ABSTRACT

CHARACTERIZATION OF THE 3.58 ALDOLASE INTERMEDIATE.AS A
DIMEB.AND ANALYSIS FOR DIMER CATALYTIC ACTIVITY

by Stanley P. Blatti

The first major aim of this research was to study the
mechanism of subunit association of native rabbit muscle
aldolase. Two initial problems had to be resolved: (1) the
subunit structure of aldolase had to be clarified; the native
enzyme was thought to have either 3 or 6 subunits, and (2)
the 3.58 obligate intermediate in the acid dissociation and
reassociation of aldolase had to be characterized by physical
techniques.

A detailed subunit molecular weight analysis of rabbit
muscle aldolase using the dissociation medium of 5.6M
guanidine HCl and 0.1M B-mercaptoethanol produced, upon
extrapolation to zero protein concentration, a weight-average
molecular weight (MS) of h2,000 for the subunits. This value,
together with the characterization of the 3.55 intermediate
as a dimer of aldolase subunits, provided support for the
four subunit model.

In the pH dissociation of rabbit muscle aldolase, the
best conditions for production of the 3.58 intermediate were
found to be pH 3.35. low salt (0.01M NaCl), and 0°-2°. The
extrapolated sedimentation coefficients and diffusion coef-
ficients under these conditions were calculated to be

330." = 3.455 and Dgo,w = 3.73 x 10"7 cmZ/second. reSpec-

Stanley P. Blatti
tively. These values yielded a molecular weight of 83,000-
86,000. However, some aggregate was found to be present;
correction for this would reduce the molecular weight by h%
to 80,000-83,000, a value which is most consistent with a
dimer of aldolase subunits (u2,000). This was further
supported by the fact that the intrinsic viscosity of 1Q.0
cc/g for the intermediate was found to be almost mid-way
between that for the native enzyme (4 cc/g) and that for the
acid dissociated subunits (24.0 cc/g).

Thus the most likely mechanism for subunit associa-
tion into tetrameric aldolase was eXpected to involve two
consecutive bimolecular association reactions: (1) monomer
association into dimers, followed by (2) dimer association
into tetramers.

The discovery and proof that the 3.58 intermediate
was a dimer, together with the fact that aldolase contained
at least 3 catalytic sites, led to the second major aim of
this research: to answer the question of whether aldolase
dimers could possess catalytic activity independent of their
'partner dimer' in the native tetramer. But before this
question could be examined, a study was conducted to find
the proper conditions for complete reversibility of activity
from pH 3.35 dimers. The following conditions were found to
give 100% recovery of activity: (1) pH 7.9, (2) 0.1M B-
mercaptoethanol, (3) 1 hour incubation at 0° followed by a
2 hour incubation at 20°, and (4) reversal concentrations

between 0.013 mg/ml and 0.065 mg/ml. The rate of activity

Stanley P. Blatti
recovery was found to be second-order with reSpect to dimer
concentration (0.09 mg/ml to 0.18 mg/ml), and the second-
order rate constant was calculated to be k = 1.35 x 10LP
liters/mole-second at 16°. These results suggested that the
pH 3.35 dimer was inactive and had to associate into tetramers
before a catalytic activity was regained.

In the course of these studies, it was discovered that
FDP inhibited dimer association at pH 5.0 and inhibited
activity reversal in the assay at pH 7.5. At 50% inhibition
of activity recovery, the dissociation constant for FDP was

found to be k s u x 10'3M; this was 500 times the binding

FDP
constant of PEP for the native enzyme.

Since dimers at pH 3.35 were inactive, the possibility
of producing active dimers under conditions of higher pH
values was investigated. When the pH of pH 3.35 dimers was
raised to pH 4.0 or pH 5.0, association was inhibited: how-
ever, the reactivation which occurred in the assay showed
essentially zero initial rates so that dimers at these pH
values must have also been inactive. In contrast, the pH
5.5 dimers, which were incubated and separated by sucrose
density centrifugation sedimentation velocity (SDSV), gave
activity immediately with little or no lag times. For this
reason all subsequent studies were completed at pH 5.5.

