molecular mass determination

Purified wild type E. coli ADP-Glc PPase and the mutants Ec-insl 17, E01323 +
EC323431, and Ec1-323 were applied to a column Pharmacia Superdex 200 HR 10/30
equilibrated with buffer A and run at 0.25 ml/min. Peaks were followed by absorbance at

280 nm, tested for enzymatic activity or confirmed by immunoblot (E01423). For

50

calibration, ferritin (440 kDa), E. coli ADP-Glc PPase (200 KDa), aldolase (158 kDa),

and Hemoglobin (67 KDa) were used as standard proteins. The calibration line and the
interval confidence (90%) were obtained after plotting the data as log (molecular mass)
vs. elution volume in the program Origin® 5.0. Molecular mass was expressed with a

confidence interval of 90%.

2.3.13. Kinetic characterization

Kinetic data were plotted as specific activity (umol min'l mg") versus substrate or
effector concentration. Kinetic constants were acquired by fitting the data to the Hill
equation with a nonlinear least square formula using the program Origin TM 5.0. Hill plots
were used to calculate the Hill coefficient rm and the kinetic constants that correspond to
the activator, substrate or inhibitor concentrations giving 50% of the maximal activation

(A05), velocity (S05), and inhibition (10.5).

51

2.4. RESULTS AND DISCUSSION

Random insertions of a single 15-bp fragment into a recombinant E. coli glgC
gene were produced to study structure-function relationships on the ADP-Glc PPase (M.
A. Ballicora and J. Preiss, unpublished results). One of the insertion mutants generated,
Ec-insl 17, had a stop codon after residue 323 in the translated polypeptide of the E. coli
glgC gene. Surprisingly, the polypeptide still possessed ADP-Glc PPase activity although
the stop codon would truncate 108 residues of the C-terminus. Colonies of BL21(DE3)
cells expressing Ec-insl 17 or the wild type enzyme, but not a control plasmid (pET24a),
stained brown in presence of iodine vapors (unpublished results). This indicated high
levels of glycogen in the cell and suggested the presence of ADP-Glc PPase activity

(194). For this reason, the properties of this putatively truncated form were characterized.

2.4.1. Purification and characterization of Ec-ins 117

Despite the lower expression level of Ec-insl 17 as observed by immunoblot and
SDS-PAGE (not shown), the pyrophosphorolysis activity of this mutant was 0.31 U/mg

in the crude extracts. That was two orders of magnitude lower than the over-expression of
the wild-type ADP-Glc PPase but two orders of magnitude higher than pET24a, the
plasmid control without insert (not shown). Attempts to isolate this ~37 kDa truncated
protein resulted in the co—purification of a 12 kDa peptide even after five different
chromatographic columns (Fig. 2.2). Despite the differences in molecular size, both

polypeptides comigrated in a high-resolution gel filtration column (Superdex 200 HR).

52

100 —"

 

Figure 2.2. SDS—PAGE of E. coli wild-type ADP-Glc PPase and Ec-insll7
mutant. Lane 1, E. coli wild-type ADP—Glc PPase; lane 2, pre-stained molecular
mass markers; lane 3, Ec-insl l7 mutant (0.5 ug); and lane 4, Ec-insl 17 mutant
(1-5 pg)

53

The N-terminal sequence of the 12 kDa band, obtained after SDS-PAGE and
blotting, was MTLNSLVSGG, which indicated that the polypeptide corresponds to the
C-terminus of the E. coli ADP-Glc PPase with residue 328 as the N-terrninal methionine
(103 amino acids). In addition, N-terminal sequencing of the purified Ec-insl 17 in
solution confirmed the presence of two polypeptides, suggesting that both the C-terminal
fragment together with the N-terminus of the enzyme (37 kDa band) were being
expressed. It also indicated that the stoichiometry of the 37 and 12 kDa polypeptides was
1:1 (Fig. 2.3). Based on the sequence data, the polypeptide of 12 kDa appeared to be the
product of an open reading frame that starts 24 bp downstream of the inserted stop codon,

328

at Met (Fig. 2.4). The lower expression of this hetero-oligomer form, compared to the

wild type expression level, could be explained by the absence of a proper ribosome-

binding site before Metm.

Kinetic properties of the Ec-insl 17 mutant purest sample were determined in the
pyrophosphorolysis direction (Table 2.2). Apparent affinities for the substrates
(ADP-Glc, PP;) and cofactor (Mg2+), were similar to those of the wild type enzyme. The
main difference was the apparent affinity for the activators. The A05 for F BP was almost

6-fold higher for Ec-insl 17 compared to the wild type.

54

 

Residue Number

 

Amino
Acid 2 3 4 5 6 7
pmol

D 3.11 2.35 2.27 2.88 2.21 2.66 10.19
N 0.72 0.51 0.87 9.09 1.44 8.08 2.78
S 3.24 11.10 2.04 1.96 8.85 1.53 2.32
T 1.41 11.40 0.49 0.57 0.35 0.22 0.77
E 0.49 0.50 0.63 9.98 2.59 1.33 1.24
M 14.60 0.33 0.20 0.23 0.27 0.36 0.43
V 14.90 1.89 1.03 1.03 1.21 1.78 11.90
K 0.49 0.47 ND 0.66 13.39 3.24 1.47
L 0.42 1.39 29.18 3.31 ND 13.09 2.98

Sequences deduced
1 V S L E K N D
2 M T L N S L V

 

Figure 2.3. N-terminal sequence analysis of the Ec—insl l7 mutant. Automated
sequencing of the Ec-insl 17 protein was performed directly from the purest
fraction. The values are the picomoles of amino acid derivative found in each
cycle of degradation. In each cycle of cleavage, major peaks are in bold.
Residues Q, G, H, A, R, Y, P, W, F, I. were negligible in all these cycles and
were omitted for simplicity. ND: Not detectable.

55

32:22 8 95.83 2:32. we. 5 2: :5 .334 8:23.. w:_m_.anov «9 RI 2: 22:3ng 95 no cog—men:
:5 5:518:82 2: wfifiuocow .39 2662 “a 3268:: 33 20:8 5on < 25:52: 3-2 2: 2 357.022 .m.~ =0:on E
“32:58: 3 235232.: mama—88-32:: .3 .0252: 83 .23: 3.2 m 52>» E .2 Exam 2: 8 826. 2: :5... 08mm BOA—06.

256:3 2: 9 2253250 3:258 5%: 2:. .3535 ofivifi :3 .N 2.. 5 h: .8352: 2: He 922.com .v.N 0.53..—

mmm mmm

 

 

.. moem ..
.. o m > a m z A H z o z m 0 m m 0 1 4 > m D O > m 2..
..ooooouunmouomouommuuooommumoomomooomuoooouoooumonommmuuumumoummmmoouooupmmm..

AL 2.. «:53: 50:32..

KHNmCH
mmm mmm ham
.. w m > A m z a H E 0 m m o m m a O > m 2..
..ooooopuummuomouommuuooomoummomomouomummoopomoumommoouooupmmm..

on? 2:5

56

Table 2.2. Apparent affinity for the different effectors of the mutant Ec-insl l7 and
wild-type ADP-Glc PPase. The assays were performed in the pyrophosphorolysis

 

 

 

direction
Enzyme
Effectors
Wild type Ec-insl 17 Ratio a

Substrates S05 (uM)

ADP-G10 166 :t 26 241 :t: 25 1.45

PPi 81 d: 14 452 13 0.55
Cofactor 50.5 (pM)

Mg2+ 142082310 2147186 1.51
Activators A05 (11M)

FBP 37.8 i 5.2 214 i 53 5.7

PLP 0.045 i 0.004 0.156 i 0.001 3.5

PEP 282 _+. 15 > 2000 b > 7.1
Inhibitor 105 (pM)

AMP 41i4° 21i8d 0.51

 

a This is the value of the kinetic constant of Ec-insl 17 divided the wild type.

b .
Saturation was not reached.

c To observe AMP inhibition, activator should be added. In this case, F BP was 40 pM
to avoid saturating concentrations of activator and allowed the competitive inhibition

by AMP. That concentration is about the A05 value.

d FBP was 300 11M, which is about the A05 value.

57

2.4.2. Purification and characterization of C-terminal truncated forms

It was important to determine whether the truncated enzyme was active by itself
or if the co-expression with the C-terminal 12 kDa polypeptide contributed to the
reconstitution of the activity. To ensure the expression of a single polypeptide comprising
residues 1-323 (Ec1-323), a Sacl restriction site was introduced immediately downstream
of the stop codon by site directed mutagenesis. Only the E01323 coding fragment was
subcloned in pMABS, expressed, and purified as described in section 2.3. With this
approach, most of the expressed enzyme was insoluble, whereas the soluble fraction had
a peptide of about 37 kDa and some lower bands recognized by immunoblotting, which
indicated that it was highly proteolyzed (data not shown). Ec.-323 (~37 kDa) was purified '
to ~50 % purity where the main contaminants were lower bands still recognized by
immunoblot. The activity was negligible (Table 2.1) and did not increase significantly
with higher concentrations of FBP, up to 10 mM (data not shown). Gel filtration showed

that the sample was highly aggregated (data not shown).

The total loss of activity after the removal of the C-terminus indicated that this
domain is involved in more than regulation. In the inactive Ec1-323 mutant, the C-terminal
deletion yielded a form highly susceptible to proteolysis. In an attempt to study which
specific regions of this domain were important, deletions of 20 to 90 residues (every 10)
from the C-terminus were evaluated. The truncated enzymes were expressed in small-
scale systems and were visualized by SDS-PAGE and western blot at their expected

molecular sizes (data not shown). The recombinant products were found mainly in the

58

insoluble fraction of the sample and the pyrophosphorolysis ADP-Glc PPase activity was
negligible in all crude extracts (<0.0001 U/mg). These results suggest most of, or maybe

all of the ~100 residues of the C-terminus are necessary for maintaining a functional

E. coli ADP-Glc PPase. This may be due to the role of specific residues or to the overall

structure of the domain that is necessary for the integrity of the enzyme.

2.4.3. Characterization of Ec1_323 + Ec328_431

The results obtained with Ec1.323 strongly suggested that the properties of
Ec-ins117 are due to the presence of both the 37 kDa and the 12 kDa polypeptides. To
verify this possibility, the E. coli glgC fragment encoding the 103 C-terminal residues of
ADP-Glc PPase (Ecmsm) was subcloned into a compatible expression vector (pMAB6).
Co-expression of Ec.-323 and EC323.431 was performed in AC7OR1-504 cells, yielding
extracts with activity (39 U/mg) comparable to the wild type (24 U/mg) (168). That level
of expression was about IOO-fold higher than the Ec-insl 17. After the same purification
procedure of the wild-type enzyme, Ecl-323 and Ecmsm co-purified, indicating a very
strong interaction. The 37-kDa fragment did not appear as proteolyzed as when it was
expressed alone (data not shown). In addition, the specific activity values were
comparable to that of the wild-type enzyme (Table 2.1). The molecular mass determined
by gel filtration in native conditions of the co-expressed Ec1-323 + EC323.431 (227 i 32 kDa)
and Ec-insl 17 mutant (213 i 30 kDa) were not distinguishable from the wild-type
enzyme (200 kDa). These results, together with the 1:1 stoichiometry (Fig. 2.3), indicate

that the structure of both Ec.-3z3 + Ec32343. and Ec-ins117 are a’4m4, rather than 0.4 as the

59

wild type (a’ and 0) being the polypeptides Ecl-323 and Ec323431, respectively). The
apparent affinities for the substrate ATP (4.7-fold higher 805) and for the activator FBP
(35-fold higher A05 in the synthesis direction) were altered compared to those of the wild
type (Table 2.3). The inhibition by AMP was not significantly changed and the apparent

affinity for GlclP was identical (Table 2.3).

60

mm“— 21 om_ mo 00:000.: 5 :0:to.t0a 0>.=_0 3:32:— 2

mm: 21 om :0 00:03:: :_ :2583: 02:0 3:55:— a

 

 

 

 

 

.memmom
30 3 z. 2 Ni: 3 2 .32 S S .80.: am 5.23 +253
Z 12 2 303.... Q 38 3 S 3.2 3 z 0 SM 252;?
A25 A25 A25 A25 9::
It : méN It a mg: == 06‘ :2 m6” Ifi new; oars-m—
:s? .5... 8:20 0:

 

.2338.» + mM-—0m 5.: 28 v :5 25 2:5 2: 2: 2.: N 33 amn— ho 3:823:00 2: .m0>._:0 $0.0
:5 n:.< 2: .5: .:o:238 23:0 9 2.: m 0.03 .mvwmmom + mM-—0m 2: :8 9:. mo 3:223:00 2: .0023 “Ex 2: :8 308$

.83.— .ianofl + nun-_0m .5: 093 2:5 03.: 052—3 .500 .N no 5509:: 2355?. 05 E E80893: 0505M .m.N 030,—.

61

2.5. CONCLUSIONS

Important structural information can be inferred from these results: (i) the stop codon
inserted in the Ec-insl 17 mutant, which generated a “nick” in the protein sequence, was
in a non-critical position for the 3D structure of the protein. (ii) These data also support
structure predictions and the hypothesis that the E. coli ADP-Glc PPase is organized in at
least two distinct domains (8). According to the proposed model of domain organization
(8, 9, 167), the stop codon was inserted in a loop that separates the two putative domains
(Fig. 2.5). The data also agree with the hypothesis that the C-terminal domain is ~100
residues long and linked to the catalytic domain by a long loop (~30 residues) (8).
Removal of this domain yielded an inactive enzyme rather than a shorter unregulated
version as other NDP-sugar PPases of 30-40 KDa (195-198). Even the removal of 20
residues at the C-terminus rendered a form without activity. It is possible that the absence
of even a small part of the C-tenninal domain leaves the enzyme in a misfolded or
insoluble form. On the other hand, a “nick” separating the C-terminus and the catalytic
domain, obtained by the co-expression of both the N- and the C-terminal polypeptides,
produced an enzyme with very similar properties to the wild type. Only the apparent
affinity for the activator was slightly modified. Possibly, the lack of a covalent bond
between the N- and C-domains yields a more relaxed enzyme that favors less the
activated conformation. A slightly higher concentration of activator overcame this
problem, 3.5- and 5.7-fold in the synthesis and pyrophosphorolysis direction,
respectively. The possibility that the residues removed in the “nick” (S323GSHG327) are

the reason for the decrease in activator affinity cannot be ruled out. However, in other

62

ADP-Glc PPases, those residues are not conserved or they are deleted, together with
Met328 (8). Most likely, they are part of an area that connects the two major domains and

does not play a specific role (Fig. 2.5).

Partial proteolysis has been a classical procedure to detect domain boundaries
(190, 191, 199). In this work, molecular biology techniques were used to probe whether
the E. coli ADP-Glc PPase comprises two domains. After co-expression of the C- and the
N-terminal domains, they remained non-covalently bound, with a tight interaction that
cannot be disturbed even after passage through several chromatography columns.
Presence of the C-terminal domain stabilizes the enzyme and prevents proteolytic
degradation of the N-terminal domain. These results provide a structural insight for the
differential roles of these regions of the ADP-Glc PPase. The hypothesis is that they are
two strongly interacting domains that have to be expressed together to obtain a fully
functional enzyme. According to previous analysis (8, 167), the catalytic domain is
located in the 37 kDa fragment (similar to most NDP-sugar PPases) and the responsibility

for the regulation is shared by both the 12 and the 37 kDa fragments.

63

 

 

 

       

Glycine rich

. FBP ATP Catalysis
region G|c1 P
Q>',.-®---..K39 Y114 D142 K195

\

   

 

Figure 2.5. Domain organization of the ADP-Glc PPase from E. coli. The secondary structure
is based on the prediction from references (8) and (9). Arrows and cylinders represent B-sheets
and a—helices, respectively. Dark circles are important residues that interact with effectors or

6

. . . . 33 295 .
part1c1pate 1n catalySIS. When Gly and Pro are replaced, allostenc mutants are generated

(8). Afier insertion of a stop codon, residues between Ser323 and Met328 were “nicked” (see Fig.

2.4).

64

CHAPTER 3

The structural role of the Escherichia coli ADP-glucose

pyrophosphorylase’s N-terminus in allosteric regulation2

 

2This work presented in this chapter has been published: Bejar, C.M., Ballicora M.A., Iglesias, A.A.,

Preiss, J. (2006) ADP-glucose pyrophosphorylase’s N-terminus: structural role in allosteric regulation.
Biochem. Biophys. Res. Commun, 343: 216-221.

65

3.1. ABSTRACT

The functional role of the Escherichia coli ADP-glucose pyrophosphorylase’s
N-terminus in allosteric regulation, and the particular effects caused by its length, were
studied and presented in this chapter. Small truncated mutants were designed, and those
lacking up to 15-residues were active and highly purified for further kinetic analyses.
NA3 and NA7 did not change the kinetics parameters respect to the wild-type. NAll and
NAl 5 enzymes were insensitive to allosteric regulation and highly active in the absence
of the activator. Co-expression of two polypeptides corresponding to the N- and
C-termini generated an enzyme with lower activation properties than the wild type (see
Chapter 2). In this section, the characterization of a NA] 5 co-expression mutant, in
which the allosteric regulation was restored to wild-type levels, is presented. Unusual
allosteric effects caused by either an N-terrninal truncation or co-expression of individual
domains may respond to structural changes favoring an up-regulated or a down-regulated

conformation rather than to the disruption of specific activator or inhibitor sites.

66

3.2. INTRODUCTION

The relevance of starch as a natural polymer and raw material for food and non-
food industries makes ADP-Glc PPase an enzyme with particular interest in
biotechnology (7). It has been clearly established that manipulation of the starch content
in plants requires managing levels of activity of the enzyme within plastids in specific
tissues (7, 60). For these technological purposes, the understanding of the enzyme’s

regulatory mechanisms is a key issue.

Characterization of amino acids involved in binding of activator for the ADP-Glc
PPase from E. coli and A grobacterium tumefaciens performed by chemical modification
and site-directed mutagenesis allowed for pointing out the relevance of the protein’s
N-terminus (N -t) in allosteric regulation (175, 176, 184, 185). More recent studies
performed with chimeric enzymes from these bacteria have indicated the involvement of
the N-t together with the C-terminus (C-t) domain (and the interaction between them) in
determining the specificity and affinity for the activator (168). Also, the role of the C-t
domain in regulation and its tight interaction with the catalytic region of the protein has

been evidenced for the E. coli enzyme (reported in Chapter 2).

The importance of the N-terminal region for the allosteric activation and
inhibition of the E. coli ADP-Glc PPase was evidenced by characterization of truncated
enzymes (18 7, 188). The enzyme lacking 10 to 13 amino acids from the N-t and 2 amino

acids from the C-t afier treatment with proteinase K is almost independent of the need of

67

FBP for maximal activity and is insensitive to inhibition by AMP (187). These results
were confirmed by engineering a recombinant truncated enzyme with 11 amino acids
deleted from the N-t and 2 from the C-t (188). The unusual regulatory properties of this
truncated ADP-(31c PPase justified a more detailed characterization of the actual role the
N-t extension has in enzyme activation and its interaction with the C-t of the protein. The
motivation for the work presented in this chapter was to analyze the structural bases
determining the enzyme’s regulatory properties and how the length of the N-t affects the
protein ability to arrange specific 3D conformations. Another question addressed here is
how the N-t communicates with other domains involved in regulation. As a result, the
present work provides us with a more insights into the relationship between structure,

function and regulation of the ADP-Glc PPases.

68

3.3. MATERIALS AND METHODS

3.3.1. Construction of N-terminal truncated enzymes

E. coli N-terminal truncated forms (EcNA3, EcNA7, EcNAl 1, EcNAlS, EcNA19
and EcNA22) were encoded by pETEC (168) derivatives and were obtained using the
following forward primers, which introduced an Ndel site (in italics):

EcNA3: 5’ ATG GTT CAT A TO GAG AAG AAC GAT CAC T 3’
EcNA7: 5’ GAO AAG CA TA TG CAC TTA ATG TTG GCG C 3’
EcNA11:5’ CAC TTA CATATG GCG CGC CAG CTG CCA 3’
EcNAlS: 5’ GCG CGC CA TATG CCA TTG AAA TCT GTT 3’
EcNAl9: 5’ CCA TTG CA TA TG GTT GCC CTG ATA CTG G 3’

EcNA22: 5’ TCT GTT CATATG ATA CTG GCG GGA GGA C 3’

The T7 terminator was used as reverse primer. The fragments were sub-cloned in
pET24a vector between Ndel-Sacl sites to form pETEC-NAB, pETEC-NA7,

pETEC-NM 1, pETEC-NA] 5, pETEC-NA19, and pETEC-NAZZ.

EcNAl 5-CA108 coding DNA was amplified by PCR from pETEC-NA] 5 with the
downstream primer used to construct EcCA108 (see Chapter 2), which introduced a Sad
site right after the codon corresponding to amino acid 323. The amplified coding region
was then cloned as Ndel-Sac] fragment in pMABS. The plasmid coding for E. coli

ADP-Glc PPase C-terminal residues 328-431, pMAB6-Ec32843l, was previously obtained

69

(see Chapter 2). All constructs used (Fig. 3.1) were verified by DNA sequencing.

3.3.2. Expression of the recombinant enzymes

N-terminal truncated enzymes encoded by pETEC derivative vectors, were

expressed in E. coli BL21(DE3) cells as previously described (168). EcNAl 5-CA108 +
139328431 were co-expressed in an ADP-Glc PPase deficient E. coli B strain, AC70R1-504
(148), as described for EcCA108 + Ec328431 (Chapter 2). In small-scale expression
assays, single colonies of the pETEC-derivatives transformants were grown in 50 ml
Luria broth medium at 37 °C with 50 pig/ml kanamycin up to OD600 = 0.8. The
expression was induced at room temperature (23-25 °C) for 4 h by addition of 1 mM final
concentration of IPTG. For the co-expression of the pMAB-derivative plasmids, cells
were grown in the same conditions but adding 50 ug/ml kanamycin plus 70 ug/ml
spectinomycin and carrying out induction with 1 mM final concentration of IPTG and 5
pig/ml nalidixic acid for 16 h. In both cases, after induction, cells were chilled on ice and
harvested by centrifugation. Cell pastes were resuspended in 3 ml buffer A (50 mM
HEPES pH 8.0, 5 mM MgC12, 0.1 mM EDTA, and 10% w/v sucrose). All subsequent
protein purification steps were conducted at 0-4 °C. Cells were disrupted by sonication,
centrifuged for 15 min at 15,000 X g and the supematants (crude extracts) were stored at —
80 °C. Dilutions of the samples (2-fold and 10-fold) were tested for activity in the

pyrophosphorolysis direction as described below.

