mes:

.. 431;.
E. m“

 

 

 

 

4..

“H‘s“

:3

 

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIII

lllllllllllllllllllllllllllllllllllllllllllallljllll

This is to certify that the

dissertation entitled

THE éLUCose—I-pnos‘pHME awn/Ma 5:75 AND
THE MECHANISM oF £5011ch v5 ACTIVATION 0F
POTATO TUBER ADP-51.116055 PYFOPHOSPHOR YLASE

presented by

Wit/63M! Fu

has been accepted towards fulfillment
of the requirements for

BA” 0 degree in Eioéfiemfih}

 

 

AQJ‘QW

Major professor

Date%_&% ’79 8

 

 

 

LIBRARY

Michigan State
University

 

 

 

PLACE iN RETURN Box
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

MTE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/93 WWW“

THE GLUCOSE-i-PHOSPHATE BINDING SITE AND
THE MECHANISM OF REDUCTIVE ACTIVATION OF
POTATO TUBER ADP-GLUCOSE PYROPHOSPHORYLASE
By

YINGBIN FU

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry
1 998

ABSTRACT
THE GLUCOSE-l-PHOSPHATE BINDING SITE AND
THE MECHANISM OF REDUCTIVE ACTIVATION OF
POTATO TUBER ADP-GLUCOSE PYROPHOSPHORYLASE
By

YINGBIN FU

ADP-glucose pyrophosphorylase (ADPGlc PPase) from higher plants is a
heterotetramer composed of two different subunits. The expression of a putative full-
length cDNA encoding the small subunit of potato tuber ADPGlc PPase in Escherichia
coli results in a catalytically active enzyme. Its specific activity is similar to that of the
purified holoenzyme when assayed at a saturating concentration of activator. The activity
of the large subunit is negligible. The homotetrameric small subunit differs from the
heterotetrameric enzyme only in its regulatory properties. Thus, the small subunit is
proposed to be mainly involved in catalysis, while the large subunit is mainly involved in
modulating the regulatory properties of the small subunit. The N—terminus of the small
subunit is most likely involved in the regulatory properties and heat stability of the
holoenzyme since these properties were shown previously to be altered in a recombinant
enzyme that contains the truncated small subunit (at the N-terminus) and the large
subunit.

'95 in the homotetrameric ADPGlc PPase from E. coli was Shown to be

Lys
involved in the binding of the substrate, glucose-l-phosphate in a previous study [Hill,
M. A., Kaufmann, K., Otero, J., and Preiss, J. (1991) J. Biol. Chem. 266, 12455-12460].
This residue is highly conserved in the ADPGlc PPases. Replacing the conserved Lys'98

of the small subunit of the potato tuber enzyme with arginine, alanine, and glutamic acid

decreased the apparent affinity for glucose-l-phosphate of the wild-type enzyme 135- to

550-fold. Mutagenesis of the conserved Lysm with arginine, alanine or glutamic acid in

the large subunit had no effect on the apparent affinity for glucose-l-phosphate. A double

mutant, 8 L had a [00-fold lower apparent affinity for glucose-l-phosphate. These

Kl98R KZISR’

findings indicate that Lys'98 in the small subunit is directly involved in the binding of

”4

glucose-l-phosphate, while Lysz" in the large subunit has no effect on the apparent

affinity for the sugar phosphate substrate.

The catalytic activity of potato tuber enzyme iS activated by a preincubation with
ADP-glucose and dithiothreitol or by ATP, glucose-l-phosphate, Ca2+, and dithiothreitol.
Using 5, 5'-dithiobis (2-nitrobenzoic acid) quantitation and SDS-PAGE analysis, it was
shown that an intermolecular disulfide bridge between the small subunits of the potato
tuber enzyme was reduced during the activation. Further experiments showed that the
activation was mediated via a slow reduction and subsequent rapid conformational
change induced by ADP-glucose. The activation process is reversible by interchanging of
the enzyme between its activated dithiol form and the nonactivated disulfide form. The
intermolecular disulfide bridge was identified to be located between the Cys12 residues of
the small subunits of the potato tuber enzyme by chemical derivatization and site-directed
mutagenesis approaches. This reductive activation mechanism is explained by a two-step

model.

To my family

iv

ACKNOWLEDGEMENTS

I am grateful to those who have helped me work on this project. First, I want to
thank my advisor, Dr. Jack Preiss, for his valuable support. During the past five and half
years, I have become a mature and independent researcher from a naive and dependent
student. This transformation would not have been easy or possible to take place without
the cooperative environment provided by Dr. Preiss’s group. Especially, I want to thank
my wonderful colleague, Dr. Miguel A. Ballicora, from whom I learned not only the way
to conduct research but also the spirit of research. I can always count on him for valuable
suggestions and encouragement whenever I encountered difficulties in my work. I also
thank my committee members Drs. Robert P. Hausinger, Kenneth Keegstra, Hans Kende,
John E. Wilson for their guidance through my graduate study. Without the “push and
harshness” of them and my advisor, I will never imagine that I would have finished my
work on Chapter 4, the reductive activation mechanism, which is not in my original
research proposal.

Drs. Mary J. Laughlin and Thomas W. Okita of Washington State University
constructed plasmid pMLlO. Dr. Steven Withers of the University of British Columbia,
Vancouver, Canada, provided many sugar-phosphate analogues. I also thank Dr. Joseph
Leykam and Colleen P. Curry of the Macromolecular Structure Facility for peptide
sequencing and technical assistance with HPLC separations.

Finally, I want to give my special thanks to my girl friend, Yifan, for her love and
understanding which made my final stage of graduate study at Michigan State University

a truly enjoyable and an unforgettable event.

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. ix

LIST OF FIGURES .................................................................................. x

LIST OF ABBREVIATIONS .................................................................... xii

INTRODUCTION ................................................................................... 1
CHAPTER 1

LITERATURE REVIEW ........................................................................... 5

1. General Properties and Structure of Starch ............................................... 6

1.1. Properties ...................................................................... 6

1.2. Structure ........................................................................ 6

1.3 Application ..................................................................... 7

2. The Metabolic Parth of Carbon to Starch .............................................. 8

3. ADP-glucose Pathway Leads to Biosynthesis of Starch ................................. 8

4. ADPGlc PPase ............................................................................... 11

4.1. Subunit and Primary Structure ............................................. 11

4.2. Regulatory Property ......................................................... 13

4.3. Requirement of Two Subunits for Plant ADPGlc PPase ............... 15

4.4. Substrate Sites ............................................................... 16

4.5. Regulator Sites ............................................................... 17

4.6. Secondary Structure of ADP-Glucose Pyrophosphorylase ............ 23

References ....................................................................................... 26
CHAPTER 2

THE SMALL SUBUNIT OF POTATO TUBER ADP-GLUCOSE

PYROPHOSPHORYLASE HAS CATALYTIC ACTIVITY ....................... 33

Abstract .......................................................................................... 34

Introduction ..................................................................................... 35

Materials and Methods ........................................................................ 38

Reagents ........................................................................... 38

Plasmids ............................................................................ 38

Assay of ADPGlc PPase ......................................................... 38

Kinetic Studies .................................................................... 39

Protein Assay ..................................................................... 39

Molecular Mass Determination ................................................. 39

Expression in E. coli ............................................................. 39

Heat Treatment .................................................................... 40

Enzyme Purification .............................................................. 41

SDS-PAGE and Immunoblot ................................................... 41

Sequence Determination ......................................................... 42

Results ........................................................................................... 43

Expression of cDNAs Encoding the

Different Subunits of ADPGlc PPase ........................................ 43

Heat Stability ..................................................................... 43

Purification of ADPGlc PPases ................................................ 44

Activation by 3PGA ............................................................. 48

Inhibition by Pi ................................................................... 48

Kinetic Constants for Substrates ................................................ 51

vi

Discussion ....................................................................................... 52

References ....................................................................................... 54
CHAPTER 3
MUTAGENESIS OF THE GLUCOSE-l-PHOSPHATE BINDING SITE
OF POTATO TUBER ADP-GLUCOSE PYROPHOSPHORYLASE ...................... 55
Abstract .......................................................................................... 56
Introduction ..................................................................................... 57
Materials and Methods ........................................................................ 58
Reagents ........................................................................... 58
Bacteria] Strains and Media ..................................................... 58
Site-Directed Mutagenesis ....................................................... 58
Expression and Purification of Mutant and Wild-Type Enzymes .......... 59
Assay of ADPGlc PPase ......................................................... 61
Kinetic Studies .................................................................... 61
Sugar Phosphate Specificity ..................................................... 62
Thermal Stability .................................................................. 62
Protein Assay ...................................................................... 62
Protein Electrophoresis and lmmunoblot ...................................... 62
Results ........................................................................................... 64
Expression and Purification of Mutant Enzymes ............................. 64
Kinetic Characterization of SKlgst, Mutant Enzymes ....................... 64
Kinetic Characterization of SMLk2l3 Mutant Enzymes ....................... 67
Kinetic Characterization of SklggRme Mutant Enzyme ................... 70
Sugar Phosphate Specificity ...................................................... 70
Thermal Stability .................................................................. 71
Discussion ........................................................................................ 73
References ....................................................................................... 77
CHAPTER 4
MECHANISM OF REDUCTIVE ACTIVATION OF POTATO
TUBER ADP-GLUCOSE PYROPHOSPHORYLASE ...................................... 79
Abstract .......................................................................................... 80
Introduction ..................................................................................... 8 1
Materials and Methods ........................................................................ 82
Reagent ............................................................................. 82
Purification of Wild-type and Mutant Potato Tuber ADPGlc PPases. . .. 82
Assay of ADPGlc PPase ......................................................... 82
Reductive Activation of ADPGlc PPase ....................................... 83
Determination of Available Sulfliydryl Groups with DTNB ................ 83
Protein Assay ...................................................................... 84
Determination of the Reduction and Activation Time Course .............. 84
SDS-PAGE and lmmunoblot Analysis ......................................... 84
Determination of the Total Number of Disulfide ............................. 84
Identification of the Intermolecular Disulfide Bridge ........................ 85
Trypsin Digestion .................................................................. 86
Purification of Labeled Peptides by HPLC .................................... 86
Sequence Determination ......................................................... 86
Site-directed Mutagenesis ....................................................... 87
Results ........................................................................................... 88
I. Reductive Activation of Potato Tuber ADPGlc PPase ............................. 88
Activation of Potato Tuber ADPGlc PPase by ADPGlc and DTT ......... 88
Activation of Potato Tuber ADPGlc PPase
by ATP, Glc-l-P, Ca2+ and DTT ............................................. 88

vii

Reduction of an Intermolecular Disulfide Bridge during Activation ...... 90

Time Course of Reduction and Activation .................................... 95

Kinetics of Activation by ADPGlc and DTT ................................. 97

Reversibility of the Reductive Activation ..................................... 97

11. Identification of the Intermolecular Disulfide Bridge ........................... 104

Determination of the Total Number of Disulfide Bridge ................... 104

Direct Labeling with [”C]iodoacetic acid ................................... 104

Reverse Labeling with [”C]iodoacetic acid ................................. 107

4-vinylpyridine Labeling ...................................................... 107

Production and Purification of Mutant Enzymes ........................... 109

Activation Characteristics of Mutant Enzymes .............................. 109

Discussion ..................................................................................... l 13

References .................................................................................... l 17
CHAPTER 5

SUMMARY AND PERSPECTIVES .......................................................... 119

1. Summary .................................................................................... 120

2. Perspectives ................................................................................ 122

References ..................................................................................... 125

viii

LIST OF TABLES

CHAPTER 1
I. Sequence Alignment of the Regulatory Sites Of ADPGlc PPases
From Plants and Cyanobacteria ................................................... 20
CHAPTER 2
I. N-terminal Amino Acid Sequences of the Small Subunits of
ADPGlc PPases from Potato Tuber and Spinach Leaf ........................ 36
11. Expression Conditions of Plasmids Containing the cDNAs Encoding
the Subunits of ADPGlc PPase of Potato Tuber ................................ 40
III. Activity of ADPGlc PPases Containing Different Subunit(s) .................. 44
IV. Heat stability of ADPGlc PPases from Different Sources ...................... 45
V. Purification of the Small Subunit of Potato Tuber ADPGlc PPase ............ 46
VI. Purification of the Heterotetrameric ADPGlc PPase
(small + large subunits) of Potato Tuber ......................................... 47
VII. Kinetic Constants for 3PGA and Pi of ADPGlc PPases Purified
from Different Sources ............................................................. 50
CHAPTER 3
I . Comparison of the Apparent Affinity for Substrates of Potato Tuber
Wild-Type and Mutant ADPGlc PPases ......................................... 68
II. Kinetic Parameters of the Potato Tuber Wild-Type
and Mutant ADPGlc PPases ...................................................... 69
III. Specificity of Sugar-phosphates as Substrates
for Wild-type and Mutant SK198 ALwt ............................................. 72
IV. Conservation of the Sugar Phosphate Binding Motif
in Sugar Nucleotide Pyrophosphorylases ........................................ 76
CHAPTER 4
I. Reactivation of Potato Tuber ADPGlc PPase .................................. 100
11. Sequence Analysis of the [”C]Iodoacetic Acid Labeled Tryptic

Peptides and 4-Vinylpyridine Labeled ADPGlc PPases ..................... 106

ix

LIST OF FIGURES

CHAPTER 1
l. The route leading from sugar precursors into starch
in nonphotosynthetic storage organs ............................................... 9
2. Alignment of the amino acid sequence around the Glc-l-P site (Lys'95)
of ADPGlc PPase from E. coli with those from other sources ................ l8
3. Prediction of the secondary structure of
ADP-glucose pyrophosphorylase ................................................ 24
CHAPTER 2
1. Comparison of the regulatory properties of the homotetrameric small subunit
and heterotetrameric enzyme (small + large subunits) ........................ 49
CHAPTER 3

1. Nucleotide sequence and the encoded protein sequence of
the potato tuber ADPGlc PPase gene in the region of Lys198

in the small subunit and Lys“3 in the large subunit ............................. 60

2. Glc-l-P dependence for the wild-type and mutant enzymes ..................... 65
CHAPTER 4

1. Activation of potato tuber ADPGlc PPase by ADPGlc and DTT ............... 89
2. Activation of potato tuber ADPGlc PPase

by ATP, 610-] -,P Ca2+ and DTT ................................................ 91
3. Activation of potato tuber ADPGlc PPase by ATP, Glc-l -P,

Ca2 and DTT with different incubation times .................................. 92
4. DTNB determination of the available sulfliydryl groups of the

activated and nonactivated potato tuber ADPGlc PPase ....................... 93
5. Electrophoretic analysis of the activated and nonactivated

potato tuber ADPGlc PPase ....................................................... 94
6. Time course of potato tuber ADPGlc PPase activation and reduction ......... 96

7. Kinetics of potato tuber ADPGlc PPase activation by ADPGlc and DTT. . 98
8. Reversibility of the reductive activation of potato tuber ADPGlc PPase. . .. 101

9. HPLC separation of tryptic fragments from the activated and
l4C-carboxymethylated potato tuber ADPGlc PPase ....................... 105

10. HPLC separation of tryptic fragments of the nonactivated potato tuber
ADPGlc PPase afler reverse labeling with [”C]iodoacetic acid ............ 108

1 1. Immunoblotting analysis of wild-type and
mutant potato tuber ADPGlc PPases ........................................... 110

12. Activation kinetics of the wild-type (wt) and mutant
potato tuber ADPGlc PPases .................................................... 111

13. A proposed model for the reductive activation mechanism of
potato tuber ADPGlc PPase ...................................................... 116

xi

ADPGlc PPase
Glc— 1 -P

DTNB

DTT

FBP
Man- 1 -P
Pi

PLP

PMI/GMP

PPi
TCA
3PGA

6F—Glc- 1 -P

LIST OF ABBREVIATIONS

ADP-glucose pyrophosphorylase
glucose- 1 -phosphate

5, 5'-dithiobis (2-nitrobenzoic acid)
dithiothreitol

fructose- 1 , 6-bisphosphate

mannose- l-phosphate

orthophosphate
pyridoxal 5'-P
phosphomannose isomerase-guanosine 5’-diphospho-D-mannose

pyrophosphorylase
pyrophosphate
trichloroacetic acid
3-phosphoglycerate

6-fluoro-glucose- l-phosphate

xii

INTRODUCTION

Starch is an end product of carbon fixation and the major storage form of
photosynthate in the sink tissues of many crop plants. Research on the biosynthetic
pathway of starch not only can lead toward understanding one of the most basic anabolic
pathways but also provides the potential to manipulate the production and structure of
starch for many applications in agriculture as well as industries.

The ADP-glucose pathway is believed to be the predominant route towards starch
synthesis (1, 2). ADP-glucose pyrophosphorylase (ADPGlc PPase, EC 2.7.7.27) catalyzes
the synthesis of ADP-glucose from ATP and glucose-l-phosphate (Glc-l-P) and is the
major regulatory enzyme in the starch biosynthetic pathway (1- 5).

The goal of this thesis project is to identify important residues or domains in the
catalysis and regulation of the potato tuber ADPGlc PPase. Chapter 1 is a review of
literature that provides information about the physiological role and the structure-function
studies on ADPGlc PPases with the emphasis on plant enzymes.

Potato (Solanum tuberosum L.) tuber ADPGlc PPase has two different subunits—
the small and large subunits (6), both of which were cloned (7). Before I began to work
on this project, the cloned cDNAs were expressed together in Escherichia coli (8). Some
of the properties of the recombinant enzyme were different from the native enzyme and
were ascribed to an N-terminal truncation of the small subunit (8). In the work presented
in Chapter 2, the expression of an extended cDNA of the small subunit that includes
those truncated amino acids in E. coli yielded a homotetrameric enzyme with a specific
activity comparable to that of the heterotetrameric enzyme consisting of the small and
large subunits. The properties of the homotetrameric and heterotetrameric enzymes are
compared in detail. The roles of the small and large subunits in plant ADPGlc PPases are
discussed. The properties of the heterotetrameric enzyme containing the large and
extended small subunits are more similar to that of the native enzyme than the

heterotetrameric enzyme, which contained the large and truncated small subunits. This

work was done in collaboration with Dr. Miguel A. Ballicora. Part of the data were
published in Plant Physiol. (9).

In ADPGlc PPase from E. coli, Lys195 has been identified as part of the binding
site for Glc-l-P (10, l 1). This residue is highly conserved in the homotetrameric bacterial
enzymes as well as in both the small and large subunits of heterotetrameric plant ADPGlc
PPases (4, 12). However, no study had been performed to characterize these conserved
lysine residues in the plant enzyme. The expression system for potato tuber enzyme in E.
coli provided a usefiJl tool to serve this purpose. This study was also helpful in further
clarifying the roles of the two subunits since recent studies suggest that the small and
large subunits of the potato tuber enzyme cannot simply be classified as catalytic and
regulatory subunits (13), respectively. Chapter 3 describes the site-directed mutagenesis
studies on the putative Glc-l-P binding sites, Lys198 on the small subunit and Lys213 on the
large subunit, of the potato tuber enzyme (14).

Chapter 4 deals with the reductive activation mechanism of the potato tuber
ADPGlc PPase. It started from the observation that the catalytic activity of potato tuber
enzyme increases with time (showing nonlinear kinetics) when measured in the absence
of the activator. Further experiments suggested that the reduction of a disulfide bridge(s)
might be involved in the activation. Although metabolic regulation by means of
thiol/disulfide exchange has been well established in the light regulation of
photosynthetic enzymes in the chloroplast via thioredoxin/ferredoxin system (15, 16),
redox modulation was not observed with any of the ADPGlc PPases. The potato tuber
ADPGlc PPase was shown to be activated by dithiothreitol (DTT) (17, 18), but the
mechanism was not understood. In Chapter 4 I provide various evidence that the
activation is due to a reduction of an intermolecular disulfide bridge between two small
subunits followed by a conformational change induced by substrates. The intermolecular
disulfide bridge is identified by both chemical derivatization and site-directed

mutagenesis approaches. This work was submitted for publication in J. Biol. Chem.

Finally, Chapter 5 summarizes my thesis project and discusses the perspectives
for future study.

Yingbin Fu

East Lansing, Michigan
May I 998

PP!”

References
Preiss, J. (1991) 0ij Surv. Plant Mol. Cell Biol. 7, 59-1 14
Okita, T. W. (1992) Plant Physiol. 100, 560-564
Preiss, J. (1982) Annu. Rev. Plant Physiol. 33, 431-454
Preiss, J. (1988) in The Biochemistry of Plants (Preiss, J., Ed.), pp. 181-254, vol. 14,
Academic Press, San Diego
Preiss, J ., and Sivak, M. (1996) in Photoassimilate Distribution in Plants and Crops

(Zamski, E. and Schaffer, A., Ed.), pp 63-96, Marcel Dekker, Inc., NY

6. Okita, T. W., Nakata, P. A., Anderson, J. M., Sowokinos, J., Morell, M., and Preiss, J.

10.
ll.

12.

13.
14.
15.
16.
17.
18.

(1990) Plant Physiol. 93, 785-790
Nakata, P.A., Greene, T. W., Anderson, J. M., Smith-White, B., Okita, T. W., and

Preiss, J. (1991) Plant Mol. Biol, 17, 1089-1093

Iglesias, A. A., Barry, G. F., Meyer, C., Bloksberg, L., Nakata, P. A., Laughlin, M. J .,
Okita, T. W., Kishore, G. M., and Preiss, J. (1993) J. Biol. Chem. 268, 1081-1086

Ballicora, M. A., Laughlin, M. J., Fu, Y., Okita, T. W., Barry, G. F., and Preiss, J.

(1995) Plant Physio]. 109, 245-251

Parsons, T., and Preiss, J. (1978) J. Biol. Chem. 253, 6197-6202

Hill, M. A., Kaufmann, K., Otero, J., and Preiss, J. (1991).]. Biol. Chem. 266, 12455-
12460

Smith-White, B., and Preiss, J. (1992) J. Mol. Evo]. 34, 449-464

Ballicora, M. A., Fu, Y., Nesbitt, N. M., and Preiss, J. (1998) Plant Physio]. in press
Fu, Y., Ballicora, M. A., and Preiss, J. (1998) Plant Physiol. In press

Buchanan, B. B. (1980) Annu. Rev. Plant Physiol. 31, 341-374

Wolosiuk, R. A., Ballicora, M. A. and Hagelin, K. (1993) FASEB J. 7, 622-637
Sowokinos, J. R. (1981) Plant Physiol. 68, 924-929

Sowokinos, J. R., and Preiss, J. (1982) Plant Physiol. 69, 1459-1466

CHAPTER 1

LITERATURE REVIEW

LITERATURE REVIEW
1. General Properties and Structure of Starch
1.1. Properties

Starch is a white, granular or powdery, odorless, tasteless, complex carbohydrate.
It is almost insoluble in cold water and in alcohol, but forms a colloidal suspension in
boiling water. Starch is an end product of carbon fixation by photosynthesis and is a
major reserve polysaccharide of green plants. It is almost universally present in every
type of plant tissues, e.g., leaves, fruits, pollen grains, roots, etc., but especially abundant
in the seeds of cereal plants and in bulbs and tubers.

1.2. Structure

Starch is organized within a three-dimensional, semicrystalline structure named
“starch granule”. The starch granules are formed in plastids—mainly the amyloplast and
the chloroplast. Amyloplast is developed from proplastids and specialized in the synthesis
and long-term storage of starch in nonphotosynthetic plant tissues. In the chloroplast,
starch serves as a temporary storage of energy and carbon for photosynthetic cells. In
leaves, starch is synthesized in the chloroplasts during the day from photosynthetically
fixed carbon and is degraded by respiration at night. The major site of starch
accumulation is in storage organs including seeds, fruits, tubers, and storage roots. Starch
content can range between 65% and 90% of the total dry weight in those organs.

Starch contains two kinds of polysaccharides: amylose and amylopectin. Amylose
consists of linear chains of a-( 1-—>4)-linked glucose residues with molecular weight
around 2.5x 105. It constitutes about 20% of ordinary starch, though this number can vary
between 11 and 36% (1). In contrast to amylose, amylopectin is highly branched with an
average of one a-(1—-)6) glucosidic linkage every 20 to 26 glucosyl residues (2, 3).
Amylopectin usually constitutes about 70% of the starch granule. Its average molecular
weight is around 108. Some amylopectins, notably those fiom potato tuber, are also

phosphorylated. It has been reported that up to one phosphate per 300 glucose residues

was found in potato tuber amylopectins (4). Besides amylose and amylopectin, proteins
(including the enzymes of starch biosynthesis) and lipid constitutes the other minor
components of the starch granules.

1.3. Application

Besides its main usage as food in the world, about one third of the total
production of starch is used for many industrial processes, e.g., sizing of paper and board,
packaging, manufacture of biodegradable polymers, textile industries, etc. In the chemical
industry, starch is used as a raw material to produce acids, polyols, amino acids,
cyclodextrins, and fructose through fennentative processes.