Dimer association was found to be concentration dependent
as eXpected: only about 3% dimer association had occurred

in a 20 hour period at very low protein concentrations

(0.03 mg/ml) and -6°. On the other hand, incubation at

Stanley P. Blatti

°, and 12°) stimulated dimer asso-

higher temperatures (0°, 4
ciation. Sucrose concentrations of h%, 16%, and 20% could
effectively inhibit dimer association. In conclusion to

this section, pH 5.5 dimers subjected to SDSV separation
exhibited immediate activity upon assaying with essentially
no lag time; therefore, it was under these conditions (except
for the SDSV analysis) that the final analysis for active
dimers was investigated.

The last section then of this work was devoted
entirely to proving whether the pH 5.5 dimer at -6° was
active when assayed at pH 7.5 and 25°. To examine this ques-
tion, a kinetic analysis of activity reversal, together with
the physical analysis of the dimer-tetramer distribution in
the assay, were performed to exclude all reaction mechanisms
but one.

The physical analysis of the dimer-tetramer distribu-
tion in the assay at various times during the reversal
process showed increased tetramer formation. The kinetics
of the reversal process in the assay was shown to be first-
order at relatively high dimer concentrations; however, at
lower dimer concentrations and under improved reversal condi-
tions in the assay of 1 mg/ml BSA and 0.1M B-mercaptoethanol,
the half-times for reversal increased, indicating a shift
towards a second-order process. The rate constants for the
first-order and second-order reactions were calculated to be

4

k1 = 2.# x 10'3/second and k2 = 6.# x 10 liters/mole-second,

reSpectively. The only mechanism which was consistent with

Stanley P. Blatti
this data was the following: (1) inactive dimer association
into an inactive tetramer followed by (2) a first-order
folding reaction of the inactive tetramer into an active
tetramer. In conclusion under all the conditions studied
here, aldolase dimers were found to be inactive-~only the

tetramers had activity.

CHARACTERIZATION OF THE 3.53 ALDOLASE
INTERMEDIATE AS A DIMER AND ANALYSIS
FOR DIMER CATALYTIC ACTIVITY

By

Stanley P. Blatti

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1968

Dedicated

to my wife, Jane

and to my son, Todd.

11

ACKNOWLEDGMENTS

The author wishes to express his sincere appreciation
to Dr. W. C. Deal, Jr. for his valuable guidance, encourage-
ment and understanding throughout the course of this work.
The stimulating discussions with George Johnson, George
Stancel. Dr. S. Yang, and Dr. S. Constantinides are also
appreciated. The author also wishes to thank Dr. W. A. Wood
and Dr. P. K. Kindel for serving on his guidance committee
and Mrs. Shirley Randall for her asSistance in the prepara-
tion of this manuscript.

The author is especially grateful to his wife, Jane,
for her love and encouragement and for her hard work as a
secretary to help with the finances throughout the course of
this work. The support of a National Defense Education Act,
Title IV, Fellowship is also appreciated.

iii

VITA

Stanley P. Blatti was born in Kasson, Minnesota, on
April 28, 1942. He graduated from Kasson-Mantorville High
School on June 6, 1960, and then attended St. Olaf College
where his interest in chemistry was first stimulated by
Dr. Finholt. Mr. Blatti continued his undergraduate studies
at the University of Minnesota where his interest in chem-
istry was further stimulated by Dr. Kreevoy and Dr. Noland.
He once was asked to leave his apartment because fumes from
his makeshift laboratory were causing some discomforts to
his upstairs neighbors. He married Jane M. Kirkwood in 1962
and a year later a son, Todd, was born. They moved into a
house where Mr. Blatti built another laboratory of a more
permanent nature equipped with steam baths, gas, and even a
hood. He received a B.A. degree in 1964 from the University
of Minnesota and then worked as a research assistant for
Dr. L. M. Henderson, chairman of Biochemistry Department at
the University of Minnesota. In his work there, he synthe-
sized 5 and 6 carbon fatty acid pantothiel anhydrides using
microtechniques, the results of which were eventually pub-
lished. Mr. Blatti then attended Michigan State University
to continue his education in the field of biochemistry under
the direction of Dr. W. C. Deal, Jr. In.August of 1968 he
will have received the degree of Doctor of Philosophy with a

iv

VITA--Continued

biochemistry major.