70

.033:::0-:0: A05 0:330 09»: 2:3 2:

:2: 030— 003 8:03:00 2: :0: bEE< 3v .mmn— 8:03:00 2: .:0 :0:0:::00:00 30. :0 :0 00:02:: 2: :_ 02:00 3:: 0:03 382:0

::0:::: 2:. :3 80:33.:— :0::>:0: :2: 1:: 3:53. 80:50:00 03m.— 0_U-na< :00 .M 0:: :0 b:—:E:m .1». 0.53..—

8:82:93:

+++++

02:2

 

 

 

 

 

 

 

 

 

 

 

:88 o. T:
05:33-0

 

Add Om at:
52...... 0:256

$22.0: +
:23: 228
21.8.30 + 2:38
:3in..” 0m
8208
2428
24:23
2360
:azom
2200
228:
25-25:

 

91: 7.:
0:35:24:

3.3—9:0 3:35: 3:?— 0-023. :00 .N

3.0

71

3.3.3. Purification of N-truncatcd ADPGlc PPase mutants

Recombinant E. coli BL21(DE3) were grown in 1-2 L and induced as described
above. The cell pastes were resuspended in 15-30 ml of buffer A, sonicated, and the
lysates were cleared by centrifugation. The resulting crude extracts were applied to a
DEAE-Fractogel column (EMD Chemicals INC), and eluted with a linear NaCl gradient
(O-O.5 M). The active fractions were pooled and precipitated with a 30-60% ammonium
sulfate cut. Afier centrifugation, the pellet was resuspended in buffer A and desalted on
Bio-Rad 10 DG chromatography columns equilibrated with the same buffer. The desalted
samples were individually applied to a Mono Q HR 10/10 (FPLC, Pharmacia) column
equilibrated with buffer A and eluted with a linear NaCl gradient (O-O.5 M). The purest
fractions (as assessed by SDS-PAGE) were pooled and concentrated using Centricon-30
devices (Amicon Inc.), After this step, ECNA3, ECNA7, and EcNAlS were >95% pure.
EcNAll MonoQ active fractions were pooled, concentrated by 80% ammonium sulfate
precipitation, applied to a phenyl-Superose (FPLC, Pharmacia) column, and eluted with a
decreasing gradient of ammonium sulfate (1.2-0.0 M). The active fractions were pooled

and concentrated with Centricon-3O devices.

3.3.4. Purification of co-expressed EcNAlS— CA108+ Ec32843.

Over-expression and purification were carried out as described for

EcCA108+Ec328431 (see Chapter 2).

72

3.3.5. Protein concentration assays

Protein concentration in the crude extracts and in the subsequent purification steps
was measured using the bicinchoninic acid reagent (200) (Pierce Chemical Co.), with
BSA as the standard. Protein concentration of the purified enzymes was determined by

UV absorbance at 280 nm using an extinction coefficient of 1.0 ml mg'l cm'I (201).

3.3.6. Protein Electrophoresis and lmmunoblotting

Purification of the recombinant proteins was monitored by SDS-PAGE as
described by Laemmli (202), using 4-15% Tris-HCI pre-cast gradient polyacrylamide
gels (Bio-Rad). PerfectTM protein markers were used as molecular mass standards.
Following electrophoresis, protein bands were either visualized by staining with
Coomassie brilliant blue R-250 or electroblotted onto a ProtranTM nitrocellulose
membrane. The membrane was then treated with affinity purified anti-E . coli B strain
AC7OR1 ADP-Glc PPase IgG (201). The resulting antigen-antibody complex was
visualized by alkaline phosphatase-linked goat anti-rabbit IgG, and then stained with BM

purple AP-substrate precipitating reagent (Roche Molecular Biochemicals).

3.3.7. Enzymatic activity assays

Assay A: Pyrophosphorolysis - Formation of [32P]ATP from [32P]PPi in the

direction of pyrophosphorolysis at 37 °C was determined by the method previously

73

described ( 123). Unless otherwise indicated, the reaction was carried out for 10 min in a
mixture that contained 50 mM HEPES (pH 8.0), 7 mM MgC12, 1.5 mM [32P]PPi (1500-

2500 cpm/nmol), 2 mM ADPGlc, 2 mM FBP, 4 mM NaF, and 0.05 mg/ml BSA, plus

enzyme in a total volume of 250 pl.

Assay B: Synthesis — Formation of [14C]ADP-glucose from [”C]GlclP in the
synthesis direction was determined at 37 °C by the method of Yep and coworkers (193).
The reaction was carried out for 10 min in a mixture that contained of 50 mM HEPES
(pH 8.0), 7 mM MgClz, 0.5 mM [‘4C]GlclP (~1000 dpm/nmol), 1.5 mM ATP, 2 mM

FBP, 0.0015 units/pl pyrophosphatase, and 0.2 mg/ml BSA, plus enzyme in a total

volume of 200 pl, unless specifically stated variations.

One unit of enzymatic activity is equal to 1 pmol of product, either [32P]ATP or

['4C]ADP-Glc, formed per min at 37 °C.

3.3.8. Kinetic characterization

Kinetic data were plotted as specific activity (nmol min'l mg") versus substrate or
effector concentration. Kinetic constants were acquired by fitting the data to the Hill
equation with a nonlinear least square method using the program Origin W 5.0. Hill plots
were used to calculate the Hill coefficienth and the kinetic constants that correspond to
the activator, substrate or inhibitor concentrations giving 50% of the maximal activation

(A05), velocity (50,5), and inhibition ([05).

74

3.4. RESULTS AND DISCUSSION

3.4.1. Purification and characterization of E. coli ADPGlc PPase N-tcrminal

truncated forms

To evaluate in detail the structural role of the N—t extension in allosteric
regulation, small deletion mutants of the ADP-Glc PPase from E. coli were designed
(Fig. 3.1) and tested. EcNA3, ECNA7, EcNAl l, EcNAlS, EcNA19 and EcNAZZ truncated
proteins were over-expressed in E. coli BL-21(DE3) and purified in small-scale. EcNA3,
EcNA7 and EcNAl 5 were enriched to >95%, and EcNAll to 60-70 % purity (Table 3.1).
As shown in Table 3.1, the four mutants had specific activities comparable to that of the
purified wild-type enzyme. Thus, the N-t comprising at least 15 amino acid residues of
the E. coli ADP-Glc PPase is not essential for catalytic activity. Conversely, deletions of
19- and 22-residues from the N-t significantly decreased the activity of ADP-Glc PPase
in the crude extracts. In fact, enzymatic activity of EcNA19 measured by the
pyrophosphorolysis assay was two orders of magnitude lower than that of the wild-type

(also in the crude extract), whereas activity of EcNA22 was negligible (Table 3.1).

Results in Table 3.1 were consistent with the predicted secondary structures of
ADP-Glc PPase (16 7) and with the 3D model previously proposed (8). This model was
also validated with the data derived from the crystal structure of an inhibited form of the
homotetrameric (small subunit) enzyme from potato tuber (1 70). In the structural model

the N-terminal tail is a loop preceding a B strand starting at Ser19 (Fig. 3.2).

75

0:00:08 0:30 8 :0:0:_8:0:0Q a

$2: 0:00:08 0:30 :_ 388.209 0

 

 

 

- - - 8.? wood v 0N~<Zom

- - - 8-2 2 .3200

- - - 00.00. am .2300: am.

m: - - RA 3: m_<Z0m_

o: - - cn-o: mo _ 220m
9mm 0N 00m on 0.3 maA m: 0349-:
Wmm x: 9m 0” mNm EA .3 m<Z0m
mdm :m 0.: H :.:m co :m _ 0:b-::3 0m

=2 :23 2:. 0x: A085:
:0:0>:0< 8:080: :0:0>:00 mm": 5:8: 3:501: 0500mm 0.:80m

 

0.0m mas—mam 80.: 8:08:00 003 003800 2: :0 DE:— 20000 mam—03:38:85: 0.: .3 :0:_::.:0:0: 0:03 00:080. 8:02:00
-8... 05 803:8 oEooam 32:: 0.0-22 :8 .m 0385:: 03:833.: 0.5 25.00.. :e 88.88 2:800 .3. 030:.

76

1 j KN'i-I #13 1j 1:27
N-terminus MVS E L ARQ PLK ALILAG‘BRGTRLKDL. . .

‘

 

 

Figure 3.2. N-terminal E. coli ADP-Glc PPase sequence and predicted
secondary structures. Residues l to 18 are predicted to be a loop. Serl9 is
predicted to be at the N-terminal end of a B strand in which Leu22 resides. The
Gly-rich loop (in light grey) is similar to P-loop-like motif present in protein

kinases and nucleotide-binding sites.

77

Following that B-strand is a glycine-rich loop similar to the P-loop motif present in
protein kinases and nucleotide binding sites. The 19- and 22-residue N-terminal deletions
may cause destabilization of the local secondary structure (including the glycine-rich
loop) that propagates to other important regions in the overall protein structure. Kinetic
analysis of EcNAB, EcNA7, EcNAll and EcNAl 5 showed that removal of up to
7-residues from the N-terminus does not significantly alter the FBP-activation properties
of the enzyme. EcNAB and EcNA7 mutants behaved in ways similar to the wild-type
enzyme, which is activated 20- to 30-fold by the specific effector (Fig. 3.3 and Table
3.1). On the other hand, EcNAll behaved as reported previously (187, I88) and as
EcNAl 5. Both mutant enzymes were highly active even in the absence of F BP, showing
almost no activation (Fig. 3.3 and Table 3.1). Additionally, EcNAlS was tested for its
ability to be inhibited by AMP (Fig. 3.4). Because it has been demonstrated that
inhibition of the E. coli ADP-Glc PPase by AMP requires the presence of at least traces
of FBP in the medium (44), inhibition kinetics were performed at two different activator
concentrations: 30 and 130 uM (Fig. 3.4.A and .B, respectively). A similar lack of
sensitivity of EcNAl 5 for this inhibitor was also exhibited by the EcNAll mutant (187,
188). Results suggested that deletion of l 1- or 15- residues from the N-t induces spatial
arrangements favoring a more active conformation of the enzyme, and possibly

disrupting the AMP site.

78

 

 

 

 

-

 

Specific Activity (Units / mg)

4

 

 

Jul 1

O n L 1 I L II n
0 1 00 200 300 1000 2000

FBP (uM)

Figure 3.3. FBP-activation of wild-type and the N-terminal truncated E. coli
ADP-Glc PPases. (0) Wild type, (0) NA3, (A) NA7, (I) NAll, (D) NAIS.

Saturation plots were determined by the pyrophosphorolysis reaction.

79

120*

 

 

 

 

 

 

 

 

80 .
40
:3
:>5 0:) .0 t . i
\ \ B
> 13 £3
80- -
4o -
D
0 . 1 L J . 5
0 100 200 300
AMP (uM)

Figure 3.4. AMP inhibition kinetics of wild-type and mutants E. coli
ADP-Glc PPases. All determinations were obtained by synthesis assay in
presence of(A) 30 or (B) 130 pM FBP. (0) Wild type, (D)NA15, (A)
ECCA108+EC32343|, (Cl) EcNAl 5CA108 + EC323431.

80

3.4.2. Characterization of EcNAlS-CA108 + Ec32843,

To evaluate the opposite effects on F BP activation caused by an N-terminal
deletion (187, 188) and by a nick separating the catalytic domain and the C-t as described
in Chapter 2, EcNAl 5C-A108 + Ecm,431 were co-overexpressed and purified as
described in section 3.2. As illustrated in Table 3.2, after the Green A affinity
chromatography the co-expressed proteins were purified to at least 90%. The
EcNAl 5C-A108+Ec328431 co-expression product showed a specific activity comparable
to that of the EcCA108 + Ec323-34- (see Chapter 2), the EcNAlS, and the wild-type
enzymes (Table 3.2). Two protein bands of the expected molecular sizes (i. e., ~35 kDa
and ~12 kDa) were observed in SDS-PAGE and western blot (data not shown) of purified
EcNAl 5C-A108 + E03234“. Interestingly, the kinetic properties of this co-expressed
protein were comparable to those of the wild-type enzyme regarding affinity for the

substrate ATP, affinity for the activator FBP and inhibition by AMP (Table 3.3).

Saturation kinetics for the FBP activation (Fig. 3.5) and AMP inhibition (Fig. 3.4)
indicated that the EcNAl SC-AIOS + E0328_431 co-expressed enzyme recovered the
sensitivity for the allosteric effectors that was respectively lost or decreased in the
EcNAlS or EcC-A108 + 13cm,431 mutant enzymes. Thus, the opposite effects in FBP-
activation caused by an N-terminal deletion and by a nick separating the C-t and the
catalytic domain seemed to compensate when these structural modifications were
combined in the same protein. Assuming that an allosteric enzyme in solution is present

as a mixture of conformations with different levels of activities, a 15 N-terminal residue

81

0.2.5: a 250 0 002000: .8 2%.: .0 sass: 58: Basso 0:00 a

 

 

 

ca 00 00 503: $203220.
00A we 02 0222”:
00A 0:. mm: a 2108: +0028:
00-00 soodv 80.0 v a 020%:
ca .00 :2 a 25.0.3
:5 0:02:§m 0:030:0::00::0§m
0.:80m
008 A085: 0302 2.08:0

 

.0_0w mOaaE-mam 80¢ :0:08::00 003 003800 2: :0 .35: 00—0800
BEE: 8 388.20: 003 00:3:00 05025 .8053 0_G-ma< :00 ”Q 3:33: ::0 oar-2:3 :0 00:33:00 05025 .Nd 030,—.

82

000.50,: a 0:50 0: 002000: 0.0 0:00: .0 03:30 as: 02.800 00:0 H.

£310. :0 :0:0:::00:00 :21 com 0: a: 3:000 :23 32030 003 5:328 0: a:

388.20: :0: 0::

 

 

 

 

 

 

- :e - .E 0.0 0.0 a 0.0 0: 0e 0:020:
0.0 :.0 a 000 0.: 0.0 a 0.0 0.: :.0 a 0.0:. 0.: 0.00 a 0.00:. 5-0080: + 00200220:
0.: 0.0 a 0.0:. 0.: 0.: a 0.0 0.0 0.0: a .200 0.0 :0 0. 0.0:: a 5.0080: + 00200:
0.0 0.0 a. 0.00 :.: 0.0 a 0.0 :.0 0.0 a :.00 0.: :.: a :.:00 a 000 2:3 o:
:2 32:: 0.0: :2 A21: 0.0: :2 :23 00:0 :0 32:: 0.00.

0:: :21 00: 0:: :2: 00 0800::
00:03:00 0:: .5:
80:02:: 0:20.

 

.3 :08030 0:03 0:0:0:_:::0:0: :< .8000: 20$: :30 .m 0055::— ::0 0:56:03 :0

$0000 002:8?

0:308:00: 030:5: .Gm 030:.

83

K
x

120

 

\
\

 

 

 

v / vMAX (%)

 

 

 

l 1 I J]! 1 I

200 400 600 800 1000 10000 4000
FBP (uM)

Figure 3.5. FBP activation kinetics of wild-type and mutants E. coli ADP-Gk
PPases. All determinations were obtained by synthesis assay. (0) Wild type, (D)
NA15,(A)ECCA108 + EC323.43|,(D)ECNA15CAIO8 + E0323431.

84

truncation may induce the same structural change as the allosteric activator, which drives
the equilibrium between enzyme species towards the more active one. In contrast, the
structure arrangement generated when the catalytic domain and the C-t are co-expressed
as individual polypeptides may favor a less active conformation. Combination of these
two modifications in EcNAl 5-CA108 + EC323.43| might have compensated their opposite
kinetic effects, making this mutant enzyme more similar to the wild-type enzyme. It is
noteworthy that this mutant was also sensitive to AMP inhibition, eliminating the above
suggestion of a possible AMP site disruption caused by deletion of 15 N-terminal
residues. It is more likely that EcNAl 5 is locked in a more active conformation that can

not be reversed by the inhibitor.

Integration of these biochemical results involving the N-terminal extension with
actual structural data remains to be done since the only ADPGlc PPase crystal structures
available lack a solved diffraction pattern corresponding to the first 11 N-terminal amino
acids. The structural data points to the flexible character of this region of the enzyme
which agrees with its suggested regulatory role as “allosteric switch”. It would be very
valuable to obtain direct structural information of all the mutants analyzed in the present
work. They would provide with a series of snapshots of the various conformations
achieved by the enzyme in the different activated states, allowing for a better

understanding of the allosteric regulatory mechanism.

85

CHAPTER 4

Molecular architecture of the glucose 1-phosphate site in

ADP-glucose pyrophosphorylases3

g

3
This work has been accepted for publication: C. M. Bejar, X. J in, M. A. Ballicora, and J. Preiss (2006)

Molecular architecture of the glucose l-phosphate site in ADP-glucose pyrophosphorylases. J. Biol. Chem.
1" Press.

86

4.1. ABSTRACT

The work presented in this chapter was intended to answer questions regarding
the three-dimensional organization of the Escherichia coli ADP-Glc PPase and to use that
information as a tool for the identification of potentially important residues in the active
site. Thus, a homology modeling approach to generate the monomeric structure of this
enzyme complexed with ADP-Glc and the results from a detailed study of the substrate
(glucose 1-phosphate) binding site are presented here. A set of amino acids in the model
has been identified to be in close proximity to the glucose moiety of the ADP-Glc ligand
and found to be conserved within the ADP-Glc PPase family. The role of these amino
acids, Glum, Ser2 12, Tyr2'6, Aspm, Phe‘m, Trpm, and Aspm, was studied by site
directed mutagenesis, through the characterization of the kinetic properties and thermal
stability of the designed mutants. All purified alanine mutants had one or two orders of
magnitude lower apparent affinity for GlclP, compared to the wild type, indicating that
the selected set of amino acids plays an important role in their interaction with the
substrate. Those amino acids were subjected to conservative and non-conservative
mutations to investigate the effect of size, hydrophobicity, polarity, aromaticity, or charge
on the affinity for GlclP. In this study, the architecture of the GlclP binding site is
characterized. The homology model overlaps with the GlclP site of other PPases, such as
the Pseudomonas aeruginosa dTDP-Glc PPase and the Salmonella typhi CDP-Glc PPase.
Therefore, the data reported here may have implications for other members of the

nucleotide-diphospho-glucose PPases family.

87

4.2. INTRODUCTION

The first ADP-Glc PPase crystal structure available was that of the
homotetrameric Solanum tuberosum (potato tuber) small subunit in its allosterically
inhibited form at a resolution of 2.1 A (I 70), reported in 2005. The same study also
reported the structural determination of the enzyme complexed with either ATP or ADP-
Glc at 2.6 and 2.2 A, respectively. Attempts to obtain information on the E. coli enzyme
structure through X-ray crystallography were unsuccessful. The potato tuber small
subunit has only about 33% sequence identity with the E. coli enzyme, but the similar

predicted secondary structure profiles, together with available biochemical data, suggest

that they share a common three-dimensional fold (8).

Previous chemical modification (172) and site directed mutagenesis studies (1 74)
on the E. coli ADPGlc PPase enzyme identified Lys195 as an important residue for GlclP
interaction. Replacement by other amino acids generated 100-10,000-fold increases in the
S05 for this substrate while showing all other kinetic constants at wild type levels. Later,
F u et al. reported similar results analyzing the homologous Lys'98 in the potato tuber
catalytic subunit (1 77). The proposed role of this amino acid is to form an ionic bond
between the e-amino with the negatively charged phosphate of the GlclP. Results with

hexose l-phosphate analogues, differing from the GlclP in their hydroxyl groups,

suggested that other residues in the active site participate in substrate binding (177).

88

The objective in this work has been to obtain structural information on the E. coli
ADP-Glc PPase by building a homology model and to probe a set of highly conserved
residues in the N-terminus domain possibly involved in GlclP binding. We studied the
role of Glum, Serm, Tyrm, Asp239, Phem, Trpm, and Asp276 by means of site directed
mutagenesis, kinetic characterization of the mutant enzymes and their thermal stability.
All residues were replaced by alanine and other amino acids to evaluate the importance of

size, charge or hydrophobicity on the effects observed in substrate interaction.

Because these residues are highly conserved among ADP-Glc PPases, it was of
interest to address the question of whether they are present in other pyrophosphorylases
that use GlclP as substrate. The observations made by comparison of the putative GlclP
site from our E. coli ADP-Glc PPase model and the reported crystal structures of two
pyrophosphorylases, the P. aeruginosa dTDP-Glc PPase leA (164) and the S. typhi
CDP-Glc PPase (169), have led us to pr0pose that the results presented here have

implications beyond the family of the ADP-Glc PPases.

89

4.3. MATERIALS AND METHODS

4.3.1. Materials

Oligonucleotides were synthesized and purified by the Macromolecular Facility at
Michigan State University. [32P]PPi was purchased from Perkin Elmer Life Sciences and
['4C]GlclP from ICN Pharmaceuticals Inc. NaPPi, ATP, ADP-Glc, AMP, and inorganic
pyrophosphatase were purchased from Sigma Chemical Co. Pfu DNA polymerase was

purchased from Stratagene (La Jolla, CA, USA). All other reagents were of the highest

quality available.

4.3.2. Homology modeling

Comparative (homology) modeling of the E. coli ADP-Glc PPase (residues 12-

431) was carried out with the program Modeller6vl (203-205) using the atomic
coordinates of the S. tuberosum ADP-Glc PPase small subunit chain B complexed with
ADP-Glc (Protein Data Bank code lYPZ, (1 70)) as template. Sequence alignment was
performed manually to match functionally conserved residues, predicted secondary
structures, and hydrophobicity profiles. Secondary structures were predicted using the
PHD (http://wwwl .embl-heidelberg.de/ServicesBanderQredidfproteinD and the PSI-
PRED programs (http://bioinf.cs.ucl.ac.uk/psipred/). The models were assessed by the
VERIFY_3D program (206, 207)

(11 ://shannon.mbi.ucla.edu/DOE/Services/Verifv_3D/).

90

4.3.3. Multiple sequence alignment

A multiple sequence alignment was generated using the server Clustal W
(http://wwwebi.ac.uk/clustalw) with representative ADP-Glc PPases belonging to
different bacterial and plant taxa. The sequences used were from Escherichia coli B
(NCBI accession number POA6Vl) (208), Agrobacterium tumefaciens (P39669) (209),
Synechococcus sp. WH 8102 (NP_89721 1) (210), T hermotoga maritima (Q9WY82)
(211), Streptococcus pneumoniae (Q97QS7) (212), Vibrio cholerae (Q9KLP4) (213),
Clostridium cellulolyticum (Q9L3 85) (direct submission), Geobacillus
stearothermophilus (008326) (120), Mycobacterium tuberculosis (005314) (214),
Deinococcus radiodurans (Q9RTR7) (215), Anabaena sp PC C 71 20 (P3052l) (145),
Arabidopsis thaliana -APSl small subunit- (P55228) (216), Solanum tuberosum -small
subunit- (P23509) (147), Zea mays -small subunit, endosperm- (AAK69627) (217),

Chlamydomonas reinhardtii -small subunit- ( AAF 75 83 2) (218).

4.3.4. Site directed mutagenesis

Site directed mutagenesis was done by overlap extension PCR (219). The
template was plasmid pMAB3, with the E. coli ADP-Glc PPase gene between Ndel-Sacl
sites, previously obtained in our lab (Ballicora et al., unpublished). The flanking primers
annealing with the T7 promoter and the SacI site (underlined) were
5'-TAATACGACTCACTATAGGG-3' and 5'-GATATCTGAA'ITCGAGCTC-3',

respectively. Table 4.2 shows the overlapping primers for each mutant. The final PCR

91

products were gel purified, digested with Ndel and SacI, and subcloned to obtain the
different pMAB3-single mutant plasmids. Plasmid pETEC-NA] 5-D276N was obtained
using pETEC-NA15 (see Chapter 3), as template, with the T7 promoter and T7
terminator as flanking primers and the same mutated overlapping primers used for
pMAB3-D276N (Table 4.1). All plasmids were sequenced at the MSU Genomics Facility

to confirm incorporation of only the desired mutation.

4.3.5. Bacterial strains and expression of the recombinant ADPGlc PPases

E. coli AC7ORl-504 cells that lack endogenous ADP-Glc PPase activity were
used for expression of the wild type and pMAB3-mutant enzymes as described

previously for pMLlO (147). EcNAl 5-D276N was expressed as EcNAl 5 (see Chapter 3).