Since starch is the major storage form of photosynthate of many crops,
enhancement of its biosynthesis can lead to an increase of the overall yield of crops.
Many important functional properties of the foods and other processed materials derived
from starch are affected by the structure and composition of starch, e.g., starch granule
shape, amylose-to-amylopectin ratio, crystallinity, phosphorylation, and lipid content, etc.
Therefore, there is great interest in increasing the yield and modifying the structure of
starch, especially to create novel starches by genetic manipulation of crops. Much effort
has been invested in understanding the structure-function relationship of the enzymes
directly involved in the starch biosynthetic pathway—ADPGlc PPase, starch synthase,
and branching enzyme (1-3, 5-7). Currently, starch is commercially extracted from
limited sources, such as maize and potato. Genetic engineering may allow the
modification of these starches toward the properties of specialized starches for specific
purposes. This review focuses on the studies involving structure-function relationships of
ADPGlc PPase, which is a key regulatory enzyme in starch biosynthesis. For discussions
on starch synthase and branching enzyme, a number of recent reviews may be referred to

(8-11).

2. The Metabolic Path of Carbon to Starch

In actively photosynthesizing leaves, it is well established that the flow of carbon
into starch occurs exclusively at the chloroplast (l l). The triose phosphate formed in the
Benson-Calvin cycle is converted to Glc-l-P via gluconeogenesis to serve as substrate for
ADPGlc production, which in turn leads towards the starch synthesis pathway (see
below). In young developing leaves that still behave as sink tissues (rather than sources),
the carbon source for starch synthesis is believed to be sucrose. In the cytoplasm, sucrose
is either hydrolyzed by invertase or by the sequential action of sucrose synthase and
UDPGlc PPase (12) to form monosaccharides, which can then enter glycolysis to be
converted into triose-phosphates. Triose-P/Pi translocator then transports the triose-
phosphates (13, 14) into the chloroplasts where they can be metabolized further for starch
synthesis.

In nonphotosynthetic tissues, starch synthesis occurs in the amyloplast. Unlike the
autotrophic chloroplasts of leaf tissues, amyloplasts are dependent on the cytoplasm for
both energy and carbon. The most accepted pathway for the transport of carbon from the
cytoplasm into arnyloplast is shown in Figure l (15, 16). In the cytoplasm, sucrose is
metabolized into hexose-Ps (Glc—6-P and Glc-l-P), which are then transported into the
amyloplasts by specific translocators (12, 17-20) for starch synthesis. An alternative
pathway of carbon flow into starch in the amyloplasts has been proposed (21). ADPGlc is
synthesized directly from sucrose in the cytoplasm and then transported via an ADPGlc
translocator into the amyloplast. However, this pathway was believed to play little, if any,
role in starch synthesis due to insufficient evidence (16).

3. ADP-glucose Pathway Leads to Biosynthesis of Starch

In the 19503, the discovery of nucleotide-diphosphate-sugars by Leloir and

Cardini (22-24) led to the uncovering of the metabolic routes of polyglucan synthesis.

The biosynthesis of starch was first shown by Leloir’s group (24, 25) to occur via the

   
  
 
 

Cytoplasm Amyloplast

Sucrose Starch

i
1
t 8. 9:
fructose + UDPGlc ADP Glc
2 3
l t J
fmctose 6-P glucose 1-P glucose 1.}:

6

4; :5

glucose 6-P

 

> glucose 6-P

Figure 1. The route leading from carbon into starch in nonphotosynthetic storage
organs. In the cytoplasm, sucrose is metabolized into glucose 6-P and glucose l-P, which
are then translocated into the amyloplast via specific translocators (indicated as ovals).
Inside the amyloplast, hexose-P is converted into ADPGlc, which in turn enters the starch
biosynthesis pathway. The enzymes involved are: 1, sucrose synthase; 2, fructokinase; 3,
UDPGlc PPase; 4, cytosolic phosphohexoseisomerase; 5, cytosolic phosphoglucomutase;
6, plastidal phosphoglucomutase; 7, ADPGlc PPase; 8, 9, starch synthase and branching
enzyme. This illustration is based on a figure from ref. 10.

transfer of the glucosyl portion of UDP-glucose to an existing starch primer.
Subsequently, ADP-glucose was reported as a better glucosyl donor than UDP-glucose
(26). Since then, many other groups have confirmed those findings (for reviews, see
ref. 8, 11, 27). In most cases, ADP-glucose is the specific glucosyl donor (28).

It is well accepted that the ADP-glucose pathway is the main pathway leading to
starch synthesis. This pathway consists of the following three reactions that are required
for the synthesis of the a-(l—-)4) and a-(1—>6) glucosidic linkages in amylose and in
amylopectin:

1. ATP + or-glucose-l-phosphate <:> ADP-glucose + pyrophosphate

2. ADP-glucose + or-glucan :> ADP + or-(l—>4)-glucosyl-glucan

3. Linear glucosyl chain of a-glucan :> branched chain of a-glucan with or-(l ——)6)

linkage branch points

In the first step, the glucosyl donor, ADP-glucose, is synthesized from ATP and
Glc-l-P in a reaction catalyzed by ADPGlc PPase (EC 2.7.7.27). The produced
pyrophosphate (PPi) is removed by inorganic pyrophosphatase and this removal makes
the reaction irreversible under physiological condition. In the second step, the glucosyl
unit of ADP-glucose was transferred to the nonreducing end of a preexisting
or-l, 4-glucan chain by starch synthase (EC 2.4.1.21) (24). In the third reaction, branching
enzyme (EC 2.4.1.18) catalyzes the formation of a—(l—)6) linkages found in amylopectin.
The three reactions constitute the so-called ADPGlc pathway of starch synthesis. In
bacteria, the synthesis of glycogen shares the same pathway (28).

ADPGlc PPase catalyzes the main regulatory step in this pathway. The supporting
biochemical and genetic evidence has been reviewed extensively (8, 11, 15, 16). All the
data demonstrated a direct causal relationship between the activity of the ADPGlc PPase

and starch accumulation. Some of the most compelling evidence is listed below:

10

In photosynthetic tissues, two mutants ofArabidopsis that were defective in starch
synthesis were isolated by Lin et a1. (29, 30). One mutant had less than 2% of the wild-
type ADPGlc PPase activity resulting in starch accumulation of less than 2% of wild-type
level in the leaf. The other mutant was deficient in the large subunit of ADPGlc PPase
causing a 93% reduction of the enzyme activity and an accumulation only 26% as much
starch in the leaf as compared to the wild-type.

In nonphotosynthetic tissues, ADPGlc PPase was also reported to play a major
role in starch synthesis via genetic studies. Two mutants of maize endosperm, shrunken 2
(sh2) and brittle 2 (bt2), which are deficient in ADPGlc PPase activity (5-10% of wild-
type), also have deficient levels of starch content (ZS-30% of wild-type) (31, 32). In
potato tuber, the expression of an antisense construct for the small subunit of ADPGlc
PPase causes a decrease in enzymatic activity to 2 to 5% of the wild-type level and
almost no starch formation (33).

Debranching enzyme has also been suggested to be involved in the formation of
amylopectin (34). From this proposal and the information on the various isozymes of
branching enzyme and starch synthase, Preiss et a1. (6) proposed a pathway for the
synthesis of amylopectin and amylose.

4. ADPGlc PPase

The synthesis of ADPGlc by the action of ADPGlc PPase was first reported by
Espada in 1962 (35). ADPGlc PPase was later found distributed mainly in the chloroplast
and amyloplast of plant tissues (36-38). Nevertheless, two reports proposed that it might
also be present in the cytoplasm (39, 40).

4.1. Subunit and Primary Structure

ADPGlc PPase from all sources is tetrameric in structure. The bacterial enzyme is
a homotetramer and each subunit has molecular mass between 50 and 55 kDa (28). The
plant enzyme consists of two types of subunits: small and large subunits. The small

subunits have molecular masses between 50 and 54 kDa, whereas the large subunits have

11

molecular masses between 51 and 60 kDa (8). Interestingly, the cyanobacterial ADPGlc
PPase is homotetrameric similar to the bacterial enzyme, but is regulated by 3PGA and Pi
like the plant enzyme (41-43).

The structural properties of the spinach leaf enzyme are understood best among
the plant ADPGlc PPases (44-46). This enzyme consists of two different subunits with
molecular masses of 51 and 54 kDa. Besides the molecular mass, the two subunits are
also different in the following properties: N-terrninal sequences, amino acid composition,
tryptic peptide maps, and antigenic properties (44). Therefore, the two subunits are
structurally different and very likely to be the products of two genes.

The potato tuber ADPGlc PPase has been purified to near homogeneity by Okita
et a1. (47). They observed two different subunits that could be distinguished by their
slight differences in the molecular mass (50 and 51 kDa) and in net charge by two-
dimensional PAGE. Antibody prepared against the spinach leaf small subunit crossreacts
only with the potato tuber small subunit. However, antibody prepared against the spinach
leaf large subunit does not react significantly with either subunit of the potato tuber
enzyme. The cDNAs of the two subunits were isolated by using antibody against the
potato tuber holoenzyme (48-50). The cloned cDNAs were expressed together in an E.
coli strain devoid of ADPGlc PPase activity (51). Most properties of the recombinant
enzyme were similar to the native enzyme purified from potato tuber except that the
recombinant enzyme is not heat stable and less sensitive to Pi inhibition. Those
differences were linked to the use of an incomplete cDNA for the small subunit, which
results in a possible truncation at the N-terminus.

The structural genes for many other plant ADPGlc PPase subunits (small or large)
have also been cloned, e.g., those from rice endosperm (49, 52, 53), maize endosperm
(54), Arabidopsis thaliana (B. Smith-White and J. Preiss, unpublished results), and wheat
leaf and endosperm (55). Since 1991, many other genes of plant ADPGlc PPases have

been isolated and are not listed here.

12

Smith-White and Preiss (56) did the first systematic comparison of the primary
structures of ADPGlc PPases from different sources. They have found that the small
subunit is highly conserved among different plant sources and tissues with an amino acid
sequence identity varying from 85 to 95%. The large subunit shares less sequence
identity, ranging between 50 and 60%. The identity between the large and small subunits
varies between 40 and 60%. Therefore, the conservation of the amino acid sequence of
the small subunit from various plant sources is higher than that of the large subunits. On
the basis of the identity between the two subunits of the plant ADPGlc PPase and their
identity with the homotetrameric bacterial enzyme (~30%), it has been proposed that the
two plant subunits may have evolved from the same ancestor (l 1, 15, 56).

4.2. Regulatory Property

The first report regarding the regulatory property of plant ADPGlc PPase was
from Ghosh and Preiss (57). They found that the spinach leaf enzyme was activated by
3PGA and inhibited by Pi. Subsequently, the same regulatory properties were also
observed in ADPGlc PPases from other higher plants (1 1), green algae (15, 58), and
cyanobacteria (41, 42). In most cases, 3PGA also increases the affinity for the substrates
(ATP and Glc—l-P) and reverses the inhibition by Pi. Based on these findings, it was
proposed that the ratio of concentrations for 3PGA and Pi regulates the activity of
ADPGlc PPase from plants, which in turn controls the level of starch synthesis (59-62).
This ratio fluctuates in response to different physiological conditions during dim'nal
growth in leaf tissues. For example, CO2 fixation during photosynthesis in the light leads
to the increase of 3PGA concentration, while ATP generation during
photophosphorylation leads to the decrease of Pi concentration, indicating that both
energy and carbon are in excess. In the dark, the opposite occurs resulting in a decrease of
the ratio. Therefore, 3PGA and Pi concentrations within the chloroplast are good
indicators of the energy and carbon status. In this regard, the allosteric regulation of

ADPGlc PPase provides a good mechanism to control the flux of carbon into starch.

l3

Many studies on both isolated chloroplast systems and intact leaves confirm the existence
of this regulation mechanism for plant ADPGlc PPase (for review see ref. 27).

In non-photosynthetic tissues, ADPGlc PPase is also believed to play a dominant
role in controlling starch synthesis through the same allosteric regulatory properties as
those in photosynthetic tissues. The enzymes derived from potato tuber and maize
endosperm are activated by 3PGA 20- to 30-fold (63-64). The enzyme from cassava root
is almost completely dependent on 3PGA for the catalytic activity (65). All three
enzymes are inhibited by Pi (63-65). However, ADPGlc PPases from pea embryo (66),
barley endosperm (67), and bean cotyledon (68) show relative insensitivity to 3PGA
activation (1.5- to 3- fold). Preiss and Sivak (8) suggested that proteolysis may be one
reason behind the allosteric insensitivity found in those enzymes, which was shown to be
the case for the maize endosperm enzyme (64). Furthermore, since the 3PGA activation
curve for ADPGlc PPase is sigrnoidal, even small changes in the concentrations of 3PGA
and Pi produce big changes in the synthetic rate of the enzyme (10). Nevertheless, it is
still unknown how the availability of carbon and energy for starch synthesis is signaled in
non-photosynthetic systems.

Many in vitro and in vivo experiments suggest that the allosteric regulation of
ADPGlc PPase by 3PGA and Pi are physiologically important for starch synthesis (11,
15, 16, 27, 69). Ball et al. (58) provided important in vivo evidence. A starch-deficient
mutant of Chlamydomonas reinhardtii was found to contain a defective ADPGlc PPase
that was less responsive to activation by 3PGA. Other supporting evidence was provided
by Stark et al. (70). They increased starch content up to 60% in potato and 10-fold in
cultured tobacco cells by transformation with a mutant E. coli ADPGlc PPase gene. The
mutant gene encoded an enzyme that can synthesize glycogen at a rate three times higher
than the wild-type enzyme in E. coli and was insensitive to the regulatory effectors of the
plant enzymes. In contrast, no increase of starch was observed when wild-type E. coli

ADPGlc PPase gene was used.

14

Neuhaus and Stitt (62) used a mutant Arabidopsis thaliana to measure the flux
control coefficient for ADPGlc PPase by the Kacser-Bums control analysis method (71,
72). In this method, the relationship between the rate of a metabolic process and the
activity of an enzyme of interest was calculated to determine the relative importance of
that enzyme in regulating the metabolic pathway. It was found that ADPGlc PPase
exercises the most significant effect in controlling starch synthesis compared with
plastidal phosphoglucomutase and phosphoglucoisomerase.

A light-mediated activation mechanism has been proposed by Kaiser and
Bassharn (73). They observed that high concentration of DTT could stimulate by 5.7-fold
the ADPGlc synthesis in the crude extract of spinach chloroplast. However, neither the
crude extract nor the purified ADPGlc PPase could be activated by DTT or by various
spinach leaf thioredoxins (27). Furthermore, no light activation could be observed on
ADPGlc PPase fi'om intact chloroplasts of either the pea or the spinach (L. B. Anderson,
T. Ashton, R. Scheibe, and J. Preiss, unpublished results). Since ADPGlc synthesis was
measured from Glc-6-P, it is possible that the phosphoglucomutase activity, which is
needed to convert Glc-6-P to Glc-l-P, was actually activated by DTT resulting in the
activation of ADPGlc synthesis (27). Nevertheless, ADPGlc PPase purified from potato
tuber has been shown to be activated by DTT (63, 74). However, the mechanism of DTT
activation was not clear.

4.3. Requirement of Two Subunits for Plant ADPGlc PPase

Many experiments indicate that both the small and large subunits are required for
optimal activity of the plant enzyme. As already mentioned, the bt 2 and sh 2 mutations
in maize endosperm reduce 90 to 95% of ADPGlc PPase activity, causing up to 75%
reduction of starch (31, 32, 75, 76). Immunoblotting experiments showed that the bt 2
and sh 2 mutants lacked the small and large subunits, respectively (1 1, 76, 77). In pea,
mutation of the rb locus affects a gene encoding the large subunit causing 90% or greater

reduction of ADPGlc PPase activity and a concomitant up to 62% drop in starch

15

accumulation (66, 78). Li and Preiss (79) have reported that mutation of the large subunit
in Arabidopsis resulted in an ADPGlc PPase that has activity only in the presence of high
concentrations of the activator. Other studies have also showed that the absence of one
subunit results in very low ADPGlc PPase activity (29, 31, 51, 80).

Multiple genes were reported to encode the large subunits of many plant ADPGlc
PPases, e.g., potato (48), barley (81, 82), Arabidopsis (83), rice (52), and wheat (55). The
expression of those genes show strong source-specificity, i.e. being restricted to different
plant tissues (e.g., leaf, stem, guard cells, tuber, endosperm, root) (55, 56, 81, 82). Thus,
it is possible that the different large subunits confer different regulatory properties for the
heterotetrameric ADPGlc PPases from different plant species and/or tissues. Further
discussion about the function of the two subunits is presented in Chapter 2.

4.4. Substrate Sites

Substrates were shown to bind ADPGlc PPases in an ordered mechanism—ATP
binds first, followed by the binding of Glc—l-P (84-86). Product release is ordered with
PPi released first, followed by ADPGlc (84). Many studies on the substrate and
regulatory site of ADPGlc PPase were started with the bacterial enzyme (87-93). These
studies prove to be quite relevant for later studies on the plant counterpart because
specific regions, e.g., the substrate and regulator sites, are highly conserved in bacterial
and plant enzymes.

Studies to target the substrate sites were mainly pursued by the methods of
chemical modification (by using substrate analogs) and site-directed mutagenesis. For
example, Pyridoxal 5'-P (PLP), a reagent for modification of lysine residues of proteins,
has been used successfully with the E. coli ADPGlc PPase. Modification of E. coli
ADPGlc PPase with [3H]PLP resulted in labeling of two lysine residues, Lys39 and Lys‘”
(91, 92). Lys39 was shown later to be located in the activator site (91). Modification of
Lys”5 resulted in a loss of catalytic activity that could be prevented by inclusion of

ADPGlc and MgClz. Therefore, Lys'95 is required for either substrate binding or for

16

catalytic activity. Further study by site-directed mutagenesis showed that LysI95 was

specifically involved in the binding of one substrate, Glc-l-P, but has no role in the rate-
deterrnination step (88). Both the charge and size of this lysine are important for proper
binding according to kinetic studies on a series of Lys195 mutants (88). Figure 2 shows the
alignment of the amino acid sequences of the Lys'95 region in the E. coli ADPGlc PPase
with sequences of ADPGlc PPases from other sources. This lysine residue and its
surrounding sequences are highly conserved in the bacterial enzymes as well as in both
the small and large subunits of plant ADPGlc PPases, suggesting they may perform the
same function. Chapter 3 presents the study on the conserved lysine residues in the small
and large subunits of ADPGlc PPase of potato tuber.

Tyr‘” has been identified as being involved in the binding of the other substrate,
ATP, in E. coli ADPGlc PPase (87, 93). It is located in a region predicted to form a
Rossman-fold supersecondary structure, which is a common nucleotide-binding motif in
proteins (94). This region is highly conserved in all ADPGlc PPases sequenced to date (8)
except that tyrosine is substituted by phenylalanine in the cyanobacterial and plant
enzymes. On this basis, it was proposed that those conserved sequences were also
involved in the binding of ATP. However, recent site-directed mutagenesis experiment
indicated that the equivalent phenylanaline in the cyanobacterial enzyme was not
involved in the binding of ATP (.1. Sheng and J. Preiss, unpublished results). Therefore,
further experiments are necessary to clarify the role of this conserved region. One
possibility is that other residues in this region are actually involved in the binding of ATP
in cyanobacterial and plant enzymes.
4.5. Regulatory Sites

Due to the importance of the regulatory properties of ADPGlc PPase to starch
synthesis, much effort has been directed toward understanding the structural basis for the
allosteric regulation of this enzyme (11, 15, 95). The first structural study on the
regulatory site of plant ADPGlc PPase was performed on the enzyme from spinach leaf.

l7

Prokaryotes
E. coli

S. typhimurium
A nabaena
Synechocystis

Small subunit of the plant enzyme
Potato tuber (50 kDa)

Spinach leaf (51 kDa)

Rice seed

Maize endosperm (54 kDa)
Arabidopsis thaliana

Wheat endosperm

Large subunit of the plant enzyme
Potato tuber (51 kDa)

Spinach leaf (54 kDa)

Maize endosperm (60 kDa)

Wheat endosperm

Figure 2. Alignment of the amino acid sequences around the Glc-l-P site (Lys
of ADPGlc PPase from E. coli with those from other sources (after Preiss and Sivak,
ref. 8). The numbers on top of all the sequences correspond to Lys-195 of the E. coli
enzyme. The positions of the equivalent lysine residues in the small and large subunit of

potato enzyme are also indicated. The conserved residues are underlined in all the

195
IIEFVEEP-AN
**D*****-**
V*D*S*:*KGE
*TD*S***QGE

198
****A***QGE
****A***KGE
*V**A*:*KGE
****A*:*KGE
****A*:*KGE
****A*:*KGE

213
VVQ*A*:*KGF
VLS*S*:*KGD
VLQ*F*:*KGA
VVQ*S*Q*KGD

sequences. *signifies the same amino acid as in the E. coli enzyme.

18

By using an activator analogue, PLP, four lysine residues [one on the small
subunit (45) and three on the large subunit (46, 96)] were found to be potential activator
sites (Table I). The activator (3PGA) prevented the modification of all those lysine
residues, while the inhibitor (Pi) prevented the modification of two lysine residues: site-l
lysine of the small subunit and site-2 lysine of the large subunit. It is believed that site 1
and site 2 play the most important role in the binding of the activator (8).

The sequence alignment shows that there is high conservation of both site 1 and
site 2 regions in the cyanobacterial enzyme and the two subunits of plant enzymes,
though the site 1 region shares higher homology (Table 1). Studies on the conserved
residues (Lys4'9-site l; Lysm-site 2) in the homotetrameric Anabaena enzyme confirmed
that these residues were involved in the binding of the activator (89, 90). Interestingly,
recent mutagenesis studies indicate that Lys404 (site 2) and Lys441 (site 1) on the small
subunit of the potato tuber enzyme are more important for the binding of 3PGA than their
counterparts (Lys4'7-site l and Lysm-site 2) in the large subunit (97). Substitution of
Lys404 and Lys‘“' on the small subunit to alanine decreased the apparent affinity for 3PGA
3090- and 54-fold, respectively. Furthermore, the apparent affinity for the inhibitor (Pi)
was also affected by the two substitutions, resulting in a more than 400-fold decrease. In
contrast, when both Lys‘“7 (site 2) and Lys455 (site 1) on the large subunit were mutated to
alanine, the apparent affinity for the activator decreased only 9- and 3-fold, respectively.
These mutations gave negligible effects on the apparent affinity for Pi. It appears that
both site 1 and site 2 of the small subunit are important for the binding of the activator as
well as the inhibitor, but their counterparts on the large subunit are relatively
unimportant.

Regarding inhibitor (Pi) site, Arg294 of the Anabaena enzyme was shown to be
specifically involved in the binding of Pi (98). This residue is conserved in the ADPGlc

PPases from cyanobacteria and plants, but not in the enteric bacterial enzymes. Both the

19

TableI
Sequence Alignment of the Regulatory Sites of ADPGlc PPases
From Plants and Cyanobacteria
(*) signifies that the sequence has been determined by amino acid sequencing
(43). (**) signifies that the same sequence in this region has been reported by Park and

Chung, accession number, u85496. Afier Ballicora, et al., ref. 97.