Upon completion of his work at Michigan State Univer-
sity the author plans to attend the animal microbiology
course offered by Dr. Phillip Marcus at the Cold Spring
Harbor Laboratory of Quantitative Biology during the late
summer of 1968. He then plans to continue his research
training with Dr. Marcus at Albert Einstein College of
Medicine on the general problem of regulation of protein
synthesis by translational inhibitory proteins in such
cellular systems as the viral interferon-induced cells, the
mitotic phase of cell division, and eventually the unferti-
lized egg.

Mr. Blatti was awarded a National Defense Education
Act, Title IV, Predoctoral Fellowship in 1965 to complete
his graduate work and a National Institutes of Health
Post-doctoral Fellowship to continue his training in bio-
chemistry with Dr. Marcus. Mr. Blatti was also accepted

for membership to the American Chemical Society.

 

TABLE OF CONTENTS

INTRODUCTION ‘0 C O O O O O O O O O O O 0 O O O O O 0

LITERATURE REVIEW . . . . . . . . . . . . . . . . .

I.
II.

III.

IV.
METHODS
I.

II.

III.
IV.
V.

General Properties of Aldolases . . . . . .

Physical Properties of Rabbit Muscle
AldOlaSe O O O O O O O O O 0 O O O O O O

A. Native Enzyme . . . . . . . . . . . . .
B. Subunits O O O O O 0 O O O O 0 O O O 0

Chemical and Catalytic Properties of Rabbit
Muscle Aldolase . . . . . . . . . . . . .

A. C-Terminal Analysis . . . . . . . . . .
B. N-Terminal Analysis . . . . . . . . . .
C. Sulfhydryl Content . . . . . . . . . .
Isozymes of Aldolase . . . . . . . . . . .

Preparation and Assay of Rabbit Muscle
Aldolase................

Ultracentrifugal Analysis . . . . . . . . .

A. Sedimentation Velocity and Diffusion
EXperlmentS.............

B. Sedimentation Equilibrium and.Archibald
Molecular Weight Analysis . . . . . .

V180081tyooooooooooooocoo.
Bio-Gel P-150 C01‘umn Preparation 0 o o o o

Sucrose Density Gradient Centrifugation . .

vi

11
13
13
14

15
18

18
19

19

21
23
25
25

TABLE OF CONTENTS-~Continued
Page
RESULT S C O O O O O O O O O O O O O O O O O O O O O O 2 7

I. The.Approach to the Problem and Initial
StUdieS o o o o o o o o o o o o o o o o o 27

A. Determination of Rabbit Muscle Aldolase
Subunit Molecular Weight in Guanidine
H01 and B-M8rcaptoethan01 o o o o o o 28

B. Determination of the pH Range for Forma-
tion of the 3.58 Aldolase Intermediate 31

II. The Nature of the 3.58 Aldolase Intermediate 38

A. Analysis for a Rapid Equilibrium by an
Investigation of the Concentration
Dependence of Sedimentation Coeffi-
cient and Molecular Weight by the
Archibald Technique . . . . . . . . . 43

B. Sedimentation Coefficient, Diffusion
Coefficient, and Molecular Weight
M(S/D) of the 3.58 Intermediate at
pH BOuo O O O O O 0 O O O O O O O O O “8

C. Sedimentation Coefficient, Diffusion
Coefficient, and Molecular Weight
M(8/D) of the 3.58 Intermediate at
PH 3035 O O O O O O O 0 O O O O O O O 50

D. Determination of the Intrinsic Viscos-
ity of the 3.58 Intermediate of
Aldolase and Comparison with That
for the Native and SUbunits o o o o o 61

E. Analysis for Incomplete Dissociation
and Non-Specific Aggregation of
Dimers: Attempts to Separate Dimers
From Higher Molecular Weight Material
and Estimate Dimer Purity . . . . . . 70

1. Effect of Salt and Temperature on
Homogeneity of Dimer Preparations 70

2. Attempts to Separate Higher Molecu-
lar Weight Material from Dimers ... 74

3. An Estimation of the Purity of the
Dimer Preparation by Comparing
the Areas of the Dimer and Native
Enzyme Peaks in the Sedimentation
VGIOCIty Patterns 0 o o o o o o o 80

vii

TABLE OF CONTENTS--Continued

III. Characterization of the Reactivation Reac-
tion from pH 3.3 Dimers o o o o o o o o o

A. Effect of B-Mercaptoethanol on Native
Enzyme and Reversal of Catalytic
Activity..............

B. Effect of pH on Reassociation from pH
3.3 Dimers at Various Times of Incu-
batlonatooandZOo.......o

C. Effect of Dimer Concentration and Time
of Incubation in the Dissociation
Medium: Attainment of Conditions for
90%-100% Recovery of Activity . . . .