4.3.6. Purification of pMAB3-single mutants

One-liter cultures of AC7OR1-504 cells transformed with the pMAB3-single
mutant plasmids, or BL21(DE3) cells transformed with pETEC-NAI 5-D276N, were
grown in 25 ug/mL kanamycin-Luna broth (1 liter) at 37 °C up to an A600 of
0.8. Induction was initiated by the addition of IPTG(1 mM final concentration), with
subsequent incubation at 25 °C for 16 h. After induction, cells were harvested, and crude
extracts were obtained as described previously (14 7). After centrifugation, the precipitate

Was resuspended in bufler A (50 mM Hepes pH 8.0, 5 mM MgC12, 0.1 mM EDTA, 10%

Sucrose). The samples were individually applied onto a DEAE-Fractogel column (EMD

92

Chemicals), and eluted with a linear NaCl gradient (0-0.5 M). The active fractions were
pooled and desalted. After this step, samples were 60-70% pure and suitable for
performing kinetic analysis. Mutants E194A/Q/D, D276A/N, W274A, Y216F and
D239N were resuspended in buffer B (bufler A plus 1.2 M ammonium sulfate), applied to
a Phenyl-Superose (FPLC, Pharmacia) column equilibrated with bufler B and eluted with
a 1.2-0.001 M linear gradient of ammonium sulfate. Further purification of the rest of the
mutants and the wild type was performed by applying the DEAE pool samples to a
MatrexTM gel Green A affinity chromatography column (Amicon Corporation, USA) and
eluting with a linear gradient of 0-2 M NaCl. The purest fractions of each enzyme were
pooled, desalted and concentrated, and after these steps the proteins were >95% pure, as

assessed by SDS-PAGE (not shown).

4.3.7. Protein methods

Protein assay, electrophoresis (SDS-PAGE) and immunoblotting were performed

using protocols previously described (167). Samples were desalted and concentrated with

Centricon-30 devices (Amicon Inc.).

93

Table 4.1. Complementary oligonucleotides used to introduce single mutations. Site directed

mutagenesis were performed by the overlap extension polymerase chain reaction. Triplets coding

for the replacing amino acid are shown in bold.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mutation Primer S’-3’ nucleotide sequence Plasmid obtained
E194A Forward CGAATTCGTTGCCAAA CCTGC pMAB3-El94A
Reverse GCAGGTTTGGCAACGAATTCG
E194D Forward CGAATTCGTTGACAAA CCTGC pMAB3-El94D
Reverse GCAGGTTIGTCAACGAATFCG
E194Q Forward CGAATTCGTTCAAAAA CCTGC pMAB3-El94Q
Reverse GCAGGT’ITTTGAACGAATTCG
812 l A Forward TCTCTGGCGGCGATGGGTATC pMAB3-S 12 l A
Reverse GATACCCATCGCCGCCAGAGA
812 l V Forward TCTCTGGCGGTGATGGGTATC pMAB3-S l 2 l V
Reverse GATACCCATCACCGCCAGAGA
S l 2 l T Forward TCTCTGGCGACCATGGGTATC pMAB3-Sl21T
Reverse GATACCCATTGGCGCCAGAGA
S l 2 I Y Forward TCTCTGGCGTATATGGGTATC pMAB3-S 121Y
Reverse GATACCCATATACGCCAGAGA
Y216A Forward ATGGGTATCGCCGTCTTTGACGCC pMAB3-Y216A
Reverse GGCGTCAAAGACGGCGATACCCAT
Y216F Forward ATGGGTATCTTCGTCTTTGACGCC pMAB3-Y216F
Reverse GGCGTCAAAGACGAAGATACCCAT
D239A Forward TCCAGCCACGCCTITGGCAAAG pMAB3-D239A
Reverse CTI'TGCCAAAGGCGTGGCTGGA
D239N and Forward TCCAGCCACAACTTTGGCAAAG Bgfiafi'gfifly
ECNA l 5-D276N Reverse CTTTGCCAAAGTTGTGGCTGGA D276N
D239E Forward TCCAGCCACGAATTTGGCAAAG pMAB3-D239E
Reverse CTI‘TGCCAAATTCGTGGCTGGA
F240A Forward TCCAGCCACGACGCGGGCAAAGA pMAB3-F24OA
Reverse TCTI‘TGCCCGCGTCGTGGCTGGA
F240M Forward AGCCACGACATGGGCAAAGAT pMAB3-F24OM
Reverse ATCTTTGCCCATGTCGTGGCT
W274A Forward GAGCCGTACGCGCGCGATGTG pMAB3-W274A
‘ Reverse CACATCGCGCGCGTACGGCTC ‘
W274F Forward GAGCCGTACTTTCGCGATGTG pMAB3-W274F
_ Reverse CACATCGCGAAAGTACGGCTC
W274L Forward GAGCCGTACTTGCGCGATGTG pMABB-W274L
__ Reverse CACATCGCGAACGTACGGCTC
D276A Forward TACTGGCGCGCGGTGGGTACC pMABB-D276A
\ Reverse GGTACCCACCGCGCGCCAGTA
D276N Forward TACTGGCGCAATGTGGGTACC pMAB3-D276N
\ Reverse GGTACCCACATTGCGCCAGTA
D276E Forward TACTGGCGCGAGGTGGGTACC pMABB-D276E
Reverse GGTACCCACCTCGCGCCAGTA

\_

 

94

4.3.8. Enzymatic assays

Assay A: Pyrophosphorolysis direction - Formation of [32P]ATP from [32P]PPi
was determined by the method of Morell et al.(123). The reaction was carried out for 10
min at 37 °C in a mixture that contained 50 mM Hepes pH 8.0, 10 mM MgCl2, 1.5 mM
[32mm (1500-2500 dpm/nmol), 4 mM ADP-Glc, 4 mM NaF, 2 mM FBP and

0.05 mg/mL BSA, plus enzyme in a total volume of 0.25 mL.

Assay B: Synthesis direction — Formation of ADP-[”C]Glc from [”C]GlclP was
determined by the method of Yep et al. (193). The reaction was carried out for 10 min at
37 °c in a mixture that contained ['4C]GlclP (~400 dpm/nmol), ATP, Mng and FBP in
varying concentrations according to the mutant enzyme assay, 50 mM Hepes pH 8.0, 1.5
units/mL pyrophosphatase, and 0.2 mg/mL BSA, plus enzyme in a total volume of 0.20

mL. One unit of enzymatic activity is one pmol of product, either [32P]ATP or ADP-

[14C]Glc, formed per min at 37 °C.

4.3.9. Kinetic characterization

Kinetic data were plotted as specific activity (U mg") vs. substrate or effector
concentration. Kinetic constants were acquired by fitting the data to the Hill equation
With a nonlinear least square formula using the program Origin TM 5.0. Hill plots were
Used to calculate the Hill coefficient m; and the kinetic constants that correspond to the

Substrate or activator concentrations giving 50% of the maximal velocity (S05) or

aetivation (A 0.5).

95

4.3.10. Thermal stability

Enzyme samples were in buffer A supplemented with BSA to 1 mg/mL in a final
volume of 100 uL. Half of the sample (50 uL) was incubated in a water bath equilibrated
at 60°C for 5 min and placed on ice immediately after. The remaining 50 [1.1. were kept on
ice as control. The enzyme activity for both, the heat-treated and control samples, were

determined in the ADP—Glc synthesis direction, as described in Assay B.

96

4.4. RESULTS

4.4.1. Homology modeling

We obtained a three-dimensional model of the E. coli ADP-Glc PPase by
comparative modeling, using the coordinates of the recently solved crystal structure of
the potato tuber small subunit ADP-Glc PPase (PDB code 1YP2) as template, as
described in Experimental Procedures (Fig. 4.1.A). While modeling is generally
guaranteed to be successful if residue identity is >40%, for lower percentages errors can
be reduced employing an accurate sequence alignment (220-222). Our two enzymes
shared only 33% residue identity, therefore the alignment was manually edited

incorporating information such as conservation of functional residues and prediction of

secondary structures.

Using Modeller6vl , we generated 143 models after several iterative refinements
of the alignment to accommodate gaps, deletions and insertions of the query sequence
respect to the template in the best possible way. We assessed the models with the
program VERIFY_3D (206, 20 7), as described in Experimental Procedures, which
evaluates the compatibility of a given residue (1D) in a certain environment (3D). A
Score below zero for a given residue means that the conformation adopted by that residue
in the model is not compatible with its surrounding environment. In our study, we
Considered only those models with all 1D-3D averaged scores above zero and, among

them, we chose the one with most similar profile to that generated by the template’s

97

Figure 4.1. Structural model of E. coli ADP-Glc PPase. A. Cartoon
representation of the monomer. The N-terminus presents a Rossmann-like fold
and holds the ADP-Glc molecule in the active site (carbons are green, all other
atoms are colored by type). The C-terminus adopts a B-helix fold and is
connected to the N-terminus by a long loop. Loops of low structural reliability,
comprising residues F90-E97 and K259-P271, are colored in black and pointed
by arrows. B. Verify_3D profile obtained from the assessment of the E. coli
ADP-Glc PPase structural model. The profile in black corresponds to our model
and the one in grey to the template crystal structure. Gaps in the template profile
correspond to gaps in the sequence alignment with the E. coli enzyme and to
stretches of amino acids not solved in the crystal structure. The two big
depressions in the E. coli profile pointed by the arrows are regions of low
structural reliability and correspond to the F90—E97 and K259-P27l loops. C.
Close-up view of the modeled active site, with a bound ADP-Glc molecule

4 . . .
I 2 '95 (white carbons), Wthh are involved

(carbons in green). Asp , Tyrl '4 and Lys
in catalysis (167), binding‘of ATP (173) and binding of'GlclP (174),

respectively, are in the active site and close to the ligand.

98

 

 

crystal structure (Fig. 4.1.8). The two profiles followed the same general trend except for
two specific regions, both corresponding to residues located in or adjacent to loops that
are not present in the template structure (indicated by arrows in Fig. 4.1.A). The first,
encompassing residues F 90-E97 in the E. coli enzyme, aligns with a region in the potato
tuber enzyme that is disordered in the crystal structure. The second loop, containing
amino acids K259-P271, is an insertion in the bacterial enzyme. Therefore, the final
conformation of these two loops in the model, which might also affect immediately
adjacent secondary structures, accounted for the differences with the template structure’s
profile. According to the model, these loops are not part of the active site nor do they

contain important conserved residues.

In agreement with the biochemical results obtained before and described in
Chapter 2, the modeled monomer shows a two domain structural organization (Fig.
4.1.A). The N-terminus of ~300 residues, presents a B-a-B motif arranged in an open
twisted B-sheet, surrounded by a-helices. It resembles the Rossmann fold typically
present in nucleotide-binding domains (223). Residues important for catalysis, Asp142
(167), and for substrates binding, Tyrl '4 for ATP (173), and Lys'95 for GlclP (174), are
located in the active site pocket in close proximity to the ADP-Glc molecule (Fig. 4.1.C),
observations that further validate the quality of our model. The C-terminus is a separate
domain folded as a B-helix and linked to the N-terminus by a long loop. The two domains
are in intimate interaction through extensive hydrophobic contacts, which supports the
requirement of a full-length polypeptide to obtain normal enzymatic activity and

regulation (see Chapter 2).

100

4.4.2. Selection of residues for analysis

The three-dimensional model complexed with ADP-Glc shows the ligand placed
in a well-defined pocket in the active site (Fig. 4.1.A) and several direct interactions
between the ligand and the enzyme are evident (Fig. 4.2). Lys'95 makes a salt bridge with
the glucose phosphate interaction that has been biochemically probed by Hill et al. (174)
in the E. coli ADP-Glc PPase and by F u et al. (1 77) analyzing the homologous residue
(Lys'gs) in the potato tuber enzyme. Additionally, the hydroxyl groups of the glucosyl
moiety of the ligand are involved in a complex net of hydrogen bonds with the enzyme.
Side chains of Glam, Asp276 and Serm, and the backbone of the latter, participate in such

interactions.

We performed a multiple sequence alignment using the catalytic subunits of
fifteen ADP-Glc PPases from several sources, each of them representative of a different
taxonomic group. Figure 4.3 depicts part of the aligned sequences, comprising residues
located in and around the putative GlclP binding domain in the N-terminus of the
protein. The residues that, in the model, appear interacting through hydrogen bonds with
the glucosyl moiety of the ligand are absolutely conserved among all ADP-Glc PPases
analyzed, suggesting that they are involved in a conserved role, such as substrate binding.
According to our structural model, other conserved residues in this region are also located

239

in the substrate-binding pocket. Based on our observations, we selected Tyrm, Asp ,

Phe240 and Trp274 to be characterized together with Glum, Serz'2 and Aspm.

101

 

Figure 4.2. E. coli ADP-Glc PPase- substrate interaction. Stereo

representation of the putative GlclP binding site showing the residues studied in
this work (white carbon) and their proposed hydrogen bonding interactions
(dashed blue lines) with the bound ADP-Glc molecule (green carbons).

102

Figure 4.3. Sequence alignment of the E. coli ADP-Glc PPase and its
homologues. The primary structures of representative bacterial ADP-Glc
PPases and catalytic subunits from photosynthetic organisms were aligned.
The region showed here encompasses residues located at and near the
putative GlclP binding site, according to our homology model. Sources
abbreviations are: Eco, Escherichia coli B; Atum, A grobacterium
tumefaciens; Synech, Synechococcus sp. WH 8102; Tmar, Thermotoga
maritima; Spneu, Streptococcus pneumoniae; Vcho, Vibrio cholerae;
Ccell, Clostridium cellulolyticum; Bstear, Geobacillus stearothermophilus;
Mtub, Mycobacterium tuberculosis; Drad, Deinococcus radiodurans; Ana,
Anabaena sp. PCC7120; Atha, Arabidopsis thaliana; Stub, Solanum
tuberosum -small subunit-; Zmay, Zea mays -small subunit, endosperm-;
Crein, Chlamydomonas reinhardtii -small subunit-. NCBI accession
numbers and references are detailed in MATERIALS AND METHODS.
Prediction of secondary structure is depicted on top (E: [i-strand; C: coil;
H: a—helix). A consensus sequence at the bottom has also been included.

Residues studied here are highlighted in grey.

103

 

Bacterial EEEBEEC -------------- CCCCCCCCCCCCEEEEEEEBEECHHHHHHHHHH

 

 

 

 

 

 

 

 

 

 

Eco 19o IEFVEKP -------------- ANPPSMPNDPSKSLASMGIYVFDADYLYELLEE 229
Atum 183 IDFIEKP -------------- ADPPGIPGNEGFALASMGIYVFHTKFLMEAVRR 222
Synech 174 KBFREKPKGDSLLEMAVDTSRFGLSANSAKERPYLASMGIYVFSRDTLFDLL-- 225
Tmar 174 voreaxp -------------- AKP ------- RSNLASLGIYVFNYEFLKKVLIE 206
Spneu 175 VEFEEKP -------------- AQP ------- KSTKASMGIYIFDWQRLRNMLVA 207
VCho 173 Tcrvaxp -------------- ADPPCIPNRPDHSLABMGIYIFNMDVLKKALTE 212
Ccell 176 ysraaxp -------------- KNP ------- KSTLASMGVYIFTWSTLREYLIK 208
Bstear 175 vsrAaxp -------------- AEP ------- KSNLASMGIYIFNWPLLKQYLQI 207
.Mtub 155 RSFVBKP —————————————— LEPPGTPDDPDTTFVSMGNYIFTTKVLIDAIRA 194
Drad 178 TEFHaxv -------------- PDPPTIPGQADLSLTSMGNYIFSRRALEELLEA 217
Ana 173 IDFSEKPKGEALTKMRVDTTVLGLTPEQAASQPYIASMGIXVFKKDVLIKLLK- 225
Plant

Atha 262 IEFAEKPKGEHLKAMKVDTTILGLDDQRAKEMPFIASMGIYVVSRDVMLDLLR- 314
Stub 193 IEFAEKPQGEQLQAMKVDTTILGLDDKRAKEMPFIASMGIYVISKDVMLNLLR- 245
Zmay 217 IEFAEKPKGEQLKAMMVDTTILGLDDVRAKEMPYIASMGIYVFSKDVMLQLLR- 269
Crein 257 IEFABKPKGEALTKMRVDTGILGVDPATAAAKPYIASMGIYVMSAKALRELLL- 309
CONSENSUS * ** .*:* *:. : . :
Bacterial CCCCCCCCCCCHHHHHHHHHHCC--CEEEEEECCEEEEEEECCC-EEBEECCCH

Eco 230 DDRDENSSHDFGKDLIPKITEAG--LAYAHPFPLSCVQSDPDAE-PYIRDVGTL 280
Atom 223 DAADPTSSRDFGKDIIPYIVEHG--KAVAHRFADSCVRSDFEHE-PYWRDVGTI 273
Synech 226 ~-DSNPGYKDPGKEVIPEALKRGD-KLKSYVFDD ------------ YflEorGTr 264
Tmar 207 DENDPNSSHDFGKDVIPRILRENLGSLYAFRFDG ------------ YWRDVGTL 248
Spneu 208 AEKSKVGMSDFGKNVIPNYLESGB-SVYAYEFSG ------------ Ynmovcrr 248
VCho 213 DAEIEQSSHDFGKDVIPKLIATG--SVFAYSFCSG--KGRVARD-CYNRDVGTI 261
Ccell 209 DNECSDSVNDFGKNIIPAMLGDGK-SMWAYQYSG ------------ YWRDVGTI 249
Bstear 208 DNANPHSSHDFGKDVIPMLLREKK-RPFAYPFEG ------------ yuxnvcrv 248
.Mtub 195 DADDDHSDHDMGGDIVPRLVADG--MAAVYDFSDNEVPGATDRDRAYURDVGTL 266
Drad 218 SISGQETGYDFGHNVIPRALSDGY-HVQAYDFHKNPIPGQ-ERPNTYWRDVGTL 269
Ana 226 ——EALERT—Q§GKEIrP-DAAKDH—NVQAYLFDD ------------ Ygagrcrr 262
Plant

Atha 315 -—NQFPGANfiicssv1PGATSLGL-RVQAYLYDG ------------ YFEDIGTI 253
Stub 246 --DKFPGANDFGSEVIPGATSLGM-RVQAYLYDG ------------ YHEDIGTI 284
Zmay 27o --EQFPEANDFGSEVIPGATSIGK-RVQAYLYDG ------------ rumorcrr 308
Crein 31o --NRMPGANDPGNEVIPGAKDAGF-KVQAFAFDG ------------ ngEDIGTv 348
CONSENSUS *:* :::* . : **.*:**:

 

104

4.4.3. Expression and purification of pMAB3—single mutants

All selected amino acids were mutated to alanine to analyze their potential role in
GlclP interaction. We performed additional mutations to investigate if the observed
effect on the affinity for GlclP was due to their shape, size, charge or aromaticity. E. coli
wild type and mutant ADP-Glc PPases were expressed and purified as described in
Experimental Procedures. They had the expected molecular weights in SDS-PAGE and
they were recognized by the anti AC70R1 E. coli ADP-Glc PPase antibody in
immunoblots (not shown). Y216A either failed to be expressed or rendered the protein
completely susceptible to proteolysis since no band 210 kDa was detected by
immunoblotting in either the soluble or the insoluble fractions of the expression cell
lysates. After the first chromatographic step, all enzymes were 60-70% pure and suitable
for kinetic characterization assays. An additional chromatographic step yielded >95%

pure enzymes, which allowed for the proper determination of their specific activities.

4.4.4. Kinetic characterization

The kinetic characteristics of the mutant enzymes were compared to those of the
wild type. All alanine mutations decreased the apparent affinity of the enzyme for the
substrate GlclP, as $0.5 values for all the mutants were one or two orders of magnitude
larger than that of the wild type (Table 4.2). The most important increments in this
kinetic parameter were observed with mutations in mutations on Glu'94 and Serm. GlclP

Saturation curves obtained for the Glu194 mutants are shown in Fig.4 as an example to

105

0.0 0 .03: 800 000..

 

106

 

 

0: 000 000 0 0000: 0. : 00 a. 0.00: .. 000::
000 :.0 0 a 0:. 0.0 0.:. a 0.0:: 00000
0.0 00 :0: 1. 0:.:.: 0.000: 00.0 a 00.0 20000
0.0 00: 00: a 000: 0.000: :0.0 a 00.0 «.0000:
00:. :0 00 a 000 0.: 0.0 n 00:0 #003
0000 0 0 n 00 :..: 0.0 a 0.000 0:003
0:0: 00 0: a 000 0.: :.00 0 0.:.00 $.00?
0000 0 :. a. 00: 0.0 0.00 a 0.00:. 20:00
0:.0 0: 0: .0 :.00 0.0 0.00 n 0.:: «.0000
0:.0 0: 00 a :.00 0.0 0.00 n 0.00: 20000
00:0 0: 00 a. 00: :.: 0.0: 0. 00:0 0000
00 :0 00: .4. :.00 0.:: 0.0 n 0.00 .0000
00 0:. 00 a 000 0.0: 0.: a 0.00 00:00,
0: 0 0 H 00 0.:00 :.0 0 0.: >000
00 :.00 :.00 n 000:. :.0 0.0 a 0.00: 00:00
:. 000 000 a 20 0.0: 0.: n :.00 >0: 00
03.: :.: :.0 .0 ::.0 :.: :.:. a 0:00 0:00
00 00 000 a :3: 0.:. 0.0 a 0.00 000:0
:.: 000 00:: a 0000 0.:. 0.:: a 0.00 90:0
0 00: 00: a 0:00 0.:.0 00.0 a 0:..0: 50:0:
000: 0 : 0 .4. 0: 0.: :..:.: 0 0.000 2:0: 003
3.25.00: 0:00- :23 0.0.0 0:00- :00:

08.0sz

65.800 520 :30

 

0000000000 020. 00500000 :.: 80:80:. a :00: 0050::
.00 00 00> 05 .3 003000: 05 .0 00:00.00 0:005:00 05 5 005.0000 0.50 :53 005050 0:03 00030580000Q .8550 008...:
0:0 2:0 0:3 .030 .m 05 00 020—0 80.5300 0:: new .5350 2.0.3.50 0:0 0053000 0502—0 «0 eaten—E00 .N... 030,—.

I]
II

 

l

100

80

60

 

40

(V/ vmax ) %

20]

 

 

   

 

 

---//
7/

:0:!

0 -- .......... . 1 . 1 .
0 100 200 30 5000 10000 15000
GIC‘l P, 11M

Figure 4.4. Steady-state kinetic measurement for GlclP dependence for wild
type and E194A, E194D and E194Q mutant enzymes. Initial velocities were
determined in the ADP-Glc synthesis direction using Assay B. For wild type (:1),
E104A (o), E194D (A) and 51040 (I), Vmax were 111.0, 4.6, 27.8 and 24.2
U/mg, respectively. Reactions for each enzyme were performed in presence of

saturating concentrations of ATP, MgClz and F BP.

107

'94 mutant enzymes. The E194A

illustrate the shift in the S05 between wild type and Glu
mutant showed a 165-fold increase (Table 4.2) compared to that of the wild type.
Therefore, we made substitutions to aspartic acid and glutamine to evaluate the
importance of the charge and the side chain size in such effect. Mutation to glutamine
increased the S05 for GlclP 85-fold, pointing out the importance of the negative charge
for substrate binding. However, mutation to aspartic acid, which also bears a negative
charge, caused a larger negative effect in this kinetic parameter (Table 4.2) highlighting
the significance of the side chain size. These two mutations, E194D and E194Q, caused a
4- and 5-fold reduced Vmax with respect to the wild type, whereas the E194A mutation
decreased it 24-fold. The apparent affinities for ATP, Mg2+, and the activator F BP were
not significantly affected by any of these mutations on Glu194 (Table 4.3). Our results
validate the hydrogen bonds observed in the structural model (Fig. 4.3) and strongly

suggest that Glu'94 is playing a role in GlclP binding.