20

Table I

 

 

Accession Source Tissue Site Slte
Number
C yauobacteria] Site 2 Site 1
3 8 2 4 1 9
211539 Anabaena - DTIIRRAIIDKNARIG IVWLKNAVITDGTII
M83556 Synochocystis - GTTIRRAI IDmARIG IWVIKNVTIADGTVI
Small subunits
4 0 4 4 4 1
Solanum tuberosum Tuber NCI-I I IGZAI IDKNARIG IV'I'VIKDALI PSGI I I
x83500 Spinacia oleracea Leaf NSHIKRAI IDKNARIG IVTVIKDALIPSGTV I
L33648 Solanum tuberosum. R. Burbank NCHIKRAI IDxNARIG IV'I'VIKDALIPSGIVI
x61 186 Solanum tuberosum. R. Burbank Tuber NCH I KRAI IDKNARIG IVTVIKDALI PSGI I I
x551 55 Solanum tuberosum. Desiree Tuber NCHI KRAI IDKNARIG IV'I'VIKDALI PSGIVI
x55650 Solanum tuberosum. Desiree Tuber NCH I KRAI IDKNARIG IV'I'VIKDALI PSGIV I
x91736 C hlam ydomonas reinhardti - NSVITNAI IDKNARVG ILVIDKDALVPTG'I'I‘I
x76941 Viciafaba (VfAGPP) cotyledons NSHIRRAIIDKNARIG IVTVIKDALIPSGTVI
x76940 Viciafaba (VfAGPC‘) Cotyledons NSHIKRAIVDKNARIG IVTIIKDALIPSGTVL
x96764 Pisum Sativum Cotyledons NSHIKRAI IDKNARIG IV'I'VIKDALI PSGTV I
X96765 Pisum Sativum Cotyledons NSHIKRAIVDKNARIG IVTI IKDALI PSGTVI
279635 lpomoea batatas (ps T]. I) Tuberous Root/ Leaf N SHIKRAI IDKNARIG IV'I‘I IKDALIPSGTI I
279636 Ipomoea batatas (ps 77.2) Tuberous Root/ Leaf NSHIKRAI IDKNARIG IVTVIKDALIPSGTVI
L41126 Lycopersicon esculentum Fruit NCLYKRAI IDKNARIG IVTVIKDALIPSGIVI
J04960 Oryza sativa Endosperm NCHIRRAI IDKNARIG IVTVIKDALLLAEQLYB
M31 616 01720 sativa Leaf NCHIRRAI IDKNARIG IVTVIKDALLLAEQLYE
X66080 Triticum aestivum Leaf NSHI KRAI IDKNARIG IVTVIKDALLPSGTV I
248562 Hordeum V ulgaris Starchy endosperm NSHIKRAI IDKNARIG IVTVIKDALLPSGTVI
248563 Hordeum V ulgaris Leaf NSHIKRAI IDKNARIG IV'I'VIKDALLPSGTV I
Zea Mays. (brittle-2) Endospenn NSC IRRAI IDKNARIG IV'I'VIKDALLPSGTVI
$72425 Zea Mays Leaf NSHIRKAI IDKNARIG IVTV IKDALLPSGTV I
Large Subunits
4 1 7 4 5 S
Solanum tuberosum Tuber NTKIRKCIIDINAKIG IIIILEKATIRDGTV I
- Spinacia oleracea ( *) Leaf . . . IKDAI IDKNAR . . IN I FINATIKDGW
x761 36 Solanum tuberosum. Desiree Tuber NTRI KDCI IDKNARIG I'I'VI SKNSTI PDG'I'VI
x61 1 87 Solanum tuberosum. R. Burbank Tuber NTKIRKCI IDKNAKIG I I I ILEKATIRDGTVI
X74982 Solanum tuberosum. Desiree Leaf NTKI QNCI IDKNAKIG ITVIMINATIKDGTVI
u81033 Lycopersicon esculentum NTKIRKCI IDKNAKIG I I I I SEKATIRDGTVI
u88089 L ycopersicon esculentum NTKIRKCIIDKNAKIG II IIAEKATIRDGTVI
U81 034 Lycopersicon escu lentum ( " ‘) NTKI QKCI IDKNAKI G I‘I'VIMINATIK‘DGTVI
X96766 Pisum Sativum Cotyledons NTKIKNCI IDKNAKIG ITI IMEXA‘I‘IEDGTVI
X78900 Beta vulgaris, Zuchtlinie Tap root NTKI KNCI IDKNAKIG ITI I LKNATIQDGLVI
X14348 Triticum aestivum. cv. Mardler Leaf NTS IQNCI I DKNARIG ITVVLKNSVIADGLVI
X14349 Triticum aestivum Endosperm NTKISNCI IDMNARIG IWIQKNATIKDGTW
221969 Triticum aestivum Dev. Grain NTKISNCIIDMNARIG IvaQxNA'rIrcDGTvv
x14350 Triticum aestivum Endosperm NTKISNCIIDMNARIG IWIQKNATIKDGTW
2381 1 1 280 Mays Embryo NTKISNCI IDMNCQGW IVWLKNATIKDGTVI
$48563 Zea Mays, (Shrunken-Z) Endosperm NTKIRNCIIDMNARIG IvaLxNA'rINECLVI
u66041 Oryza sativa Endospenn NTKIRNCI IDMNARIG IWILICNATNATITGIGTVI
u66876 Hordeurn vulgaris Leaf NTSIQNCI IDKNARIG ITWLKNSVIADGLVI
x671 51 Hordeum vulgaris Starchy endosperm NTKI SNC I IDMNARIG IWIQKNATIKDGTVV

 

21

cyanobacterial and plant enzymes are inhibited by Pi but not by AMP, an inhibitor of the
bacterial enzymes. In the spinach leaf enzyme, one or more arginine residues were
suggested to be involved in the allosteric regulation. However, the specific residue(s) has
not been identified (99).

Besides the approach of chemical modification and site-directed mutagenesis,
random mutagenesis has also been used to target the regulator sites in the large subunit of
potato tuber enzyme. In one study, Pro55 (52 in ref. 100) was replaced by leucine and
resulted in a mutant enzyme with 45-fold lower affinity for 3PGA (100). However, since
the mutation occurred in a loop region, the effect could be caused by a conformational
change. In another study, Asp‘“6 (413 in ref. 101) was mutated to alanine resulting in a
mutant enzyme whose affinity for 3PGA decreased about 6- to 10- fold. This effect is
similar to that observed when a neighboring residue, Lys‘m, was mutated to alanine (97).
The moderate effects of the mutations on these two residues suggest that they may not be
critical for the binding of the activator.

An in vivo, transposon mediated mutagenesis system was used by Giroux et al.
(103) to introduce short insertions to the C-terminal region of sh 2, a gene encoding the
large subunit of ADPGlc PPase in the maize endosperm. A mutant line was isolated with
an 11-18% increase of seed weight. Further analysis showed that the starch content and
several seed components were increased. The mutant ADPGlc PPase was less sensitive to
Pi inhibition, but with normal response to 3PGA activation. Sequence analysis revealed
that two amino acids (tyrosine and serine) were inserted at the N-tenninal side (about 10
amino acids) of the site-l lysine. Insertion of tyrosine and serine at the corresponding
position of the potato large subunit resulted in a heterotetrameric recombinant enzyme
that is insensitive to Pi inhibition. Therefore, the insertion region on the large subunit

comprises part of the allosteric domain.

22

4.6 Secondary Structure of ADP-Glucose Pyrophosphorylase

As mentioned in the preceding text, various mutagenesis and chemical
modification methods have provided much useful information on the structure-function
relationship of the substrate and activator sites of ADPGlc PPases. However, a three-
dimensional structure is needed so that we can fit all those “jigsaw pieces” together to
thoroughly uncover the catalytic and regulatory mechanism of this enzyme. The E. coli
enzyme has been crystallized (104) but the crystals are of poor diffraction quality and
sensitive to x-ray exposure.

Despite lack of knowledge regarding the three-dimensional structure, a general
secondary structure of ADPGlc PPases was predicted by Ballicora et a]. (102). In the first
step, a hydrophobic cluster analysis method (105) was applied to ADPGlc PPases from E.
coli, Anabaena, and potato tuber. It was found that all of these enzymes share identical
hydrophobic clusters in many regions and are different only in some insertions/deletions.
The insertions/deletions contain many hydrophilic amino acids, suggesting they are not
part of the "core" structure of the protein. These observations indicate that ADPGlc
PPases share a common folding pattern in spite of their different quaternary structures
(homotetrameric in bacteria and heterotetrameric in plants) and different specificity for
the regulators.

In the second step, the amino acid sequences of E. coli and Anabaena ADPGlc
PPases together with the two subunits of potato tuber enzyme were analyzed by the PHD
program (106), and a general secondary structure was obtained (Figure 3). The enzyme
has five major domains. Domains 1, 2 and 4 consist of a/B structure while domain 3 and
5 mainly consist of [3 sheets. Good agreement was observed between this model and
biochemical data, especially those from the limited proteolysis studies with the E. coli
(M. -X. Wu and J. Preiss, manuscript submitted) and Anabaena enzymes (Y. -Y. Chamg
and J. Preiss, unpublished results; J. Sheng and J. Preiss, unpublished results). All the

23

on: 33% m we 56% mm can 358506 on 8980 N 98 _ 32mm: 52.33 .3288 BE. .85 owmzwofi 038on 853?:
8355. ANS «8 ..E 8 Baez—am 6%.: omflbonmmosmoba 3839mm? mo ocaonbm Eva—80w 05 mo cocacom .m DBMS

- - u
=8 .m 5 8E3... ‘

J

W nus—n v5. canon-5 5
U. 82 a 5 v2.85 5...:

 

5633‘s 3: §u>=u< .... A
2 2.523. . _ .

\

       

no~ A—

“ LOO.— OZOJV

.

      

4

ESE—0.70
900..

i ;
2+

J_U__ __.._._.._._,

T—TN-

/L ...._ , ._
\\_ . .... U.. ....

   

a £5 a: v. 02 U

 

 
 

 

 

usage—AV c... 8: 53>=u< . \ .363 E ,. x . “cw“.uflwn—UEW
836:2 _
n N
q .- .
. U _ asset
, . l _ .. I U . . . U ,r U 355.2
. U . U . _ m I . A W 984
U _ . A . _ U . . . _ _ .
. U U m m . _ _
.2 U. _ _ w m . U U n
a . .. . ,. n A V .i., 1 .
25E55E35 / ¢ ff \ K V \ _n i; / \
t 38.3
#—
ein—2:25 moo.— mo_v_ 2.“ a? D ....l-,....l.iI
. _

’ SEA—.0

ml. .532
. 5.: EEO _

lililill . ll.

24

protease digestions occur at the predicted loop regions with one exception, which lies in
an or helix insertion in domain 5 of the Anabaena and plant enzymes. However, since this
helix is predicted to be exposed, it is most likely to be a part of a loop region. Most of the
conserved residues that are important for the binding of substrates (Tyr114 and Lys'” in E.
coli) and activators (Lysz'9 in E. coli; Lys4'9 and Lys382 in Anabaena representing sites 1
and 2, respectively) are located in loops, or very close to loops. Pro'95 and Gly336 of the E.
coli enzyme that are important for the regulatory properties (107, 108) are also located in
the loop regions.

Domain 2 contains a Rossmann fold supersecondary structure, in which the ATP
binding site of E. coli enzyme is located. It is suggested that domains 1, 2, and 3 form a
catalytic core with substrates binding on the top (according to the orientation depicted in
Figure 3).

Recently, ADPGlc PPase from E. coli was treated with proteinase K to
characterize different structure domains (M. -X. Wu and J. Preiss, manuscript submitted).

The cleavage sites close to the Glc-l-P binding site (Lys'95

in E. coli enzyme) are
protected by ADP-glucose, Mg”, and F BP (activator). When the E. coli enzyme was
treated with proteinase K in the presence of ADP-glucose, Mg”, and FBP, 11-13 amino
acids from the N-terrninus and 2 amino acids from the C-terrninus were cleaved. The
truncated enzyme retained maximal activity, but was no longer sensitive to inhibition and

did not require the activator for increase in activity. Thus, the N-terrninal loop was

suggested to be involved in the allosteric regulation of the E. coli enzyme.

25

10.

ll.
12.

References
Detherage, W. L., MacMasters, M. M., and Rist, C. E. (1955) Trans. Am. Assoc.
Cereal Chem. 13, 31-42
Manners, D. J. (1985) in Biochemistry of Storage Carbohydrates in Green Plants
(Dey, P. M., and Dixon, R. A., Eds.) pp 149-203, Academic Press, NY
Morrison, W. R., and Karkalas, J. (1990) in Methods in Plant Biochemistry (Dey, P.
M., Ed.) Vol. 2, pp 323-353, Academic Press, NY
Takeda, Y., and Hizukuri, S, (1982) Carbohydr. Res. 102, 312-327
Smith, A. M., and Martin, C. (1993) in Biosynthesis and Manipulation of Plant
Products, Plant Biotechnology Series (Grierson, D., Ed.) Vol. 3, pp 1-54, Glasgow:
Blackie Academic and Professional Publishers, London, UK
Preiss, J., Ballicora, M. A., Laughlin, M. J ., Fu, Y., Okita, T. W., Barry, G. F.,
Guan, H. P. and Sivak, M. N. (1995) in Current Topics in Plant Physiology
(Madore, M. A., and Lucas, W.J., Eds.), Vol. 13, pp 91-99, Rockville, MD
Okita, T. W., Nakata, P., Ball, K., Smith-White, B., and Preiss, J. (1993) In Gene
Conservation and Exploitation (Gustafson, J. P., Appels, R., and Raven, P., Eds.)
pp 161-191, Plenum Press, NY
Preiss, J ., and Sivak, M. (1996) in Photoassimilate Distribution in Plants and Crops
(Zamski, E., and Schaffer, A., Eds.), pp 63-96, Marcel Dekker, Inc., NY
Smith, A. M., Denyer, K., and Martin, C. (1997) Ann. Rev. Plant Physio]. Plant
Mol. Bio]. 48, 67-87
Sivak, M., and Preiss, J. (1998) Starch: Biochemistry and Molecular Biology
Academic Press, NY, in press
Preiss, J. (1991) Oxf Surv. Plant Mol. Cell Biol. 7, 59-114
ap Rees, T., and Entvvistle, G. (1989) in Physiology, Biochemistry and Genetics of
Non-Green Plastids (Boyer, C. D., Shannon, J. C., Hardison, R. C., Eds.), pp 49-
62, Rockville, MD.

26

l3.
14.
15.

l6.
l7.

l8.
19.
20.
21.

22.
23.
24.
25.

26.
27.
28.
29.

30.

31.
32.

Heber, U., and Heldt, H. W. (1981) Ann. Rev. Plant Physio]. 32, 139-168

Heldt, H. W., Flugee, U. -I., and Borchert, S. (1991) Plant Physio]. 95, 341-343
Preiss, J. (1988) in The Biochemistry of Plants (Preiss, J ., Ed.), pp. 181-254, vol. 14,
Academic Press, San Diego

Okita, T. W. (1992) Plant Physiol. 100, 560-564

Keeling, P. L., Wood, J. R., Tyson, R. H., and Bridges, I. G. (1988) Plant Physio].
87, 311-319

Tyson, R. H., and ap Rees, T. (1988) Planta 175, 33-38

Viola, R., Davies, H. V., and Chudeck, A. R. (1991) Planta 183, 202-208

Hill, L. M., and Smith, A. M. (1991) Planta 185, 91-96

Pozueta-Romeo, J., Ardila, F., and Akazawa, T. ( 1991) Plant Physio]. 97, 1565-
1572

Leloir, L. F., and Cardini, C. E. (1955) J. Biol. Chem. 214, 157-165

Cardini, C. E, Leloir, L. F ., and Chiriboga, J. (1955) J. Biol. Chem. 214, 149-155
De Fekete, M. A. R., Leloir, L. F., and Cardini, C. E. (1960) Nature, 187, 918-919
Leloir, L. F., De Fekete, M. A. R., and Cardini, C. E. (1961) J. Biol. Chem. 236,
636-641

Recondo, E., and Leloir, L. F. (1961) Biochem. Biophys. Res. Commun. 6, 85-88
Preiss, J. (1982) Annu. Rev. Plant Physiol. 33, 431-454

Preiss, J. (1984) Annu. Rev. Microbiol. 38, 419-458

Lin, T. -P., Caspar, T., Somerville, C., and Preiss, J. (1988) Plant Physiol. 88, 1175-
1181

Lin, T. -P., Caspar, T., Somerville, C., and Preiss, J. (1988) Plant Physio]. 86, 113 l-
1135

Tsai, C. Y., and Nelson, 0. E. (1966) Science 151, 341-343

Dickinson, D. B., and Preiss, J. (1969) Plant Physiol. 44, 1058-1062

27

33.

34.

35.
36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

Mfiller-Rdber, B., Sonnewald, U., and Wilmitzer, L. (1992) EMBO J. 11,
1229-1238

Ball, S., Guan, H. -P., James, M., Myers, A., Keeling, P., Mouille, G., Buléon, A.,
Colonna, P., and Preiss, J. (1996) Cell 86, 349-352

Espada, J. (1962) J. Biol. Chem. 237, 3577-3581

Okita, T. W., Greenberg, E., Kuhn, D. N., and Preiss, J. (1979) Plant Physiol. 64,
187-192

ap Rees, T. (1984) in Storage Carbohydrates in Vascular Plants (Lewis, D. H., Ed.)
pp 53-73, Cambridge University Press, Cambridge, UK

Kim, W. T., Franceschi, V. R., Okita, T. W., Robinson, N. L., Morell, M., and
Preiss, J. (1989) Plant Physiol. 91, 217-220

Thorbjornsen, T., Villand, P., Denyer, K., Olsen, O.-A., and Smith, A. M. (1996)
PlantJ. 10, 243-250

Denyer, K., Dunlap, F., Thorbjernsen, T., Keeling, P., and Smith, A. M. (1996)
Plant Physiol. 112, 779-785

Levi, C., and Preiss, J. (1976) Plant Physiol. 58, 753-756

Iglesias, A. A., Kakefuda, G., and Preiss, J. (1991) Plant Physiol. 97, 1187-1195
Chamg, Y., Kakefuda, G., Iglesias, A. A., Buikema, W. J., and Preiss, J. (1992)
Plant Mol. Bio]. 20, 37-47

Morell, M., Bloom, M., Knowles, V., and Preiss, J. (1987) Plant Physiol. 85,
182-187

Morell, M., Bloom, M., and Preiss, J. (1988).]. Biol. Chem. 263, 633-637

Ball, K., and Preiss, J. (1994).]. Biol. Chem. 269, 24706-24711

Okita, T. W., Nakata, P. A., Anderson, J. M., Sowokinos, J ., Morell, M., and Preiss,
J. (1990) Plant Physiol. 93, 785-790

Miiller-ROber, B., Kossmann, J., Hannah, L. C., Wilmitzer, L., and Sonnewald, U.
(1990) Mol. Gen. Genet. 224, 136-146

28

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.
61.

62.
63.
64.

Anderson, J. M., Okita, T. W., and Preiss, J. (1990) in Molecular and Cellular
Biology of the Potato (Vayda, M. E., and Park, W. D., Eds.) pp 159-180, CAB
International, Wallingford, UK

Nakata, P. A., Greene, T. W., Anderson, J. M., Smith-White, B., Okita, T. W. and
Preiss, J. (1991) Plant Mol. Biol, 17, 1089-1093

Iglesias, A. A., Barry, G. F., Meyer, C., Bloksberg, L., Nakata, P. A., Laughlin, M.
J., Okita, T. W., Kishore, G. M., and Preiss, J. (1993) J. Biol. Chem. 268, 1081-
1086

Krishnan, H. B., Reeves, C. D., and Okita, T. W. (1986) Plant Physiol. 81, 642-645
Anderson, J. M., Hnilo, J., Larson, R., Okita, T. W., Morell, M., and Preiss, J.
(1989).]. Biol. Chem. 264, 12238-12242

Barton, C., Yang, L., Galvin, M., Sengupa-Gopalan, C., and Borelli, T. (1986) in
Regulation of Carbon and Nitrogen Reduction and Utilization in Maize (Shannon,
J. C., Kinevel, D. P., and Boyer, C. D., Eds.) pp 363-365, Rockville, MD

Olive, M. R., Ellis, R. J ., and Schuch, W. W. (1989) Plant Mo]. Biol. 12, 525-538
Smith-White, B., and Preiss, J. (1992) J. Mol. Evo]. 34, 449-464

Ghosh, H. P., and Preiss, J. (1965) J. Biol. Chem. 240, 960-961

Ball, 8., Marianne, T., Dirik, L., Fresnoy, M., Delrue, B., and Decq, A. (1991)
Planta 185, 17-26

Heldt, H. W., Chon, J., Maronde, D., Stan Kovic, Z. S., Walker, D. A., Krarniner,
A., Kirk, M. R., and Heber, V. (1977) Plant Physiol. 59, 1146-1155

Pettersson, G., and Ryde-Pettersson, U. (1989) Eur. J. Biochem. 179, 169-172
Neuhaus, H. E., Kruckeberg, A. L., Feil, R., and Stitt, M. (1989) Planta 178,
110-122

Neuhaus, H. E., and Stitt, M. (1990) Planta 182, 445-454

Sowokinos, J. R., and Preiss, J. (1982) Plant Physiol. 69, 1459-1466

Plaxton, W. C., and Preiss, J. (1987) Plant Physio]. 83, 105-112

29

65.
66.
67.

68.
69.

70.

71.
72.

73.
74.
75.
76.

77.

78.

79.

80.

81.

82.

83.

Hawker, J. S., and Smith, G. M. (1982) Aust. J. Plant Physio]. 9, 509-518

Hylton, C., and Smith, A. M. (1992) Plant Physiol. 99, 1626-1634

Kleczkowski, L. A., Villand, P., Lathi, E., Olsen, 0. -A., and Preiss, J. (1993) Plant
Physio]. 101, 179-186

Weber, H., Heim, U., Borisjuk, L., and Wobus, U. (1995) Planta 195, 352-361
Preiss, J., and Levi, C. (1980) in The Biochemistry of Plants (J. Preiss, ed.), Vol. 3,
pp 371-423, Academic Press, NY

Stark, D. M., Timmerman, K. P., Barry, G. F ., Preiss, J ., and Kishore, G. M. (1992)
Science 258, 287-292

Kacser, H., and Burns, J. A. (1973) Symposium Soc. Exp. Biol. 27, 65-107

Kacser, H. (1987) in The Biochemistry of Plants (Davies, D. D., Ed.) Vol. 11, pp
39—67, Academic Press, NY

Kaiser, W. M., and Bassharn, J. A. (1979) Plant Physiol. 63, 109-113

Sowokinos, J. R. (1981) Plant Physio]. 68, 924-929

Bae, J. M., Giroux, J., and Hannah, L. C. (1990) Maydica 35, 317-322

Bhave, M. R., Lawrence, S., Barton, C., and Hannah, L. C. (1990) Plant Cell 2,
581-588

Preiss, J., Danner, S., Summers, P. S., Morell, M., Barton, C. R., Yang, L., and
Nieder, M. (1990) Plant Physiol. 92, 881-885

Martin, C., and Smith, A. M. (1995) Plant Cell 7, 971-985

Li, L., and Preiss, J. (1992) Carbohydr. Res. 227, 227-239

Hannah, L. C., and Nelson, 0. E. (1976) Biochem. Genet. 14, 547-560

Villand, P., Olsen, O.-A., Kilian, A., and Kleczkowski, L. A. (1992) Plant Physiol.
100, 1617-1618

Villand, P., Aalen, R., Olsen, O.-A., Lfithi, E., LOnneberg, A., and Kleczkowski, L.
A. (1992) Plant Mo]. Bio]. 19, 381-389

Villand, P., Olsen, O.-A., and Kleczkowski, L. A. (1993) Plant Mo]. Biol. 23, 1279-

30

84.
85.

86.
87.

88.

89.
90.
91.
92.
93.
94.
95.

96.

97.
98.
99.

100.

101.

1284

Paule, M. R., and Preiss, J. (1971) J. Biol. Chem. 246, 4602-4609

Kleczkowski, L. A., Villand, P., Preiss, J., and Olsen, 0. -A. (1993) J. Biol. Chem.
268, 6228-6233

Haugen, T. H., and Preiss, J. (1979).]. Biol. Chem. 254, 127-136

Kumar, A., Tanaka, T., Lee, Y. M., and Preiss, J. (1988).]. Biol. Chem. 263, 14634-
14639

Hill, M. A., Kaufrnann, K., Otero, J., and Preiss, J. (1991) J. Biol. Chem. 266,
12455-12460

Chamg, Y., Iglesias, A. A., and Preiss, .l. (1994) J. Biol. Chem. 269, 24107-24113
Sheng, J ., Charng, Y. -Y., and Preiss, J. (1996) Biochemistry 35, 3115-3121
Parsons, T. F., and Preiss, J. (1978) J. Biol. Chem. 253, 7638-7645

Parsons, T. F., and Preiss, J. (197 8) J. Biol. Chem. 253, 6197-6202

Lee, Y. M., and Preiss, J. (1986) J. Biol. Chem. 261, 1058-1064

Rossmann, M. G., Moras, D., and Olsen, K. W. (1974) Nature 250, 194-199

Preiss, J. (1997) in Engineering Improved Carbon and Nitrogen Resource Use
Efficiency in Higher Plants (Foyer, C., and Quick, P., Eds.) pp 81-104, Taylor and
Francis Publishers, London and Washington, D. C.

Preiss, J., Ball, K., Chamg, Y., and Iglesias, A. A. (1992) in Research in
Photosynthesis (Murata, N., Ed.) Vol. 3, pp 697-700, Kluwer Academic Publishers,
Dordrecht, Netherlands

Ballicora, M. A., Fu, Y., Nesbitt, N. M., and Preiss, J. (1998) Plant Physiol. in press
Sheng, J, and Preiss, J. (1997) Biochemistry 36, 13077-13084

Ball, K., and Preiss, J. (1992) J. Protein Chem. 11, 231-238

Greene, T. W., Chantler, S. E., Kahn. M. L., Barry, G. F., Preiss, J., and Okita, T.
W. (1996) Proc. Nat]. Acad. Sci. USA 93, 1509-1513

Greene, T. W., Woodbury, R. L., and Okita, T. O. (1996) Plant Physiol. 112, 1315-

31

102.

103.

104.

105.

106.
107.

108.

1320

Ballicora, M. A, Fu, Y., Wu, M., Sheng, J., Nesbitt, N. M., and Preiss, J. (1996) in
Regulation and Manipulation of Starch and Sucrose Metabolism in Plants
(Nakamura, Y., Ed.) pp 5-11, Tsukuba, Japan

Giroux, M. J., Shaw, J., Barry, G., Cobb, B. G., Greene, T., Okita, T., and Hannah,
L. C. (1996) Proc. Natl. Acad. Sci. USA 93, 5824-5829

Mulichak, A. M., Skrzypczak-Jankum, E., Rydel, T. J., Tulinsky, A, and Preiss, J.
(1988).]. Biol. Chem. 263, 17237-17238

Lemesle-Varloot, L., Henrissat, B., Gaboriaud, C., Bissery, V., Morgat, A., and
Mornon, J. P. (1990) Biochimie. 72, 555-574

Rost, B., and Sander, R. (1993) Proc. Nat]. Acad. Sci. USA 90, 7558-7562

Meyer, C. R., Ghosh, P., Remy, E., and Preiss, J. ( 1992) J. Bacteria]. 174,
4509-4512

Kumar, A., Ghosh, P., Lee, Y. M., Hill, M. A., and Preiss, J. (1989) J. Biol. Chem.
264, 10464-10471

32

CHAPTER 2
THE SMALL SUBUNIT OF POTATO TUBER ADP-GLUCOSE
PYROPHOSPHORYLASE HAS CATALYTIC ACTIVITY

Part of the data were published in the issue of Plant Physiol. (1995) 109, 245-251
by Ballicora, M. A., Laughlin, M. J., Fu, Y., Okita, T. W., Barry, G. F., and Preiss, J.