D. Recovery as a Function of Time at Two
Temperatures . . . . . . . . . . . .

E. Kinetics of Reversal Process at Two
Concentrations . . . . . . . . . . .

F. Effect of Concentration on Half-Time of
Reactivation in the Assay (16°) . . .

G. Effect of FDP on Reactivation of pH 3.3
Dimers in the Assay . . . . . . . . .

IV. A Study Preliminary to the Analysis for
Active Dimers of Aldolase: Attainment of
Aldolase Dimers Under Conditions Where
Native Tetramers are Active . . . . . . .

A. Test for Active Dimer Production by
Dissociation of Tetramers . . . . . .

B. Test for an Active Dimer by Studying
the Effect of Various Conditions on
Selectively Folding (Without Associ-
ation) the pH 3.35 Unfolded Dimer at
Higher PH values 0 o o o o o o o o o

1. Effect of pH on the Reassociation
of Dimers to Tetramers after a
15 Minute Incubation . . . . . .

2. Concentration De endence of Reasso-
ciation at pH .0 and pH 5.5 . .

a. Effect of Dimer Concentration
on Reassociation at pH 4.0
after a 15 Minute Incubation

viii

Page

82

84

87

9O

94

97

100

.103

107

108

112

116

119

119

TABLE OF CONTENTS-~Continued
Page

b. Effect of Dimer Concentration
on Reassociation at pH 5.5
after 3.5 Hours . . . . . . . . 120

3. Effect of Incubation Time on Asso-
ciation at Various Concentrations
atpH5.5.............123

4. Effect of Sucrose Concentration on
Dimer Association at pH 5.5 . . . . 128

5. Effect of Incubation at 0° and -60
for Various Times at 0.03 mg/ml
and pH 5.5 o o o o o o o o o o o o 137

6. Effect of Incubation at a Higher
Temperature (12°) on Dimer
Association . . . . . . . . . . . . 141

7. Effect of Substrate on Dimer Asso-
ciation at pH 5.0 After a 4 Hour
Incubation . . . . . . . . . . . . 144

V. Analysis for Active Dimers . . . . . . . . . 149

A. Test for Dimer Association After the
SDSV Analysis but Before the Assay
Procedure . . . . . . . . . . . . . . . 151

1. Effegt of Post- SDSV Incubation at
4°, and 12° on Activity; An
IndireCt Test 0 O O O O O O O C O O 151

2. Direct Test for Possible Tetramers
in the Dimer Peak . . . . . . . . . 155

B. Test for Dimer Association in the Assay . 158

1. Kinetics of Activity Reversal from
pH 5.5 Dimers (0°) Not Subjected
to SDSV Analysis: Determination
of the Rate Constant and the Order
of Activity Recovery From pH 5.5
Dimers at 25° and pH 7.5 . . . . . 161

2. Direct Test for Possible Dimer Asso-
ciation in the.Assay: SDSV Analy-
sis After Incubation of Dimers in
the Assay . . . . . . . . . . . . . 168

C. Proof that Dimers are Not Catalytically
Active . . . . . . . . . . . . . . . . 173

ix

TABLE OF CONTENT8--Continued
Page

1. Stimulation of Reactivation in the
Assay with Bovine Serum Albumin
and/or B-Mercaptoethanol to Aid
Later Kinetic Analysis at Lower
Concentrations . . . . . . . . . . 173

2. Kinetic Analysis of Reversal in the
Assay Over 150-Fold Concentration
Range (0.00016 mg/ml to 0.025 mg/
ml): Demonstration that Dimer
Association into Tetramers, Fol-
lowed by a Tetramer Conformational
Change, Precedes Catalytic Activ-
ity O O O O O O O O O O O O O O 0 177