Our structural model proposes that Serzl2 binds O3 and 04 of the sugar moiety of
the ligand through hydrogen bonds with the side chain and backbone, respectively (Fig.
4.2 and 4.6). Here, we probed the role of the side chain in GlclP binding. All Ser2'2
mutations maintained apparent affinity properties for ATP, Mg2+ and FBP at wild type
levels (Table 4.3). S212A also showed values of km similar to the wild type but it
displayed a 14-fold increased S05 for GlclP (Table 4.2). $212V and S212T caused
dramatic effects in the apparent affinity for GlclP, with 377- and 274-fold increased S05
(Table 4.2). Mutation to valine decreased the km ~16-fold, whereas mutation to threonine

did it ~2-fold, compared to the wild type. Surprisingly, replacement of the Serz'2 to a

108

tyrosine only increased the apparent affinity for GlclP 5-fold. The km in this mutant,
however, was 23l-1ower than the wild type. These results strongly suggest that Serz'2 is
located in the GlclP binding pocket and it contributes to the enzyme’s affinity for this
substrate.

276 was replaced by alanine, asparagine and glutarnic acid. The three

Asp
mutations decreased the apparent affinity for GlclP 100-, 85-, and 24-fold, respectively
(Table 4.2). Our results point out the importance of Asp276 for GlclP binding and the
significance of both the negative charge and the size of its side chain on such effect. The
analyses of these mutants suggest an additional role for Asp276 besides GlclP interaction
given that other kinetic parameters were also affected. D276A and D276N had ~1000-
fold lower Vmax than the wild type (Table 4.2), and 3.4 and ~4-fold higher S05 for ATP,
respectively (Table 4.3). Instead, D276E displayed a 3-fold decreased Vmax respect to the
wild type (Table 4.2) but a bigger change on the apparent affinity for ATP, characterized
by an 8-fold increased S05 for this substrate (Table 4.3). On the other hand, all three
mutations decreased the apparent affinity for Mg2+ ~4- to 6-fold (Table 4.3). These
results would correlate with the role of Mg“ ion chelator proposed for the homologous

280

residue, Asp , in the S. tuberosum enzyme (170).

Furthermore, the three Asp276 mutants had 5- to 15-fold higher A05 for F BP than
that of the wild type (Table 4.3). To investigate whether this residue was involved in the
activator site, we studied another mutant. Previous reports (see Chapter 3 and (188)

showed that deletions of 11 and 15 residues from the N—terminus of the E. coli

109

0.0 0 .03: 80:: 0:00 0

 

 

 

 

 

0.0 0 0 :0 0.: :.0 a. :..0 0.0 :0.0 « 0:.0 0 000::
o.m wév H w._wN Wm Nd H N.m_ Tm cod H 2.6 wohmfl
w.N_ 0.90 H vdch N6 Wm H m.m_ QM _.o H m.N 75an
OS v.0”. H 060v mé v.0 H N._ _ Wm Ned H mod (firmn—
w.m o8 H hdmm N._ 86 H :er. wd mod H wed 43.35
o._ No H Won ad _.o H YN We 56 H «Nd n—vnmkr
_.m o._ _ H o.vom M.N Nd H od w._ vod H Xv.— (JAN;
N._ _.N H 9N0. NN md H Md 0mm 3 .o H wag Seem..—
w._ .dv H 0.02 04 N6 H N40 a; c. .c H E .— (Sewn
m._ 0N. H 0.00. N.N md H 06 ad mod H end Zammfl
o.N 06— H Nu: ON .6 H NW 0; mod H 00.0 mamND
”N Cd H odo— v._ m6 H “mm m.o mod H 26 «5me
_.N OS H O62 QM 0.0 H wfi 0.0 Nod H mmd 05— N>
t ON. 9: H o. _N_ .0; mod H Nam v.0 mod H mvd >N_Nm
9o Nd H mfim m._ Nd H Wm 5.0 vod H :06 HN—Nm
9o vé H Nwm N; mod H _.m We no.9 H and >N_Nm
v; N._ H mfiw o.N Nd H _.m N._ nod H wed <N _ Nm
To «N H odm 0..— Nd H We m6 mod H :.o Oven,—
m._ 3.: H 0.0.x o.N mod H mam wd 86 H ovd Ova—m
in _.wv H o. _Nm EN 06 H 0.0. 9N 3.0 H cm:— <vam
0.. WV H Vow A: mod H 9N o._ mod H @md 090 0:3
00:00:05 223 00:00:05 9205 00:00:05 32::
20:- 0 0v. 20.0- 0 0% 200- 0 0m. 08.00:”,—
00: +002 :5.

 

110

00:00:05 :0 00:00:00: 05 .«0

0000:0006 0w0:0>0 0:: H 0000:5090 303003 00:5 :0 03: 00 0w0:0>0 05 0:000:00: 80n— dQQmRNS. QZV 04053.20: 5 0005000 00 Am 00003

00:02:: 0:005:00 05 5 38:00:00 0:03 00050000: 0000.: 0.0.0:? 2.5::— 000 00.»: 0:3 .000...» .N 05 00 0:30.520:— 03000— .m... 030,—.

ADP-Glc PPase rendered activated enzymes even in the absence of FBP with all other
kinetic parameters similar to those of the wild type. Based on these results, we combined
both the N-terminal deletion and the single mutation D276N to create EcNAlS-D276N.
The activity of the partially purified double mutant was 0.027i0.003 U/mg, similar to the
partially purified D276N single mutant (data not shown), whereas the A05 for FBP was
51 uM, similar to that of the wild type (Table 4.3). This strongly suggests that Asp276 is
not directly involved in activator binding but is a pivotal residue for the correct
interaction of the substrates with the enzyme influencing the resulting conformational
changes upon their binding.

27‘ was studied by substituting it

The role of the size and aromaticity of Trp
with alanine, leucine and phenylalanine. Mutation to alanine was characterized by a 22-
fold decrease in the apparent affinity for GlclP and did not have significant effect in the
Vmax of the enzyme (Table 4.2) or in the apparent affinities for ATP, Mg”, and FBP
(Table 4.3). We obtained similar results when a leucine was placed in that position.
Instead, all parameters remained almost unchanged compared to wild type levels when

274

Trp was replaced by a phenylalanine, indicating that aromaticity is required in that

position for proper interaction of the GlclP with the enzyme.

Tyr216 is conserved not only among ADP-Glc PPases (Fig. 4.3) but also in the
leA -Tyr'76- (Fig. 4.6.A and B) (164). Mutation to phenylalanine allowed us to study
the role, if any, of the side chain hydroxyl group in this position. The Y216F mutant

displayed a 46-fold lower apparent affinity for GlclP (Table 4.2) and it showed small

111

variations in the apparent affinities for ATP, Mg2+, and FBP, from 1 to 3-fold increase in
the respective kinetic constants (Table 4.3). However, this substitution, in which the OH
was removed caused a 10-fold decrease in the Vmax of the mutant enzyme. Our structural
model is not showing any direct interaction between this residue and the bound ADP-Glc
(Figure 4.3). Our biochemical data, however, suggest that the side chain OH plays a role
in GlclP interaction, possibly by driving the correct positioning of the substrate in the

pocket, which also affects the concomitant catalytic reaction.

Asp239 and Phe240 are also conserved residues located in close proximity to the
ligand that do not show any evident interaction with it in the three-dimensional model.
However, the D239A mutation decreased 31-fold the apparent affinity for GlclP and 11-
fold the Vmax without significant change in the other kinetic constants. Likewise, D239N
and D239E increased the S05 for GlclP 16- and lO-fold, respectively, compared to that of
the wild type. The Vmax, instead, was 2-fold lower than the wild type in the D239N
mutant and was not affected by the D239E substitution (Tables 4.2 and 4.3). Replacement
of Phe240 to alanine and methionine affected the apparent affinity for GlclP, displaying
S05 12- and 7-fold, respectively, higher than that of the wild type. No significant changes
in the Vmax and all other kinetic parameters analyzed here were observed with F240A and
F24OM (Tables 4.2 and 4.3). Together, these results suggest that Asp239 and Phe240 are
239

important residues for GlclP interaction. They also point out the significance of Asp

negatively charge side chain for proper catalytic activity.

112

4.4.5. Thermal stability

The enzymes were also studied reSpect to their thermal stability as explained in
Experimental Procedures. Wild type enzyme and all Glum, Tyrz'6 and Asp276 mutants, as
well as SZIZA, 8212V, F24OM and D239E, showed ~80-85% activity after heat
treatment (Table 4.4).

40 -
2 to alanine caused the thermal

It is interesting to note that mutation of Phe
stability of the protein to decrease 50% under the assayed conditions. This result suggests
that, in position 240 of the E. coli ADP-Glc PPase, not only hydrophobicity but also size
of the side chain is important for the enzyme to adopt a proper and heat-stable folding. A

274. Replacement of this residue by alanine and

similar situation was observed with Trp
leucine, two hydrophobic but small side chain amino acids, rendered enzymes with less
than 1% residual activity after heat treatment, whereas a Phe in that position allowed the

mutant enzyme to retain at least 50% of the activity (Table 4.4).

Mutants SZlZT and SZlZY retained 60% and 70% of their initial activities. These
mutations not only affected the apparent affinity for Glcl P, but also they affected the kw,
suggesting a structural distortion of the active site. It is possible that this side chains are
also misplacing significant structural determinants or disrupting important stabilizing
interactions in the protein. Mutations in Asp239 rendered enzymes with 2%, 59% and 84%
residual activity after heat treatment when replaced to alanine, asparagine and glutamic
acid, respectively. A negative charge is necessary in position 239 to guarantee the

stability of the enzyme at temperatures higher than the optimum for activity.

113

Table 4.4. Thermal stability of the wild type and single mutants. Enzyme activity was

measured after heat treatment (5 min at 60 °C) or control (0 °C) conditions as described in
M4 T ERIALS AND METHODS.

 

 

 

Enzyme kc... (s") % initial
actwrty
Control 60 °C
Wild type 370.0 :1: 14.4 315.8 i 12.2 85
E194A 15.43 i 0.07 13.0 i 1.3 84
E194D 92.7 i 11.7 75.0 d: 2.3 81
El94Q 80.7 i: 2.0 66.3 :1: 6.3 82
S212A 371.2 :1: 4.1 335.7 2+: 25.9 90
8212V 22.4 :t 1.9 18.0 :t 1.4 81
$212T 179.0 a: 0.7 107.0 at 11.3 60
SZlZY 1.6 :i: 0.1 1.1 :1: 0.1 70
Y216F 29.0 i 1.0 24.7 i: 1.3 85
D239A 32.7 :t 0.2 0.5 :1: 0.2 2
D239E 347.7 i 15.9 304.3 :1: 20.4 84
D239Na 152.0 :t 1 1.7 90.3 i 2.7 59
F240A 111.6 :h 2.3 56.7 :1: 9.0 51
F240Ma 254.4 :1: 13.4 207.0 :1: 3.7 86
W274A 384.0 :t 23.4 < 0.02 < 0.04
W274F 266.0 i 8.0 133.] i 5.2 50
W274L 247.5 :t 0.6 < 0.003 0.6
D276A 0.36 i: 0.01 0.307 at 0.003 85
D276N 0.37 :t 0.03 0.30 d: 0.03 82
D276E 1 12.7 :1: 4.5 89.4 a: 0.5 79

 

a: enzymes were ~70-80% pure

114

4.5. DISCUSSION

In this work, we report the first detailed characterization of the sugar-phosphate
site and the three-dimensional structure of the E. coli ADP-Glc PPase, the GlclP site. We
selected a set of residues implicated in shaping this substrate pocket by examination of
theprimary sequences of several ADP-Glc PPases and the three-dimensional structural of
the E. coli enzyme complexed with ADP-Glc obtained by homology modeling. The role
of the selected residues in binding GlclP was probed by means of site directed
mutagenesis and steady state kinetics. The kinetic characterization of the individual

mutants revealed the importance of the replaced amino acids.

Knowledge on the three-dimensional structure of the E. coli ADP-Glc PPase was
essential to understand the complex net of interactions established between the protein
and the substrate for proper binding. The first published ADP-Glc PPase crystal structure
is that of the homotetrameric potato tuber small subunit solved by J in et al. (I 70), which
we used as template to build a model of the E. coli enzyme. The sequence identity
between these two proteins is 33%, which is close to the lowest range of accepted
homology for performing modeling (222). However, the functional similarity between
our query and template proteins and a careful inspection of the sequence alignment,
which included information on predicted secondary structures and functional conserved

residues, increased the probabilities of obtaining a reliable model.

115

The E. coli ADP-Glc PPase has been subject of numerous structure-function
relationship studies, including those aimed to elucidate the functional role of individual
amino acids. Previously, Lys195 was studied (1 74) and showed a very specific effect in
GlclP interaction. The reported mutations in this residue increased 100- to 10,000- fold
the S05 for GlclP without affecting other kinetic constants. To illustrate this, data
reported for mutant K195Q have been included here in Tables 4.2 and 4.3. Those results
are consistent with a very specific role of LysI95 in the binding of GlclP, probably by
ionic interaction between the positively charged side chain e-amino and the negative
phosphate group of the GlclP (174, I 77). It is possible that the rest of the amino acids in
the substrate pocket, which are the subject of this work, interact with the sugar hydroxyls
to increase the affinity of the binding and to provide the correct positioning of the ligand

for catalysis.

The three-dimensional model of the ADP-Glc PPase that we obtained here
allowed us to visualize the spatial arrangement of a set of conserved residues potentially
involved in the interaction between the enzyme and the substrate GlclP. It has been
reported that although proteins can bind carbohydrates in many different ways, certain
amino acids show high propensity to be in a sugar binding site (224, 225). Some
examples of this are the aromatic rings that can pack against the hydrophobic face of the
sugar (226), and carboxylates that can form bidentate hydrogen bonds with two adjacent

hydroxyls of the saccharide (225). In our model, we identified Trpm, Tyr2'6 and Phem,

194 239 276

as well as Glu , Asp and Asp , some of which were showing direct contacts with

the modeled ligand ADP-Glc.

116

We also performed a close examination of the reported three-dimensional
structures of enzymes that catalyze very similar reactions as the ADP-Glc PPase. These
enzymes are the P. aeruginosa leA (164) (PDB code 1623) and the S. tyhpi CDP-Glc

PPase (I69) (PDB code 1TZF). We inspected closely their active sites and we identified

194 195 276 274

a Trp 9 as well as the catalytic ASp142

residues homologous to Glu , Lys , Asp
(1 6 7) of the ADP-Glc PPase in their active sites (Fig. 4.5.A). Interestingly, the CDP-Glc
PPase is a trimeric enzyme with three active sites formed in the interface of adjacent

monomers (169). Most of the residues contributing to the architecture of the GlclP site

178 194

belong to one of the subunits except for Glu and Lysm, homologous to Glu and

Lys195 in the E. coli ADP-Glc PPase, that are provided by the neighboring subunit (169).

The ADP-Glc PPase structural model shows the hexose moiety of ADP-Glc
largely engaged in hydrogen bonds to surrounding residues (side chains of Lys'gs, Glum,
Ser’m and Aspm, Fig. 4.2) and protein backbone (Serz'2 and Glym, Fig. 4.6). Lys195 is
interacting with the B-phosphate of the ADP-Glc molecule. This observation is validated

by the biochemical characterization reported by Hill et al. (I 74) and discussed here.
GluI94 is proposed to interact with O2 and 03 of the sugar ring by a bidentate

hydrogen bond. The Glu194 mutants displayed the biggest changes in GlclP apparent

affinity when substituted by other residues (Table 4.2). Removal of the negative charge,

117

Figure 4.5. Superposition of the amino acids in the GlclP site from
three NDP-Glc PPases. A. Superposition of residues from the E. coli
ADP-Glc PPase model with carbons in white (this work) and the crystal
structures of the P. aeruginosa leA (PDB code 1023) with carbons in
magenta(164), and the S. typhi CDP-Glc PPase (PDB code 1TZF) with
carbons in cyan (169). The ADP-Glc PPase model has a RMSD of 1.9 with

the leA, and of 2.2 A with the CDP-Glc PPase. We show the CDP-Glc

PPase active site as it is in the active enzyme, with residues Aspm, Trp232

7 ' 179‘
l 8 from the

and Asp236 belonging to one subunit and Glu and Lys
neighboring monomer. The ADP-Glc molecule was modeled in the
ADP-Glc PPase enzyme, and the Mg2+ is present in the CDP-Glc PPase
crystal structure. B. Residues homologues to Alamo, SerZI2 and TyrZ'6 in E.
coli ADP-Glc PPase (white carbons) overlay with amino acids Tyrm,
Asn188 and PheI92 in the CDP-Glc PPase (cyan carbons) and with Leu'os,
Val172 and Tyr176 in the leA (carbons in magenta), respectively. C.
Amino acids Asp239 and Phe240 in the E. coli ADP-Glc PPase (white

‘ carbons), overlay with homologous residues Thr208 and Trp209 in the CDP-

198 199 -

Glc PPase (cyan carbons), and Glu and Ile m the leA (carbons in

magenta).

118

A 0276/D236/D226

 

    
 
     

ADP—Glc

Al40/L109/
Y129

Y216/Yl77/
F192

 

as observed with the glutamine mutant, caused a large decreased in this kinetic parameter
(85-fold), suggesting its importance for substrate interaction. Still, the size of the side
chain is also essential given that substitution by an aspartic acid decreased the apparent
affinity for GlclP more than 380-fold. Given that the distance between two atoms
engaged in a hydrogen bond is crucial for the establishment of such interaction, the effect
observed with a shorter side chain in position 194 supports the existence of a hydrogen
bond between the ligand and Glum. In addition, the enzyme activity seems to be affected

by modifications at this position. It is possible that Glu194

plays a key role in positioning
the substrate in the correct orientation for catalysis, which also agrees with a critical
contribution of the size in the functionality of this residue. Our results support the central
role of Glu'94 in GlclP binding and explain the absolute conservation of this amino acid

in the ADP-Glc PPase family (Fig. 4.3) and other NDP-glucose pyrophosphorylases, such

us the leA and the CDP-Glc PPase (Fig. 4.5.A).

Ser212 may bind the GlclP through hydrogen bonds between the side chain and
the backbone and O3 and 04 of the sugar ring, respectively (Fig. 4.2 and 4.6). We probed
the role of the side chain OH group in this interaction by making conservative and non-
conservative mutations. Although in various degrees, all mutants in Ser2'2 affected the
apparent affinity for GlclP. Homology modeling of the Serz'2 mutants active site
residues complexed with ADP-Glc show that the interaction predicted in the wild type
enzyme model between LysI95 s-amino and the phosphate of the ligand, is disrupted

when Ser212 is replaced by other amino acid (Fig. 4.7). The 14-fold increase in GlclP S05

120

 

Figure 4.6. Hydrogen bond network involving Ser212 in the E. coli ADP-Glc
PPase GlclP site. Ser212 interacts with adjacent secondary structures through a
complex network of hydrogen bonds (dashed lines) involving its side chain and
backbone. This network of interactions might be important for the correct

positioning of other key residues for GlclP binding, such as Glu194 and Lysws.

121

Figure 4.7. Stick representation of the Ser212 mutants modeled active sites.
(A) Modeled active site of the wild type enzyme, depicting the interactions
proposed between Ser2'2 and the ADPGlc molecule. The side chain hydrogen
bond is not longer predicted in the modeled $212A (B), 8212Y (C), SZl2T (D),
and $212V (E) mutants. The hydrogen bond between the Lysl95 e-amine is only
predicted in the wild type (A).

122

 

123

caused by the $212A mutant might be explained by the disruption of one hydrogen bond
between the side chain and 03 of the glucose moiety of the ligand. Surprisingly, the
effect of the side chain OH provided by threonine is counteracted by the presence of an
additional methyl group, by comparison with serine. A similar situation is observed with
valine in position 212. This extra methyl group is largely disrupting the proper
conformation of the binding pocket. The model predicts that Ser212 is spatially close to

'94 and Lys'95 and as previously indicated, important

secondary structures containing Glu
in GlclP interaction. Set”2 is also largely engaged in a hydrogen bond network with
these structures (Fig. 4.6). These observations might explain that some of the mutations
in Serm2 affect the apparent affinity for GlclP as mutations in Glu194 and Lys195 did.
Surprisingly, substitution of Set212 by a bulky side chain amino acid, tyrosine, caused a
slight change in the apparent affinity for this substrate specifically. It is possible that, as
the homology model predicts the preferred orientation for the tyrosine side chain in
position 212 is the one directing the phenyl group away to the GlclP pocket, burying the
side chain against other hydrophobic side chains and stabilizing this position by a
hydrogen bond between the phenyl OH and an adjacent backbone (not shown). It is
possible that the burying of the phenyl group is causing structural arrangements, which
probably extend to other parts of the active site affecting specifically an important
catalytic residue. This would be explained by the dramatic reduction in the kw displayed

212

by the mutant SZl2Y. Therefore, the side chain of Ser might contribute to the overall

affinity for GlclP by making direct interactions with the 03 of the sugar ring and with
adjacent backbones containing important residues for the positioning of this substrate. On

12

the other hand, the model shows Ser2 peptide carbonyl group binding 04 of the hexose

124

through a hydrogen bond (Fig. 4.2 and 4.6). This interaction can also be observed in the
crystal structures of the other NDP-Glc PPases leA (1 64) and CDP-Glc PPase (165).
The peptide carbonyl groups of Val'72 in the leA and AsnI88 in the CDP-Glc PPase,
homologous to Ser212 in ADP-Glc PPase (Fig. 4.6), also make hydrogen bonds with the
substrate, implying that this interaction is important for the correct geometry of the GlclP
in the binding pocket. Apart from the specific interactions, the size of the side chain is

important for the proper architecture of the GlclP binding site.

Asp276 is important for the enzyme interaction with GlclP and it may bind the 06
of the hexose through a hydrogen bond (Fig. 4.2). Substitutions by other residues affected
the apparent affinity for this substrate ~25- to 100-fold, supporting this hypothesis.
However, Asp276 might have a broader role rather than exclusively interacting with the
GlclP molecule since the Vmax and the apparent affinity for the other substrates were also
affected by the studied mutations (Table 4.2 and 4.3). Asp276 is spatially close to the
catalytic Asp142 (167), and its homologous Asp280 in the potato tuber ADP-Glc PPase has
been proposed as Mg"+ chelator (1 70). These observations would explain why the

276

different substitutions in Asp also affected other kinetic parameters besides the GlclP

apparent affinity. In contrast to the mutations on other residues in the GlclP site, the

276 mutants (Table 4.2). The results obtained

activation by FBP was also altered in the Asp
with mutant EcNAl 5-D276N strongly suggest that this amino acid is not directly

participating in activator binding.

125

276 may be located in a hinge-like region of the active site between the ATP

Asp
and the GlclP sub-domains. Apart form interacting with the sugar ring and the Mg”, it
may also contact other residues from adjacent secondary structures establishing a
network of interactions that drives the conformational changes experienced upon binding
of the substrates. Comparison of the potato tuber ADP-Glc PPase crystal structures
complexed with ATP or ADP-Glc illustrates such sub-domain movement (1 70). The
observations drawn by Haugen and Preiss (201) would also contribute to explain the
negative effects in all the kinetic properties of the enzyme when Asp276 was mutated.
They demonstrated that: (a) ATP alone displays half-site occupancy in the
homotetrameric enzyme, (b) GlclP does not bind to the enzyme unless Mng and ATP
are present, and (c) in the presence of GlclP, ATP displays full—site occupancy. A
synergistic effect in the binding of FBP and ATP was also reported. Thus, the cooperative
properties and the heterotropic interactions between substrates and effectors (201) would

also explain the broad effect on the kinetic properties of the enzyme when this

strategically located residue is modified in its physiochemical properties.