33

Abstract

The different properties between a recombinant and native ADP-glucose
pyrophosphorylase of potato tuber were ascribed to a possible truncation at the N-
terrninus of the small subunit based on the conservation of the N-terminal sequence
between the potato tuber and spinach leaf small subunit. In this work, the cDNA of the
truncated small subunit was extended by ten amino acids at its N-terminus. Significantly,
expression of this extended cDNA of the small subunit in E. coli yielded a catalytically
active enzyme. After purifying to homogeneity, its specific activity (assayed at high
concentration of activator) is similar to that of the purified heterotetrameric enzyme
containing both the small and large subunits. The activity of the large subunit alone is
negligible. The homotetrameric small subunit differs from the heterotetrameric enzyme
only in its regulatory properties. Therefore, the small subunit is proposed to be mainly
involved in catalysis, while the large subunit is mainly involved in modulating the
regulatory properties of the small subunit of plant enzymes. The properties of the
heterotetrameric enzyme containing the large and extended small subunits are more
similar to those of the native enzyme than the heterotetrameric enzyme, which contained
the large and truncated small subunits. The N-terrninus of the small subunit may be

involved in establishing the regulatory properties and conferring the heat stability of the

holoenzyme.

34

Introduction

In spite of the importance of starch synthesis in reserve tissues, e. g., potato tuber,
no study has been done on the protein structure of the potato tuber enzyme due to the
difficulty in obtaining large amounts of pure enzyme. The genes of both the small and
large subunits of the potato tuber enzyme have been cloned (1). The cloned cDNAs have
been expressed together using two compatible vectors in an E. coli mutant devoid of
ADPGlc PPase activity (2). The recombinant enzyme is similar to the native enzyme
purified from potato with respect to most properties. However, two major differences
exist between them: the recombinant enzyme exhibits low heat stability and less
sensitivity to Pi inhibition (2). ADPGlc PPases from different sources have been reported
to be stable at 60-70 °C for 5 min (3-5). Heat treatment was used as a purification step for
the native enzyme from potato tuber (3), but it was not possible to use it as a purification
step for the recombinant enzyme (2). By comparison with the mature small subunit from
spinach leaf, it was found that the cloned small subunit from potato tuber lacked 10
amino acids of the N-tenninus (Table I). The spinach leaf enzyme is the only ADPGlc
PPase whose small subunit N-terminal sequence has been determined by protein
sequencing (4). Previous attempts to sequence the N-terminus of the subunits of the
native potato tuber enzyme were unsuccessful. The overall amino acid sequence of the
potato tuber small subunit shows 95% identity to that of the spinach leaf small subunit
and their N-tenninal sequences are almost the same (1, 6). It was proposed that the reason
to account for the different properties between the recombinant and native enzyme was
due to the truncation of about 10 amino acids at the N terminus of the potato tuber small
subunit (2).

In this work another plasmid was constructed that contains a cDNA encoding the
small subunit of the potato tuber enzyme with 10 more amino acids at the N-terminus.

The extended polypeptide was named small subunit to distinguish from the truncated

35

Table I
N-terminal Amino Acid Sequences of the Small Subunits of ADPGlc PPases

from Potato Tuber and Spinach Leaf "

 

 

source of small subunit amino acid sequence

potato tuber CDN A clone ETVSPKAVSDSQNSQTCLDPDA...
pMON 1 73 3S MALDPDA...
pMLiO MAVSDSQNSQTCLDPDA...
Spinach leaf VSDSQNSQTCLDPEA...

 

" The amino acid sequence of the small subunit from potato tuber cDNA clone (1)

was deduced from nucleotide sequence, so were those encoded by pMON17335 (2) and

pMLlO. The N-terrninal sequence of the small subunit encoded by pMLlO was finther

verified by protein sequence as described in “Materials and Methods”. The sequence of

the small subunit from spinach leaf was determined by protein sequencing (4). The

position number of the methionine residue in the small subunit from potato tuber cDNA

clone is according to Nakata et a1. (1). The underlined amino acids are those added to the

N-terminus of the truncated small subunit.

36

small subtmit used in the previous study. Significantly, the expression of this extended
small subunit yielded a homotetrameric enzyme with catalytic activity comparable to the
heterotetrameric holoenzyme when determined in the presence of high concentrations of
3PGA, whereas the large subunit had negligible catalytic activity. The possible roles of
the small and large subunits of the plant ADPGlc PPases are discussed. The recombinant
enzyme that contained the large and extended small subunits is heat stable and is as

sensitive to Pi inhibition as the native enzyme purified from potato tuber.

37

Materials and Methods

Reagents

ATP, ADPGlc, Glc-l-P, and 3PGA were purchased from Sigma. [”C]Glc-l-P and
32PPi were purchased from DuPont-New England Nuclear. All other reagents were of the
highest available commercial grade.
Plasmids

The construction of plasmids pMONl7335 and pMONl7336, which contain
cDNAs encoding the truncated small subunit and large subunit of ADPGlc PPase of
potato tuber, respectively, was described before (2). For generation of the extended
cDNA for the small subunit, a methionine was engineered at position 71 (Table I) of the
cDNA clone of the potato tuber small subunit to introduce an NcoI site using the
following oligonucleotide: 5’ -GGGGTCGCCCATGGCTGT‘TTCTGATTCG-3’ . After
amplification by PCR, the NcoI/Kpnl fragment was ligated to pMONl7335 digested with
NcoI/Kpnl to form pMLlO. The entire coding region was verified by sequencing. The
newly generated cDNA encoding a polypeptide with 10 more amino acids at the N-
terminus than the truncated small subunit encoded by the cDNA in pMONl7335
(Table I).
Assay of ADPGlc PPase

(A) Assay I. In the pyrophosphorolysis direction, enzyme activity was assayed
according to a previously described method (4). The reaction mixture contained 80 mM
glycylglycine, pH 8.0, 2 mM ADPGlc, 5 mM MgC12, 3 mM DTT, 1.5 mM 32PPi (1,000-
2,000 cpm/nmol), 3 mM 3PGA, 10 mM NaF, 200 pg mL'1 of BSA, and enzyme in a total
volume of 250 uL.

(B) Assay II. In the ADPGlc synthesis direction, enzyme activity was measured
according to a previously described method (7). The reaction mixture contained 100 mM
Hepes-NaOH, pH 8.0, 0.5 mM ["C]Glc-1-P (1,000-3000 cpm/mnol), 1.5 mM ATP, 5

mM MgC12, 3 mM DTT, 200 pg mL'1 of BSA, 0.3 unit of inorganic pyrophosphatase, and

38

enzyme in a final volume of 200 uL. In the activated condition, the concentration of
3PGA used in Assay II was 3 mM for the heterotetrameric enzyme and 24 mM for the
homotetrameric small subunit.
Kinetic Studies

For determination of kinetic parameters, the concentration of the substrate or
effector tested was systematically varied with the other substrates and effectors fixed at a
saturating concentration as described in Assay I or [1. Kinetic data were plotted as initial
velocity versus substrate or effector concentration. The kinetic constants, A0,, S0,, and
10.5, which correspond to the concentration of activator, substrate, and inhibitor giving
50% of maximal activation, velocity, and inhibition, respectively, were obtained from a
computer program using nonlinear iterative least-squares fitting to a modified Michaelis-
Menten equation (8).
Protein Assay

Protein concentration was measured using the Pierce bicinchoninic acid (BCA)
reagent (9) with BSA as the standard.
Molecular Mass Determination

The molecular weights of the recombinant heterotetrameric enzyme and the
homotetrameric small subunit were determined on a 5 to 20% sucrose density gradient in
50 mM Hepes, pH 7.5, 2 mM EDTA (10). The purified small subunit and the
recombinant enzyme containing both the extended small and large subunits had
essentially the same molecular masses of 201 i 13 and 202 i 20 kDa, respectively. Since
the large and small subunits have molecular masses of 51 kDa and 50 kDa, respectively,
both enzymes have tetrameric structure. The heterotetrameric enzyme was shown to have
two large subunits and two small subunits (11).
Expression in E. coli

Competent cells of E. coli mutant strain AC70R1-504 were transformed either

with two compatible plasmids (pMLlO or pMONl7335 with pMONl7336) or with one

39

plasmid (pMLlO, pMONl7335, or pMONl7336). The transformed bacteria were grown
in LB medium containing 25 pg/mL kanamycin and/or 70 pg/mL spectinomycin at 37 °C
on a rotary shaker. Cells were grown until OD600=1.2-1.4 and then transferred to room
temperature to add 500 uM IPTG and/or 5 ug/mL nalidixic acid for induction (Table II).
Table 11
Expression Conditions of Plasmids Containing the cDNAs encoding the Subunits of

ADPGlc PPase of Potato Tuber

 

 

plasmid subunit antibiotics inducer

pML l 0 small subunit kanamycin IPTG
pMONl7335 truncated small subunit kanamycin IPTG
pMONl7336 large subunit spectinomycin nalidixic acid

 

After ~ 40h of expression, cells were harvested at 3, 000 g for 15 min at 4 °C. The
pellet was resuspended in buffer A [50 mM Hepes-NaOH, pH 8.0, 5 mM MgC12, 1 mM

EDTA, and 20% (w/v) sucrose] at a ratio of 5 mL buffer/g of cell. Cells were then

disrupted by sonication in a Heat Systems Ultrasonic Sonicator model W-220F and then
centrifuged at 12,000 g for 15 min. The supernatant was immediately subjected to a heat
treatment procedure as indicated below. With this improved procedure for expression, the
specific activity of the recombinant enzyme (large + small) in the crude extract was at
least 20 times higher (1.4-2.0 units/mg) than that reported before (2).
Heat Treatment

Crude extract (0.2 mL) was heated for 5 min at the indicated temperature in a
1.5-mL microcentrifuge tube and then placed immediately in ice water. The treated
sample was centrifuged at 14,000 g for 10 min at 4 °C to collect supernatant. Aliquots

were withdrawn from the supernatant to assay activity in the pyrophosphorolysis

direction (Assay I). The sample without treatment was used as the control. When used as

a purification step, the crude extract was heated in a water bath at 65°C for the

40

heterotetrameric enzyme (small + large subunits) and 70 °C for the small subunit in an
Erlenmeyer flask with continuous agitation. The sample was kept for 5 min when the
desired temperature was reached, then immediately placed in an ice-water bath. The
supernatant was collected by a centrifugation step at 15,000 g for 15 min before being
frozen in liquid nitrogen. The heat-treated sample was kept at -80°C until further
purification steps.
Enzyme Purification

The recombinant enzyme containing the large subunits and the truncated small
subunits was purified as previously described (2). The small subunit and the recombinant
enzyme containing both the small and the large subunits were purified following a

described procedure with some modifications (3). The buffers did not contain glutathione

and the enzymes were loaded onto the column in the presence of 1.2 M ammonium
sulfate instead of 1.3 M potassium phosphate. For purification of the enzyme with both
the small and large subunits, an ammonium sulfate precipitation step (50% saturation)
was added after the heat treatment step. Then the pellet was dissolved in a minimal
volume of buffer A and dialyzed against the same buffer overnight. The dialyzed enzyme
was centrifuged at 20,000 g for 15 min at 4°C to remove insoluble materials before being

loaded onto the DEAE Fractogel column. The heat treatment step was used for the
purification of the small subunit and recombinant enzyme containing both the small and
the large subunits, but was not used for the recombinant enzyme containing the large
subunit and the truncated small subunit. Assay I was used to follow the activity through
the purification steps.
SDS-PAGE and lmmunoblot Blot

SDS-PAGE was performed according to Laemmli (12) on 10% polyacrylarrride
gels. The purified small subunit (8 ug) was subjected to electrophoresis in SDS-PAGE
and subsequently blotted to a PVDF membrane (Applied Biosystems Inc.). After staining

41

with Coomassie blue R250, the protein band was cut for sequencing as described
previously (13). For immunoblot analysis, proteins were transferred to a nitrocellulose
membrane and treated with affinity-purified rabbit anti-spinach leaf ADPGlc PPase IgG.
The antigen-antibody complex was visualized by treatment with alkaline phosphatase-
linked goat anti-rabbit IgG followed by staining with BM purple alkaline phosphatase
-substrate precipitating reagent (Boehringer Mannheim GmbH).
Sequence Determination

Protein sample was applied to Procise (Applied Biosystems model 494A)
automated sequencer for amino acid sequence analysis. The N-terminal sequence for the
small subunit was: AVSDSQNSQT(X)L This sequence confirms that the putative mature
small subunit contains the 10 extra amino acids (Table I). The first methionine was

processed. Residue 11, which should be cysteine, was not identified.

42

Results
Expression of cDN As Encoding the Different Subunits of ADPGlc PPase

The expression of the cDNAs of ADPGlc PPase was confirmed by resolving the
crude extract proteins on SDS-PAGE. Potato tuber ADPGlc PPases were identified by
immunoblotting with antibody against spinach leaf ADPGlc PPase. It had been shown
that the small subunit of the potato tuber ADPGlc PPase could cross-react significantly
with the anti-spinach leaf antibody, while the large subunit only reacts weakly (4). Three
enzymes (small subunit, truncated small subunit and the heterotetramer containing both
the small and large subunits) were produced at a similar level based on the result of
immunoblotting (data not shown). Their apparent sizes were the same at about 50-51
kDa.

In the crude extract, expression of pMONl7336 (encoding the large subunit)
yielded an enzyme with undetectable ADPGlc PPase activity (Table III). Expression of
both pMONl7336 and pMLlO, which encoded both the small and large subunits, yielded
an enzyme with maximal activity. Significantly, expression of pMLlO (encoding the
small subunit) resulted in an enzyme with catalytic activity (1.2 units/mg) comparable to
that of the heterotetrameric enzyme with the small and large subunits (1.8 units/mg). In
contrast, expression of pMON17 335 (encoding the truncated small subunit) gave
negligible activity, which was consistent with the previous study (2).

Heat Stability

As indicated in Table IV, the heterotetrameric enzyme containing both the
extended small subunit and large subunits is heat stable at both 60 °C (89% recovery) and
65 °C (83% recovery). This is similar to the heat stability of the native enzyme from
potato tuber (90% recovery at 60 °C and 95% recovery at 65 °C). In contrast, the
heterotetrameric enzyme containing the truncated small subunit and large subunit was not
stable at 60°C (24% recovery) and was almost completely inactivated at 65°C as

previously reported (2).

43

Table III
Activity of ADPGlc PPases Containing Diflerent Subunit(s) "

 

 

subunit composition of the enzyme enzyme activity (units/mg)
none <0.0003
truncated small subunit <0.002

large subunit <0.002

large + truncated small subunits 0.86

large + small subunits 1.8

small subunit 1.2

 

a The activity was assayed in the pyrophosphorolysis direction (Assay I) in the

crude extract.

The homotetrameric small subunit had a 93% recovery of activity in a 5-min
treatment at 70 °C. Under the same condition, the heterotetrameric enzyme containing the
large and extended small subunits lost more than 90% of its activity (data not shown).
Therefore, the small subunit is more stable than the heterotetrameric enzyme.
Purification of ADPGlc PPases

As shown in Table V, the small subunit was purified 43-fold with 46% yield,
resulting in an enzyme with a specific activity of 51.6 units/mg. The heterotetrameric
enzyme containing two subunits was purified 33-fold with a specific activity of 63.5
units/mg (Table VI), which is Slightly higher than that of the small subunit. The specific
activities of both enzymes were close to that of the highly purified native enzyme from
potato tuber (56.9 units/mg) (14).

Both enzymes were purified to apparent homogeneity estimated fi'om 4 pg of

protein on SDS-PAGE. The heat treatment step appears to give a better purification for

44

Table IV
Heat stability of ADPGlc PPases from Different Sources "

 

 

actfiity recovery (%)
Enzymes

60 °C 65 °C
homotetramer (small subunit) 98 102
heterotetramer (large + small subunits) 89 83
Heterotetramer (large + truncated small subunits) 24 1.4
native enzyme b 90 95

 

" Heat treatments were done with crude extracts for 5 min at the indicated
temperature. One hundred percent activity corresponds to 0.19, 0.17, and 0.04 unit/mg in
the homotetramer (small subunit), heterotetramer (large + small subunits), and
heterotetramer (large + truncated small subunits) enzymes, respectively. b Data is obtained

from ref. 3.

45

632.2 55 Because. 28 :8 ...:o m a: see :5 8E8 as gossameé .

 

 

2. on. 0.: 8w 8 mm o 82:
mo mew fimm 3 : mm mm 5:28 MO
K 56 o. g _ onfi S _ an Omocmnaom-m<ma
co me Nd VNE m _ m mm Go oS 525me “no:
oo. o; m:— ozw— em— 2: 88:6 25.8
o\o 30K. M2522: .32: ME Q5
23% 6:85:36 53:8 oEoomm brace 38 5886 0823 new

 

a 3.53 ONDKQV Bag SSokKo “Egan. =th 3:6 meteomxizm

> 2an

46

552.2 as. 82:20th as, Engage £8 :8

.m .B a 68 see :5 Base

83 connects;

U

 

 

9: Wm». meme 9% m.m w O 256
f I 5% <2 NA 4.8 : 5:28 .0
n: :2 63 m8 3: S aoafiadéo
9mm .3 5: :K 9% 2 .0432 $3-8
now am 3. we: as. o: 6. m8 BEBE .8;
QB o._ a; vvom 602 o: Seuss £56
ARC 36X. ME\&.E= $.33 ME .NE
22% cougar—E 33:8 058% brace :38 5896 2:29» 93m

 

.. x335 o~c~e&\e 3335. 633 + :6me 36km ONUKQV 6.2653666ch 65th concomxtam

~> 2an

47

the homotetrameric small subunit than the heterotetrameric enzyme since it was
performed at a higher temperature (70 °C) for the homotetrameric enzyme than the
heterotetramer (65 °C).

Activation by 3PGA

Interestingly, although the specific activity of the purified homotetrameric small
subunit was similar to that of the purified heterotetrameric enzyme, their allosteric
properties toward 3PGA activation were strikingly different (Figure 1A). Unlike the
heterotetrameric enzyme, no activity of the small subunit could be detected in the absence
of 3PGA (data not shown). The A05 for the homotetrameric enzyme was 15 times higher
in the synthesis direction (Table VII) and 96 times higher in the pyrophosphorolysis
direction (data not shown), indicating that the small subunit has a much lower affinity for
the activator than the heterotetrameric enzyme.

When determined in the synthesis direction, the A05 for the recombinant
heterotetrameric enzyme containing the complete small subunit was almost 3 times
higher than that of the recombinant heterotetrameric enzyme with the truncated small
subunit, resulting in an enzyme more similar to the native enzyme (Table VII).

Inhibition by Pi

The homotetrameric small enzyme was much more sensitive to inhibition by Pi.
In the presence of 3.0 mM 3PGA, 0.5 mM Pi completely inhibited the activity of the
small subunit but only inhibited 40% of the activity of the heterotetrameric enzyme
(Figure 13). At this concentration of 3PGA, the 10.5 value (80 pM) of the small subunit
was about 8 times lower than that of the heterotetrameric enzyme (Table VII).

In the presence of 0.25 mM 3PGA, the 10.5 value (0.07 mM) of the
heterotetrameric enzyme containing the large and extended small subunits was more
similar to that of the native enzyme (0.12 mM) than that of the heterotetrameric enzyme

(0.68 mM) containing the large and truncated small subunits (Table VII).

48

 

  

 

 

 

 

120 ' I 7 T ''''''' I flr'I'T'IT

100- A Fl
gg 80‘ i
t 6°— 1
,9 .' ‘
8 4°,
2 m,

0‘ ' I ' I ' I ' I Y III/I'I ''''''

0 1 2 3 4 5 “31520251”
3PGAOmM)
1mm 8

 

 

 

OAT ' .flrlbI ? I ' I V '7
00 04 08 12 16 20
Pl (NM)

4|

 

Figure 1. Comparison of the regulatory properties of the homotetrameric small
subunit (0) and heterotetrameric enzyme (small + large subunits) (I). Panel A and B
show the activation by 3PGA and inhibition by Pi of the two enzymes, respectively.
100% activity corresponds to 6.3 and 5.9 nmol/min in A, 5.8 and 11.3 mnol/min in B, for
the homotetramer and heterotetramer, respectively. Assay was determined in the
ADPGlc synthesis direction (Assay II) as described under Materials and Methods. Pi

inhibition was determined in the presence of 3 mM 3PGA.

49

.0\0O~H Cmfiumg OH Q—Dmofl—VOHQQH 0.8 mug Domau «mmo— um DEOU EODQ

00: 3008:0008 2t. .m we 80b 000830 mm 809 0 0000008 0000 was 000 N «8 80¢ 005050 E 309 2 00580200 000 .-- ..

 

 

mmd Nfio #6 0 05300 033:
.. mod bmod BAmzcsnnm =«Em 080055 + ems: 0080030080:
mod 36 E .o 3:503 :95 + 0903 0080080080:
mod n .. 0N 9553 2083 006003080:
28 28 28
<9? . -. .
28 QM a a 80 m .2 <9: 25 mg a a as . 2 6% 80 S... 8885

 

moan—om 0:80th 80¢ BEES momma; 05mQ< («0 E 0:0 <Omm 00% 3003000 03009
:> 050,—.

50

Kinetic Constants for Substrates
The $0.5 values for the substrates Mg”, ATP, and Glc-l-P were 2.0 mM, 0.120
mM, and 0.040 mM for the heterotetramer and 2.2 mM, 0.200 mM, and 0.029 mM for the

homotetrameric small subunit, respectively. No difference of more than 2-fold was

observed.

51

Discussion

The most important observation of this study is that the fiill-length small subunit
of potato tuber ADPGlc PPase is highly active (51.6 units/mg) in the presence of
relatively high concentration of 3PGA. Its specific activity is very close to that of the
purified native (56.9 units/mg; ref. 14) and cloned (63.5 units/mg) heterotetrameric potato
tuber enzyme (14). In contrast, the activity of the large subunit remains undetectable
(<0.001 units/mg) even if the concentrations of Mg”, ADPGlc and 3PGA were raised to
10 mM, 8 mM and 20 mM, respectively. The functional small subunit must exist as a
homotetramer since its native form has a molecular mass of 201kDa. The major
differences between the homotetrameric small subunit and the heterotetrameric enzyme
are their allosteric regulatory properties. The small subunit requires a much higher
concentration of 3PGA for maximal activation and is more sensitive to inhibition by Pi.
This is in agreement with a previous study that showed that an Arabidopsis mutant
ADPGlc PPase lacking the large subunit had activity but with lower affinity for 3PGA
(15). The kinetic constants for the substrates are similar between the small subunit and
the heterotetrameric enzyme. These observations suggest that: l. The small subunit is
mainly involved in catalysis; 2. The large subunit serves to modulate the allosteric
regulation of the small subunit since it increases the affinity for the activator and
simultaneously decreases the inhibition by the inhibitor. Regarding the role of the large
subunit, Ballicora et a]. recently showed that the putative activator sites in the large
subunit are not as important as the overall interaction of the large subunit with the small
subunit (16). It would be of interest to know if this is a general mechanism for all the
plant ADPGlc PPases.

This study provides a basis to answer two important questions: what are the roles
of the two subunits in the plant ADPGlc PPases? Why does the plant enzyme have a
two-subunit structure versus the one-subunit structure of the bacterial counterpart? Since

the small subunit is mainly involved in catalysis and all the ADPGlc PPases share the

52

same catalytic mechanism, it becomes apparent that the small subunit is under high
selection pressure during evolution (6). The two-subunit structure has the advantage that
by altering the large subunit it confers different sensitivity of the small subunit toward
allosteric regulation in different plants and different tissues or during different
development stages of the same plant, therefore, to allow the plants to accommodate
different physiological needs.

By inclusion of 10 amino acids in the N-terminus of the truncated small subunit, a
recombinant enzyme was obtained which is more similar to the native enzyme with
respect to heat stability and allosteric properties than the recombinant enzyme containing
the large and truncated small subunits. The kinetic constants for the substrates were very
similar between the recombinant heterotetrameric enzyme, which contains the large and
extended small subunit, and the native enzyme (data not shown). Therefore, it can be
concluded that the N-terminus of the small subunit is involved in maintaining the native
conformation of the protein that is related to the regulatory properties and heat stability.
As a matter of fact, chapter 4 shows that an intermolecular disulfide bridge between the
N-terminus of the two small subunits is responsible for the heat stability of the

heterotetrameric enzyme.

53

10.
ll.

12.

l3.

14.

15.

16.