DISCUSSION 0 O O O O O O O O O O O O O O O O O O 0 O 1 81

BIBLIOGRAPHY O O O O O O O O O O O O O 0 O O O O O O 191

LIST OF TABLES

Table Page

I. Physical Properties of 3.58 Aldolase Inter-
mediate O O O O O O O O O O O O O 0 O O 0 O O 60

II. Physical Properties of Aldolase Monomers,
Dimers and Tetramers . . . . . . . . . . . . 69

xi

LIST OF FIGURES

Figure Page

1. Extrapolation of the apparent weight-average
P) and apparent z-average (Mapp)
moIecular weights to zero concengration . . . 30

2. The effect of pH on the structure of aldolase . 34

3. Sedimentation velocity patterns of aldolase
for the pH range between pH 3.1 and pH 3.5 . 36

4. The most likely mechanisms by which a tetramer
or trimer native enzyme could give rise to
8.3.531ntermedlate.............1+0

5. Sedimentation coefficients vs. molecular
weights for perfect spheres and globular
proteins . . . . . . . . . . . . . . . . . . 42

6. Extrapolation of sedimentation coefficients
to zero protein concentrations at pH 3. 45
and pH 3. no 0 C O O O O C O O O O O C C O O O ”6

7. Extrapolation of the diffusion coefficients to
zero protein” concentration at pH 3. 45 and
PH 3. O O O O O O O O O O O O O O O 52

8. Extrapolation of the sedimentation coefficients
to zero protein concentration for the
rabbit muscle aldolase intermediate at
pH 3.35 O O O O C O O O O O O O O O O O O O O 56

9. Extrapolation of the diffusion coefficients to
zero protein concentration for the rabbit
muscle aldolase intermediate at pH 3.35 . . 58

10. Effect of dimer formation on reduced viscosity . 64

11. Extrapolation of the reduced viscosity to zero
protein concentration for native enzyme,
dimers, and monomers at the indicated pH
values00000000000000.000067

12. Sedimentation velocity patterns of different
dimer preparations eXposed to higher tem-
peratures and higher salt concentrations . . 73

xii

LIST OF FIGURES-Continued

Figure
13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

P-150 Bio-Gel filtration of an aldolase sample
atPHBOBSandzsmg/mlooococo...

Effect of B-mercaptoethanol on native enzyme
and reversal of catalytic activity from pH
303d1mers00.000.000.000...

Effect of pH on reversal of pH 3.3 dimers at
various times of incubation at 0°C and 20°C

Effect of dimer concentration and time of
incubation in the dissociation medium:
attainment of conditions for 90%-100%
recovery of activity . . . . . . . . . . .

Recovery as a function of time at two temper-
atures . . . . . . . . . . . . . . . . . .

The reversed of pH 3.3 dimers as a function
of time at two enzyme concentrations . . .

Effect of dimer concentration on half-life of
reversal in the assay at 16°C and pH 7.5 .

Inhibition of dimer reversal under assay con-
ditions by FDP O O O O O O O O O O O O O 0

Test for dissociation of tetramers at pH 5.0
into active dimers . . . . . ,,, . . . . .

This model shows the approach of producing a
dimer at pH 3.4, raiSing the pH to a higher
value, and testing for dimer association or
dimer catalytic activity . . . . . . . . .

Effect of pH on dimer association after a 15
minute incubation at each pH . . . . . . .

The effect of dimer concentration on dimer
association at pH 4.0 after a 15 minute
incubation0000000000000...

Effect of dimer concentration on the reasso-
ciation of dimers into tetramers after a
3.5 hour incubation at pH 5.5 . . . . . . .