Aromatic residues, typically Trp or Phe, are key components of several
saccharide-binding sites (226). Usually, these aromatic rings have been found involved in
stacking interactions against the face of a sugar (225 ). However, in our structural model,
none of the three aromatic residues in close proximity to the glucosyl moiety of the
ligand is orienting its side chain parallel to the sugar ring. The great conservation of

274

Trp observed among ADP-Glc PPases (Fig. 4.3) and other pyrophosphorylases (Fig

4.6.A.) might be explained by its structural role within the GlclP site. Substitution by

126

short aliphatic side chain amino acids, like alanine and leucine, affected not only the
apparent affinity for GlclP (Table 4.2), but also decreased greatly the thermal stability of

274 was mutated to a

the enzyme (Table 4.4). These effects were lesser when Trp
phenylalanine, suggesting that aromaticity is important in that position. This amino acid
might provide the necessary stacking interactions to shape the GlclP site correctly, while

establishing the proper hydrophobic interactions that increase the thermal stability of the

protein.

Tyr216 is also located close to the ligand but no evident interaction is observed
between the sugar ring and the side chain OH group. We evaluated the role of such
group in GlclP binding with the Y216F mutation, which lower the Vmax 10-fold and the
apparent affinity for this substrate 46-fold (Table 4.2). Tyrz'6 is conserved in all ADP-Glc
PPases studied so far (Fig. 4.3) and is present in the leA -Tyr'76- (Fig. 4.6.A and B).

The CDP-Glc PPase, instead, bears a phenylalanine -Phe192

- in the homologous position
but Tyrm, located in an adjacent B-strand, orients its side chain so that the OH group
overlaps with that of Tyr216 in the ADP-Glc PPase and of TyrI76 in the leA (Fig. 4.6.B).
Given the conservation of the aromatic ring in that position, it is possible that Tyr216 is
playing a structural role in the GlclP site architecture. On the other hand, the OH group
could also be making a hydrogen bond with a water molecule in direct contact with the
substrate, as observed with Tyr176 in the leA (164). This interaction might be crucial to
drive the correct positioning of the GlclP molecule for the enzymatic reaction since not

only the apparent affinity for this substrate but also the catalytic activity was affected

with the removal of the side chain OH group.

127

We also analyzed the possible roles of Asp239 and Phe240 as part of the GlclP site.
Our results with mutant D239E showed that the change in size significantly affected the
apparent affinity for GlclP and that a negative charge in position 239 is necessary to
maintain significant enzymatic activity and thermal stability (Tables 4.2 and 4.4). On the
other hand, substitutions for asparagine and alanine caused the biggest alteration in
apparent affinity for GlclP, catalytic activity (Table 4.2), and thermal stability (Table
4.4). The structures of leA and CDP-Glc PPase show other hydrogen bond donors in
the position homologous to Asp239, a glutamic acid (Glum) and a threonine (Thrzos),

'98 interacts with 02 of the

respectively (Fig. 4.6.C). Moreover, in the leA structure Glu
dTDP-Glc molecule through a bridging water molecule. Hydrogen bonds and ion pairs
with ordered water molecules are considered important interactions that increase the
thermal stability of the protein (227) and the binding affinity and specificity for the
substrate (225). Another possibility is that Asp239 indirectly interacts with the GlclP

through a solvent molecule, which is crucial for substrate positioning and enzymatic

activity.

The data obtained with Phe240 mutants demonstrate that a hydrophobic bulky
residue is needed to maintain the properties of the enzyme at wild type levels. The role of

Phe24o might be merely structural and the effects on GlclP apparent affinity may be a

cons uence of the close roximit to As 239. In the three-dimensional model, Phe240 is
eq P Y P

surrounded by a hydrophobic environment, and it is probably necessary to anchor the

239 240

100p containing Asp in the correct position. Phe is conserved in most of the ADP-

Glc PPases (Fig. 4.3), except in those from the Mycobacterium sp. taxonomic group,

128

which bear a methionine in the homologous position. Similarly, the CDP-Glc PPase
shows a tryptophan (Trpzog), whereas the leA has a smaller hydrophobic residue (Ile‘99)
(Fig. 4.6.C). These observations together with our biochemical results support the role of

Phe240 as an important structural component of the GlclP site.

In this work, we present data supporting that key amino acids in the ADP-Glc
PPase have a role in the enzyme’s affinity for GlclP. Whether establishing direct
hydrogen bonds with the hydroxyls in the sugar ring or solvent molecules, or properly
shaping the substrate pocket, they all have an important role in determining the
architecture of the GlclP site. This is the first thorough biochemical characterization of
the kind performed on the ADP-Glc PPases. The combination of biochemical data and
the information from the three-dimensional model allowed us to hypothesize on the
structural basis of the substrate binding. Comparison of our model with other NDP-Glc
PPases reveals remarkable similarities, suggesting that the architecture of the GlclP site
is conserved. Biochemical data involving the examined amino acids have not been
reported on other pyrophosphorylases to date. We believe that the results reported in this
work can be extended to other members of the NDP-glucose pyrophosphorylase family

providing new insights towards the understanding of the evolution of these enzymes.

129

CHAPTER 5

Conclusions

130

The first ADP-Glc PPase characterized was that isolated from soybean (41).
Afterwords, the homologous enzymes from several prokaryotes and photosynthetic
eukaryotes were reported (8, 9, 44, 228, 229). Understanding ADP-Glc PPases has been
of central interest due to their key role in the pathway of a-l ,4-polysaccharide
biosynthesis in bacteria (glycogen) and plants (starch). Characterization of the enzyme
from the different sources pointed out that ADP-Glc PPase is a regulatory enzyme, with
its activity being allosterically modulated by key intermediate metabolites of central
routes for carbon and energy in the corresponding cell. Thus, the study of the enzyme has
also been significantly valuable to understand protein structure-to-function relationships
as well as enzymatic regulation by allosterism, a mechanism that is present in every
metabolic system. In addition, ADP-Glc PPases proved helpful to address questions
concerning enzyme evolution, particularly those integrating structure, function and
regulation of the protein and the relationship with metabolic pathways operating in the

different organisms.

Of all the isolated ADP-Glc PPases, the one from E. coli has been the best
characterized and has served as a working model system to give insights into the rest of
the members of the ADP-Glc PPases. Despite the advance thus reached to understand the
structure and functioning of the enzyme, several queries still remained unsolved at the
time that this thesis work started. For instance [1] information on the tertiary structure of
the enzyme and how it assembled in the quaternary structure was not yet available. On
the other hand, it has been suggested that part of the N-terminal region of ADP-Glc

PPases adopt a similar fold than that of other known PPases, referred as the catalytic or

131

PPase domain (8, 9, 167). However, not only do these related enzymes either have
smaller C-terminal regions (164, 165) or their C-termini have a different enzymatic
activity (163), but ADP-Glc PPases generally present more extended N-terminal ends as
well. Considering that ADP-Glc PPases are allosterically regulated but the other NDP-
sugar PPases are not (8, 9, 163), another intriguing question has been [2] whether the role
of the N- and C-terminal regions in ADP-Glc PPases would be determining the allosteric
properties of these enzymes, which are not present in the other, unregulated and smaller,
PPases. Of special interest has been also the investigation of the structural basis of [3] the
intimate communication between the N- and C-terminal regions to regulate enzyme
activity (168) and [4] the allosteric properties determined by a short stretch of amino

acids comprising ~15-17 N-terminal residues (147, 154, 186-188).

ADP-Glc PPases (158-160), and other NDP-sugar PPases (163-165), catalyze
their respective chemical reactions following a sequential ordered kinetic mechanism,
with binding of the first substrate (a nucleoside-triphosphate) which allows for the
binding of the second substrate (a sugar l-phosphate). Individual amino acids involved in
catalysis or interacting with substrates and allosteric modulators have been identified and
their respective roles characterized in the ADP-Glc PPase (8, 9). However, [5] a more
complete analysis of the structural basis responsible for ATP and GlclP binding
(especially at the three-dimensional level) and determining affinity and specificity for

these substrates remained to be investigated in detail.

The work reported in this thesis has been motivated by questions presented, with

132

respect to the ADP-Glc PPase family, using the E. coli ADP—Glc PPase as a model
system. Because of a certain degree of structural similarity between this family of
enzymes and other PPases, some of the results presented are potentially applicable to the

family of the NDP-sugar PPases.

5.1. Structural and functional role of N- and C-termini in allosteric regulation

The work reported in Chapter 2 presents solid evidence that envisage the domain
organization of the monomer in the enzyme even when a direct atomic visualization, like
the one provided by a crystal structure, was not yet available. The ADP-Glc PPase
monomer from E. coli comprises two independent folding domains which are capable of
reconstituting an active enzyme as the wild type even when they are expressed as
independent polypeptides. The N-terminal domain, accounting for 3/4 of the total length
of the whole monomer, has been predicted to include a Rossmann-like fold (1 6 7, 181)
and contains all the important amino acids with proposed roles in catalysis (16 7, 181) or
substrate binding (167, 173, 1 74, 181). The type of fold that the C-terminus might adopt
has been more difficult to predict since related NDP-sugar PPases of known structure are
either smaller polypeptides consisting of only a conserved PPase N-terminal domain
(164, I 65), or have a C-terminus playing a different function. An example of the latter is
the case of the UDP-N-acetylglucosamine PPase (GlmU), where the C-terminus contains
the active site for N-acetyl transferase activity exhibited by this bifunctional enzyme

(163).

133

In addition, results in Chapter 2 strongly suggest that the integrity of the
C-terminus is needed to maintain a functional ADP-Glc PPase. Removal of 10 to ~100
residues from the C-terminal end yielded inactive, unfolded proteins, which were also
highly susceptible to proteolysis. One possible explanation for the requirement of a
full-length C-terminal domain to maintain a functional ADP-Glc PPase was suggested by
comparison with the related enzyme GlmU, in which the C-terminus has a structural role
in preserving the quaternary organization of the homotrimeric enzyme (163). This idea
was ruled out after the potato tuber small subunit ADP-Glc PPase crystal structure was
reported (1 70). Surprisingly, the C-terminus in the ADP-Glc PPase adopts the same fold
as in GlmU, a B-helix, but it has a shorter length (Fig. 1.2). In addition, the
oligomerization arrangement of the solved tetrameric ADP-Glc PPase structure was
reported to be unique and unprecedented (Fig. 5.1) (1 70). If the enzyme is viewed as a
dimer of dimers with each monomer labeled as A, A', B, and B', A and B make extensive
stacking contacts in an end-to-end fashion between their B-helical C-terminus, but not
along this structure as observed in the GlmU (163). Also, the big loop connecting the N-

and the C-terminus from A and B makes a big interface.

When the two domains were expressed as independent polypeptides, as reported
here in Chapter 2, the ADP-Glc PPase structure would be viewed as having a gap or nick
of five amino acids in the loop linking the two domains. Even though the primary
structure was interrupted, the enzyme was still fully active and the molecular mass of the
native enzyme determined by size exclusion chromatography was similar to that of the

wild type enzyme, indicating that the oligomeric arrangement had not been altered.

134

    
 

  

     
 
 
 
    

if". 32., ,,
hi“ my};
‘C ,

   

L \
a . £1603»! as

'11“ 11W .

     

Figure 5.1. Three-dimensional structure of the potato tuber ADP-Glc PPase.
Crystal structure of the tetrameric small subunit ADP-Glc PPase from potato
tuber (residues 10-451) (1 70). Monomers A, A’, B and B’ are colored in blue,

magenta, yellow and green, respectively. Amino acids 10- 40 in monomers A and
A’ are colored in black.

135

Instead, the activity and stability of the enzyme was highly compromised with truncations
from the C-terminal end, which possibly affected the interface between this and the
N-terminal pyrophosphorylase domain. Therefore, inter-subunit interactions other than
the end-to-end contacts between B-helices from neighboring monomers and those
between their loops connecting both domains contribute to oligomerization. Moreover,
the close contacts between the C- and N-terminus are important for enzyme activity,
possibly due to the immediacy to the ATP site in the pyrophosphorylase N-terminal

domain (I 70).

In ADPGlc PPases, the C-terminus has been also implicated in allosteric
regulation. Several key residues with activator binding roles have been identified in this
domain in the enzymes from spinach leaves (Lys440 in the small subunit) (179), potato
tuber (Lys404 and Lys“‘) (157), and from Anabaena (Lys382 and Lys‘m) (181, 182). Later
studies with chimeric enzymes between A. tumefaciens and E. coli established that the
interaction between this domain and the N-terminus is critical to determine affinity and
specificity for the activators (168). Therefore, it is possible that the activator site in these
enzymes is arranged three-dimensionally by residues belonging to both domains. These
bacterial enzymes had been subject of several chemical modification and site directed
mutagenesis studies which identified residues involved in activator binding that were
located in the N-terminus (Lys39 in E. coli and Arg32, Arg33 and Arg45 in A. tumefaciens).
Therefore, it is possible that residues from the C-terminus are positioned in close

proximity to those identified in the N-tenninus in order to shape the architecture of the

136

activator site. Amino acids playing a similar role located in this domain have not been
proposed in any plant enzyme. But the sequences of several ADP-Glc PPases show that a
Lys or an Arg is in a position homologous to the E. coli enzyme's Lys39, such as Lys40 in
potato tuber small subunit. It would be interesting to investigate whether these residues
are also involved in activator binding. Thus, the combination of the N- and C-terminus to
shape the activator site would be a common structural feature applying to other members
of the ADP-Glc PPase family and the specificity for the different allosteric activators in
each organism could be determined by specific N- and C-terminal amino acid side chains
present in that site. The structural components of the activator site in ADP-Glc PPases
from different organisms could have evolved together with the type of metabolic

pathways of carbon assimilation operating in the cell.

The allosteric properties of ADP-Glc PPases are not only established by local
structural determinants, such as the amino acids in the putative activator binding site.
Studies on the enzymes from potato tuber (186) and E. coli (187, 188) and a detailed
analysis performed on this bacterial enzyme which is reported in Chapter 3 of this thesis,
have implicated the N-terminal tail (~l-l7 amino acids) in allosteric regulation. The
enzyme became fully activated and insensitive to allosteric inhibition when those
residues were removed. However, their direct role in inhibitor binding was ruled out, as
discussed in Chapter 3. Instead, this short stretch of amino acids seem to influence the
conformations adopted by the enzyme, which range from more activated to less activated

forms, as the binding of activators and inhibitors may do.

137

The potato tuber ADP-Glc PPase small subunit Cys”, located in this short region,
forms an inter-subunit disulfide bond whose reduction is required so to respond to the
allosteric activators (154, 189). Figure 5.1. depicts the way in which the monomers are
arranged in the quaternary structure allowing the disulfide bond between monomers A
and A' to be formed in the center of the tetramer (1 70). The Cys is not conserved among
the family of the ADP-Glc PPases and the mechanism of reductive activation has been

described only in the potato tuber and Arabidopsis leaves enzymes.

Even though the N-terminal tails of the ADP-Glc PPases have been predicted as
loops, the amino acid sequence in this region is not conserved. All together, the presented
evidence points out that in this stretch of residues it is the structure rather than specific
amino acids that determines its “allosteric switch” role in regulation. The mechanism is
still unclear and direct evidence on the orientation or positioning of this region is yet
unavailable since the electron density map in the potato tuber enzyme was disordered
(1 70). However, in the three-dimensional model of the E. coli ADP-Glc PPase monomer
which comprises residues 12-431 (Fig. 5.2), the N-terminal tail (amino acids 12-19) and
the following secondary structures (residues 20-40) span along the monomer, from one
side to the other. This stretch of amino acids includes the Gly-rich loop, which is in close
proximity to the adenine ring of the modeled ADP-Glc molecule, and that was proposed
as part of the ATP binding site (170). Lys”, located at the opposite end respect to the N-
terrninal tail, is structurally close to the C-terminus and the putative activator binding site.

A direct relationship between the N-terminal tail, the ATP binding pocket and the

putative activator binding site in the interface between the C- and the N-terminal PPase

138

 

Figure 5.2. Three-dimensional model of E. coli ADP-Glc PPase monomer.
(A) Front and (B) top view of the monomer (residues 12-431) complexed with

ADP-Glc in blue. N- and C-terminus are two distinctive domains making
intimate contacts among each other. Amino acids 10—40 are in black. Lys39 side

chain is shown (carbon atoms in white and nitrogen in blue).

139

domain is observed. Thus, it is possible that the effect on the flexibility of the N-terminal
tail propagate through the continuous secondary structures affecting the active site
similarly to the effect produced by the activator binding at the other end of this secondary
structures stretch. Also, in view of the positioning of the N-terminal tails in the tetramer
(Fig. 5.1) it is likely that they interact with each other, such as in the oxidized state of the
potato tuber ADP-Glc PPase, to lock the enzyme in a less flexible structure. These types
of interactions may prevent the conformational changes that drive the equilibrium
towards a more activated state of the enzyme. Contrarily, an increase in flexibility of the
tails by reduction of the disulfide bond may allow the enzyme to undergo conformational
changes to a more active state upon the binding of the activators. Complete removal of
lS-residues in this N-terminal tail might also promote the conformational changes
shifting the equilibrium to a more active state that cannot be reversed by the allosteric

inhibitors (Chapter 3).

The only available crystallographic structure ADP-Glc PPase reported to date. was
attained in its allostericaly inhibited form. It would be beneficial for the elucidation of the
allosteric regulation mechanism to obtain crystalline structures of the enzyme in other
states of activation. For this purpose, crystallization of the wild type enzymes from the
different sources and in the presence of the various and specific allosteric activators
would be the goal. Instead, the allosteric mutants already characterized, such as the NA] 5

variant reported in Chapter 3, would result in a valuable alternative.

140

5.2. Structural basis of substrate binding

In order to answer questions regarding the three-dimensional organization of the
E. coli ADP-Glc PPase, and because all previous X-ray crystallography attempts had
been unsuccessful with this particular enzyme, a three-dimensional model was predicted
using sequence homology-based computational methods and presented in Chapter 4 of
this thesis. The template employed was the small subunit from the potato tuber ADP-Glc
PPase because, despite having only about 33% sequence identity, these enzymes have

been suggested to adopt a common three-dimensional fold (8).

Three forms of the potato tuber small subunit crystal structure have been reported:
the apo form of the enzyme, the one complexed with ATP, and the one complexed with
ADP-Glc (PDB codes 1YP2, 1YP3 and 1YP4, respectively) (1 70). In that work, only
two out of four monomers of the ATP-bound form, A and A’, bind the substrate,
observation that is in agreement with previous substrate binding studies showing that
ATP is the first substrate to bind displaying only half occupancy in the absence of the
second substrate, GlclP (159). Instead, monomers A, A’ and B are binding ADP-Glc in
the 1YP4 structure. Also, some regions in the protein, mainly loops, show great mobility
when the structures from the three forms are compared, suggesting conformational
changes upon substrate binding. To illustrate this, homology models of the apo and ATP
bound E. coli enzyme were also created here, using the methodology described in
Chapter 4 for the modeling of the ADP-Glc bound structure, and using the respective

potato tuber small subunit crystal structures as templates, 1YP2 and 1YP3. Figure 5.3

141

Figure 5.3 Overlap of the three E. coli ADP-Glc PPase models: non-ligand-,
ATP- and ADP-Glc-bound forms. Apo form of the enzyme is in red, ATP-
bound form in orange and ADP-Glc-bound form in yellow. The insets show a

170 212
-Ser ,close to the

close up of the (A) region comprising residues Val
glucose-moiety of the ADP-Glc., (B) loops in the adenine vicinity undergoing
conformational changes upon ATP (blue molecule) or ADP-Glc (cyan) binding,

Gly27-Thr31 and Gln'OS-Glyl '6.

142

 

 

 

 

 

 

 

 

 

 

143

depicts an overlap of the three models and the changes undergone by these loops are

evidenced.

The overall structure of the N-terminal PPase domain in ADP-Glc PPase contains
a Rossmann-type fold, and presents a core of a mixed seven-stranded B-sheet surrounded
by six a-helices (see Fig. 4.1.A. and C.). The active center lies in a deep pocket and is
lined by a number of polar residues that are conserved in related nucleotidyltransferases,
some of which have been implicated in either catalysis or substrate binding and affinity.
The active site pocket is delimited by two lobes. The first lobe encompasses strands [31-
[34 (Leun-Tyrms, E. coli ADP-Glc PPase numbering) which interacts primarily with the
adenine ring of ADP-Glc, whereas the second lobe includes strands [35- B7 (Lys'46-

Gly278) shaping a cavity that interacts mainly with the Glc portion of ADP-Glc.

The adenine ring, in both the ATP- and the ADP-Glc- bound forms, makes
extensive interactions with two loops: the Gly-rich loop, encompassing 627GXGXRL33,
and a loop formed by amino acids Gln'OS-Glyl '6, numbered according to the E. coli
enzyme (Fig. 5.3.B). The main chain 0 carbonyl of Glyz8 and of ArgI '5 make hydrogen
bonds with N7 and N6 of the adenine ring, respectively, and the side chains of Leu25 and
Ile72 make extensive hydrophobic interactions. Also, the pentose ring 02 and O3 bind to
Lys43 e-NH and to AspI42 ODl, respectively. There is biochemical evidence suggesting a

role for Tyrl ‘4 in ATP-binding (1 73), and mutations in Trpl '3

also implied the positioning
of this residue close to the ATP site (unpublished). Comparison of the three models show

that these residues all present different orientations whether an adenine ring is bound or

144

Figure 5.4. Conformational changes in ADP-Glc PPase active site upon
ligand binding. (A) Apo form of the enzyme, before ligand binding. (B) Binding
of ATP promotes the side chain orientation change of several amino acids in the
active site: Arg”, Arg32, Glul l l, Asnl 12, Trpl '3, Tyrl 14, Aspm, Glum, Lys'gs,

239, and Asp276, are shown here. The B-phosphate of the modeled ATP -in

Asp
blue- (modeled according to the potato tuber ADP-Glc PPase crystal structure)
partially occupies the Glc-binding site. The molecule of dTTP from the leA
crystal structure (was docked in the model to illustrate the possible positioning of
the [3- and y- phosphates, interacting with Arg32. (C) Binding of ADP-Glc causes
similar changes in the orientation of the side chain as observed with ATP, except
for residues close to the Glc-portion of the ligand that come closer to it and

interact. The most important changes are experimented by Glu194 and Lyslgs.

145

(A)

(B)

 

(C)

 

 

dTTP ADP-Glc

not. The ATP- and in the ADP-Glc-bound forms show similar changes with respect to the

apo form of the enzyme in the vicinity of the adenine ring (Fig. 5.4)

The binding of GlclP in ADP-Glc PPase occurs only when ATP has at least
occupied two active sites in the native enzyme (159). Thus, the binding of the ATP might
promote conformational changes that favor the binding of the second substrate.
Comparison of the three structures (Fig. 5.3.A) depicts the large motion in the

2'2, close to the glucose-

carbohydrate-interaction lobe, encompassing residues Valm-Ser
moiety of the ADP-Glc. Specifically important in this region are residues Lys'gs, whose
role in GlclP has been reported (174), and Glum, which has been studied in this work
and presented in Chapter 4. These two residues are both part of a conserved motif in
many NDP-sugar PPases, but only information on the role of LysI95 in the E. coli ADP-
Glc PPase (and its homologous LysI98 in the potato tuber small subunit) was available to
date. Mutations in these residues decreased their apparent affinity for GlclP ~900- and
~300-fold, respectively, suggesting their critical role in binding of this substrate. Overlay
of the structures shows them as involved in the biggest motion in the molecule upon
substrate binding (Fig. 5.3, 5.4 and 5.5). These residues are shifted out of the binding
pocket in the non-ligand-bound form of the enzyme, while they are pulled more inward in

the ATP-bound form and they are pulled in significantly in the ADP-Glc-bound form,

allowing direct interaction with the substrate.