References
Nakata, P. A., Greene, T. W., Anderson, J. M., Smith-White, B., Okita, T. W. and
Preiss, J. (1991) Plant Mol. Biol. 17, 1089-1093
Iglesias, A. A., Barry, G. F., Meyer, C., Bloksberg, L., Nakata, P. A., Laughlin, M. J.,
Okita, T. W., Kishore, G. M., and Preiss, J. (1993) J. Biol. Chem. 268, 1081-1086
Sowokinos, J. R., and Preiss, J. ( 1982) Plant Physiol. 69, 1459-1466
Morell, M. K., Bloom, M., Knoeles, V., and Preiss, J. ( 1987) Plant Physiol. 85, 182-
187
Iglesias, A. A., Kakefuda, G., and Preiss, J. (1991) Plant Physiol. 97, 1187-1195
Smith-White, B., and Preiss, J. (1992) J Mol. Evol. 34, 449-464
Ghosh, H. P., and Preiss, J. (1966) J. Biol. Chem. 241, 44914504

Canellas, P. F., and Wedding, R. T. (1980) Arch. Biochem. Biophys. 199, 259-264
Smith, P. K., Krohn, R. 1., Hermanson, G. T., Mallia, A. K., Garter, F. H.,
Provenzano, M. D., Fujimoto, E. K., Goeke, N. M., Olson, B. J., and Kenk, D. C.
(1985) Anal. Biochem. 510, 76-85

Martin, R. G., and Ames, B. N. (1961) J. Biol. Chem. 236, 1372-1379

Preiss, J. (1997) in Starch: Chemistry and Technology, 3rd Ed. (Whistler, R. L., and
BeMiller, J. N., Eds.) Academic Press, NY, in press

Laemmli, U. K. (1970) Nature 227, 680-685

LeGendre, N., and Matsudaira, P. T. (1989) in A Practical Guide to Protein and
Peptide Purification for Microsequencing (Matsudaira, P., Ed.) pp 77-84, Academic
Press, San Diego

Okita, T. W., Nakata, P. A., Anderson, J. M., Sowokinos, J ., Morell, M., and Preiss, J.
(1990) Plant Physiol. 93, 785-790

Li, L., and Preiss, J. (1992) Carbohydr. Res. 227, 227-239

Ballicora, M. A., Fu, Y., Nesbitt, N. M., and Preiss, J. (1998) Plant Physiol. in press

54

CHAPTER 3
MUTAGENESIS OF THE GLUCOSE-l-PHOSPHATE BINDING SITE OF
POTATO TUBER ADP-GLUCOSE PYROPHOSPHORYLASE

This work was submitted for publication and is currently in press — Fu, Y.,

Ballicora, M. A., and Preiss, J. (1998) Plant Physiol.

55

Abstract
Lys‘” in the homotetrameric ADPGlc pyrophosphorylase from Escherichia coli
was shown previously to be involved in the binding of the substrate, glucose-l-phosphate
(Glc-l-P). This residue is highly conserved in the ADPGlc pyrophosphorylase family.
Site-directed mutagenesis was used to investigate the function of this conserved lysine
residue in the large and small subunits of the heterotetrameric potato tuber enzyme. The
apparent affinity for Glc-l-P of the wild-type enzyme decreased 135- to 550- fold by

'98 of the small subunit to arginine, alanine, or glutamic acid, suggesting that

changing Lys
both the charge and size of this residue influence the binding of Glc-l-P. These mutations
had little effect on the kinetic constants for the other substrates (ATP and Mg2+ or
ADP-glucose and PPi), activator (3-phosphoglycerate), inhibitor (Pi), and on the thermal
stability. Mutagenesis of the corresponding lysine residue (Lysm) in the large subunit had
no effect on the apparent affinity for Glc-l-P by substitution with arginine, alanine or
glutamic acid. A double mutant, Smame, was also obtained with a 100-fold reduction
of the apparent affinity for Glc-l-P. The data indicate that Lys'98 in the small subunit is

directly involved in the binding of Glc-l-P, while they appear to exclude a direct role of

Lys213 in the large subunit in the interaction with this substrate.

56

Introduction

Chemical modification and site-directed mutagenesis studies have identified that
LysI95 in ADPGlc PPase from E. coli is involved in the binding of the substrate, Glc-l-P
(l, 2). This residue is highly conserved in the bacterial enzymes as well as in both the
small and large subunits of plant ADPGlc PPases (3). However, no studies have been
done to investigate the function of these highly conserved lysine residues in the plant
ADPGlc PPases which consist of two different subunits. The expression system for
potato tuber ADPGlc PPase (see Chapter 2) provides a useful tool to characterize the role
of the corresponding lysine residues in the small subunit, Lys‘98, as well as in the large
subunit, Lysz”, and to see if they retain the same function as in the homotetrameric E.
coli enzyme. As indicated before, recent mutagenesis studies suggest that the small and
large subunits can not be simply classified as catalytic and regulatory subunit (4),
respectively, this study will also be helpful to further clarify the roles of the small and

large subunits.

57

Materials and Methods

Reagents

ATP, ADPGlc, Glc-l-P, Man-l-P, Gal-l-P, glucuronic acid-l-P, 3PGA, and PPi
were purchased from Sigma. 32PPi and [8-‘4C]ATP were purchased from DuPont NEN.
[u-“C]Glc-l-P was from ICN Pharmaceuticals, Inc. The [a-3SS]dATP and the in vitro
mutagenesis kit were from Amersham. Enzymes for DNA manipulation and sequencing
were from New England Biolabs and US. Biochemical Corp., respectively.
Oligonucleotides were synthesized and purified by the Macromolecular Structure Facility
at Michigan State University. All other reagents were of the highest available commercial
grade.
Bacterial Strains and Media

E. coli strain TGl [F 'traD36 lac!” A(lacZ)M15 proA+B+/supEA(hde-
mch)5(r,,’m,,‘Mch’) thi A(lac-proAB)] was used for site-directed mutagenesis. E. coli
mutant strain AC70R1-504 (5), which exhibited negligible ADPGlc PPase activity, was
used for expression of the potato tuber ADPGlc PPase gene (6). Both E. coli strains were
grown in LB medium.
Site-Directed Mutagenesis

For mutagenesis, the gene for the small subunit of potato tuber ADPGlc PPase
was subcloned as an EcoRI fi'agment from pMLlO, a plasmid containing the cDNA gene
of the small subunit (6), into the EcoRI site of M 13mp18RF. The gene for the large
subunit was subcloned as an Xbal fragment from pMONl7336, a plasmid containing the
cDNA gene of the large subunit (6, 7), into the Xbal site of Ml3mpl9RF. Afier
mutagenesis, in the case of the small subunit, the mutated EcoRI fragment was exchanged
with the unmutated EcoR I fragment in pMLlO; for the large subunit, the mutated Xbal
fragment was exchanged with the unmutated Xbal fragment in pMONl7336. Site-
directed mutagenesis experiments were performed according to a previously described

method (8) using the in vitro site-directed mutagenesis kit from Amersham. Three

58

heterotetrameric mutant enzymes with a single substitution of arginine, alanine, and
glutamic acid at LysI98 of the small subunit were designated as SK198RLwta SKlggALm, and
SKlggELw respectively. The mutant enzymes with substitution of arginine, alanine, and
glutamic acid at Lysz'3 of the large subunit were designated as SMLIQUR, SMLsz, and
SthKZIBEa respectively. The double mutant enzyme, in which both Lys”)8 of the small
subunit and Lys213 of the large subunit were replaced with arginine, was designated as
SKI98RLK213R. The Oligonucleotides used to create those desired mutations are shown in
Figure 1. Prior to expression of the genes for mutant enzymes, the entire coding regions
of these mutant alleles were sequenced to verify that there were no undesired mutations.
Expression and Purification of Mutant and Wild-Type Enzymes

The single mutant enzymes were obtained by co-expressing the mutated plasmid
pMLlO (or pMONl7336) with unmutated plasmid pMONl7336 (or pMLlO) in E. coli
mutant strain AC7OR1-504. The double mutant enzyme was obtained by co-expressing
the two mutated plasmids in AC7OR1-504. Mutant enzyme SKI98RLwt was expressed as
previously described (6). The other mutant and wild-type enzymes were expressed in the
same manner except that the concentration of IPTG was increased fi'om 10 1.1M to 0.5
mM for induction. An improved procedure over the one in the previous study (6) was
used for the purification of the wild-type and mutant enzymes. In the hydrophobic
chromatography step, the enzyme was loaded onto the column in the presence of 1.2 M
ammonium sulfate instead of 1.3 M potassium phosphate buffer. After the heat treatment
step, a 50% saturation ammonium sulfate precipitation step was added, afier which the
pellet was dissolved in a minimal volume of extraction buffer [100 mM Hepes-NaOH, pH
8.0, 5 mM MgC12, 1 mM EDTA, and 20%(w/v) sucrose] and dialyzed against the same
buffer overnight. The dialyzed enzyme was centrifuged at 20,000 g for 15 min at 4°C to

remove insoluble materials before being loaded onto the DEAE Fractogel column.

59

SMALL SUBUNIT

EFAEEPQ

wild-type 5' - GAA TTT GCA GAG fl CCG CAA — 3'

3'- CTT AAA CGT CTC TTT GGC GTT - 5'
SKI98RLwt 5' - GAA TTT GCA GAG gG_A CCG CAA - 3'
legsALw. 5' - GAA TTT GCA GAG gels CCG CAA - 3'
SngL... 5' - GAA TTT GCA GAG GA_A CCG CAA - 3'
LARGE SUBUNIT

FAEEPKG

wild-type 5' - AG TTT GCT GAA AAA CCA AAA GGT TT - 3'

3'- TC AAA CGA CTT m GGT TT'I‘ CCAAA - 5'
8“,,me 3' - TC AAA CGA err 995 GGT TTT CCA AA - 5'
3...me 3'- TC AAA CGA c'r'r g1; GGT TTT CCA AA - s'
SMLmE 3'- TC AAA CGA CTT gr; GGT TTT CCA AA - 5'

Figure 1. Nucleotide sequence and the encoded protein sequence of the potato
tuber ADPGlc PPase gene in the region of Lys198 in the small subunit and Lys“3 in the
large subunit. The synthetic oligonucleotides used for site-directed mutagenesis at these
positions are shown beside the corresponding mutants they created. The codons for
position 198 in the small subunit and the anticodons for position 213 in the large submit

are underlined.

60

Assay of ADPGlc PPase

(A) Assay I. In the pyrophosphorolysis direction, enzyme activity was assayed
according to a previously described method (9). The reaction mixture contained 80 mM
glycylglycine, pH 8.0, 2 mM ADPGlc, 5 mM MgC12, 3 mM DTT, 1.5 mM 32PPi (1,000-
2,000 cpm/nmol), 3 mM 3PGA, 10 mM NaF, 200 ug mL’1 of BSA, and enzyme in a total
volume of 250 uL. The assay conditions for the mutant enzymes were identical to the
wild-type except that the amounts of MgCl2 and ADPGlc were altered for some mutant
enzymes to obtain maximal activity. The amount of MgCl2 was raised to 10 mM for
S,(,.,8ALwt and SngLm, and to 20 mM for SKI98RLwt and SK198RLK213R' For the SmmLwt and
lemLwt enzymes, 3 mM ADPGlc was used.

(B) Assay II. In the ADPGlc synthesis direction, enzyme activity was measured
according to a previously described method (10). The reaction mixture contained 100
mM Hepes-NaOH, pH 8.0, 0.5 mM [u-”C]Glc-l-P (1,000-3000 cpm/nmol), 1.5 mM
ATP, 5 mM MgC12, 3.0 mM 3PGA, 3 mM DTT, 200 ug mL’1 of BSA, 0.3 unit of
inorganic pyrophosphatase, and enzyme in a final volume of 200 [AL For assay of the
an9saqu SmgALw” SngLm, and SKIWLK213R mutant enzymes, [8-"C]ATP (about 200-
500 cpm/nmol) instead of [”C]Glc-l-P was used to monitor the synthesis of ADPGlc,
and the amount of Glc-l-P and MgCl2 were increased to 30 and 20 mM, respectively, to
obtain maximal activity. For the double mutant Klgskme enzyme, the 3PGA
concentration was increased to 10 mM in addition to the changes mentioned. Due to the
high content of Glc-l-P in the assay mixture, the time for the alkaline phosphatase
digestion was extended to overnight. Control experiments showed that the product
[”C]ADPGlc was stable during the overnight digestion.

Kinetic Studies

For determination of kinetic parameters, the concentration of the substrate or

effectors tested was systematically varied with the other substrates and effectors fixed at a

saturating concentration as described in Assay I or [1. Kinetic data were plotted as initial

61

velocity versus substrate or effector concentration. The kinetic constants, A0,, 80.5, and
105, which correspond to the concentration of activator, substrate, or inhibitor giving 50%
of maximal activation, velocity, or inhibition, respectively, as well as the interaction
coefficient, n", were obtained from a computer program using nonlinear iterative least-
squares fitting to a modified Michaelis-Menten equation (11).
Sugar Phosphate Specificity

Reactions were performed in the ADPGlc synthesis direction (Assay II) with
different sugar phosphates substituted for Glc-l-P. [8-“C]ATP (about 200-500
cpm/nmol) was used to monitor the assay. The reaction time was extended to 30 min at
37 °C. Under the conditions with Glc-l-P as substrate, the production of ADPGlc was
shown to be linear for both the wild-type and mutant enzyme SKlggALM. The time for the
alkaline phosphatase digestion was extended to 48 h. The extent of reaction was
controlled by varying the amount of enzyme used in the assay.
Thermal Stability

Enzyme samples were diluted to give the same final protein concentration (0.3 mg
mL") in a final volume of 20 11L. Dilution buffer was 50 mM Hepes-NaOH, pH 8.0, 5
mM MgC12, 1 mM EDTA, 20% (w/v) sucrose, and 1 mg mL'l BSA. The enzyme samples
were heated for 5 min in a water bath equilibrated at 60 °C, then immediately placed on
ice. The enzyme activities were assayed in the ADPGlc synthesis direction, as described
in Assay 11.
Protein Assay

Protein concentration was measured using the Pierce bicinchoninic acid (BCA)
reagent (12) with BSA as the standard.
Protein Electrophoresis and lmmunoblot

SDS-PAGE was performed according to Laemmli (13) on 10% polyacrylamide
slab gels. Following electrophoresis, proteins on the gel were visualized by staining with

Coomassie Brilliant Blue R-250 or electroblotted onto a nitrocellulose membrane (14).

62

Afier electroblotting, the nitrocellulose membrane was treated with affinity-purified
rabbit anti-spinach leaf ADPGlc PPase immunoglobulin G, and the antigen-antibody
complex was visualized by treatment with alkaline phosphatase-linked goat anti-rabbit
IgG followed by staining with BM purple alkaline phosphatase -substrate precipitating
reagent (from Boehringer Mannheim GmbH).

63

Results

Expression and Purification of Mutant Enzymes

Wild-type and mutant enzymes of potato tuber ADPGlc PPase were identified by
immunoblotting with antibody prepared against the spinach leaf ADPGlc PPase. It had
been shown that the small subunit of the potato tuber enzyme cross-reacts significantly
with the anti-spinach leaf ADPGlc PPase antibody (15). Under the same conditions, the
expression level of all the mutant enzymes were similar to that of the wild-type based on
the results of immunoblotting. The apparent sizes of these mutant polypeptides were the
same as that of the wild-type, having a molecular mass about 50 kDa. Mutant enzyme
SmmLwt was purified to more than 85% homogeneity, as estimated fiom about 4 ug
protein analyzed on SDS-PAGE. All the other enzymes were purified to greater than 95%
homogeneity.
Kinetic Characterization of SKme, Mutant Enzymes

The apparent affinity for Glc-l-P decreased dramatically when Lys198 in the small
subunit was mutated to arginine, alanine, or glutamic acid. The 805 values for Glc-l-P of
the SWRLW SKIQSALW and SWELwt enzymes were about 135-, 400-, and 550- fold higher
than that of the wild-type enzyme, respectively (Table I). Substitution of Lysw8 in the
small subunit by alanine and glutamic acid resulted in such large SQ5 changes that they
could not be accurately determined (Figure 2A). The highest Glc-l-P amount used in the
assay mixture was 40 mM, which is about 2 times higher than the 80.5 of the alanine
mutant and slightly higher than the 805 of the glutamic acid mutant, due to a solubility
problem. The interaction coefficients (nH) were changed from 1.1 for the wild-type to 1.3
to 1.8 for the mutant enzymes. The changes of the Vmax value in the synthesis direction
were 2 fold or less and in pyrophosphorolysis, were less than 4 fold, except for mutant
SmSELM. This suggests that LysI98 in the small subunit has virtually no role in the rate-

determination step of catalysis. This is consistent with the observation on the

64

Figure 2. Glc-l-P dependence for the wild-type and mutant enzymes. For wild-
type (C), SK198RLW,(A), SWSALWJI), and SK,985LW,(O) (panel A), 100% activity
corresponds to 1.2, 16.6, 33.7, 39.0 nmol/ 10 min, respectively. For wild-type (O) and
8,"me (El) (panel B), 100% activity corresponds to 1.2 and 1.6 nmol/ 10 min,
respectively. Initial velocities of the enzymes were determined in the ADPGlc synthesis
direction (Assay II) as described under Materials and Methods with the concentration of
Glc-l-P being varied. The amounts of the wild-type, SKIQSRva SmgALm, SKlssstv and
SmLme proteins were about 2.5 x 10", 70 x 10", 80 X 10'3, 2.4, and 5.5 x 10'3 pg,

respectively.

65

 

120 11
I ' I V I I 1'7" I v ' I j v

 

 

relative activity (%)

 

 

 

 

 

 

 

 

 

12o . , . . 1 . fl .

_fl-Lr—W—a— --..__..~__.._'._‘.

:\°‘ 0:
.3‘

.9 _
‘3

E .-

1io ' 1'.5 ' 2:0

Glo-1-P (mM)

Figure 2.

66

corresponding Lys195 of the E. coli ADPGlc PPase (2). Replacement with glutamic acid
not only caused the largest increase of 805 value for Glc-l-P, but also substantially
decreased the catalytic efficiency relative to the wild-type enzyme.

Considering the mutations’ large effect on the 805 of Glc-l-P, it was not surprising
to see that the apparent affinity of ADPGlc was affected (Table 1). Indeed, the So.5 value of
ADPGlc was increased 2.5- to 10- fold. The kinetic constant for the other substrate, PPi,
was increased 3- to 6- fold for the mutant enzymes (Table 11). However, these changes
were relatively small in comparison with the changes of the S05 value for Glc-l-P.
Overall, the various mutations at position 198 in the small subunit caused little or no
alteration in the apparent affinities for the other substrates (ATP and Mg”) and activator
(3PGA) (Table II). The data suggest that the conformations of those ligand-binding sites
are relatively unchanged. The 2- to 7- fold increase of the 10.5 value for the inhibitor, Pi, is
relatively small but possibly the inhibitor site may be in close proximity to the Glc-l-P
binding site.

Kinetic Characterization of SmLkm Mutant Enzymes

The apparent affinity for Glc-l-P was not affected when Lys213 in the large subunit
was replaced with arginine, alanine, or glutamic acid (Table I). As shown in Figure 2B,
there is no difference between the wild-type and SMLK213R enzyme in terms of Glc-l-P
dependence. This is also true for both the SMLK213 A and SMLKmE enzymes (data not
shown). These observations are in sharp contrast with the effect caused by mutations on
Lys198 in the small subunit (Figure 2A). Interestingly, although the apparent affinity of
Glc—l-P was not affected, the apparent affinity for ADP-glucose decreased 2- to 4- fold

2'3 of the large subunit caused small changes (less

(Table I). In general, mutations on Lys
than 4- fold) to the kinetic constants for the substrates (ATP, Mg”), activator (3PGA),

and inhibitor (Pi) (Table II).

67

Table I

Comparison of the Apparent A ffinity for Substrates of Potato Tuber Wild-Type and Mutant

ADPGlc PPases“

 

synthesis direction

pyrophosphorolysis direction

 

 

 

S05 (nH) for Vmax S0.S (nH) for Vmax
Glc- l -P (mM) (units/mg)b ADP-glucose (mM) (units/mg)
wild-type 0.057 t 0.003 (1 .1) 48 i- 1 0.20 i 0.01 (1.3) 55 :r 1
SmSRLw, 7.7 i 0.1 (1.3) 24 i 1 0.49 i 0.02 (1.1) 16 i 1
smme, 22.0 i 2.5 (1.5) 46 i 3 1.3 i 0.1 (1.0) 20 i 1
SngLw, 31.1 :27 (1.8) 1.7 i0.l 2.1 :02 (1.0) 1.43:0.1
8."me 0.044 i 0.002 (1 .1) 27 3r. 1 0.36 i 0.03 (1.8) 37 i 1
swim, 0.037 i 0.001 (1.0) 25 i 1 0.46 i 0.02 (1.7) 31 i 1
SMLK,13E 0.036 i 0.001 (0.9) 31 i 1 0.68 i 0.01 (1.8) 45 :1: 1
s,.,,,,.LK,,,R 5.6 i 0.1 (1 .5) 24 i 1 0.72 i 0.01 (1.9) 14 i1

 

3 Reactions were performed in either synthesis direction (Assay II) or pyrophosphorolysis

direction (Assay I) as described under Materials and Methods. Data represent the average

of two identical experiments i average difference of the duplicates. The values in

parentheses are the Hill interaction coefficients (nH). b One unit of enzyme activity is

expressed as the amount of enzyme required to form 1 umol of ADPGlc/min at 37 °C

when assayed in either synthesis or pyrophosphorolysis direction.

68

000300 0000000 03000 05 00.0 0000 00>» <93 SE 3 .000SN00
00820 0_w0mm 000 00.500 0% 0:3 05 00.0 28 m 003 0000 00000000000 <Omm 0.3. 0 0000005000 005000005 5: 05 000 0000500000
00 003? 0:0. 000002000 05 00 0000000000 0300.6 H 3000:0090 1000002 030 .00 0w80>0 05 000000000 0009 000602 000 03.00002

0000 0005000 00 Q A0003 00000000 max—000000000050 00 S A0003 00:00.00 06050.? 00500 E 000000.000 003 0000000m ..

 

 

 

 

 

@000 0000 a: 000 .2 8.: 0000.. 00.00.0000 8.0000000 5.00.0.0
3. 0 _ .1. mm a. 0 _0 A 0.0 8.8 00.0 H 00.0 8.8 0.0 H 00 0.. 0 0 0 0: 6:00.30
8.: N H 00 A0: _0 A 0.0 8.: 00.0 A 00.0 0.8 0.0 H 0.0 8. 0 _ 0 02 5.5.0.30
88 N ..n 0. 8.: _.0 H 00 8.8 00.0 H 2.0 8.8 _0 0 0.0 8.: 0 H 000 05.0.30

G. 0 00 H 00.0 a. 0 0.0 H 0.00 A0: _00 H 00.0 :0 _0 A 0.0 8.: X. A N00 .302;

3.00002 0.: 0000.0 8.: 0002.0 8.8 0000.0 8.0302 ...—5.26.0

$.08 00: 8.000000 304000000 0.8 0.03.0 8.: 0 00: .30....00
A0. 0 m H 8 0.. 0 0.0 0 E 3.8 _00 0 :0 :8 0.0 0. 0.0 6.: N 0. 00 600-003
Azmflk 908 a 02:0 .4000 02.8 .02 02.0 000

60 0.8 2.0 80 0.8 .20 80 0.8 ....< 80 00 2.0
00000000 max—000000000003 00:00.00 $0050.00

 

.60000KK 0&0me 020.33 00:0 0808 .003: 00038 880% 002.3 0.00005000m SEEM
: 0300.

69

Kinetic Characterization of 8“,“me Mutant Enzyme

When both Lys'98 in the small subunit and Lys213 in the large subunit were
replaced with arginine, the 805 value for Glc-l-P was about lOO-fold higher than that of
the wild-type enzyme (Table 1). Considering the l35-fold increase of the 805 value in the
mutant enzyme SKlggRLw the double mutation did not cause a further decrease in the
apparent affinity of Glc-l-P over the single mutation. In either direction of assay, the Vmax
of the double mutant enzyme is essentially the same as the single mutant enzyme,
SmSRLwt.

The double mutation did not cause much alteration in the apparent affinities for
Mg2+ and Pi (Table 11). However, the A05 value for 3PGA increased ll-fold relative to the
wild-type. This effect seemed to be additive since the A05 value for both anquLm and
S“,,LK213R increased 3-fold. The 6-fold increase of 805 value for PPi was similar to the
effect seen in the single mutant enzyme, 51(03an The other minor effects were 4-fold
increases of the 805 values for both ATP and ADP-glucose (Tables I and 11).
Sugar Phosphate Specificity

Since Lys'98 on the small subunit was implicated in Glc-l-P binding, it was of
interest to examine whether the specificity for the substrate of the SmgLwt mutant
enzymes had been changed. To analyze systematically the contribution of each specific
hydroxyl group in the binding of substrate to the active site of wild-type and mutant
enzymes, a variety of compounds whose sugar moieties differed from glucose
stereochemically or by substitution or elimination of hydroxyl groups at different
positions were used. In the measurement of the activity of mutant enzyme SKMALM, a
large amount of enzyme had to be used to obtain accurate measurements. The results
were summarized in Table 111. When 6-deoxy-Glc-l-P and 6F-Glc-l-P were used to
substitute for Glc-l-P, the activity of the wild-type and SKlggALw, enzyme decreased 3.1-
to 4.5- fold and l7- to 34- fold, respectively. When the other analogs were used to

substitute Glc-l-P, the activity of the wild-type and SKI98ALwt enzyme decreased more

70

than 14- and 150- fold, respectively. Thus, both enzymes showed the largest tolerance
towards the elimination or substitution of the hydroxyl group at C-6 of the Glc-l-P
molecule, followed by the changes at C-2, C-3 and 04. In no case was there a substrate
analog that showed a significant enhanced reactivity with the mutant enzyme as
compared to the wild-type. A substitution of a carboxyl group at C-6 indicates that
glucuronic acid-l-P is not a substrate. The sugar phosphate analogues also showed the
same pattern of effects for the mutant enzyme SKI98RLwt (data not shown). The results
indicate that the hydroxyl groups at C-2, C-3 and C-4 probably played much more
important roles than the C-6 hydroxyl group in the binding process. Overall, replacement
of Lys198 with alanine or arginine does not cause any broadened changes in substrate
specificity for sugar-l-P.
Thermal Stability

After heat treatment at 60 °C for 5 min, the activity of the wild-type enzyme
remained unchanged, while the SKI98ALwt9 SKl9SELwt, SMLKMA, SngLK213R enzymes
retained 104, 67, 94, and 105% activity, respectively. Thus, substitution of Lys198 with a
negatively charged glutamic acid made the protein more susceptible to heat inactivation.
Nevertheless, both residue 198 in the small subunit and residue 213 in the large subunit

are not critical for the stability of the native folded state of the potato tuber enzyme.