Effect of dimer concentration on the reasso-

ciation of dimers into tetramers after a
48 hour incubation at pH 5.5 . . . . . . .

xiii

Page

77

86

89

92

96

99

102

106

110

114

118

122

125

126

LIST OF FIGURES-~Continued
Figures Page

27 and 28. Effect of sucrose concentration on dimer
association at pH 5.5 after various incuba-
tiontimeSoo00.000000000000130-131

29. Effect of higher protein concentration and
higher sucrose concentration (16%) on dimer
8830013151011 at pH 5.5 o o o o o o o o o o 0 131+

30 and 31. Effect of incubation at 0°C and -6°C on
dimer association into tetramers at the
indicated. incubation times 0 o o o o o o o 0139-11‘1’0

32. Effect of a higher temperature (12°C) on dimer
association into tetramers . . . . . . . . . 143

33. Effect of substrate on dimer association into
tetramers after a 4 hour incubation at pH 4
5.0 O O C O O O O O O O O O O O O O O O O O 1 7

34. Effect of post-SDSV incubation at the indicated
temperatures on dimer reactivation . . . . . 154

35. Test for tetramer in dimer peak by "dimer
rerun" and "tetramer rerun" SDSV eXperiments
after the "original" SDSV separation . . . . 157

36. Test for first-order kinetics of pH 5.5 dimer
reversal in the assay (pH 7.5 and 25°C) . . 165

37. Test for second-order kinetics of pH 5.5 dimer
reversal in the assay (pH 7.5 and 25°C) . . 167

38. SDSV analysis of composition in assay at
various times after adding pH 5.5 dimers
to the assay . . . . . . . . . . . . . . . . 170

39. Effect of bovine serum albumin and 0.1M B-
mercaptoethanol on dimer reversal in the
assay after a 30 minute incubation . . . . . 176

40. Effect of dimer concentration on half-lives
of reactivation . . . . . . . . . . . . . . 179

xiv

INTRODUCTION

One of the central problems of biochemistry is to
achieve an understanding of protein biosynthesis. It is
convenient to divide protein biosynthesis into two steps:
(a) the synthesis of the primary sequence and (b) the fold-
ing and association of subunits into the native, polymeric
enzyme. It is presently impossible to study protein con-
formational changes in cell free extracts during protein
biosynthesis; however, in_zit£gireaSSociation of subunits
provides a model system which in many reSpects may simulate
this process.

Previous studies (Deal 23 31.. 1963; Stellwagen 32
21., 1962) on rabbit muscle aldolase had demonstrated rever-
sible dissociation by mild acid. More importantly, the
studies by Deal gt‘gl.. (1963) have indicated an obligate
3.58 intermediate in both the dissociation and reassociation
processes. The physical and chemical data at that time
indicated that rabbit muscle aldolase was composed of three
polypeptide chains. Assuming, when this study was begun,
this model to be correct (it has subsequently been revised
to a four subunit native enzyme, (Penhoet £2 31., 1966))
the 3.58 intermediate was expected to be either: (1) very
unfolded trimers, (2) folded monomers, or (3) a rapid
equilibrium mixture consisting of monomers and dimers.

1

2

The original purpose of this research was to determine
which of these possibilities was correct. The approach was
to physically characterize the intermediate. This 3.58
aldolase intermediate is of Special interest for its signifi-
cance to protein biosynthesis since the intermediate may also
occur in the process of subunit association in living cells.

During this investigation it became evident that none
of the previously mentioned structural models for the inter-
mediate was consistent with the physical data. This necessi-~
tated a consideration of the accuracy of the values of the
molecular weight of the native enzyme and subunits. Using
the values for the native enzyme from the literature and a
subunit molecular weight determined in this research, the
number of subunits was calculated to be mid-way between three
and four subunits. Thus, the subunit molecular weight analy-
sis, together with physical analysis of the 3.58 intermediate
as a dimer, necessitated a consideration of the tetramer model.

The discovery and proof that the 3.58 in aldolase
intermediate was a dimer from this research, together with
the fact that aldolase contained at least 3 catalytic sites
(Horecker 33.21.. 1963: Lai gt,§l.. 1964; Castellino and
Barker, 1966; Ginsburg and Mohler, 1966) led to the question
of whether aldolase dimers could possess catalytic activity
independently of their "partner dimer" in the native tetramer.
Answering this question was the second major aim of this
thesis. The question of whether the "individual subpolymers

of multi-subunit, multi-active site enzymes are active" is a

3
fundamental Question in biochemistry which needs to be
answered. The strength of the dimer-dimer interaction is
most important in this regard; if it is too strong, the con-
ditions required for breaking the interaction would probably
also unfold the resulting dimer molecules. But since aldol-
ase is so easily dissociated, it might be eXpected to have an
active dimer.