The positioning of Lys195 and GluI94 is critical for determining the affinity for the

GlclP. The structure of the potato tuber ADP-Glc PPase small subunit complexed with

147

14
Glu9

 

194

Figure 5.5. Conformational changes experienced by Glu and Lys195 upon
ligand binding. The structures of the modeled E. coli ADP-Glc PPase’s apo
(red), ATP-bound (orange), and ADP—Glc-bound forms were overlaid to illustrate
the differences in the three-dimensional positioning respect to the Glc 1P pocket

of 61ng4 and Lys195 . ATP (blue) and ADP-Glc (cyan) were the ligands modeled.

148

ADP-Glc shows that, of the three monomers binding ADP-Glc (A, A’ and B), the ligand
molecule of B is the only one with the glucose moiety oriented and accommodated to a
correct distance to properly interact with LysI95 and Glu'94. In all other subunits, these
two residues are too far out of the binding pocket to make a direct interaction with the
substrate, which might have caused the higher mobility of the glucose portion of the
ligand and its resulting disordered electron density map. Other residues might be
involved in guaranteeing the correct positioning of Lys195 and Glum. For instance,

mutations of Ser212

, reported in Chapter 4, suggested an important role in GlclP binding
and affinity. The homology model with ADP-Glc proposes, on one hand, direct
interactions between the backbone and side chain of this amino acid and the glucose

moiety (Fig. 4.3). But more importantly, Serz'2

is proposed to be involved in a complex
network of hydrogen bonds with neighboring secondary structures (Fig. 4.6), which are
those containing Lys'95 and Glum, and experiencing the big conformational changes
upon ligand binding.

The orientation of the side chain of Asp239 is also changed upon binding of ATP
or ADP-Glc. Results presented in Chapter 4 suggest that this residue is important in

GlclP binding and it is also important for enzymatic activity and thermal stability. The

modeled structures suggest that the binding of ATP and ADP-Glc promotes the re-

239 114

orientation of Asp and Tyr side chains allowing formation of a hydrogen bond. This
interaction may be involved in coupling the conformational changes occurring in the

ATP-site to those in the GlclP site and determining the stability of the enzyme.

149

The detailed mechanism underlying the conformational changes that couple
substrate binding, which are the basis for the cooperative effects observed between
substrates, still remains unclear. However, the evidence presented in Chapter 4 of this
thesis made it possible to propose biochemical roles for key residues present in the GlclP

site of the E. coli ADP-Glc PPase and highly conserved among other PPases.

5.3. Structural basis of the catalytic mechanism

Close observation of the active site and the residues likely involved in the
catalytic mechanism of reaction have been already presented and discussed for other
NDP-sugar PPases, such as the GlmU (163), leA (164), CDP-Glc PPase (165) and
ADP-Glc PPase from potato tuber (1 70). The similarity in the kinetic mechanism and
chemical reaction catalyzed, together with the structural conservation of the amino acids
involved, allow to propose a general catalytic mechanism which applies also to the E. coli

ADP-PPase.

The chemical reaction catalyzed by these enzymes has been proposed to be of the
SNZ type and proceeds by a sequential ordered mechanism (Fig. 1.2 and 5.6). In the
ADP-Glc synthesis direction of the reaction, ATP binds to the enzyme, followed by
GlclP, which acts as a nucleophile attacking the (rt-phosphate of ATP. The B- and y-
phosphates of ATP are displaced and released as inorganic pyrophosphate. As required
for SN2 type of reactions (230), the leaving group (PP,) and the attacking nuclephile (OlP

of GlclP) reside in opposite sides of the central atom in the active site of the enzyme (Fig

150

.Em_:0:00:: 080-0350 £5 E 00:00.: 00 2003 0.:: .00 0000000500 05 0:0 000:0 000—00020 A20.0

05 00 8000 00w>x0 000500000: 0 000.500 0000 Dim < .005: 20$: 05 .3 000.0500 00:000.. 30:00.5 .e.m 0.:—arm

000020-004. 0‘ 7.08020
:0 , :0
O I
0:
IO
beamV
b\ /o

‘s

’.------‘

   

O O O b.
a. as ,0.
0‘ /O‘ /O..

cams

.12 :5.

 

151

5.4). The amino acids in the Gly-rich loop, in particular Arg”, would interact with the B-
and y-phosphates of ATP, neutralizing the negative charges and guaranteeing the proper
orientation of the PP, moiety by folding them back over the nucleotide to leave the

opposite space free for GlclP binding (Fig. 5.4.B).

The ATP molecule modeled from the ATP-bound form of the potato tuber
ADP-Glc PPase small subunit (Fig. 5.4.B), shows its B-phosphate partially occupying the
Glc-binding site and represents a conformation incompetent for catalysis (1 70). In that
structure, a sulfate ion was found occupying the position of the y-phosphate of dTTP and
CTP in the dTTP- and CTP-bound forms of leA (164) and CDP-Glc PPase (165),
respectively. Here, the molecule of dTTP was modeled and overlaid to that of ATP to
illustrate the position that the B- and y-phosphates of ATP most likely adopt when

binding to the enzyme (Fig. 5.4.B).

Therefore, the flexibility of the Gly—rich loop is possibly important in allowing
binding of the ATP and departure of PP, as well as in enzyme activity. In addition, the
particularity of the loop becomes mostly important due to the linear relationship between
this region and the allosteric regulation sites, such as the N-terminal tail at one end and
the putative allosteric activator site (close to the C-terminus) at the other end. This may
be one of the ways to explain how allosteric modulators communicate with the active site
to affect catalytic activity, hence, contributing to the understanding of the structural basis

of the allosteric regulation.

152

The correct orientation of the attacking nucleophile, the OlP of GlclP is
guaranteed by interaction of Lys195 , which undergoes a marked conformational change
observed by comparison of the apo- and ADP-Glc bound forms of the enzyme (Fig. 5.3,

5 .4 and 5.5). The electrostatic repulsions between the phosphates of ATP and GlclP are
counterbalanced by a number of basic residues located in the active site, such as Arg”,
Lys195 and Lys42 and the backbone amido groups of the Gly-rich loop. The proposed roles
of Arg'8 and Lys25 in GlmU, homologous to the E. coli ADP-Glc PPase Arg32 and Lys”,
have been reported (163), and introduction of an Arg and Lys in the homologous
positions in the regulatory large subunit of the potato tuber ADP-Glc PPase converted

this non-catalytic subunit into a catalytic one, demonstrating their key role in the reaction

(127).

A divalent cation, preferably Mg“, is required in this reaction, which could
further neutralize the negative charges. The reported structures suggest that the metal
would be coordinated by a bidentate interaction utilizing a nonbridging oxygen from the
a- and B-phosphates of ATP and the side chain oxygen atoms of two conserved residues:
Asp142 and Asp276 in E. coli ADP-Glc PPase. The role of AspI42 has been studied and
mutations on this amino acid reduced the catalytic activity of the enzyme by four orders

276

of magnitude (167). The important role of Asp , suggested from observation of the

reported structures, has been analyzed and presented in Chapter 4 of this thesis. Asp276 is
within the set of conserved residues also interacting with the glucosyl moiety of ADP-Glc

and has been thoroughly investigated here. Results indicate that not only it may be

involved in binding and affinity for GlclP but it may also play a broader role in the

153

active site, since mutations to other amino acids also affected the apparent affinities for
ATP and Mg”, as well as reducing the km by three orders of magnitude.

5.4. Final remarks and future direction

In summary, the work reported in this thesis provides solid evidence contributing
to our better understanding of the biochemistry of the ADP-Glc PPase that is fundamental
in the bacterial and plant reserve polysaccharide metabolism. A detailed analysis of the
relationship between structure, fiinction and regulation of the enzyme has been presented
integrating information regarding the structural organization at the three-dimensional
level and functionality of important regions in the protein, ranging from domains to

individual residues.

Several questions regarding the biochemistry of ADP-Glc PPases still remain
unanswered and they set the direction of future research in this field. For instance, still
intriguing are the molecular determinants of ATP and Mg2+ binding and specificity and
the conformational changes underwent upon substrate binding which are the basis of
cooperative effects between substrates. Current efforts are directed to investigate them
using molecular biology and protein chemistry approaches. In addition, the crystal
structure of the first homotetrameric bacterial ADP-Glc PPase, the one from
Agrobacterium tumefaciens, has been recently reported (231 ) however the coordinates
have not yet been deposited in the Protein Data Bank. The information provided by this
structure would be very valuable to understand the tetrameric organization of the

homotetrameric enzyme and the structural basis of the cooperative effects between

154

subunits. It would also contribute greatly to the understanding of the evolution of the

ADP-Glc PPases and other NDP-sugar PPases.

Another interesting goal is the application of the information obtained in the field
thus far for biotechnological purposes such as the increase of starch content in crops.
Manipulation of the ADP-Glc PPases in plants has proved to be an efficient way of
affecting the rate of storage polysaccharide accumulation. Following this line of research,
current efforts are focused on the characterization of the GlclP site of the potato tuber
enzyme, based on the results obtained with the E. coli enzyme, and presented here in
Chapter 4. In addition, the study of transgenic plants bearing allosteric mutant ADP-Glc
PPases, as reported before(60), would be an interesting line of research to follow. The E.
coli N-terminal truncated allosteric mutants presented in Chapter 3 provide an excellent
alternative for this type of approach. In this case, a compromise between modified
allosteric properties and heat stability would have to be achieved and interesting to be

analyzed, among several other biochemical questions.

Therefore, while problems such as the mechanisms governing activity regulation
and its evolution are still puzzling, the work on the E. coli ADP-Glc PPase presented in
this thesis has paved the way for future advances in the matter and will greatly benefit not
only the research on the family of the ADP-Glc PPases but of that of other NDP-sugar

PPases as well.

155

BIBLIOGRAPHY

156

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

Stryer, L. (1995) Biochemistry, 4th ed., Walter Freeman and Company, New york.

Voet, D., Voet, J. G., and Pratt, C. W. (2001) Fundamentals of Biochemistry
Upgrade Edition, John Wiley & Sons, Inc., New York.

Iglesias, A. A., and Preiss, J. (1992) Bacterial glycogen and plant starch
biosynthesis. Biochemical Education 20, 196-203.

Preiss, J. (1991) Biology and molecular biolology of starch synthesis and
regulation, in Oxford Surveys of Plant Molecular and Cell Biology (Miflin, J .,
Ed.) pp 59-114, Oxford University Press, Oxford, United Kingdom.

Preiss, J. (2006) Bacterial glycogen inclusions: enzymology and regulation of
synthesis, in Microbiology Monographs (Shively, J. M., Ed.) pp 71-108, Springer,
Heidelberg, Germany.

Preiss, J ., and Sivak, M. N. (1998) Starch and glycogen biosynthesis, in
Comprehensive Natural Products Chemistry (Pinto, B. M., Ed.) pp 441-495,
Pergamon Press, Oxford.

Sivak, M. N., and Preiss, J. (1998) Starch: basic science to biotechnology, in
Advances In Food and Nutrition Research (Taylor, S. L., Ed.) pp 1-199,
Academic Press, San Diego.

Ballicora, M. A., Iglesias, A. A., and Preiss, J. (2003) ADPglucose
pyrophosphorylase; a regulatory enzyme for bacterial glycogen synthesis.
Microbiol. Mol. Biol. Rev. 6 7, 213-225.

Ballicora, M. A., Iglesias, A. A., and Preiss, J. (2004) ADP-glucose
pyrophosphorylase, a regulatory enzyme for plant starch synthesis. Photosynthesis
Research 79, 1-24.

Smith, A. M., Denyer, K., and Martin, C. (1997) The synthesis of the starch
granule. Annual Review of Plant Physiology and Plant Molecular Biology 48,
449-464.

Roach, P. J. (2002) Glycogen and its metabolism. Current Molecular Medicine 2,
101-120.

157

(12)

(13)

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21)

Ball, S. G., and Morell, M. K. (2003) From bacterial glycogen to starch:
Understanding the biogenesis of the plant starch granule. Annual Review of Plant
Biology 54, 207-233.

Manners, D. J. (1991) Recent developments in our understanding of glycogen
structure. Carbohydrate Polymers 16, 37-82.

Melendez, R., Melendez-Hevia, E., and Cascante, M. (1997) How did glycogen
structure evolve to satisfy the requirement for rapid mobilization of glucose? A
problem of physical constrains in structure building. Journal of Molecular
Evolution 45, 446-455.

Strange, R. E. (1968) Bacterial “Glycogen” and survival. Nature 220, 606-607.

Preiss, J. (1984) Bacterial glycogen synthesis and its regulation. Annual Review of
Microbiology 38, 419-58.

Gibbons, R. J ., and Kapsimalis, B. (1963) Synthesis of intracellular iodophilic
polysaccharide by Streptococcus mitis. Archives of Oral Biology 8, 319-29.

Belanger, A. E., and Hatfull, G. F. (1999) Exponential-phase glycogen recycling
is essential for growth of Mycobacterium smegmatis. Journal of Bacteriology 18],
6670-6678. '

Eidels, L., and Preiss, J. (1970) Carbohydrate metabolism in Rhodopseudomonas
capsulata: enzyme titers, glucose metabolism, and polyglucose polymer
synthesis. Archives of Biochemistry and Biophysics 140, 75-89.

Kiel, J. A., Boels, J. M., Beldman, G., and Venema, G. (1994) Glycogen in
Bacillus subtilis: molecular characterization of an operon encoding enzymes

involved in glycogen biosynthesis and degradation. Molecular Microbiology 11,
203-218.

Martin, M. C., Schneider, D., Bruton, C. J ., Chater, K. F., and Hardisson, C.
(1997) A glgC gene essential only for the first of two spatially distinct phases of

glycogen synthesis in Streptomyces coelicolor A3(2). Journal of Bacteriology
179, 7784-7789.

158

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

Bonafonte, M. A., Solano, C., Sesma, B., Alvarez, M., Montuenga, L., Garcia-
Ros, D., and Gamazo, C. (2000) The relationship between glycogen synthesis,

biofilm formation and virulence in Salmonella enteritidis. FEMS Microbiology
Letters 191, 31-36.

Spatafora, G., Rohrer, K., Barnard, D., and Michalek, S. (1995) A Streptococcus
mutans mutant that synthesizes elevated levels of intracellular polysaccharide is
hypercariogenic in vivo. Infection and immunity 63, 25 56-63.

Buleon, A., Colonna, P., Planchot, V., and Ball, S. (1998) Starch granules:
structure and biosynthesis. International Journal of Biological Macromolecules
23, 85-112.

Hizukuri, S. (1995) Starch: analytical aspects, in Carbohydrates in food
(Eliansson, A. C., Ed.) pp 347-429, Marcel Dekker, New York.

Morrison, W. R., and Karkalas, J. (1990) Starch, in Methods in Plant
Biochemistry (Dey, P. M., Ed.) pp 323-352, Academic Press, London.

Sachs, J. (1887) in Lectures of the Physiology of the Plants (Ward, H. M., Ed.) pp
304-325, Clarendon Press, Oxford.

Edwards, G. E., and Walker, D. A. (1983) C3, C4: mechanism of cellular and
environmental regulation of photoshynthesis, University of California Press,
Berkeley.

Bandenhuizen, I. P. (1969) The Biogenesis of Starch Granules in Higher Plants,
Appleton-Century Crofis, New York.

Outlaw, W. H., and Manchester, J. (1979) Guard cell starch concentration
quantitatively related to stomatal aperture. Plant Physiology 64, 79-82.

Ritte, G., and Raschke, K. (2003) Metabolite export of isolated guard cell
chloroplasts of Viciafaba. New Phytologist 15 9, 195-202.

Zeiger, E., Talbott, L. D., Frechilla, S., Srivastava, A., and Zhu, J. (2002) The
guard cell chloroplast: a perspective for the twenty-first century. New Phytologist
153, 415-424.

159

(33)

(34)

(35)

(36)

(37)

(38)

(39)

(40)

(41)

(42)

(43)

Cori, G. T., and Cori, C. F. (1939) The activating effect of glycogen on the

enzymatic synthesis of glycogen from glucose 1-phosphate..]. Biol. Chem. 131,
397-398.

Devlin, T. M. (1997) in Textbook of Biochemistry with clinical correlations, John
Wiley & Son, New York.

Caputto, R., Leloir, L. F., Cardini, C. E., and Paladini, A. C. (1950) Isolation of
the coenzyme of the galactose phosphate-glucose phospahte transformation.
Journal of Biological Chemistry 184, 333-350.

Cardini, C. E., Paladini, A. C., Caputto, R., and Leloir, L. F. (1950) Uridine
diphosphate glucose: the coenzyme of the galactose-glucose phosphate
isomerization. Nature 165, 191.

Leloir, L. F. (1971) Two decades in the biosynthesis of polisaccharides. Science
172, 1299-303.

Leloir, L. F., Rongine de Fekete, M. A., and Cardini, C. E. (1961) Starch and
oligosaccharide synthesis from uridine diphosphate glucose. J. Biol. Chem. 236,
636-641.

Recondo, E., and Leloir, L. F. (1961) Adenosine diphosphate glucose and starch
synthesis. Biochemical and Biophysical Research Communications 6, 86-88.

Recondo, E., Dankert, M., and Leloir, L. F. (1963) Isolation of adenosine
diphosphate D-glucose from corn grains. Biochemical and Biophysical Research
Communications 12, 204-7.

Espada, J. (1962) Enzymic synthesis of adenosine diphosphate glucose form
glucose l-phosphate and adenosine triphosphate. Journal of Biological Chemistry
237, 3577-3581.

Leloir, L. F., and Goldemberg, S. H. (1960) Synthesis of glycogen from uridine
diphosphate glucose in liver. Journal of Biological Chemistry 235, 919-923.

Sigal, N., Cattaneo, J ., and Segel, l. H. (1964) Glycogen accumulation by wild-
type and uridine diphosphate glucose pyrophosphorylase-negative strains of
Escherichia coli. Archives of Biochemistry and Biophysics 108, 440-451.

160

(44)

(45)

(46)

(47)

(48)

(49)

(50)

(51)

(52)

(53)

Preiss, J ., Shen, L., Greenberg, E., and Gentner, N. (1966) Biosynthesis of
bacterial glycogen. IV. Activation and inhibition of the adenosine diphosphate
glucose pyrophosphorylase of Escherichia coli B. Biochemistry 5, 1833-1845.

Preiss, J. (1982) Regulation of the biosynthesis and degradation of starch. Annual
Review of Plant Physiology 5 4, 431-454.

Preiss, J. (1988) Biochemistry of the starch and its regulation, in The
Biochemistry of Plants (Preiss, J ., Ed.) pp 181-254, Academic Press, New York.

Preiss, J ., and Romeo, T. (1994) Molecular biology and regulatory aspects of
glycogen biosynthesis in bacteria, in Progress in nucleic acid research and
molecular biology (Cohn, W. E., and Moldave, K., Eds.) pp 299-329, Academic
Press, New York.

Ghosh, P., Meyer, C., Remy, E., Peterson, D., and Preiss, J. (1992) Cloning,
expression, and nucleotide sequence of glgC gene from an allosteric mutant of
Escherichia coli B. Archives of Biochemistry and Biophysics 296, 122-128.

Meyer, C. R., Bork, J. A., Nadler, S., Yirsa, J ., and Preiss, J. (1998) Site-directed
mutagenesis of a regulatory site of Escherichia coli ADP- glucose
pyrophosphorylase: the role of residue 336 in allosteric behavior. Arch. Biochem.
Biophys. 353, 152-159.

Meyer, C. R., Ghosh, P., Remy, E., and Preiss, J. (1992) Cloning, expression, and
nucleotide sequence of a mutant glgC gene from Escherichia coli B. Journal of
Bacteriology 1 74, 4509-4512.

Meyer, C., Ghosh, P., Nadler, S., and Preiss, J. (1993) Cloning, expression, and
sequence of an allosteric mutant ADPglucose pyrophosphorylase from
Escherichia coli B. Arch. Biochem. Biophys. 302, 64-71.

Preiss, J ., Sabraw, A., and Greenberg, E. (1971) An ADP-glucose
pyrophosphorylase with lower apparent affinities for substrate and effector
molecules in an Escherichia coli B mutant deficient in glycogen synthesis.
Biochem. Biophys. Res. Commun. 42, 180-186.

Ball, 8., Marianne, T., Dirick, L., Fresnoy, M., Delrue, B., and Decq, A. (1991) A
Chlamydomonas reinhardtii low-starch mutant is defective for 3-

161

(54)

(55)

(56)

(57)

(58)

(59)

(60)

(61)

(62)

phosphoglycerate activation and orthophosphate inhibition of ADP-glucose
pyrophosphorylase. Planta 185, 17-26.

Neuhaus, H. E., and Stitt, M. (1990) Control analysis of photosynthate
partitioning. Impact of reduced activity of ADP-glucose pyrophosphorylase or

plastid phosphoglucomutase on the fluxes to starch and sucrose in Arabidopsis
thaliana (L.) Heynh. . Planta 182, 445-454.

Lin, T. P., Caspar, T., Somerville, C., and Preiss, J. (1988) Isolation and
characterization of a starchless mutant of Arabidopsis thaliana L. Henyh lacking
ADPglucose pyrophosphorylase activity. Plant Physiology 86, 1131-1135.

Lin, T. P., Caspar, T., Somerville, C., and Preiss, J. (1988) A starch deficient
mutant of Arabidopsis thaliana with low ADPglucose pyrophosphorylase activity
lacks one of the two subunits of the enzyme. Plant Physiology 88.

Li, L., and Preiss, J. (1992) Characterization of ADP-glucose pyrophosphorylase
from a starch-deficient mutant of Arabidopsis thaliana (L). Carbohydrate
Research 22 7, 227-239.

Giroux, M. J ., Shaw, J. R., Barry, G. F ., Cobb, B. G., Greene, T., Okita, T. W.,
and Hannah, L. C. (1996) A single mutation that increases maize seed weight.

Proceedings of the National Academy of Sciences of the United States of America
93, 5824-5829.

Smidansky, E. D., Clancy, M., Meyer, F. D., Lanning, S. P., Blake, N. K., Talbert,
L. E., and Giroux, M. J. (2002) Enhanced ADP-glucose pyrophosphorylase

activity in wheat endosperm increases seed yield. Proceedings of the National
Academy of Sciences of the United States of America 99, 1724-1729.

Stark, D. M., Timmerrnan, K. P., Barry, G. F., Preiss, J ., and Kishore, G. M.
(1992) Role of ADPglucose pyrophosphorylase in regulating starch levels in plant
tissues. Science 258, 287-292.

Geigenberger, P., Stitt, M., and Femie, A. R. (2004) Metabolic control analysis
and regulation of the conversion of sucrose to starch in growing potato tubers
Plant Cell and Environment 27, 655-673.

Smidansky, E. D., Meyer, F. D., Blakeslee, B., Weglarz, T. E., Greene, T. W., and
Giroux, M. J. (2006) Expression of a modified ADP-glucose pyrophosphorylase

162

(63)

(64)

(65)

(66)

(67)

(68)

(69)

(70)

(71)

large subunit in wheat seeds stimulates photosynthesis and carbon metabolism.
Planta (in press).

Smidansky, E. D., Martin, J. M., Hannah, L. C., Fischer, A. M., and Giroux, M. J.
(2003) Seed yield and plant biomass increases in rice are conferred by
deregulation of endosperm ADP-glucose pyrophosphorylase. Planta 216, 656-
664.

Kumar, A., C.E., L., and Preiss, J. (1986) Biosynthesis of bacterial glycogen.
Primary structure of Escherichia coli ADPglucose:a-1,4-glucan, 4-
glucosyltransferase deduced from the nucleotide sequence of the glgA gen.
Journal of Biological Chemistry 261, 14634-14639.