71

Table III

Specificity of Sugar-phosphates as Substrates for Wild-type and Mutant Sir/93,111....“-

 

 

substrates substrate concentration wild-type SK198 ALwt
mM unit/mg unit/mg

Glc-l-P 2 48 i l 1.5 i 0.2

10 55.8 i4.7 10.421209
6-deoxy- Glc-l-P 2 13.6 i 0.3 0.085 i 0.001

10 12.5 i 0.1 0.62 i 0.04
6F-Glc-1-P 2 13.0 i 0.5 0.044 d: 0.002

10 18.1 i 0.6 0.47b
2F-Glc-1-P 2 1.0" 0.007 i 0.001

10 4.0 i 0.1 0.018 i 0.03
mannose-l-P 2 1.7 i 0.1 0.010 i 0.001

10 4.1 i 0.2 0.058 i 0.003
3-deoxy-glucose-1-P 2 0.5 i 0.1 0.010 i 0.006

10 1.3 i 0.1 0.017 1- 0.002
3F-Glc-l-P 2 0.05 i 0.01 50.002

10 0.15 i 0.01 50.002
galactose—l-P 2 0.16 i 0.03 50.003

10 0.28 i 0.04 50.002
glucuronic acid-l-P 2 $0.01 $0.001

10 50.01 $0.002

 

3 Reactions were performed in the synthesis direction as described under Materials
and Methods, with the presence of sugar-phosphates as indicated. Data represent the
average of two duplications i standard deviation. The lower limit of detection of enzyme
activity for the wild-type is 0.01 unit/mg, for the mutant enzyme SKI98ALwt is 0.001
unit/mg when sufficient amount of enzyme was used in the assay as indicated in the text.

b Single determination.

72

Discussion

According to the results presented here, we can conclude that Lys198 in the small
subunit of potato tuber ADPGlc PPase is primarily involved in Glc—l-P binding. The 135-
to 550— fold increases of the SO,5 value for Glc-l-P when this residue was replaced by
other amino acids explains the high conservation of this lysine in plant and bacterial
ADPGlc PPases. This lysine residue probably is required for the proper substrate binding
to ADPGlc PPase under physiological concentrations of Glc-l-P. Although Lys'98 is
critical in interacting with Glc-l-P, it is obviously not essential for thermal stability. From
the moderate effect on Vmax values and the kinetic constants for ATP, Mg”, 3PGA, and Pi
(Tables I and II), LysI98 is probably neither involved in the rate-limiting step of the
catalytic mechanism nor responsible for maintaining the native conformation of the
enzyme. The interaction coefficients (11“) of Glc-l-P of the mutant enzymes were
increased to 1.3-1.8 compared to 1.1 for the wild-type. However, since both the SmmLWt
and SngLw, mutant enzymes had high S0,, values for Glc-l-P, it was impossible to
perform kinetic studies for them under saturated concentration of Glc-l-P. Only 50-70%
of Vmax could be attained based on the estimation from Lineweaver-Burk plot. Therefore,
the nH values could be overestimated for these two enzymes. Nevertheless, it has been
observed for some enzymes that a single mutation that caused decreased affinity for a
ligand resulted in an increase in cooperativity (16, 17). The phenomenon was explained
by a theory of preexisting cooperativity (17).

As the substitution of Lys'98 varied from basic to neutral to acidic amino acid, the
apparent affinities for Glc-l-P decreased. There seems to be a highly specific requirement
for a lysine residue in terms of its charge, size and shape to be present in the active site to
allow optimal binding of substrate. Even the most conservative substitution of an arginine
resulted in a mutant enzyme with l35-fold lower apparent affinity for Glc-l-P, suggesting

that charge alone is insufficient to account for proper interaction with the substrate.

73

Arginine, being a slightly larger amino acid than lysine, may sterically interfere with
substrate binding.

In contrast to the effects observed for the mutations of Lys")8 on the small subunit,
mutations of Lysz'3 on the large subunit had no effect on the 805 of Glc—l-P. When both
residues were replaced by arginine, the effect on the apparent affinity for Glc—l-P was
similar to that obtained with the single arginine substitution of the small subunit, ruling
out a direct role of Lys213 in the binding of the substrate. As indicated in the Introduction,
these two lysine residues and their surrounding sequences are highly conserved in the
ADPGlc PPase family. A sequence search on the large subunit of tuber ADPGlc PPase

2”. Therefore, it is

revealed no consensus sequence other than the region surrounding Lys
unlikely that Glc-l-P binds to an alternative site on the large subunit. This seems to be
consistent with the proposed function of this subunit, i.e., modulating the allosteric
regulation of the small subunit by 3PGA and Pi, with no direct role in catalysis. It is
worth noting that this lysine residue is replaced by glutamine in the large subunit of
ADPGlc PPase from wheat endosperm (WE7) (18), which may reflect the relative
unimportance of this residue in the large subunit. In one small-subunit isozyme, VfAGPP,
of ADPGlc PPase from Vicia faba L. seeds, this lysine is replaced by asparagine. In the
other isozyme, VfAGPC, the lysine residue is retained. Both small subunit genes are
expressed in identical temporal and spatial patterns (19). However, nothing is known
about the comparative kinetics of the two enzymes. Recent binding experiments on the
potato tuber enzyme by equilibrium dialysis (Y. Fu, and J. Preiss, unpublished results)
showed that ADPGlc bound to four sites per tetrameric enzyme. Unfortunately,
experiments to determine the number of binding sites of Glc—l-P were unsuccessful due
to the interference of ADP-Glc produced in the binding procedure. Still, there is a
possibility that Glc-l-P may bind to the large subunit, but with no catalysis after the

binding event. In any case, the data provide further evidence that the main function of the

74

small subunit is catalysis as suggested by a previous study (6). At this stage, fiirther
studies are necessary to clarify the precise role of the large subunit.

253 (252 in ref. 20) on the small subunit was suggested to be

Recently, Asp
involved in the binding of Glc-l-P and Asp'22 (121 in ref. 20) on the small subunit was
suggested to be involved in the binding of both Glc-l-P and ATP on the basis of results
obtained via random mutagenesis (20). However, replacement of both residues by
asparagine resulted in relatively small changes (less than lO-fold) of the S0,, of the
substrates compared to the effect seen with mutations of Lys'98. It is possible that those
mutations may affect directly or indirectly the conformation of the substrate domain
rather than disrupting specific interactions between the enzyme and the substrates.

The Lys195 region (FVEKP) of E. coli ADPGlc PPase is not only conserved in
potato tuber ADPGlc PPase (FAEKP), but also identical to the mannose-l-P binding site
of PMI/GMP (21). Furthermore, this motif or a closely related sequence (GVEEP,
IVEEY, KVIEP, FKE_I§P) is found in many enzymes with the common characteristic of
catalyzing the synthesis of a nucleotide diphosphate-sugar from sugar phosphate and
nucleotide tn'phosphate (Table IV). Therefore, FVEKP may be part of a sugar phosphate-
binding motif for this class of sugar nucleotide pyrophosphorylases. Of course, other
sequences must dictate the sugar specificity, e.g. for Man-l-P or Glc—l-P.

Various sugar- 1 -P analogs whose sugar moieties differed from Glc at each hydroxyl
group were tested as substrates for wild-type and mutant enzyme SK198ALwt (Table 111). No
broadened specificity for the mutant enzyme was observed. This probably suggests that
Lys'98 only participates in forming an ionic bond between its positively charged e-amino
group and the negatively charged phosphate group of Glc-l-P. Those hydroxyl groups
may interact with the side chains of the other residues in the active site, i.e. by hydrogen
bonding, to anchor the substrate correctly. Therefore, those analogs tested would have

similar effects on the wild-type as well as the mutant enzymes.

75

Table IV

Conservation of the Sugar Phosphate Binding Motif in Sugar Nucleotide

 

 

Pyrophosphorylases
Enzymes source sugar-P binding motif ’ references
ADPGlc PPase E. coli IIEFVEEPAN 2
PMI/GMP P. aeruginosa VQS * * * 1* DE 21
PMI/GMP X. campestri VER* * * * * LA 22
GDP-mannose PPase E. coli VRT* * * * *NL 23
GDP-mannose PPase S. typhimurium VAE* * * * *DI 24
UDP-glucose PPase E. coli PMVG* * * * KA 25
UDP-glucose PPase B. subtilis VKN* * * * * PK 26
UDP-glucose PPase Solanum tuberosum L. TLKI * * *Y* * 27
UDP-glucose PPase D. discoideum ETNK* I * *YK 28
CDP-glucose PPase Y. pseudotuberculosis VRS * K* * * KG 29

 

a Lys-195 in E. coli ADPGlc PPase and Lys-175 in P. aeruginosa PMI/GMP,
which were shown to bind to Glc-l-P and mannose-l-P respectively, are underlined.

* Signifies the same amino acid as in the E. coli ADPGlc PPase sequence.

76

10.

11.
12.

l3.

14.

15.

16.

References

Parsons, T. F., and Preiss, J. (1978).]. Biol. Chem. 253, 6197-6202
Hill, M. A., Kaufmann, K., Otero, J., and Preiss, J. (1991) J. Biol. Chem. 266, 12455-

2460
Preiss, J., and Sivak, M. (1996) in Photoassimilate Distribution in Plants and Crops

(Zamski, E., and Schaffer, A., Eds.), pp 63-96, Marcel Dekker, Inc., NY

Ballicora, M. A., Fu, Y., Nesbitt, N. M., and Preiss, J. (1998) Plant Physiol. in press
Carlson, C. A., Parsons, T. F., and Preiss, J. (1976) J. Biol. Chem. 251, 7886-7892
Ballicora, M. A., Laughlin, M. J., Fu, Y., Okita, T. W., Barry, G. F., and Preiss, J.

(1995) Plant Physiol. 109, 245-251
Iglesias, A. A., Barry, G. F., Meyer, C., Bloksberg, L., Nakata, P. A., Laughlin, M. J.,
Okita, T. W., Kishore, G. M., and Preiss, J. (1993) J. Biol. Chem. 268, 1081-1086

Sayers, J. R., Schmidt, W., and Eckstein, F. (1988) Nucleic Acids Res. 16, 791-802
Morell, M. K., Bloom, M., Knowles, V., and Preiss, J. (1987) Plant Physiol. 85, 182-
187

Preiss, J ., Shen, L., Greenberg, E., and Gentner, N. (1966) Biochemistry 5, 1833-1845

Canellas, P. F., and Wedding, R. T. (1980) Arch. Biochem. Biophys. 199, 259-264
Smith, P. K., Krohn R. I., Hermanson, G. T., Mallia, A. K., Garter, F. H., Provenzano,
M. D., Fujimoto, E. K., Goeke, N. M., Olson, B. J., and Kenk, D. K. (1985) Anal.
Biochem. 150, 76-85

Laemmli, U. K. (1970) Nature 227, 680-685

Burnette, W. N. (1981) Anal. Biochem. 112, 195-203
Okita, T. W., Nakata, P. A., Anderson, J. M., Sowokinos, J ., Morell, M., and Preiss, J.

(1990) Plant Physiol. 93, 785-790

Stebbins, J. W., and Kantrowitz, E. R. (1992) Biochemistry 31, 2328-2332

77

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

First, E. A., and Fersht, A. R. (1993) Biochemistry 32,13651-13657

Smith-White, B., and Preiss, J. ( 1992) J. Mol. Evol. 34, 449-464

Weber, H., Heim, U., Borisjuk, L., and Wobus, U. (1995) Planta 195, 352-361
Laughlin, M. J ., Payne, J. W., and Okita, T. W. (1998) Phytochemistry 47, 621-629
May, T. B., Shinabarger, D., Boyd, A., and Chakrabarty, A. M. (1994) J. Biol. Chem.
269, 4872-4877

Koplin, R., Arnold, W., Hotte, B., Simon, R., Wang, G., and Piihler, A. (1992) J.
Bacteriol. 174, 191-199

Marolda, C. L., and Valvano, M. A. (1993) J. Bacteriol. 175, 148-158

Jiang, X. M., Neal, B., Santiago, F., Lee, S. J., Romana, L. K., and Reeves, P. R.
(1991) Mol. Microbiol. 5, 695-713

Hossain, S. A., Tanizawa, K., Kazuta, Y., and Fukui, T. (1994) J. Biochem (Japan)
115, 965-972

Varné, D., Boylan, S. A., Okamoto, K., and Price, C. W. (1993) J. Bacteriol. 175,

3964-3971
Katsube, T., Kazuta, Y., Mori, H., Nakano, K., Tanizawa, K., and Fukui, T. (1990) J.

Biochem. (Japan) 108, 321-326
Ragheb, J. A., and Dottin, R. P. (1987) Nucleic Acids Res. 15, 3891-3906

Thorson, J. S., Kelly, T. M., and Liu, H. W. (1994) J. Bacteriol. 176, 1840-1849

78

CHAPTER 4
MECHANISM OF REDUCTIVE ACTIVATION OF POTATO TUBER
ADP-GLUCOSE PYROPHOSPHORYLASE

This work was submitted for publication — Fu, Y., Ballicora, M. A., Leykam, J.
F ., and Preiss, J. (1998) J. Biol. Chem.

79

Abstract

The catalytic activity of potato tuber (Solanum tuberosm L.) ADP-glucose
pyrophosphorylase is activated by a preincubation with ADP-glucose and dithiothreitol or
by ATP, glucose-l-phosphate, Ca2+, and dithiothreitol. The activation was accompanied
by the appearance of new sulfhydryl groups as determined by
5, 5'-dithiobis(2-nitrobenzoic acid) quantitation. By analyzing the activated and
nonactivated enzymes on SDS-PAGE under nonreducing condition, it was found that an
intermolecular disulfide bridge between the small subunits of the potato tuber enzyme
was reduced during the activation. Further experiments showed that the activation was
mediated via a slow reduction and subsequent rapid conformational change induced by
ADP-glucose. The activation process could be reversed by oxidation with
5, 5'-dithiobis(2-nitrobenzoic acid). Preincubation with ADP-glucose and dithiothreitol
could reactivate the oxidized enzyme. Chemical modification experiments with
['4C]iodoacetic acid and 4-vinylpyridine determined that the intermolecular disulfide
bridge was located between Cys'2 of the small subunits of the potato tuber enzyme.
Mutation of Cysl2 in the small subunit into either Ala or Ser eliminated the requirement
of DTT on the activation and prevented the formation of the intermolecular disulfide of
the potato tuber enzyme. The mutants had instaneous activation rates as found in the

wild-type in the reduced state. A two-step activation model is proposed.

80

Introduction

Plant ADPGlc PPases are mainly regulated by 3PGA and Pi. However, little
information is known on the possible activation of ADPGlc PPases via reduction. It is
well established that the activity of several chloroplastic enzymes is regulated by
reversible thiol/disulfide interchange (l, 2). During photosynthetic electron transport in
the light, covalent redox modification mediated by a redox chain, the
ferredoxin-thioredoxin system, leads to reductive activation of several stromal target
enzymes, e. g., fructose-1,6-bisphosphatase, NADP-malate dehydrogenase,
phosphoribulokinase, etc. (3). The activity of potato tuber ADPGlc PPase was found to
be stimulated by dithiothreitol (DTT) (4, 5). The authors had suggested the presence of
key sulfliydryl (-SH) groups at the catalytic and/or allosteric site. The mechanism of DTT
stimulation, however, is not known. In the current study, I present evidence that the
activation is due to synergism involving the enzyme interacting with both DTT and its
substrates. A reduction of the intermolecular disulfide bridge between Cys12 of the two

small subunits of potato tuber ADPGlc PPase is involved in the activation process.

81

Materials and Methods

Reagent

ATP, ADPGlc, Glc-l-P, 3PGA, inorganic pyrophosphate, and 4-vinylpyridine
were purchased from Sigma. 32PPi were purchased from DuPont New England Nuclear.
[”C]Glc-l-P and [l-‘4C]iodoacetic acid were from ICN Pharmaceuticals, Inc. All other
reagents were purchased at the highest available commercial grade.
Purification of Wild-type and Mutant Potato Tuber ADPGlc PPases

The wild-type and mutant ADPGlc PPase cDNAs were expressed as before (6).
By a described procedure (6), the wild-type enzyme was purified to apparent

homogeneity as estimated from about 4 pg protein on SDS-PAGE. The mutant proteins

were purified as the wild-type, except that the heat treatment and the hydrophobic

chromatography steps were eliminated, and a second Mono Q chromatography step was
added. For the added Mono Q step, protein elution was carried in 50 mM Hepes, pH 8.0,
5 mM MgCl,, 1 mM EDTA, and 20% sucrose using a linear NaCl gradient from 0.15 M
to 0.30 M over 20 min. The mutant enzymes were purified to about 50% homogeneity as
estimated by SDS-PAGE with about 3 pg of protein.
Assay of ADPGlc PPase

(A) Assay I. In the pyrophosphorolysis (reverse) direction, enzyme activity was
assayed according to the method of Morel] et a1. (7). The reaction mixture contained 80
mM glycylglycine, pH 8.0, 2 mM ADP-glucose, 5 mM MgC12, 3 mM DTT, 1.5 mM 32PPi
(about 1,000-2,000 cpm/nmol), 10 mM NaF, 200 ug/mL of bovine serum albumin, and
enzyme in a total volume of 250 pl. This assay was primarily used for measuring enzyme
activity during purification steps.

(B) Assay II. In the ADP-glucose synthesis (forward) direction, enzyme activity
was measured at 37 °C according to the method of Preiss et al. (8). Enzyme activity was

measured in this direction in the kinetic analysis unless otherwise indicated. The reaction

82

 

mixture contained 100 mM Hepes-NaOH, pH 8.0, 0.5 or 1 mM [”C]Glc-l-P (1,000-3000
cpm/nmol), 1.5 mM ATP, 5 mM MgC12, 3 mM DTT, 200 pg/mL of bovine serum
albumin, 0.3 unit of inorganic pyrophosphatase, and enzyme in a final volume of 200 pL.
The reaction time was 1 min unless otherwise indicated.
Reductive Activation of ADPGlc PPase

The enzyme was activated with mixture A (100 mM Hepes, pH 8.0, 2 mM
ADPGlc, and 3 mM DTT) at 37 °C for 30 min unless otherwise indicated. This condition
was referred as activation condition. In the control, the enzyme was incubated with
mixture B (100 mM Hepes, pH 8.0, and 2 mM ADPGlc). This condition was referred as
nonactivation condition.
Determination of Available Sulfhydryl Groups with DTNB

To determine the available sulfliydryl groups of the activated and nonactivated
potato tuber ADPGlc PPase, enzyme (36 pg) were incubated either under activation or
nonactivation conditions. Then samples were desalted into mixture C (100 mM Hepes,
pH 8.0, 5 mM MgC12, 1 mM EDTA, and 2 mM ADPGlc) with a Bio-Spin 30 column
(Bio-Rad, Hercules, CA). 0.48 mM DTNB was added to the desalted enzyme and Am”,
was measured every 15 s with a Beckman spectrophotometer model DU680. A
2-nitro-5-thiobenzoate extinction coefficient of 13, 600 M"cm’l was used for the
calculation (9). A molecular mass of 202 kDa (10) was used to calculate the amount of
enzyme used for DTNB measurement. The desalting procedure could efficiently remove
DTT fi'om the samples as judged by the fact that in the absence of enzyme both mixture
A and mixture B gave the same blank reading afier desalting. After removal of DTT, the
newly formed —SH groups of the activated enzyme were retained as determined by SDS-
PAGE under nonreducing condition. Furthermore, the activity of the activated and
nonactivated enzyme was fully retained after desalting as determined by a comparison of

the specific activity of the corresponding enzyme before the procedure.

83

'l-‘¢-I.

Protein Assay

Protein concentration was determined by the method of Smith et a1. (1 1).
Determination of the Reduction and Activation Time Course

Enzyme (48 pg) was incubated with mixture A at 37 °C in a final volume of 48
pL. Aliquots of 2.5 pL were withdrawn periodically for activity measurement in the
synthesis direction, in parallel, aliquots of 4 pL were withdrawn and immediately mixed
with 4 pL of 100 mM iodoacetamide to stop the reduction prior to SDS-PAGE analysis
as described below. The protein contents of the 100 kDa band (small subunit dimer) and
50 kDa band from each sample were quantified by scanning the stained gels with a
Molecular Dynamics Computing Densitometer. The reduction time course was obtained
by measuring the decrease of the protein content of the 100 kDa band with time.
SDS-PAGE and lmmunoblot Analysis

SDS-PAGE was done as described by Laemmi (12) on 10% polyacrylamide gel.
2-mercaptoethanol was not added to the protein samples under nonreducing condition.
The nonactivated potato tuber ADPGlc PPase (8 pg) was separated in SDS-PAGE under
nonreducing condition and subsequently blotted to a ProBlott membrane (Applied
Biosystems Inc.). After staining with Coomassie blue R250, the 100-kDa band (small
subunit dimer) and 50-kDa band were cut for sequencing as described previously (13).

For immunoblot analysis, proteins were transferred to a nitrocellulose membrane
and treated with affinity-purified rabbit anti-spinach leaf ADPGlc PPase IgG. The
antigen-antibody complex was visualized by treatment with alkaline phosphatase-linked
goat anti-rabbit IgG followed by staining with BM purple alkaline phosphatase -substrate
precipitating reagent (Boehringer Mannheim GmbH).
Determination of the Total Number of Disulfides

Potato tuber ADPGlc PPase (17 pg) was incubated with mixture A (activation
condition) or mixture B (nonactivation condition). 8M urea was added and the samples

were incubated 15 min at 50 °C then 30 min at 37 °C. Samples were brought to room

84

temperature and 20 mM iodoacetamide was added to block all the exposed free thiols.
Afier precipitation with 10% trichloroacetic acid (TCA), the pellets were washed three
times and dissolved with 8 M urea in 0.4 M NH4HCO3. 1 mM DTT was added and the
reduction was continued at 50 °C for 15 min then 37 °C for 30 min. 12.5 mM
[”C]iodoacetic acid (8,240 cpm/nmol) was added and the samples were incubated in the
dark for 30 min at room temperature. The reaction was stopped by addition of 50 mM
2-mercaptoethanol. The proteins were precipitated and washed as before; the pellets were
dissolved with 2% SDS in 100 mM Hepes, pH 8.0. Eight mL of scintillation liquid was
added and the samples were counted in a Packard liquid scintillation analyzer.
Identification of the Intermolecular Disulfide Bridge

(A) Direct Labeling with [”leodoacetic Acid. Potato tuber ADPGlc‘PPase (216
pg) was incubated either under activation or nonactivation conditions as described before.
Then 15 mM [”C]iodoacetic acid (8,240 cpm/nmol) was added and incubated for l h in
the dark at room temperature. [”C]carboxymethylation was terminated by acidification
with 10% TCA.

(B) Reverse Labeling. Potato tuber ADPGlc PPase (68 pg) was incubated either
under activation or nonactivation conditions as described before. 20 mM 4-vinylpyridine
was added to the incubated solution to stop the DTT dependent reduction of protein and
to block exposed thiols. 8 M urea was added and the samples were incubated 30 min at
50 °C then 1 h at 37 °C. Samples were brought to room temperature and incubated 2 h.
After precipitation with 10% TCA, the pellets were washed four times and dissolved with
25 pl 8 M urea in 0.4 M NH4HCO3; 5 pl of 24 mM DTT was added and samples were
incubated at 50 °C for 15 min then 37 °C for 30 min. Afier cooling down to room
temperature, 10 pl of 40 mM [”C]iodoacetic acid (8,240 cpm/nmol) was added and the
samples were incubated in the dark for 15 min. Carboxyrnethylation was stopped by
addition of 16 pl of 100 mM DTT.

85

(C) 4- Vinylpyridine Labeling. Potato tuber ADPGlc PPase (17 pg) was incubated
either under activation or nonactivation conditions in a final volume of 15 pl. Then 2.8 pl
of 200 mM 4-vinylpyridine was added and the samples were incubated at 37 °C for 3 h.
The labeled proteins were stored at —20 °C prior to sequence analysis.

Trypsin Digestion.

For potato tuber ADPGlc PPase after direct labeling with [”C]iodoacetic acid,
trypsin digestion was performed according to a described procedure (14). For digestion of
potato tuber enzyme after reverse labeling, water was added to adjust the final
concentration of urea to 2 M and then TPCK-trypsin was added at a trypsin to ADPGlc
PPase ratio of 1:50 (w/w). The digestion was allowed to proceed for 24 h at 37 °C.
Reaction was terminated by freezing of the samples at —20 °C.