Furukawa, K., Tagaya, M., Inouye, M., Preiss, J ., and F ukui, T. (1990)
Identification of lysine 15 at the active site in Escherichia coli glycogen synthase.
Conservation of Lys-X-Gly-Gly sequence in the bacterial and mammalian
enzymes. J. Biol. Chem. 265, 2086-2090.

Furukawa, K., Tagaya, M., Tanizawa, K., and Fukui, T. (1994) Identification of
Ly3277 at the active site of Escherichia coli glycogen synthase. Application of
affinity labeling combined with site-directed mutagenesis. J. Biol. Chem. 269,
868-871.

Holmes, E., Boyer, C., and Preiss, J. (1982) Immunological characterization of
Escherichia coli B glycogen synthase and branching enzyme and comparison with
enzymes from other bacteria. J. Bacteriol. 151, 1444-1453.

Holmes, E., and Preiss, J. (1979) Characterization of Escherichia coli B glycogen
synthase enzymatic reactions and products. Archives of Biochemistry and
Biophysics I 96, 436-48.

Liu, J ., and Mushegian, A. (2003) Three monophyletic superfamilies account for
the majority of the known glycosyltransferases. Protein Sci 12, 1418-1431.

MacGregor, E. A. (2002) Possible structure and active site residues of starch,
glycogen, and sucrose synthases. Journal of Protein Chemistry 21, 297-306.

Yep, A., Ballicora, M. A., and Preiss, J. (2004) The active site of the Escherichia
coli glycogen synthase is similar to the active site of retaining GT-B

163

(72)

(73)

(74)

(75)

(76)

(77)

(78)

(79)

(80)

glycosyltransferases. Biochemical and Biophysical Research Communications
316, 960-966.

Yep, A., Ballicora, M. A., Sivak, M. N., and Preiss, J. (2004) Identification and
characterization of a critical region in the glycogen synthase from Escherichia
coli. J. Biol. Chem. 279, 8359-8367.

Buschiazzo, A., Ugalde, J. E., Guerin, M. E., Shepard, W., Ugalde, R. A., and
Alzari, P. M. (2004) Crystal structure of glycogen synthase: homologous enzymes
catalyze glycogen synthesis and degradation. EMBO Journal 23, 3196-205.

Horcajada, C., Guinovart, J. J ., Fita, 1., and Ferrer, J. C. (2006) Crystal Structure
of an Archaeal Glycogen Synthase: insights into oligomerization and substrate
binding of eukaryotic glycogen synthases. J. Biol. Chem. 281, 2923-2931.

Baba, T., Nishihara, M., Mizuno, K., Kawasaki, T., Shimada, H., Kobayashi, E.,
Ohnishi, S., Tanaka, K., and Arai, Y. (1993) Identification, cDNA cloning and

gene expression of soluble starch synthase in rice (Oryza sativa L.) immature
seeds. Plant Physiology 103, 565-573.

Downton, W. J. S., and Hawker, J. (1973) Enzymes of starch and sucrose
metabolism in Zea Mays leaves. Phytochemistry 12, 1551-6.

Macdonald, F. D., and Preiss, J. (1985) Partial purification and characterization of
granule-bound starch synthases from normal and waxy maize. Plant Physiology
78, 849-852.

Ozbun, J. L., Hawker, J. S., and Preiss, J. (1971) Multiple forms of a-1,4 glucan
synthetase from spinach leaves. Biochemical and Biophysical Research
Communications 43, 631-636.

Ozbun, J. L., Hawker, J. S., Greenberg, E., Lammel, C., Preiss, J ., and Lee, E. Y.
C. (1973) Starch synthetase, phosphorylase, ADPglucose pyrophosphorylase and
UDPglucose pyrophosphorylase in developing maize kernels. Plant Physiology

51,1-5.

Edwards, A., Fulton, D. C., Hylton, C. M., Jobling, S. A., Gidley, M., Rossner,
U., Martin, C., and Smith, A. M. (1999) A combined reduction in activity of
starch synthases II and III of potato has novel effects on the starch of tubers. Plant
Journal 1 7, 251-261.

164

(81)

(82)

(83)

(84)

(85)

(86)

(87)

(33)

(39)

Jobling, S. A., Westcott, R. J ., Tayal, A., Jeffcoat, R., and Schwall, G. P. (2002)
Production of a freeze-thaw-stable potato starch by antisense inhibition of three
starch synthase genes. Nature Biotechnology 20, 295-299.

Lloyd, J. R., Landschutze, V., and Kossmann, J. (1999) Simultaneous antisense
inhibition of two starch-synthase isoforms in potato tubers leads to accumulation
of grossly modified amylopectin. Biochemical Journal 338, 515-521.

Maddelein, M. L., Libessart, N., Bellanger, F., Delrue, B., Dhulst, C.,
Vandenkoornhuyse, N., Fontaine, T., Wieruszeski, J. M., Decq, A., and Ball, S.
(1994) Toward an understanding of the biogenesis of the starch granule -
determination of granule-bound and soluble starch synthase functions in
amylopectin synthesis. Journal of Biological Chemistry 269, 25150-25157.

Visser, R. G., Somhorst, I., Kuipers, G. J ., Ruys, N. J ., Feenstra, W., and
Jacobsen, E. (1991) Inhibition of the gene for granule-bound starch synthase in
potato by antisense constructs. Molecular & General Genetics 225, 289-296.

Wattebled, F ., Buleon, A., Bouchet, B., Ral, J .-P., Lienard, L., Delvalle, D.,
Binderup, K., Dauvillee, D., Ball, S., and D'Hulst, C. (2002) Granule-bound
starch synthase: A major enzyme involved in the biogenesis of B-crystallites in
starch granules. European Journal of Biochemistry 269, 3810-3820.

Delrue, B., Fontaine, T., Routier, F., Decq, A., Wieruszeski, J. M., Van Den
Koomhuyse, N., Maddelein, M. L., Fournet, B., and Ball, S. (1992) Waxy
Chlamydomonas reinhardtii: monocellular algal mutants defective in amylose
biosynthesis and granule-bound starch synthase activity accumulate a structurally
modified amylopectin. Journal Of Bacteriology 1 74, 3612-3620.

Gao, Z., Keeling, P. L., Shibles, R., and Guan, H. (2004) Involvement of lysine-
193 of the conserved “K-T-G-G” motif in the catalysis of maize starch synthase
Ila. Archives of Biochemistry and Biophysics 42 7, 1-7.

Imparl-Radosevich, J. M., Keeling, P. L., and Guan, H. (1999) Essential arginine
residues in maize starch synthase Ila are involved in both ADP-glucose and
primer binding. F EBS Letters 45 7, 357-362.

Nichols, D. J., Keeling, P. L., Spalding, M., and Guan, H. (2000) Involvement of
conserved aspartate and glutamate residues in the catalysis and substrate binding
of maize starch synthase. Biochemistry 39, 7820-7825.

165

(90)

(91)

(92)

(93)

(94)

(95)

(96)

(97)

(98)

Nakamura, Y., Francisco, P. B. J., Hosaka, Y., Sato, A., Sawada, T., Kubo, A.,
and Fujita, N. (2005) Essential amino acids of starch synthase Ila differentiate
amylopectin structure and starch quality between japonica and indica rice
varieties. Plant Molecular Biology 58, 213-27.

Horibata, T., Nakamoto, M., F uwa, H., and Inouchi, N. (2004) Structural and
physicochemical characteristics of endosperm starches of rice cultivars recently
bred in Japan. Journal of Applied Glycoscience 51 , 303-313.

Borovsky, D., Smith, E. E., and Whelan, W. J. (1976) On the mechanism of
amylose branching by potato Q-enzyme. European Journal of Biochemistry 62,
307-312.

Baecker, P. A., Greenberg, E., and Preiss, J. (1986) Biosynthesis of bacterial
glycogen: Primary structure of Escherichia coli a-l,4-glucan: a-l,4-glucan 6-
glycosyltransferase as deduced from the nucleotide sequence of the glgB gene.
Journal of Biological Chemistry 261.

Kiel, J. A. K. W., Boels, J. M., Beldman, G., and Venema, G. (1990) Nucleotide
sequence of the Synechococcus sp. PCC7942 branching enzyme gene (glg B):
Expression in B. subtilis. Gene 89, 77-84.

Kiel, J. A. K. W., Boels, J. M., Beldman, G., and Venema, G. (1991) Molecular
cloning and nucleotide sequence of the branching enzyme gene (glg B) from
Bacillus stearothermophilus and expression in Escherichia coli and Bacillus
subtilis. Mol. Gen. Genet. 230, 136-144.

Kiel, J. A. K. W., Boels, J. M., Beldman, G., and Venema, G. (1992) The glg B
gene from the therrnophile Bacillus caldolyticus encodes a therrnolabile branching
enzyme. DNA sequence .' the journal of DNA sequencing and mapping 3, 221-
232.

Rumbak, E., Rawlings, D. E., Lindsay, G. G., and Woods, D. R. (1991)
Characterization of the Butyrivibriofibrisolvens glgB gene which encodes a
glycogen-branching enzyme with starch clearing activity. Journal of Bacteriology
I 73, 6732-6741.

Takata, H., Takaha, T., Kuriki, T., Okada, S., Takagi, M., and Imanaka, T. (1994)
Properties and active center of the thermostable branching enzyme from Bacillus
stearothermophilus. Applied and Environmental Microbiology 60, 3096-3104.

166

(99)

(100)

(101)

(102)

(103)

(104)

(105)

(106)

(107)

(108)

Takata, H., Takaha, T., Okada, S., Hizukuri, S., Takagi, M., and Imanaka, T.
(1996) Structure of the cyclic glucan produced from amylopectin by Bacillus
stearothermophilus branching enzyme. Carbohydrate Research 295, 91-101

Schwall, G. P., Safford, R., Westcott, R. J., Jeffcoat, R., Tayal, A., Shi, Y. C.,
Gidley, M. J ., and Jobling, S. A. (2000) Production of very-high-amylose potato
starch by inhibition of SBE A and B. Nature Biotechnology 18, 551-554.

Garwood, D. L., Shannon, J. C., and Creech, R. G. (1976) Starches of endosperrns
possessing different alleles at amylose-extender locus in Zea Mays L. Cereal
Chemistry 53, 355-364.

Blauth, S. L., Kim, K. N., Klucinec, J ., Shannon, J. C., Thompson, D., and
Guiltinan, M. (2002) Identification of Mutator insertional mutants of starch-
branching enzyme 1 (sbe1 ) in Zea mays L. Plant Molecular Biology 48, 287-297.

Blauth, S. L., Yao, Y., Klucinec, J. D., Shannon, J. C., Thompson, D. B., and
Guiltinan, M. J. (2001) Identification of Mutator insertional mutants of starch-
branching enzyme 2a in corn. Plant Physiol. 125, 1396-1405.

Svensson, B. (1994) Protein engineering in the a-amylase family: catalytic
mechanism, substrate specificity, and stability. Plant Molecular Biology 25, 141-
157.

Kuriki, T., Guan, H. P., Sivak, M. N., and Preiss, J. (1996) Analysis of the active
center of branching enzyme 11 from maize endosperm. Journal of Protein
Chemistry 15, 305-13.

Baba, T., Kimura, K., Mizuno, K., Etoh, H., lshida, Y., Shida, O., and Arai, Y.
(1991) Sequence conservation of the catalytic regions of amylolytic enzymes in
maize branching enzyme-I. Biochemical and Biophysical Research
Communications 181, 87-94.

Abad, M. C., Binderup, K., Rios-Steiner, J ., Ami, R. K., Preiss, J ., and Geiger, J.
H. (2002) The X-ray crystallographic structure of Escherichia coli branching
enzyme. J. Biol. Chem. 277, 42164-42170.

Kuriki, T., Stewart, D. C., and Preiss, J. (1997) Construction of chimeric enzymes
out of maize endosperrnbranching enzymes 1 and 11: Activity and properties.
Journal of Biological Chemistry 272, 28999-29004.

167

(109)

(110)

(111)

(112)

(113)

(114)

(115)

(116)

(117)

Binderup, K., Mikkelsen, R., and Preiss, J. (2000) Limited proteolysis of
branching enzyme from Escherichia coli. Archives of Biochemistry and
Biophysics 3 77, 366-371.

Binderup, K., Mikkelsen, R., and Preiss, J. (2002) Truncation of the amino
terminus of branching enzyme changes its chain transfer pattern. Archives of
Biochemistry and Biophysics 397, 279-285.

Devillers, C. H., Piper, M. E., Ballicora, M. A., and Preiss, J. (2003)
Characterization of the branching patterns of glycogen branching enzyme
truncated on the N-terminus. Archives of Biochemistry and Biophysics 418, 34-38.

Kawabata, Y., Toeda, K., Takahashi, T., Shibamoto, N., and Kobayashi, M.
(2002) Preparation of highly branched starch by glycogen branching enzyme from

Neurospora crassa N2-44 and its characterization. Journal of Applied
Glycoscience 49, 273-279.

Takata, H., Ohdan, K., Takaha, T., Kuriki, T., and Okada, S. (2003) Properties of
branching enzyme from hypertherrnophilic bacterium, Aquifex aeolicus, and its
potential for production of highly-branched cyclic dextrin. Journal of Applied
Glycoscience 50, 15-20.

Takata, H., Takaha, T., Nakamura, H., Fujii, K., Okada, S., Takagi, M., and
Imanaka, T. (1997) Production and some pmperties of a dextrin with a narrow
size distribution by the cyclization reaction of branching enzyme. Journal of
Fermentation and Bioengineering 84, 119-123.

Takata, H., Takaha, T., Okada, S., Takagi, M., and Imanaka, T. (1996)
Cyclization reaction catalyzed by branching enzyme. Journal of Bacteriology
1 78.

Preiss, J. (1999) Biosynthesis of bacterial and mammalian glycogen and plant
starch: synthesis and regulation, in Biorganic chemistry: Carbohydrates (Hecht,
S. M., Ed.) pp 489-554, Oxford University Press, Oxford.

Krisman, C. R., and Barengo, R. (1975) A precursor of glycogen biosynthesis: a-
l,4-glucan-protein. European Journal of Biochemistry 52, 1 17-123.

168

(118)

(119)

(120)

(121)

(122)

(123)

(124)

(125)

(126)

Rodriguez, I. R., and Whelan, W. J. (1985) A novel glycosyl-amino acid linkage:
rabbit-muscle glycogen is covalently linked to a protein via tyrosine. Biochemical
and Biophysical Research Communications 132, 829-836.

Iglesias, A. A., Charng, Y. Y., Ball, 8., and Preiss, J. (1994) Characterization of
the kinetic, regulatory, and structural properties of ADP-glucose
pyrophosphorylase from Chlamydomonas reinhardtii. Plant Physiology 104,
1287-1294.

Takata, H., Takaha, T., Okada, S., Takagi, M., and Imanaka, T. (1997)
Characterization of a gene cluster for glycogen biosynthesis and a
heterotetrameric ADP-glucose pyrophosphorylase from Bacillus
stearothermophilus. J. Bacteriol. 179, 4689-4698.

Igarashi, R. Y., and Meyer, C. (2000) Cloning ans sequencing of glycogen
metabolism genes from Rhodobacter spaeroides 2.4.1. Expression and
characterization of recombinant ADP-glucose pyrophosphorylase Archives of
Biochemistry and Biophysics 3 76, 47-58.

Uttaro, A. D., Ugalde, R. A., Preiss, J ., and Iglesias, A. A. (1998) Cloning and
expression of the glgC gene from Agrobacterium tumefaciens: purification and
characterization of hte ADP-glucose synthetase. Archives of Biochemistry and
Biophysics 35 7, 13-21.

Morell, M. K., Bloom, M., Knowles, V., and Preiss, J. (1987) Subunit structure of
spinach leaf ADPglucose pyrophosphorylase. Plant Physiology 85, 182-187.

Preiss, J ., and Sivak, M. N. (1998) Biochemistry, molecular biology and
regulation of starch synthesis. Genet. Eng. 20, 177-223.

Nakata, P. A., Greene, T. W., Anderson, J. M., Smith-White, B. J ., Okita, T. W.,
and Preiss, J. (1991) Comparison of the primary sequences of two potato tuber
ADPglucose pyrophosphorylase subunits. Plant Molecular Biology 17, 1089-
1093.

Frueauf, J. B., Ballicora, M. A., and Preiss, J. (2003) ADP-glucose
pyrophosphorylase from potato tuber: site-directed mutagenesis of homologous
aspartic acid residues in the small and large subunits. Plant J. 33, 503-511.

169

(127)

(128)

(129)

(130)

(131)

(132)

(133)

(134)

(135)

(136)

Ballicora, M. A., Dubay, J. R., Devillers, C. H., and Preiss, J. (2005) Resm'recting
the ancestral enzymatic role of a modulatory subunit. Journal of Biological
Chemistry 280, 10189-95.

Smith-White, B. J ., and Preiss, J. (1992) Comparison of proteins of ADP-glucose
pyrophosphorylase from diverse sources. J. Mol. Evol. 34, 449-464.

Muller-Rober, B., Nast, G., and Willmitzer, L. (1995) Isolation and expression
analysis of cDNA clones encoding a small and a large subunit of ADP-glucose
pyrophosphorylase from sugar beet. Plant Molecular Biology 27, 191 -7.

Chen, B.-Y., and Janes, H. W. (1998) Multiple forms of ADP glucose
pyrophosphorylase from tomato leaf. Physiologia Plantarum 103, 491-496.

Chen, B.-Y., Janes, H. W., and Gianfagna, T. (1998) PCR cloning and
characterization of multiple ADP-glucose pyrophosphorylase cDNAs from
tomato Plant Science 1 3 6, 59-67.

Burton, R. A., Johnson, P. E., Beckles, D. M., F incher, G. B., Jenner, H. L.,
Naldrett, M. J ., and Denyer, K. (2002) Characterization of the genes encoding the
cytosolic and plastidial forms of ADP-glucose pyrophosphorylase in wheat
endosperm. Plant Physiol. 130, 1464-1475.

Weber, H., Heim, U., Borisjuk, L., and Wobus, U. (1995) Cell-type specific,
coordinate expression of two ADP-glucose pyrophosphorylase genes in relation to
starch biosynthesis during seed development of Viciafaba L. Planta 195, 352-61.

La Cognata, U., Willmitzer, L., and Muller-Rober, B. (1995) Molecular cloning
and characterization of novel isoforms of potato ADP-glucose pyrophosphorylase.
Molecular & General Genetics 246, 538-48.

Kim, I. J., Noh, S. J., Lee, B. H., Jo, J., Kim, Y. S., and Chung, W. I. (2001)
Molecular characterization of cDNA clones for ADP-glucose pyrophosphorylase
from Citrus. Biochimica et biophysica acta 1518, 324-8.

Crevillen, P., Ventriglia, T., Pinto, F ., Orea, A., Merida, A., and Romero, J. M.
(2005) Differential pattern of expression and sugar regulation of Arabidopsis

thaliana ADP-glucose pyrophosphorylase-encoding genes. J. Biol. Chem. 280,
8143-8149.

170

(137)

(138)

(139)

(140)

(141)

(142)

(143)

(144)

(145)

(146)

Crevillen, P., Ballicora, M. A., Merida, A., Preiss, J ., and Romero, J. M. (2003)
The different large subunit isoforms of Arabidopsis thaliana ADP-glucose
pyrohosphorylase confer distinct kinetic and regulatory properties to the
heterotetrameric enzyme. Journal of Biological Chemistry 278, 28508-28515.

Preiss, J. (1973) Adenosine diphosphoryl glucose pyrophosphorylase, in The
Enzymes (Boyer, P. D., Ed.) pp 73-119, Academic Press, New York.

Preiss, J ., and Romeo, T. (1989) Physiology, biochemistry and genetics of
bacterial glycogen synthesis. Adv. Microb.Physiol. 30, 184-238.

Greenberg, E., Preiss, J. E., Van Boldrick, M., and Preiss, J. (1983) Biosynthesis
of bacterial glycogen: Activator specificity of the ADPglucose pyrophosphorylase
of Rhodopseudomonas. Archives of Biochemistry and Biophysics 220, 594-604.

Furlong, C. E., and Preiss, J. (1969) Biosynthesis of bacterial glycogen synthesis.
VII. Purification and properties of adenosine diphosphoglucose
pyrophosphorylase of Rhodospirillum rubrum. Journal of Biological Chemistry
244, 2539-2548.

Yung, S. G., and Preiss, J. (1981) Byosynthesis of bacterial glycogen: purification
and structural properties of Rhodospirillum tenue adenosine diphosphate glucose
synthetase. Journal of Bacteriology 14 7, 101-109.

Matsuno, K., Blais, T., Serio, A. W., Conway, T., Henkin, T. M., and Sonenshein,
A. L. (1999) Metabolic imbalance and sporulation in an isocitrate dehydrogenase
mutant of Bacillus subtilis. Journal of Bacteriology 181, 3382-3391.

Iglesias, A. A., and Podesta, F. E. (2005) Photosynthate formation and
partitioning in crop plants, in Handbook of photosynthesis, 2"" Ed. (Pessarakli,
M., Ed.) pp 525-545, CRC Press, Taylor & Francis Group, Boca Raton, FL.

Charng, Y. Y., Kakefuda, G., Iglesias, A. A., Buikema, W. J ., and Preiss, J.
(1992) Molecular cloning and expression of the gene encoding ADPglucose
pyrophosphorylase from the cyanobacterium Anabaena sp. strain PCC 7120.
Plant Molecular Biology 20, 37-47.

Iglesias, A. A., Kakefuda, G., and Preiss, J. (1991) Regulatory and structural
properties of the cyanobacterial ADPglucose pyrophosphorylases. Plant Physiol.
97, 1187-1195.

171

(147)

(148)

(149)

(150)

(151)

(152)

(153)

(154)

(155)

Ballicora, M. A., Laughlin, M. J ., Fu, Y., Okita, T. W., Barry, G. F., and Preiss, J.
(1995) Adenosine 5'-diphosphate-glucose pyrophosphorylase from potato tuber.
Significance of the N terminus of the small subunit for catalytic properties and
heat stability. Plant Physiol. 109, 245-251.

Iglesias, A. A., Barry, G. E, Meyer, C., Bloksberg, L., Nakata, P. A., Greene, T.,
Laughlin, M. J ., Okita, T. W., Kishore, G. M., and Preiss, J. (1993) Expression of
the potato tuber ADP-glucose pyrophosphorylase in Escherichia coli. J. Biol.
Chem. 268, 1081-1086.

Gomez-Casati, D. F ., and Iglesias, A. A. (2002) ADP-glucose pyrophosphorylase
from wheat endosperm. Purification and characterization of an enzyme with novel
regulatory properties. Planta 214, 428-434.

Hylton, C. M., and Smith, A. M. (1992) The rb mutation of peas causes structural

and regulatory changes in ADP-glucose pyrophosphorylase from developing
embryos. Plant Physiology 99, 1626-1634.

Kleczkowski, L. A., Villand, P., Luthi, E., Olsen, 0. A., and Preiss, J. (1993)
Insensitivity of barley endosperm ADP-glucose pyrophosphorylase to 3-
phosphoglycerate and orthophosphate regulation. Plant Physiol. 101, 179-186.

Plaxton, W. C., and Preiss, J. (1987) Purification and properties of nonproteolytic
degraded ADPglucose pyrophosphorylase from maize endosperm. Plant Physiol.
83, 105-112.

Rudi, H., Doan, D. N. P., and Olsen, O.-A. (1997) A (His)6-tagged recombinant
barley (Hordeum vulgare L.) endosperm ADP-glucose pyrophosphorylase
expressed in the baculovirus-insect cell system is insensitive to allosteric
regulation by 3-phosphoglycerate and inorganic phosphate. F EBS Letters 419,
124-130.

Fu, Y., Ballicora, M. A., Leykam, J. F ., and Preiss, J. (1998) Mechanism of
reductive activation of potato tuber ADP-glucose pyrophosphorylase. J. Biol.
Chem. 273, 25045-25052.

Tiessen, A., Hendriks, J. H., Stitt, M., Branscheid, A., Gibon, Y., Farre, E. M.,
and Geigenberger, P. (2002) Starch synthesis in potato tubers is regulated by post-
translational redox modification of ADP-glucose pyrophosphorylase: a novel

regulatory mechanism linking starch synthesis to the sucrose supply. Plant Cell
14, 2191-2213.

172

(156)

(157)

(158)

(159)

(160)

(161)

(162)

(163)

(164)

Hendriks, J. H. M., Kolbe, A., Gibon, Y., Stitt, M., and Geigenberger, P. (2003)
ADP-glucose pyrophosphorylase is activated by posttranslational redox-
modification in response to light and to sugars in leaves of Arabidopsis and other
plant species. Plant Physiology 133, 83 8-849.

Ballicora, M. A., Fu, Y., Nesbitt, N. M., and Preiss, J. (1998) ADP-Glucose
pyrophosphorylase from potato tubers. Site-directed mutagenesis studies of the
regulatory sites. Plant Physiol. 118, 265-274.

Paule, M. R., and Preiss, J. (1971) The kinetic mechanism of adenosine
diphosphoglucose pyrophosphorylase from Rhodospirillum rubrum. Journal of
Biological Chemistry 246, 4602-4609.

Haugen, T. H., and Preiss, J. (1979) Biosynthesis of bacterial glycogen. The
nature of the binding of substrates and effectors to ADP-glucose synthase. .1. Biol.
Chem. 254, 127-136.

Kleczkowski, L. A., Villand, P., Preiss, J ., and Olsen, 0. A. (1993) Kinetic
mechanism and regulation of ADP-glucose pyrophosphorylase from barley
(Hordeum vulgare) leaves. J. Biol. Chem. 268, 6228-6233.

Lamerz, A.-C., Haselhorst, T., Bergfeld, A. K., von Itzstein, M., and Gerardy-
Schahn, R. (2006) Molecular cloning of the Leishmania major UDP-glucose
pyrophosphorylase, functional characterization, and ligand binding analyses using
NMR spectroscopy. J. Biol. Chem. 281, 16314-16322.

Sheu, K. F., and Frey, P. A. (1978) UDP-glucose perphosphorylase.
Stereochemical course of the reaction of glucose 1-phosphate with uridine-5'[1-
thiotriphosphate]. J. Biol. Chem. 253, 3378-3380.

Brown, K., Pompeo, F., Dixon, S., Mengin-Lecreulx, D., Cambillau, C., and
Bourne, Y. (1999) Crystal structure of the bifunctional N-acetylglucosamine 1-
phosphate uridyltransferase from Escherichia coli: a paradigm for the related
pyrophosphorylase superfamily. EMBO Journal 18, 4096-4107.

Blankenfeldt, W., Asuncion, M., Lam, J. S., and Naismith, J. H. (2000) The
structural basis of the catalytic mechanism and regulation of glucose-l-phosphate
thymidylyltransferase (leA). EMBO Journal 19, 6652-6663.

173

(165)

(166)

(167)

(168)

(169)

(170)

(171)

(172)

(173)

Zuccotti, S., Zanardi, D., Rosano, C., Sturla, L., Tonetti, M., and Bolognesi, M.
(2001) Kinetic and crystallographic analyses support a sequential-ordered Bi Bi
catalytic mechanism for Escherichia coli glucose-l-phosphate
thymidylyltransferase. Journal of Molecular Biology 313, 831-843.

Koropatkin, N. M., Cleland, W. W., and Holden, H. M. (2005) Kinetic and
structural analysis of {alpha}-D-glucose-1-phosphate cytidylyltransferase from
Salmonella typhi. J. Biol. Chem. 280, 10774-10780.

Frueauf, J. B., Ballicora, M. A., and Preiss, J. (2001) Aspartate residue 142 is
important for catalysis by ADP-glucose pyrophosphorylase from Escherichia coli.
J. Biol. Chem. 276, 46319-46325.

Ballicora, M. A., Sesma, J. I., Iglesias, A. A., and Preiss, J. (2002)
Characterization of chimeric ADPglucose pyrophosphorylases of Escherichia coli
and Agrobacterium tumefaciens. Importance of the C- terminus on the selectivity
for allosteric regulators. Biochemistry 41, 9431-943 7.

Koropatkin, N. M., and Holden, H. M. (2004) Molecular structure of {alpha}-D-
Glucose-l-phosphate cytidylyltransferase from Salmonella typhi. J. Biol. Chem.
279, 44023-44029.

Jin, X., Ballicora, M. A., Preiss, J ., and Geiger, J. H. (2005) Crystal structure of
potato tuber ADP-glucose pymphosphorylase. EMBO Journal, 694-704.

Lee, Y. M., Mukherjee, S., and Preiss, J. (1986) Covalent modification of
Escherichia coli ADPglucose synthetase with 8- azido substrate analogs. Archives
of Biochemistry and Biophysics 244, 585-595.

Lee, Y. M., and Preiss, J. (1986) Covalent modification of substrate-binding sites
of Escherichia coli ADP-glucose synthetase. Isolation and structural
characterization of 8- azido-ADP-glucose-incorporated peptides. Journal of
Biological Chemistry 261, 1058-1064.

Kumar, A., Tanaka, T., Lee, Y. M., and Preiss, J. (1988) Biosynthesis of bacterial
glycogen. Use of site-directed mutagenesis to probe the role of tyrosine 114 in the
catalytic mechanism of ADP-glucose synthetase from Escherichia coli. J. Biol.

Chem. 263, 14634-14639.

174

(174)

(175)

(176)

(177)

(178)

(179)

(180)

(181)

(182)

Hill, M. A., Kaufmann, K., Otero, J ., and Preiss, J. ( 1991) Biosynthesis of
bacterial glycogen. Mutagenesis of a catalytic site residue of ADP-glucose
pyrophosphorylase from Escherichia coli. J. Biol. Chem. 266, 12455-12460.

Parsons, T. F., and Preiss, J. (1978) Biosynthesis of bacterial glycogen.
Incorporation of pyridoxal phosphate into the allosteric activator site and an ADP-
glucose- protected pyridoxal phosphate binding site of Escherichia coli B ADP-
glucose synthase. .1. Biol. Chem. 253, 6197-6202.

Parsons, T. F., and Preiss, J. (1978) Biosynthesis of bacterial glycogen. Isolation
and characterization of the pyridoxal-P allosteric activator site and the ADP-
glucose-protected pyridoxal-P binding site of Escherichia coli B ADP-glucose
synthase. J. Biol. Chem. 253, 7638-7645.

Fu, Y., Ballicora, M. A., and Preiss, J. (1998) Mutagenesis of the glucose-l-
phosphate-binding site of potato tuber ADP- glucose pyrophosphorylase. Plant
Physiol. 11 7, 989-996.

Ball, K., and Preiss, J. (1994) Allosteric sites of the large subunit of the spinach
leaf ADPglucose pyrophosphorylase. Journal of Biological Chemistry 269,
24706-2471 1.

Morell, M., Bloom, M., and Preiss, J. (1988) Affinity labeling of the allosteric
activator site(s) of spinach leaf ADP-glucose pyrophosphorylase. J. Biol. Chem.
263, 633-637.

Sheng, J ., and Preiss, J. (1997) Arginine(294) is essential for the inhibition of
Anabaena PCC 7120 ADP-glucose pyrophosphorylase by phosphate.
Biochemistry 36, 13077-13084.

Charng, Y. Y., Iglesias, A. A., and Preiss, J. (1994) Structure-function
relationships of cyanobacterial ADP-glucose pyrophosphorylase. Site-directed
mutagenesis and chemical modification of the activator-binding sites of ADP-
glucose pyrophosphorylase from Anabaena PCC 7120. J. Biol. Chem. 269,
24107-24113.

Sheng, J ., Charng, Y. Y., and Preiss, J. (1996) Site-directed mutagenesis of
lysine3 82, the activator-binding site, of ADPglucose pyrophosphorylase from
Anabaena PCC 7120. Biochemistry 35, 3115-3121.

175

(183)

(184)

(185)

(186)

(187)

(188)

(189)

(190)

(191)

Frueauf, J. B., Ballicora, M. A., and Preiss, J. (2002) Alteration of inhibitor
selectivity by site-directed mutagenesis of Arg(294) in the ADP-glucose
pyrophosphorylase from Anabaena PCC 7120. Arch. Biochem. Biophys. 400, 208-
214.

Gardiol, A., and Preiss, J. (1990) Escherichia coli E-39 ADPglucose synthetase
has different activation kinetics from the wild-type allosteric enzyme. Archives of
Biochemistry and Biophysics 280, 175-180.

Gomez-Casati, D. F., Igarashi, R. Y., Berger, C. N., Brandt, M. E., Iglesias, A. A.,
and Meyer, C. R. (2001) Identification of functionally important arnino-terminal
arginines of Agrobacterium tumefaciens ADP-glucose pyrophosphorylase by
alanine scanning mutagenesis. Biochemistry 40, 10169-10178.

Laughlin, M. J ., Chantler, S. E., and Okita, T. W. (1998) N- and C-terminal
peptide sequences are essential for enzyme assembly, allosteric, and/or catalytic
properties of ADP-glucose pyrophosphorylase. Plant J. 14, 159-168.

Wu, M. X., and Preiss, J. (1998) The N-terminal region is important for the
allosteric activation and inhibition of the Escherichia coli ADP-glucose
pyrophosphorylase. Arch. Biochem. Biophys. 358, 182-188.

Wu, M. X., and Preiss, J. (2001) Truncated forms of the recombinant Escherichia
coli ADP-glucose pyrophosphorylase: The importance of the N-terminal region
for allosteric activation and inhibition. Arch. Biochem. Biophys. 389, 159-165.

Ballicora, M. A., Frueauf, J. B., Fu, Y., Schurmann, P., and Preiss, J. (2000)
Activation of the potato tuber ADP-glucose pyrophosphorylase by thioredoxin. J.
Biol. Chem. 275, 1315-1320.

Guerin, M., and Parodi, A. J. (2003) The UDP-glucosezglycoprotein
glucosyltransferase is organized in at least two tightly bound domains from yeast
to mammals. J. Biol. Chem. 278, 20540-20546.

Hammet, A., Pike, B. L., Mitchelhill, K. 1., Teh, T., Kobe, B., House, C. M.,
Kemp, B. E., and Heierhorst, J. (2000) FHA domain boundaries of the Dunlp and
Rad53p cell cycle checkpoint kinases. FEBS Letters 4 71, 141-146.

176

(192)

(193)

(194)

(195)

(196)

(197)

(198)

(199)

(200)

Hallet, B., Sherratt, D. J ., and Hayes, F. (1997) Pentapeptide scanning
mutagenesis: random insertion of a variable five amino acid cassette in a target
protein. Nucleic Acids Research (Online) 25, 1866-7.

Yep, A., Bejar, C. M., Ballicora, M. A., Dubay, J. R., Iglesias, A. A., and Preiss,
J. (2004) An assay for adenosine 5’-diphosphate (ADP)-glucose
pyrophosphorylase that measures the synthesis of radioactive ADP-glucose with
glycogen synthase. Analytical Biochemistry 324, 52-59.

Govons, S., Vinopal, R., Ingraharn, J ., and Preiss, J. (1969) Isolation of mutants of
Escherichia coli B altered in their ability to synthesize glycogen. Journal of
Bacteriology 97, 970-972.

Lindqvist, L., Kaiser, R., Reeves, P. R., and Lindberg, A. A. (1994) Purification,
characterization, and high performance liquid chromatography assay of

Salmonella glucose-l-phosphate cytidylyltransferase from the cloned rbe gene.
J. Biol. Chem. 269, 122-126.

Ning, B., and Elbein, A. D. (2000) Cloning, expression and characterization of the
pig liver GDP-mannose pyrophosphorylase. Evidence that GDP-mannose and
GDP-Glc pyrophosphorylases are different proteins. European Journal of
Biochemistry 26 7, 6866-6874.

Peneff, C., Ferrari, P., Charrier, V., Taburet, Y., Monnier, C., Zamboni, V.,
Winter, J ., Harnois, M., F assy, F., and Bourne, Y. (2001) Crystal structures of two
human pyrophosphorylase isoforms in complexes with UDPGlc(Ga1)NAc: role of
the alternatively spliced insert in the enzyme oligomeric assembly and active site
architecture. EMBO J. 20, 6191-6202.

Weissbom, A. C., Liu, Q., Rumley, M. K., and Kennedy, E. P. (1994) UTP:
alpha-D-glucose-l-phosphate uridylyltransferase of Escherichia coli: isolation
and DNA sequence of the gal U gene and purification of the enzyme. Journal of
Bacteriology I 76, 2611-2618.

Westerholm-Parvinena, b. A., Vemosa, 1., and Serranob, L. (2000) Kinesin
subfamily UNC104 contains a FHA domain: boundaries and physicochemical
characterization. FEBS Letters 486, 285-290.

Smith, P. K., Krohn, R. I., Herrnanson, G. T., Mallia, A. K., Gartner, F. H.,
Provenzano, M. D., Fujimoto, E. K., Goeke, N. M., Olson, B. J ., and Klenk, D. C.
(1985) Measurement of protein using bicinchoninic acid [published erratum

177

(201)

(202)

(203)

(204)

(205)

(206)

(207)

(208)

(209)

(210)

appears in Anal Biochem 1987 May 15;163(1):279]. Analytical Biochemistry [50,
76-85.

Haugen, T. H., Ishaque, A., and Preiss, J. (1976) Biosynthesis of bacterial
glycogen. Characterization of the subunit structure of Escherichia coli B glucose-

1-phosphate adenylyltransferase (EC 2.7.7.27). Journal of Biological Chemistry
251, 7880-7885.

Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the
head of bacteriophage T4. Nature 22 7, 680-685.

Fiser, A., Do, R. K., and Sali, A. (2000) Modeling of loops in protein structures.
Protein Science, 1753-1773.

Marti-Renom, M. A., Stuart, A., Fiser, A., Sénchez, R., Melo, F., and Sali, A.
(2000) Comparative protein structure modeling of genes and genomes. Annu. Rev.
Biophys. Biomol. Struct, 291-325.

Sali, A., and Blundell, T. L. (1993) Comparative protein modelling by satisfaction
of spatial restraints. J. Mol. Biol, 779-815.

Luthy, R., Bowie, J ., and Eisenberg, D. (1992) Assessment of protein models with
three-dimensional profiles. Nature 356, 83-5.

Bowie, J ., Luthy, R., and Eisenberg, D. (1991) A method to identify protein
sequences that fold into a known three-dimensional structure. Science 253, 164-
70.

Baecker, P. A., Furlong, C. E., and Preiss, J. (1983) Biosynthesis of bacterial
glycogen. Primary structure of Escherichia coli ADP-glucose synthetase as
deduced from the nucleotide sequence of the glgC gene. J. Biol. Chem. 258, 5084-
5088.

Uttaro, A. D., and Ugalde, R. A. (1994) A chromosomal cluster of genes encoding
ADP-glucose synthetase, glycogen synthase and phosphoglucomutase in
Agrobacterium tumefaciens. Gene 150, 117-122.

Palenik, B., Brahamsha, B., Larimer, F. W., Land, M., Hauser, L., Chain, P.,
Lamerdin, J ., Regala, W., Allen, E. E., McCarren, J ., Paulsen, I., Dufresne, A.,

178

(211)

(212)

(213)

(214)

(215)

Partensky, F., Webb, E. A., and Waterbury, J. (2003) The genome of a motile
marine Synechococcus. Nature 424 1037-1042.

Nelson, K. E., Clayton, R. A., Gill, S. R., Gwinn, M. L., Dodson, R. J ., Haft, D.
H., Hickey, E. K., Peterson, J. D., Nelson, W. C., Ketchum, K. A., McDonald, L.,
Utterback, T. R., Malek, J. A., Linher, K. D., Garrett, M. M., Stewart, A. M.,
Cotton, M. D., Pratt, M. 8., Phillips, C. A., Richardson, D., Heidelberg, J ., Sutton,
G. G., Fleischmann, R. D., Eisen, J. A., White, 0., Salzberg, S. L., Smith, H. O.,
Venter, J. C., and Fraser, C. M. (1999) Evidence for lateral gene transfer between

Archaea and bacteria from genome sequence of Thermotoga maritima. Nature
399, 323-329.

Tettelin, H., Nelson, K. E., Paulsen, I. T., Eisen, J. A., Read, T. D., Peterson, 8.,
Heidelberg, J ., DeBoy, R. T., Haft, D. H., Dodson, R. J ., Durkin, A. S., Gwinn,
M., Kolonay, J. F., Nelson, W. C., Peterson, J. D., Umayam, L. A., White, 0.,
Salzberg, S. L., Lewis, M. R., Radune, D., Holtzapple, E., Khouri, H., Wolf, A.
M., Utterback, T. R., Hansen, C. L., McDonald, L. A., Feldblyum, T. V.,
Angiuoli, S., Dickinson, T., Hickey, E. K., Holt, 1. E., Loftus, B. J ., Yang, F.,
Smith, H. O., Venter, J. C., Dougherty, B. A., Morrison, D. A., Hollingshead, S.
K., and Fraser, C. M. (2001) Complete genome sequence of a virulent isolate of
Streptococcus pneumoniae. Science 293, 498-506.

Heidelberg, J. F., Eisen, J. A., Nelson, W. C., Clayton, R. A., Gwinn, M. L.,
Dodson, R. J ., Haft, D. H., Hickey, E. K., Peterson, J. D., Umayam, L., Gill, S. R.,
Nelson, K. B, Read, T. D., Tettelin, H., Richardson, D., Ermolaeva, M. D.,
Vamathevan, J ., Bass, 8., Qin, H., Dragoi, 1., Sellers, P., McDonald, L.,
Utterback, T., Fleishmann, R. D., Nierrnan, W. C., White, 0., Salzberg, S. L.,
Smith, H. O., Colwell, R. R., Mekalanos, J. J ., Venter, J. C., and Fraser, C. M.
(2000) DNA sequence of both chromosomes of the cholera pathogen Vibrio
cholerae. Nature 406 477-483.

Fleischmann, R. D., Alland, D., Eisen, J. A., Carpenter, L., White, 0., Peterson, J.
D., DeBoy, R. T., Dodson, R. J ., Gwinn, M. L., Haft, D. H., Hickey, E. K.,
Kolonay, J. F ., Nelson, W. C., Umayam, L. A., Ermolaeva, M. D., Salzberg, S. L.,
Delcher, A., Utterback, T. R., Weidman, J. F ., Khouri, H. M., Gill, J ., Mikula, A.,
Bishai, W., Jacobs, W. R., Jr., Venter, J. C., and Fraser, C. M. (2002) Whole-
genome comparison of Mycobacterium tuberculosis clinical and laboratory
strains. J. Bacteriol. 184, 5479-5490.

White, 0., Eisen, J. A., Heidelberg, J ., Hickey, E. K., Peterson, D., Dodson, R. J .,
Hafi, D. H., Gwinn, M., Nelson, W. C., Richardson, D., Moffat, K. S., Qin, H.,
Jiang, L., Parnphile, W., Crosby, M., Sheng, M., Vamathevan, J ., Lam, P.,
McDonald, L., Utterback, T., Zalewski, C., Makarova, K. S., Aravind, L., Daly,

179

(216)

(217)

(218)

(219)

(220)

(221)

(222)

(223)

(224)

M. J ., Minton, K. W., Fleischmann, R. D., Ketchum, K. A., Nelson, K. E.,
Salzberg, S. L., Smith, H. O., Venter, J. C., and Fraser, C. M. (1999) Genome

sequence of the radioresistant bacterium Deinococcus radiodurans R1. Science
286, 1571-1577.

Wang, S. M., Lue, W. L., Yu, T. 8., Long, J. H., Wang, C. N., Eimert, K., and
Chen, J. (1998) Characterization of ADGl , an Arabidopsis locus encoding for
ADPG pyrophosphorylase small subunit, demonstrates that the presence of the
small subunit is required for large subunit stability. Plant Journal 13, 63-70.

Hannah, L. C., Shaw, J. R., Giroux, M. J ., Reyss, A., Prioul, J. L., Bae, J. M., and
Lee, J. Y. (2001) Maize genes encoding the small subunit of ADP-glucose
pyrophosphorylase. Plant Physiology 127, 173-183.

Zabawinski, C., Van Den Koomhuyse, N., D'Hulst, C., Schlichting, R., Giersch,
C., Delrue, B., Lacroix, J .-M., Preiss, J ., and Ball, S. (2001) Starchless mutants of
Chlamydomonas reinhardtii lack the small subunit of a heterotetrameric ADP-
Glucose pyrophosphorylase. J. Bacteriol. 183, 1069-1077.

Ho, S. N., Hunt, H. D., Horton, R. M., Pullen, J. K., and Pease, L. R. (1989) Site-
directed mutagenesis by overlap extension usign the polymerase chain reaction.
Gene 15, 51-59.

Sanchez, R., and Sali, A. (1997) Advances in comparative protein-structure
modelling. Current Opinion in Structural Biology 7, 206-214.

Sénchez, R., and Sali, A. (2000) Comparative protein structure modeling, in
Protein Structure Prediction: Methods and Protocols (Webster, D. M., Ed.) pp
97-129, Humana Press Inc., Totowa, NJ.

Tramontano, A. (1998) Homology modeling with low sequence identity. Methods
14, 293-300.

Rossmann, M. G., Liljas, A., Branden, C. 1., and Bansazak, L. J. (1975)
Evolutionary and structural relationships among dehydrogenases, in The Enzymes
(Boyer, P. D., Ed.) pp 61-102, Academic Press, New York.

Taroni, C., Jones, 8., and Thornton, J. M. (2000) Analysis and prediction of
carbohydrate binding sites. Protein Engineering, Design and Selection 13, 89-98.

180

(225)

(226)

(227)

(228)

(229)

(230)

(231)

Vyas, N. K. (1991) Atomic features of protein-carbohydrate interactions. Current
Opinion in Structural Biology 1, 732-740.

Sujatha, M. S., Sasidhar, Y. U., and Balaji, P. V. (2004) Energetics of galactose-
and glucose-aromatic amino acid interactions: Implications for binding in
galactose-specific proteins. Protein Science 13, 2502-2514.

Vogt, G., Woell, S., and Argos, P. (1997) Protein thermal stability, hydrogen
bonds, and ion pairs. Journal of Molecular Biology 269, 631-643.

Greenberg, E., and Preiss. (1965) Biosynthesis of bacterial glycogen. II.
Purification and properties of the adenosine diphosphoglucose: glycogen
transglucosylase of Arthrobacter species NRRL B1973. Journal of Biological
Chemistry 240, 2341 -2348.

Shen, L., and Preiss, J. (1964) The Activation and Inhibition of Bacterial
Adenosine-Diphosphoglucose Pyrophosphorylase. Biochemical and Biophysical
Research Communications 1 7, 424-429.

Morrison, R. T., and Boyd, R. N. (1990) Organic Chemistry, Fifth Edition,
Addison-Wesley Iberoamericana, Wilmington, Delaware.

Cupp-Vickery, J. R., Igarashi, R. Y., and Meyer, C. R. (2005) Preliminary
crystallographic analysis of ADP-glucose pyrophosphorylase from A grobacterium
tumefaciens. Acta crystallographica. Section F, Structural biology and
crystallization communications 61, 266-8.

181