Purification of Labeled Peptides by HPLC

The tryptic digests were applied to a C18 Vydac RP column (4.6 mm x 250 mm).
Peptides were eluted with a linear gradient formed by mixing solvent A (0.1%
trifluoroacetic acid) and solvent B (90% acetonitrile, 0.1% trifluoroacetic acid). A 5 to
60% solvent B gradient was used at a flow rate of 1.0 mL/min over 117 min. The
radioactive fractions were pooled and acetonitrile was evaporated with a Speed-Vac prior
to further separation. For fiirther purification of the pooled radioactive fractions from
direct and reverse labeling, a l35-min linear gradient (5 to 45% solvent B) and a 90-min
linear gradient (5 to 30% solvent B) was used, respectively, at a flow rate of 40 pl/min.
Both HPLC separations were performed on a microbore C,8 Vydac RP column (0.8 mm x
250 mm).

Sequence Determination
Peptides from HPLC and whole protein samples were applied to Procise (Applied

Biosystems model 494A) automated sequencer for amino acid sequence analysis.

86

Site-directed Mutagenesis.

The mutant enzymes with Ser and Ala substitutions at residue 12 of the small
subunit of potato tuber ADPGlc PPase were designated as SwsLwt and 302wa
respectively. The SmSLwt mutant was obtained by using the QuickChangeTM site-directed

mutagenesis kit (Stratagene, La Jolla, CA) (15); the SanLm mutant was obtained by

sequential polymerase chain reaction steps (16). Then the NcoI-KpnI fragment containing

the Ala substitution was transferred into the expression vector. The sequence of the whole

NcoI-KpnI fragment was verified by double-strand sequencing. The oligonucleotides used
for mutagenesis are shown below with the underlined bases introduced to replace the
codon initially encoding Cysl2 in the small subunit:

Sclzstr 5' -CGCAGAATTCACAGACATCTCTAGACCCAGATGC- 3 '
3'—GCGTCTTAAGTGTCTGTAGAGATCTGGGTCTACG- 5'
SmALw, 5 . -ATTCACAGACAGCTCTAGACCCAGATGCTAGCCGGAGT- 3 '

3 ’ -TAAGTGTCTGTCGAGATCTGGGTCTACGATCGGCCTCA- 5 '

87

Results
1. Reductive Activation of Potato Tuber ADPGlc PPase

The catalytic activity of potato tuber ADPGlc PPase was found to increase with
time (showing nonlinear kinetics) when measured in the absence of activator, 3PGA.
Various combinations of effectors were tested for their ability to activate the enzyme
during a preincubation at 37 oC.
Activation of Potato Tuber ADPGlc PPase by ADPGlc and DTT

As seen in Figure 1, both ADPGlc and DTT were required to give about lO-fold
activation of the potato tuber enzyme. In the absence of DTT, ADPGlc could slightly
activate the enzyme (close to 2-fold). However, when DTT was included in the
preincubation mixture in the absence of ADPGlc, about 70% activity was lost afier
30-min preincubation. To further examine the effect of DTT, the rate of ADPGlc
synthesis was measured in the presence or absence of DTT. When DT'T was eliminated
from the assay mixture, the enzyme was kept in a low activity form. The conversion to a
high activity form only took place when DTT was present (data not shown). This may
suggest that reduction of a disulfide bridge(s) is involved in the activation process.
Activation of Potato Tuber ADPGlc PPase by ATP, Glc-l-P, Ca2+ and DTT

Different combinations of ATP, Glc-l-P, Ca2+, and DTT were also tested for their
effect on the activation of potato tuber ADPGlc PPase. DTT was included in all the
combinations since it was required for the activation. Since catalysis would take place
when the three effectors (ATP, Glc-l-P, and Mgfi) were present together, Ca 2+ was used
as a substitute for Mg2+ to separate the activation process from the catalysis. Experiment
showed that Ca2+ could replace Mg“ as a cofactor for the potato tuber ADPGlc PPase at
about 1/10 of the rate seen with Mg", and the affinity of Ca2+ for the enzyme (S0_5= 1.8
mM; Y. Fu, and J. Preiss, unpublished results) is similar to that of Mg2+ (30.5: 2.0 mM).

These findings suggest that Ca2+ binds to the same site as Mg2+ but with lower catalytic

88

 

activity (nmol/Zm'n)

 

 

 

.. + + - + + - - 2 mM ADPGlc
- - + + - + + - 3 mM DTT
- - - - + + + + 5 mM Mg2+

.8
8
0

Figure 1. Activation of potato tuber ADPGlc PPase by ADPGlc and DTT.
Enzyme (1.7 pg) was incubated with 100 mM Hepes, pH 8.0, 0.2 mg/mL BSA, and
different additional effectors in a final volume of 8 pl for 30 min at 37 °C. The synthesis
reaction was started by adding 192 pl assay mixture into the incubated solution and

continued at 37 °C for 2 min. Control experiment was carried out without adding

additional effectors and the preincubation step was omitted.

89

efficiency. After the incubation, EGTA and Mgz+ were added to start the assay. Since
EGTA has a very high affinity for Ca2+ and a very poor affinity for Mg”, Ca“ in the
assay mixture was efficiently chelated and Mg2+ would be the cation for the reaction. This
metal exchange method was successfully used in the study of activation of chloroplast
fructose-1,6-bisphosphatase (2, 17). As shown in Figure 2, the enzyme was only activated
when all three effectors were present at the same time. With DTT present in cases where
the three effectors were not present together, the enzyme activity actually decreased from
the control value afier a 30-min preincubation.

Since 9.6 nmol of ADPGlc was produced when all three effectors were present in
the preincubation (see Figure 2 legend), another experiment was conducted to
differentiate the activation from the formed ADPGlc and that fiom ATP, Glc-l-P and
Ca2+. Figure 3 indicates that even before ADPGlc was produced in the preincubation (2-6
min), the enzyme was already activated. This demonstrates that ATP, Glc—l-P, Ca2+ could
activate potato tuber ADPGlc PPase without prior formation of ADPGlc.

Reduction of an Intermolecular Disulfide Bridge during Activation

To determine if a reduction occurred in the activation process, DTNB was used to
quantitate the available sulfhydryl groups of the potato tuber enzyme under activated or
nonactivated condition. As shown in Figure 4, the activation of potato tuber ADPGlc
PPase was accompanied by the appearance of about 2.5 new sulfhydryl groups per
tetrameric enzyme. This difference correlates with the reduction of a disulfide bridge in
the activation.

Since proteins with disulfide bridges ofien exhibit altered migration on SDS-
PAGE under nonreducing conditions, both the activated and nonactivated potato tuber
ADPGlc PPases were subjected to SDS-PAGE under nonreducing condition (Figure 5,
lanes a and b). The activated protein migrated as a single band with molecular mass

about50 kDa (Figure 5, lane a). This was in agreement with previous studies on native

90

 

_L
C) (D O
J . r A r r

.h
1 .

J

 

activity (nmol/2min)

N

 

O

- + - + + - + 1.5 mM ATP
0.5 mM GIc-1-P
_ - - - + + + 2 mM Ca2+

é

- - + + - + +
o
0

Figure 2. Activation of potato tuber ADPGlc PPase by ATP, Glc-l-P, Ca2+ and
DTT. Enzyme (2 pg) was preincubated with a mixture that contained 80 mM
glycylglycine, pH 8.0, 0.2 mg/mL BSA, 3 mM DTT and different combinations of
additional effectors in a final volume of 80 pl for 30 min. The synthesis reaction was
started by adding 60 pl incubated solution to 140 p1 of assay mixture and continued at
37 °C for 2 min. A control experiment was carried out without adding additional effectors
and the activation step was omitted. Since the enzymatic reaction would slowly take
place when three effectors were present in the preincubation, the value (10.3 nmol/2min)
was obtained after the subtraction of 9.6 nmol of ADPGlc produced in the 30-min

preincubation. Whenever Ca2+ was used, 1.6 mM EGTA was also included in the assay

mixture.

91

 

 

O.)

O
.44

8

N
.h
R

63
\

.3

m

ADPGlc formed (nmol/5min)
S
\.
K3
ADPGlc formed In the control
(nmol)

. control ,

Gab—o—o—Q , . . I . u 0
0 6 12 18 24 30

activation time (min)

 

 

 

 

Figure 3. Activation of potato tuber ADPGlc PPase by ATP, Glc-l-P, Ca2+ and
DTT with different incubation times. Enzyme (22.1 pg) was activated with a mixture
containing 80 mM glycylglycine, pH 8.0, 0.2 mg/mL BSA, 3 mM DTT, 1.5 mM ATP, 2
mM CaCl2 and 0.5 mM ['4C]G1c-l-P (942 cpm/nmol) at 37 °C in a final volume of 1,360
pl. An 80 pl aliquot was withdrawn periodically to measure the activity in the synthesis
reaction for 5 min, in parallel, an 80 p1 aliquot was withdrawn to determine the amount of
ADPGlc produced during activation as a control (0). The amount of product formed (0)

for the incubated enzyme was obtained after the subtraction of the amount of ADPGlc

produced during activation. 1.6 mM EGTA was included in the reaction mixture.

92

 

 

 

 

 

10 e I
9: l
81 1
a 7- -
E - .
9 6- ~
"' 1 1
3 5- .
o. . .
c3 4. 4
c» ‘ - ‘
I 3- nonactivated ‘
"P 2- -
1 ..
01 fi r V l ' I v r v I v T if
0 5 1o 15 20 25 30 35

time (min)

Figure 4. DTNB determination of the available sulflrydryl groups of the activated
and nonactivated potato tuber ADPGlc PPase. The experiment was performed as

described under MATERIALS AND METHODS.

93

kDa

150—

100— _-
75—

ill

50— — — fl '. w
35*- u
25— "'

[enonreducingal p—reducing—al

Figure 5. Electrophoretic analysis of the activated and nonactivated potato tuber
ADPGlc PPase. Enzyme (36 pg) was incubated under either activation or nonactivation
conditions as described under MATERIALS AND METHODS. About 1.5 pg of
activated (lanes a and d) or nonactivated enzymes (lanes b and e) were subjected to 10%
SDS-PAGE under nonreducing (lanes a and b) or reducing (lanes (1 and e) conditions.

The positions of molecular mass standards (lane 0) are indicated.

94

and cloned ADPGlc PPase (10, 18) that showed the molecular masses of the small and
large subunit were 50 and 51 kDa, respectively. Apparently, the small and large subunits
were too close to be distinguishable. The nonactivated enzyme migrated as two bands
corresponding to molecular masses of 50 and 100 kDa (Figure 5, lane b). Both bands
were transferred to a ProBlott membrane and their N-terminal sequences were
determined. For the 100 kDa band, it was: AVSDSQN; for the 50 kDa band, AVSVIT'T‘.
The former was the same as the N-terrninal sequence of the small subunit (10), and the
latter was the same as that of the large subunit deduced from cDNA sequence (18) except
that the first methionine was processed in both cases. Thus, the 100-kDa band was the
dimer of the small subunit. This result indicated the existence of an intermolecular
disulfide bridge between the small subunits of the potato tuber enzyme, which was
reduced during activation. Under reducing conditions, both the activated and nonactivated
enzyme migrated as a single band of 50 kDa (Figure 5, lane (1 and e).
Time Course of Reduction and Activation

Since a reduction step was involved in the activation of potato tuber ADPGlc
PPase, it was of interest to compare the rate of reduction with the rate of activation.
Because the reduction resulted in a shift of the dimer of the small subunit (100 kDa) to its
monomer position (50 kDa) (Figure 5), it was possible to determine the time course of the
reduction by measuring the decrease of the protein content of the 100-kDa band with
time. The result was shown in Figure 6. The rate of enzyme activation matched closely
the rate of reduction. It shows a correlation between the reduction and the activation
process, and indicates that the reduction of the intermolecular disulfide is the rate-limiting
step in the activation of potato tuber enzyme. A faint band just below the 100-kDa band

was due to protein degradation from the small subunit during storage.

95

 

 

 

 

I T I l f T I
c 1 -i O -
.Q 00 I//
*5 i K
3 /Z’ .
E) 80‘ ///’” "
5 . 0],! .

_ -I’ _

.5 60 /
/ _
8 , .
To 1/
.E 20* 1’ / ‘
a i o
E / .
ox° 0‘ '

I ' I ' I ' T ' I

0 5 10 15 20

incubation time (min)

“few ‘ ‘MW

 

 

Figure 6. Time course of potato tuber ADPGlc PPase activation (0) and reduction
(I). Activation was performed in mixture A (100 mM Hepes, pH 8.0, 2 mM ADPGlc,
and 3 mM DTT). Aliquots were withdrawn periodically for activity measurements and, in
parallel, for SDS-PAGE analysis. The enzyme activities corresponding to 0 and 100%

activation were 1.7 and 9.8 nmol/min, respectively.

96

Kinetics of Activation by ADPGlc and DTT

As both ADPGlc and DTT were required for the activation of potato tuber
ADPGlc PPase, an experiment was done to differentiate the effect of these two
compounds. As shown in Figure 7A, the potato tuber enzyme was first incubated with
DTT at 37 °C for 30 min, then DTT was removed by a rapid desalting step. Addition of
ADPGlc resulted in instaneous activation. The enzyme reached 80% of the maximal
activity within 10 sec. This is in agreement with the result in Figure 6, which shows that
the reduction is the rate-limiting step of the activation process. During the first incubation
with DTT, the enzyme activity decreased as seen in Figs. 1 and 2. When the enzyme was
first incubated with ADPGlc, subsequent addition of DTT could not activate the enzyme
in the absence of ADPGlc (Figure 78). Still, both ADPGlc and DT'T were required to
activate the enzyme. The results suggest that reduction is a prerequisite for the activation.

In Table I, when both ADPGlc and DTT were removed from the incubation
mixture, the activity of the activated enzyme decreased to below that of the nonactivated
enzyme (control). A second incubation either with ADPGlc or with ADPGlc plus DTT
reactivated the enzyme. Addition of DTT showed no effect in the reactivation process
indicating that the intermolecular disulfide bridge was already reduced. This was
confirmed by SDS-PAGE analysis (under nonreducing condition) showing that the
desalted enzyme remained reduced after the first incubation. The data indicate that the
ADPGlc induced conformational change of the reduced enzyme is reversible.
Reversibility of the Reductive Activation

When the activated (reduced) form (Figure 8A, lane b) of enzyme was incubated
with DTNB, a mobility shifi of the 50-kDa band to 100-kDa position could be observed
on SDS-PAGE under nonreducing condition (Figure 8A, lane c). Oxidation by DTNB
was accompanied by a decrease of the enzyme activity (Figure 88), suggesting the
reformation of the intermolecular disulfide bridge. The DTNB treated enzyme

97

Figure 7. Kinetics of potato tuber ADPGlc PPase activation by ADPGlc and DTT.
In panel A, potato tuber enzyme (34 pg) was first incubated with 100 mM Hepes, pH 8.0,
0.2 mg/mL BSA, and 3 mM DTT (I) in a final volume of 160 pL at 37 oC. Aliquots of
2.5 pg enzyme were withdrawn periodically to measure the activity in the synthesis
direction. After 30 min, DTT was removed by desalting the enzyme rapidly into mixture
D (100 mM Hepes, pH 8.0, 5 mM MgC12, and 1 mM EDTA). In the second incubation,
the desalted enzyme was incubated with 100 mM Hepes, pH 8.0, 0.2 mg/mL BSA, and 2
mM ADPGlc (c) at 37 °C. The enzyme activity was measured at different times. In panel
B, enzyme (42.5 pg) was first incubated with 100 mM Hepes, pH 8.0, and 2 mM
ADPGlc (A) in a final volume of 100 pL at 37 °C. Aliquots of 2.1 pg enzyme were
withdrawn periodically to measure activity as before. Afier 30 min, ADPGlc was
removed by desalting the enzyme into mixture D. In the second incubation, 7/12 of the
desalted enzyme was incubated with 100 mM Hepes, pH 8.0, 2 mM ADPGlc, and 3 mM
DTT (9); the rest was incubated with 100 mM Hepes, pH 8.0, and 3 mM DTT (0). Both
incubations were done at 37 °C. The enzyme activity was measured at different

incubation times. 100% activity corresponds to 16 nmol/min in A and 7.5 nmol/min in B.

98

 

 

 

 

 

 

 

 

99

16 ' I T f f I ' I ' I I
144 A -
.. /.\."—.\ .4
A 12- f 0 ~
.E .
E 101 r
E” ‘ ‘
5 8? -
a? 6 1 -
'5 i ADPGlc
g 4 * /D1T desalting -
2 -r -
T‘- -\I\
0 I f I f I ' I f I m T I
0 1o 20 30 4o 50 60
time (min)
I I ' T I I T
7 - -
. B /. .
6 -1 /./. “
’E o ‘
-§ 5 ~ / -1
= . f
o desalting /
E 4‘ -
5 1
2‘ 3 1 ‘
3; . o ADPGlc + DTT
~5- 2 j /ADPG|c { -
cu
‘ x
i ‘——*v
1 1L/‘ "/ EDTT ,0 7
1 o/ \\\ J
0 ‘0 '0
I I ‘ I ' I I T
o 10 20 30 40 50 60
time (min)
Figure 7.

Table I
Reactivation of Potato Tuber ADPGlc PPase
In the first incubation, enzyme (25 pg) was incubated with 2 mM ADPGlc and 3
mM DTT at 37 °C for 30 min. Then the reagents were removed by desalting the enzyme
into a solution containing 100 mM Hepes, pH 8.0, 5 mM MgC12, and 1 mM EDTA. In the
second incubation, the desalted enzyme was incubated either with 2 mM ADPGlc or 2
mM ADPGlc plus 3 mM DTT at 37 °C for 30 min. 100 mM Hepes, pH 8.0 was included
in all the incubations. Aliquots of 2.1 pg of enzyme were withdrawn after each treatment

to measure the activity.

 

 

 

Treatment Activity (nmol/min)
control 1.1 d: 0.1
first incubation 13.6 i 0.1
desalting 0.7 :1: 0.1
second incubation (ADPGlc) 8.6 :1: 0.4
second incubation (ADPGlc+DTT) 8.7 i: 0.1

 

100

Figure 8. Reversibility of the reductive activation of potato tuber ADPGlc PPase.
In the control, 38 pg of enzyme was mixed with 100 mM Hepes, pH 8.0 and 2 mM
ADPGlc in a final volume of 80 pl. Two pl of 100 mM DTT was added to initiate the
first incubation. After 30 min at 37 °C, DTT was removed by desalting the enzyme into
mixture C (100 mM Hepes, pH 8.0, 5 mM MgC12, 1 mM EDTA, and 2 mM ADPGlc).
Then 2 mM DTNB was added and the oxidation was allowed to proceed at room
temperature for 30 min. DTNB was removed by desalting as before. Reactivation was
started by incubating the desalted enzyme with 2 mM ADPGlc and 3 mM DTT in a final
volume of 45 p1 at 37 °C for 30 min. Aliquots of about 1.2 pg enzyme were withdrawn
at different stage to measure the activity in the synthesis direction (B), in parallel,
aliquots of about 4.8 pg enzyme were withdrawn for SDS-PAGE analysis under

nonreducing condition (A).

101

kDa
100— -. —.{-

50_ .

 

 

C)
1
a:

   

activity (nmol/2min)
4b

 

 

control lst incubation DTNB 2nd incubation
(ADPGlc+D1T) treatment (ADPGlc+DIT)

Figure 8.

102

could be reactivated by a second preincubation with ADPGlc and DTT. Upon

reactivation, the intermolecular disulfide bridge was reduced again (Figure 8A, lane d).

103

II. Identification of the Intermolecular Disulfide Bridge

Determination of the Total Number of Disulfide Bridge

There are a total 28 cysteine residues in potato tuber ADPGlc PPase (6 in the
small subunit and 8 in the large subunit). Prior to locating the intermolecular disulfide
bridge between the small subunits, the total number of disulfide bridges of potato tuber
ADPGlc PPase was determined by ["C]iodoacetic acid labeling. The nonactivated
(oxidized) enzyme was first denatured with urea to expose all free sulfhydryl groups,
which were blocked by subsequent addition of iodoacetamide. Then the protein was
reduced with DTT before being labeled with [”C]iodoacetic acid. In this way, only the
oxidized (disulfide) groups would be labeled. It was found that 2.7 —SH were labeled per
tetrameric protein. When this procedure was applied to the activated enzyme, 0.7 -SH
was labeled per tetrameric protein, apparently from nonspecific labeling. This indicates
that there is only one disulfide bridge in the potato tuber ADPGlc PPase.
Direct Labeling with [“C]iodoacetic acid

To determine the location of the intermolecular disulfide bridge, both the
activated and nonactivated enzyme were labeled with [”C]iodoacetic acid and then
digested with trypsin. The digests were separated by reversed-phase HPLC. As shown in
Figure 9, one major radioactive fraction (peak A, 60% of total radioactivity) was obtained
for the activated enzyme. After further purification by HPLC, sequence analysis showed
that its N-terminal sequence corresponded to Alaz-Ser‘8 in the small subunit of potato
tuber ADPGlc PPase (A in Table II). One labeled carboxyrnethylcysteine was identified
at cycle 11. For the nonactivated enzyme, the overall labeling was low (data not shown),
suggesting most sulfhydryl groups were buried in the protein. The sequences for three

other minor radioactive peaks were not determined due to their low labeling (Figure 9).

104

 

 

 

 

0.20 1000
A
‘ -300
0.15 A
E
2 —600 3
"e .E‘
5 0.10 ,2
r .-
05" Q
g —400 8
‘6
L as
0.05 (I
-200
0.00 ' . - . J o

 

 

Retention time (min)

Figure 9. HPLC separation of tryptic fragments from the activated and
l4C-carboxymethylated potato tuber ADPGlc PPase. Elution was monitored by the UV
absorption at 214 nm (solid line), and the radioactivity was determined by withdrawing
200 pl aliquots from each fraction and measured by liquid scintillation counting (dotted
line). Fractions subjected to further HPLC purification for the purpose of Edman
degradation are indicated by bars and capital letters, with further details provided in

Table II.

105

 

Table II
Sequence Analysis of the [”leodoacetic Acid Labeled T ryptic Peptides and
4- Vinylpyridine Labeled ADPGlc PPases”

 

 

 

Peptides or proteins Amino acid sequences

2

A AVSDSQNSQTCm-CLDPDAS

B” AVSDSQNSQTCm-CLDP
AVSDSQNSQTPE-CLD

C AYSVITTENDT QT
AVSDSQNSQTXC L

D AYSVITTENDTQ

 

" A and B refer to purified labeled fractions of the HPLC chromatograms of Figs.
9 and 10; C and D refer to the activated and nonactivated and 4-vinylpyridine labeled
potato tuber ADPGlc PPases, respectively. Both C and D contain two N-terminal
sequences corresponding to the small and large subunits. The sequence corresponding to
each subunit was deduced from the known sequences of the two subunits (10, 19).
Boldface sequences were assigned to the small subunit. The position number of the first
amino acid of peptide A refers to the sequence of the small subunit (10). The labeled
cysteine residues were underlined: Cm—C, carboxymethylcysteine; PE-C,
s-B-(4-pyridylethyl)-cysteine. Cm-C and PE-C were identified by comparison with the
elution times of the two standard cysteine derivatives as determined separately. b Besides
the major sequence shown in the table, there was also a minor sequence corresponding to
Alam-Lys208 in the small subunit. ‘ X indicates that no amino acid could be detected in

the cycle.

106

Reverse Labeling with ["Cliodoacetic acid

To eliminate the possibility that the labeling of A was due to the unmasking of
buried sulflrydryl groups in the activated enzyme, a reverse labeling experiment, which
would specifically label the oxidized (disulfide) groups, was performed for the
nonactivated enzyme. As shown in Figure 10, only one major radioactive peak was
obtained. Its N-terrninal sequence was determined after further HPLC separation. The
major sequence corresponded to Alaz-Pro‘s in the small subunit (B in Table 11). One
carboxyrnethylcysteine was identified at cycle 11. This is in agreement with the result
from Figure 9. There was also a minor sequence present in B corresponding to
Ala'96-Lys208 in the small subunit. It did not contain any cysteine residue. When this
procedure was performed on the activated enzyme, no significant labeling was observed
(data not shown).
4-vinylpyridine Labeling

In order to avoid any ambiguity, sequence analysis was performed on
4-vinylpyridine labeled whole protein since the N-terrninal cysteines of the small subunits
were implicated in forming the disulfide bridge. The results were expected to show two
sequences corresponding to both the small and large subunit. Based on the known
sequences of the two subunits, the sequence of each subunit could be deduced. As shown
in Table II, Cysl2 in the small subunit of the activated enzyme was labeled by
4-vinylpyridine. On the contrary, sequence analysis did not detect any residue at the same
position in the small subunit of the nonactivated enzyme, suggesting the presence of
nonderivatizable sulfhydryls (unmodified cystine is not stable during Edman
degradation). The data confirm the results obtained by direct labeling (A in Table II) and
reverse labeling (B in Table II) by [”C]iodoacetic acid. These results demonstrate that the

cysteine residues located at position 12 of the two small subunits are linked together by a

107

 

 

 

 

0.08 1000
0.06‘ ”800 E
A G.
<3: 3
V ' l i '600
$0.04- El“ u A;
I l ' l 2.:
w B ‘le O
U) ' CU
g ‘ ‘400 Q
0.02- . '; 3
L200
0.00" .
-..l' '.," ".....w‘l I ..... [HI-1' 0
20 40 60 80 100

Retention time (min)

Figure 10. HPLC separation of tryptic fragments of the nonactivated potato tuber
ADPGlc PPase afier reverse labeling with [”C]iodoacetic acid. The details are the same

as described in the legend of Figure 9.

108

disulfide bridge in the nonactivated (oxidized) potato tuber ADPGlc PPase. The disulfide
was cleaved during the activation.
Production and Purification of Mutant Enzymes

The expression of mutant ADPGlc PPase cDNAs was confirmed by resolving the
crude extract proteins on SDS-PAGE. Potato tuber ADPGlc PPases were identified by
immunoblotting with antibody against spinach leaf ADPGlc PPase that has been shown
to be reactive with the potato tuber enzyme (18). The two mutant enzymes, SCIzst, and
SmALM, were produced at a similar level to the wild-type based on the result of
immunoblotting. Their apparent sizes were the same as that of the wild-type.

To determine if the mutations prevented the formation of the intermolecular
disulfide between the small subunits, the mutant and wild-type enzymes were subjected
to SDS-PAGE under reducing and nonreducing conditions and transferred to a
nitrocellulose membrane. Immunoblotting results showed that the mutant proteins
migrated as a single band under nonreducing condition, while the wild-type migrated as
two bands corresponding to molecular masses of 50 and 100 kDa (Figure 11). Under
reducing condition, the mutant and wild-type enzymes migrated as single bands. Thus,
mutagenesis indeed eliminated the intermolecular disulfide in potato tuber enzyme. This
observation confirms the results obtained from chemical modification approaches.
Activation Characteristics of Mutant Enzymes

As shown in Figure 12, substitution of Cysl2 in the small subunit by either Ser or
Ala eliminated the requirement of DTT on the activation of potato tuber ADPGlc PPase.
Another striking difference between the mutant enzymes and wild-type was the time
course of activation. The wild-type needed 17.5 min to reach maximal activity, the two
mutant enzymes were fully activated within 10 seconds. Thus, the mutant enzymes show

the same activation characteristics as the reduced wild-type (Figure 7A)

109

            

              

      
  

....»

4:

than”.

an

if"...

a).

Punt. ..w

a... .

. .....“xi wage."
.....n........3...m..t m

. «day
uvrrr ls v
we .u! to. vo-

.. m5.

                                 

3...... "new"

         

  

v

 

   

.... not". .
an. arm-.443“.
. a... . .
.4... .....w. Iv . to n
1.5.. ......«5...

r .q #1 5.4.

.. ... .. 51}
. a «P.,»... .
... .14....

..em... 5» .. 3

w I r. ...
at“...

..q v J.
. . .

Inksrowa...“

2..ka

. ... .3. r .. ..

. .... . 1.4 l

 

   

.
a. In J
. . r I . . .
. ......» franc: . v
......m..... x 1 ....»
.... a ..1.
. Ems. .
v 45......
53.53% .
£63....va .
...v in... .... a
«3...?! r.. s y .
tyrann-

         

           

..."
3"...»

         

 

     

         
        

                         
          

 

   

    

       

. .
mud?
Err. Hat-awn

V

1(- nonreducing -)|

ing —>I

1(— reduc

type and mutant potato tuber ADPGlc

is of wild

tting analys

Figure l 1. Immunoblo

, and SanLm (lanes b and e)

Scnstt (lanes a and d)

’

PPases. Wild-type (lanes c and f)

were subjected to SDS-PAGE under reducing (lanes a-c) and nonreducing (lane d-f)

rve SpCCICS were

trocellulose membrane. Immunoreact'

ni

conditions and transferred to a

leaf ADPGlc PPase.

rnach

detected using antibody against sp

110

Figure 12. Activation kinetics of the wild-type (wt) and mutant potato tuber
ADPGlc PPases. The untreated enzymes were incubated at 37 °C for l min, then the
activation medium (2 mM ADPGlc or 2 mM ADPGlc + 3 mM DTT) was added and the
activity was measured in the synthesis reaction on aliquots of 20 pl as a function of time.
100% activity of the wild-type, SUZSLM, and SmALw, enzymes corresponds to 16.6, 8.7,

and 10.3 nmol/min, respectively. Arrows indicated the addition of ADPGlc or ADPGlc
plus DTT.

lll

100 I ' l ' I v I . I

 

   
 

 

 

 

 

 

 

 

 

 

 

 

 

< -/-
80 - / A (wt) .
60 -
—o—— ADPGlc
40 ‘ —-— ADPGlc + DTT
20 - / .D_D_D____D D
‘ l
0 «Fit i : : . : . :
100 "‘ ‘ #9
o 0 -
§ 30 ‘ 0- A0 8 (scrzstr)
3‘ . l/
I; 60 -
g - —o— ADPGlc
a) 40 ~ *0 ADPGlc + DTT
>
iii
E) 20 -
0 $7" : i : : : i = l
100 l o A ‘/\
. /0\ ' ' , .
80 - V \3Z ° °
. i c (S....L..)
60 -
40 ‘ —<>-- ADPGlc
' 0 ADPGlc + DTT
20 -
o i 3"" . . . .
0 5 10 15 20
time (min)
Figure 12.

112

Discussion

The present work shows that the activation of the potato tuber ADPGlc PPase
proceeds via a reduction of the intermolecular disulfide bridge between its small subunits
and subsequent conformational change induced by substrates. The observation that all
three ligands ATP, Glc-l-P and Ca2+ (Mgzi) are required to be present to have an
equivalent activation effect as ADPGlc on the enzyme is consistent with an ordered
binding mechanism as previously shown for ADPGlc PPases from E. coli (20), R. rubrum
(21), and barley leaf (22). ATPzMg2+ binds first, and then Glc-l-P binds. Mg” was
required for the binding of ATP, but not for the binding of ADPGlc (20). Therefore, all
ligands, ATP, Glc-l-P and Ca2+ (Mgz’) are needed to bind the catalytic sites in contrast to
ADPGlc. In this regard, it seems that both the ATP site and Glc-l-P site are required to
be occupied in order to induce the conformational change following the reduction step.

Reduction of an intermolecular disulfide bridge resulted in a shift of the dimer
band of the small subunit to its monomer position. By analyzing protein samples
withdrawn from different time points during the activation, the reduction time course
could be visualized. This technique can be conveniently used to follow the reduction
course of intermolecular disulfide bridges in proteins. Both Figure 6 and Figure 7 indicate
that the activation of the potato ADPGlc PPase is mediated via a slow reduction and a
rapid conformational change. Moreover, the enzyme needs to be reduced first in order for
the conformational change to take place. For some enzymes such as chloroplast
fi'uctose-l, 6-bisphosphatase, the rate of the reduction process is strongly accelerated by
specific conformational changes induced by modulators (23). However, inclusion of
ADPGlc did not accelerate the reduction rate of potato tuber ADPGlc PPase (data not
shown). This further indicates that the activation of potato tuber enzyme proceeds in a
sequential order with the reduction occurring first.

When used to measure the sulfhydryl groups in proteins, DTNB can result in

formation of disulfides in proteins (9). DTNB treatment reversed the activation of potato

113

tuber ADPGlc PPase by reoxidizing the reduced intermolecular disulfide bridge (Figure
8). Addition of ADPGlc and DTT reactivated the enzyme. These results suggested that
the activated dithiol form and the nonactivated disulfide form of the enzyme could be
interconverted. The intermolecular disulfide bridge seems to act as a regulator for the
activation process. Although DTNB treatment resulted in formation of a disulfide in
ADPGlc PPase, the stoichiometry that 1 mole of 2-nitro-5-thiobenzoate anion was
formed per mole of protein sulflrydryl still applied.

It is important to distinguish the activation induced by the substrates plus DTT
from that induced by the physiological activator, 3PGA. All the studies on the reductive
activation of potato tuber ADPGlc PPase were performed in the absence of 3PGA. When
3PGA was included in the reaction mixture, the enzyme showed linear kinetics and the
specific activity was about 13-fold higher in the ADPGlc synthesis reaction and 2.5-fold
higher in the pyrophosphorolysis reaction than that from the reductive activation.

Intermolecular disulfide bridges are often involved in maintaining the quaternary
structure of proteins. Besides its involvement in the activation of potato tuber ADPGlc
PPase, the intermolecular disulfide bridge between the small subunits is apparently also
involved in maintaining the enzyme in a correctly folded state. The potato tuber ADPGlc
PPase became unstable when DTT was present in the preincubation mixture (Figs. 1, 2
and 3). Once the disulfide bridge was reduced, either ADPGlc or ATP, Glc-l-P plus Ca2+
are needed to protect the enzyme from inactivation. Consistent with this observation, the
reduced wild-type and mutant enzymes were thermal labile at 60 °C for 5 min, while the
wild-type enzyme was stable at this condition (10).

Mutation of Cys12 in the small subunit into either Ala or Ser yielded mutants with
instantaneous activation rates as the wild-type in the reduced state. This suggested that
the role of Cysl2 was neither related to its hydrophobicity nor its hydrogen-bonding
capacity but specifically to its ability to form a disulfide bridge. Sequence alignment of

all the plant ADPGlc PPases available indicates that Cys'2 and its surrounding sequence,

114

-S-Q-T-C-L-D-P-, is conserved in the small subunit of enzymes from all dicot plants, e.g.
spinach leaf, Vicia faba, Beta vulgaris, Pisum sativum, Ababidopsis thaliana, etc. It is
also conserved in the small subunit of one monocot plant enzyme, that from barley leaf
(B. Smith-White, and J. Preiss, unpublished results). However, ADPGlc PPase from
spinach leaf could not be activated by ADPGlc and DTT. By analyzing the spinach leaf
ADPGlc PPase on SDS-PAGE under nonreducing condition, it was also found that an
intermolecular disulfide bridge existed between its small subunits (Y. Fu, and J. Preiss,
unpublished results). Reduction of this disulfide bridge made the spinach leaf enzyme
heat labile as the case of the potato tuber enzyme. Information regarding the reductive
activation of the other ADPGlc PPases with the conserved Cys is not available.

Several chloroplastic enzymes are regulated by reversible thiol-disulfide
interchange mediated by light controlled ferredoxin-thioredoxin system (2). Interestingly,
the same potato ADPGlc PPase small subunit gene is expressed both in tubers
(non-photosynthetic tissue) and leaves (photosynthetic tissue) (24). However, the
expression level in leaves is significantly lower than that in tubers. It is not clear if the
same potato tuber ADPGlc PPase is also expressed in potato leaves. Reduced thioredoxin
from Spirulina was substituted for DTT to activate the potato tuber ADPGlc PPase,
however, no significant effect could be observed. There is still a possibility that a wrong
type of thioredoxin was used. The physiological importance of the reductive activation
phenomenon in this enzyme is still unclear as in vivo, the enzyme may be continuously
exposed to the activator 3PGA. Nevertheless, the possibility cannot be discarded that an
indigenous reductant plays a role in the fine regulation of the potato tuber enzyme.

In summary, the activation mechanism of potato tuber ADPGlc PPase is shown in
Figure 13. The intermolecular disulfide bridge between the small subunits locks the
protein in the nonactivated conformation. Reduction frees the enzyme, and subsequent
binding of ADPGlc induces a rapid conformational change of the enzyme to the activated

state. Removal of ADPGlc converts the enzyme to its nonactivated dithiol form, while

115

reoxidation of the intermolecular disulfide bridge converted the enzyme back to its
nonactivated disulfide conformation. For clarity, only the small subunits are shown, but it
must be kept in mind that the reduction of the intermolecular disulfide bridge probably

leads to a rearrangement of the small and large subunits during the activation.

 

 

     

 

Reductant +ADPGlc
(DTT) - =
____' V
Reduction -ADPGIc
(810W) Conformational
changa (fast)
Nonactivated Nonactivatad Activated
dlsultlda form dithiol form dlthlol tor-m
Oxldatlon
(DTN B)

Figure 13. A proposed model for the reductive activation mechanism of potato
tuber ADPGlc PPase. Only the small subunits are shown. The amino-terminal extensions

are shown on the outside of each subunit.

116

P9!”

10.

ll.

12.
13.

14.

15.
16.

References
Buchanan, B. B. (1980) Annu. Rev. Plant Physio]. 31, 341-374
Wolosiuk, R. A., Ballicora, M. A., and Hagelin, K. (1993) FASEB J. 7, 622-637
Scheibe, R. (1991) Plant Physiol. 96, 1-3
Sowokinos, J. R., and Preiss, J. (1982) Plant Physiol. 69, 1459-1466
Sowokinos, J. R. (1981) Plant Physiol. 68, 924-929
Fu, Y., Ballicora, M. A., and Preiss, J. (1998) Plant Physiol. In press
Morell, M. K., Bloom, M., Knowles, V., and Preiss, J. ( 1987) Plant Physiol. 85, 182-
187
Preiss, J ., Shen, L., Greenberg, E., and Gentner, N. (1966) Biochemistry 5, 264
Habeeb, A. F. S. A. (1972) in Methods in Enzymology (Hirs, C. H. W., and
Timasheff, S. N., Eds.) Vol. 25, pp 457-464, Academic Press, New York
Ballicora, M. A., Laughlin, M. J., Fu, Y., Okita, T. W., Barry, G. F., and Preiss, J.
(1995) Plant Physiol. 109, 245-251
Smith, P. K., Krohn R. 1., Hermanson, G. T., Mallia, A. K., Garter, F. H., Provenzano,
M. D., Fujimoto, E. K., Goeke, N. M., Olson, B. J., and Kenk, D. K. (1985) Anal.
Biochem. 150, 76-85
Laemmli, U. K. (1970) Nature 227, 680-685
LeGendre, N., and Matsudaira, P. T. (1989) in A Practical Guide to Protein and
Peptide Purification for Microsequencing (Matsudaira, P., Ed.) pp 77-84, Academic
Press, San Diego
Stone, K. L., and Williams, K. R. (1989) in A Practical Guide to Protein and Peptide
Purification for Microsequencing (Matsudaira, P., Ed.) pp 43-69, Academic Press,
San Diego
Papworth, C., Brarnan, J ., and Wright, D. A. (1996) Strategies 9 (1): 3-4
Ausubel, M. F ., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J.
A., and Struhl, K. (1997) Current Protocols in Molecular Biology, John Wiley &

117

Sons, Inc., New York

17. Hertig, C. M., and Wolosiuk, R. A. (1983) J. Biol. Chem. 258, 984-989

18. Okita, T. W., Nakata, P. A., Anderson, J. M., Sowokinos, J ., Morell, M., and Preiss, J.
(1990) Plant Physiol. 93, 785-790

19. Nakata, P. A., Greene, T. W., Anderson, J. M., Smith-White, B., Okita, T. W. and
Preiss, J. (1991) Plant Mo]. Bio]., 17, 1089-1093

20. Haugen, T. H., and Preiss, J. (1979) J. Biol. Chem. 254, 127-136

21. Paule, M. R., and Preiss, J. (1971).]. Biol. Chem. 246, 4602-4609

22. Kleczkowski, L. A., Villand, P., Preiss, J ., and Olsen, O. (1993) J. Biol. Chem. 268,
6228-6233

23. Ballicora, M. A., and Wolosiuk, R. A. (1994) Eur. J. Biochem. 222, 467-474

24. Nakata, P. A., Anderson, J. M., and Okita, T. W. (1994) J. Biol. Chem. 269, 30798-
30807

118

CHAPTER 5

SUMMARY AND PERSPECTIVES

119

SUMMARY AND PERSPECTIVES
1. Summary

The expression of the full-length cDNA encoding the small subunit of potato
tuber ADPGlc PPase in E. coli results in a catalytically active enzyme. After purifying to
homogeneity, its specific activity (51.6 units/mg) is close to that of the purified native
(56.9 units/mg; ref. 1) and cloned (63 units/mg) heterotetrameric potato tuber enzyme. In
contrast, the activity of the large subunit is negligible (<0.001 units/mg). The major
differences between the homotetrameric small subunit and the heterotetrameric enzyme
are their allosteric regulatory properties by 3PGA and Pi. The small subunit requires
much higher concentrations of 3PGA for maximal activation and is more sensitive to
inhibition by Pi. The kinetic constants for the substrates are similar between the small
subunit and the heterotetrameric enzyme. These observations suggest that the small
subunit is mainly involved in catalysis and the large subunit functions to modulate the
sensitivity of the small subunit toward allosteric regulation.

The N-terminus of the small subunit is likely involved in the regulatory properties
and heat stability of the heterotetrameric enzyme since these properties are altered in a
recombinant enzyme that contains the truncated small subunit (at the N-terminus) and the
large subunit (2). As a mater of fact, it is the N-terminal intermolecular disulfide bridge
between the small subunits that is responsible for the heat stability of the enzyme
(Chapter 4). The expression system with the full-length small subunit and large subunit
provides a useful tool to identify important residues or domains in the catalysis and
regulation of the holoenzyme.

The lysine residues corresponding to Lys198 in the small subunit and Lys213 in the
large subunit of the heterotetrameric potato tuber enzyme are highly conserved in the
homotetrameric bacterial enzymes, and in both the small and large subunits of plant
ADPGlc pyrophosphorylases. The conserved lysine residue (Lys'gs) in the

homotetrameric E. coli enzyme was shown to be involved in the binding of the substrate,

120

Glc-l-P (3). The functions of the conserved lysine residues in the two subunits of the
heterotetrameric potato tuber enzyme were investigated by site-directed mutagenesis
methods. The apparent affinity for Glc-l-P of the wild-type enzyme decreased 135-, 400-,

'98 of the small subunit with arginine, alanine, and glutamic

and 550— fold by replacing Lys
acid, respectively. The results suggest that both the charge and size of this residue
influence Glc-l-P binding. Those mutations had little effect on the kinetic constants for
the other substrates (ATP and Mg” or ADP-glucose and PPi), activator (3PGA), inhibitor
(Pi), and on the thermal stability. Experiments to substitute Glc-l-P with various
sugar-l-P analogs indicates that the hydroxyl groups at C-2, C-3 and C-4 of the glucose
unit played much more important roles than the C-6 hydroxyl group in the binding

8 with either alanine or arginine does not cause any

process. Replacement of Lysl9
broadened changes in substrate specificity for sugar-l-P. These observations suggest that
Lys'98 probably only participates in forming an ionic bond between its positively charged
a-amino group and the negatively charged phosphate group of Glc-l-P.

In contrast to the effects seen from mutations on the conserved residue (Lys'98) of
the small subunit, mutagenesis of the conserved residue (Lysm) in the large subunit had
no effect on the apparent affinity for Glc—l-P by substitution with arginine, alanine or
glutamic acid. A double mutant, SKI98RLK213R’ was also obtained with a 100-fold reduction
of the apparent affinity for Glc-l-P. This effect is similar to that from the single arginine
mutation on the small subunit. These findings indicate that Lys'98 in the small subunit is
directly involved in the binding of Glc-l-P, while Lys213 in the large subunit is not. This
is consistent with the proposed function of the two subunits as mentioned in the
preceding text.

Chapter 4 focuses on the study of the mechanism of reductive activation of
heterotetrameric potato tuber enzyme. The intermolecular disulfide bridge between the

small subunits locks the protein in the nonactivated conformation. Reduction fi'ees the

enzyme, and subsequent binding of ADPGlc induces a rapid conformational change of

121

the enzyme to the activated state. Removal of ADPGlc converts the enzyme to its
nonactivated dithiol form, while reoxidation of the intermolecular disulfide bridge
converted the enzyme back to its nonactivated disulfide conformation. Therefore, the
activated dithiol form and the nonactivated disulfide form of the enzyme could be
interconverted.

Both chemical modification and site-directed mutagenesis approaches were used
to locate the intermolecular disulfide bridge between the small subunits. Analysis of the
tryptic digest of the activated and MC-carboxymethylated potato tuber enzyme revealed
that the radioactive label was mostly incorporated in Cys'2 of the small subunit. The same
sulflrydryl was found to be labeled for the nonactivated enzyme by a reverse labeling
procedure. These results were further confirmed by N-tenninal sequence analysis of 4-
vinylpyridine labeled potato tuber ADPGlc PPase. S-B-(4-pyridylethyl)-cysteine was
identified at position 12 in the small subunit of the activated enzyme, while it was not
detected in the same position in the small subunit of the nonactivated enzyme. Mutation
of Cys12 in the small subunit into either Ala or Ser eliminated the requirement of DTT on
the activation and prevented the formation of the intermolecular disulfide of the potato
tuber enzyme. The mutants had instantaneous activation rates as the wild-type in the
reduced state. These findings indicate that the intermolecular disulfide bridge is located

between Cysl2 of the small subunits of the potato tuber ADPGlc PPase.

2. Perspectives

Although the small subunit is mainly implicated in catalysis and the large subunit
functions to modulate the regulatory properties of the small subunit, the precise roles of
the two subunits are not well understood. The C-terrninal region of the large subunit of
the enzymes from both maize endosperm and potato tuber was found to be important for
allosteric regulation (4). However, recent mutagenesis studies showed that the putative

activator sites in the small subunit play more important roles in the interaction of the

122

activator with the potato tuber enzyme than the homologous sites in the large subunit (5).
Therefore, the small and large subunits do not simply act as catalytic and regulatory
subunits, respectively, in the holoenzyme. Although mutagenesis experiments have
provided very useful information concerning the structure-function relationships of
ADPGlc PPase, the best approach is to crystallize the enzyme and determine its three-
dimensional structure. By comparison of the three-dimensional structure of the
homotetrameric small subunit with that of the heterotetrameric holoenzyme, the function
and the interaction of the two subunits can be understood to a much greater extent. It can
also give insight into the regulatory mechanism of the enzyme. This can be quite
beneficial since the structure of the potato tuber enzyme can be viewed as a model of
those of the other plant ADPGlc PPases. It is reasonable to expect that the current basic
research will be transformed into future improvement of starch content in plants,
especially in the storage organs of the major crops

Many enzymes known to be activated by DTT were later found to be modulated
by thioredoxin, for example, NADP—glyceraldehyde-3-phosphate dehydrogenase,
fructose- l ,6-bisphosphatase, sedoheptulose- l ,7-bisphosphatase, phosphoribulokinase,
and NADP-malate dehydrogenase (for review, see ref. 6). Although reduced thioredoxin
from Spirulina could not activate the potato tuber enzyme, this may be simply due to a
specificity problem. Different types of thioredoxin, e.g., thioredoxin-f or thioredoxin-m,
from different sources could still be tested on the activation of the potato tuber enzyme.

In vivo, ADPGlc PPase is mainly under the allosteric regulation by 3PGA and Pi.
It would be of interest to characterize the kinetic constants for different effectors of the
Cys'2 mutants (SCIZSLwr and Saul-m) and compare them with those of the wild-type in the
oxidized state. There is a possibility that reduction can make the enzyme work more
efficiently, for example, by increasing the affinity for the activator rather than increasing
the Vmax in the presence of activator. On the contrary, oxidation can decrease the affinity

of the enzyme for the activator. This might be physiologically relevant since the enzyme

123

is believed to operate in a highly inhibited environment (7). Under this circumstance,
sensitivity to the activator is important for the function of the enzyme.

The physiological importance of the reductive activation phenomenon in the
potato tuber enzyme is unclear. This issue is also complicated by the fact that the potato
tuber is a nonphotosynthetic tissue and little information is available regarding redox
changes inside the amyloplast. Nevertheless, since Cys12 and its surrounding sequence is
conserved in the small subunit of ADPGlc PPases from all dicot plants sequenced to date,
it will be interesting to test if a similar reductive activation mechanism also exists in
photosynthetic tissues. Determination of the intermolecular disulfide bridge will also be
important to understand the quaternary structure of ADPGlc PPase. It can provide very
helpful information to solve the three-dimensional structure after getting the x-ray
diffraction pattern of the enzyme, which is the focus of current research to understand the

structure basis for the function of this enzyme.

124

References
Okita, T. W., Nakata, P. A., Anderson, J. M., Sowokinos, J ., Morell, M., and Preiss,
J. (1990) Plant Physiol. 93, 785-790
Iglesias, A. A., Barry, G. F., Meyer, C., Bloksberg, L., Nakata, P. A., Laughlin, M.
J., Okita, T. W., Kishore, G. M., and Preiss, J. (1993) J. Biol. Chem. 268, 1081-
1086
Hill, M. A., Kaufmann, K., Otero, J., and Preiss, J. (1991) J. Biol. Chem. 266,
12455-12460
Giroux, M. J., Shaw, J., Barry, G., Cobb, B. G., Greene, T., Okita, T., and Hannah,
L. C. (1996) Proc. Nat]. Acad. Sci. USA 93, 5824-5829
Ballicora, M. A., Fu, Y., Nesbitt, N. M., and Preiss, J. (1998) Plant Physiol. in press
Buchanan, B. B. (1980) Annu. Rev. Plant Physiol. 31, 341-374
Nakata, P. A., and Okita, T. W. ( 1994) in Molecular and Cellular Biology of the
Potato, 2nd Ed. (Belknap, W. R., Vayda, M. E., and Parks, W. D., Eds.), pp 31-44,
CAB lntemational, Wallingford, UK

125

l'l 9N STQ

TE UNIV. LIBRARIES
111111 1111111111111111111111111
29 0 1 688 753 5

1

111