The effect of breaking this dimer bond on the other
parts of the molecule, eSpecially the active site, could be
tested by the analysis of possible catalytic activity of the
dimers. For example, if the integrity of the active site had
remained intact after dissociation, as if the dimers were
hard spheres, then the dimers would be catalytically active.
However, any small diSplacement of a side chain resulting
from dissociation could conceivably destroy the catalytic
site,.and the test for activity would be negative.

. But how does one approach this problem? The approach
of this investigator was to try to produce dimers under con-
ditions in which some native enzyme can remain as a tetramer
and can retain its catalytic activity. Obviously, there are
many complicating problems in this kind of approach: (1) The
tetramers may lose activity as the solvent conditions are
altered. (2) The dimers may tetramerize and the activity
assays may be complicated by this activity in the tetramer
form. (3) The dimers may associate into aggregates of higher
order which will further complicate this problem. Since non-

Specific aggregation is usually a major problem in subunit

4

reassociation studies, a thorough study of the reactivation
of the pH 3.35 dimers into tetramers was first conducted.
After this study, conditions were investigated which affected
dimer association and reversal of activity at higher pH
values (pH 5.5 and pH 7.5). Finally. the question of dimer
catalytic activity was examined, and an answer to this ques-

tion was obtained.

LITERATURE REVIEW

I. General Properties 23 Aldolases

Fructose diphOSphate aldolases have been divided into
two classes, I and II; this division is based upon their
reSpective physio-chemical properties, catalytic prOperties,
and their biological origins (for a detailed review, see
Rutter, 1964). What follows is a general description of
class I aldolases and class II aldolases, in that order.

Class I aldolases have been isolated from animals,
plants, protozoans, and green algae. This class of aldolase,
until recently, was thought to be composed of three subunits--
in fact, it was thought to be the only well-documented case
of a three subunit enzyme. But it now appears that the
enzyme has four subunits (Tanford and Kawahara, 1966;

Rajkumar gtflgl.. 1966; Penhoet gt al.. 1966). Three isozymes,
A, B, and C, have been discovered in mammals (Rajkumar, 1966).
The enzymes of this class (I) are inhibited by metals, whereas
class II aldolases are stimulated by monovalent or divalent
metals. Removal of the C-terminal tyrosine from class I
enzymes by carboxypeptidase results in a greatly reduced rate
of fructose diphoSphate (FDP) cleavage but not fructose 1-
phoSphate (F-l-P) cleavage; however, the binding of FDP or

F-1-P is not affected by carboxypeptidase treatment.

6
Dihydroxyacetone phOSphate (DHAP) has been reduced onto the
lysine residue of the enzyme by sodium borohydride, indicat-
ing a Schiff base intermediate in the catalysis (Horecker
§£_§;.. 1963). The pH profiles of FDP cleavage and hydrogen
exchange upon DHAP binding are quite broad for class I
enzymes (Richards and Butter, 1961a).

Class II aldolases have been isolated from bacteria,
blue-green algae, yeast and fungi. Aldolases in this class
have a molecular weight of 70.000 to 80,000 in contrast to
150,000 to 160,000 for the class I enzyme, and probably con-
sists of two identical subunits (Richards and Butter, 1961
1961b). The possible role of lysine in catalysis of class II
aldolases has not been clearly defined, since there is no
inactivation of the enzyme in the presence of FDP with addi-
tion of sodium borohydride (Rutter, 1964). A high sulfhydryl
lability of this enzyme is demonstrated by the requirement of
sulfhydryl reagents in the isolation procedure (Rutter gtwgl..
1966). The pH profiles of the hydrogen exchange with DHAP
and of FDP cleavage are very sharp for class II enzymes in
contrast to class I enzymes (Richards, and Butter, 1961b).

Both class I and class II aldolases have been discovered
in Euglena and Chlamydomonas. In some cases mutants have been
found in which one of the aldolases is absent (Rutter, 1964).
Both aldolases must be controlled by the nuclear DNA because
their formation is not correlated with chloroplast formation
(Rutter, 1967).

Since this research was concerned with rabbit muscle

7
aldolase, a class I enzyme, the rest of the literature review
will discuss the past work on the class I enzymes. Except
for the recent isozyme work and the eXperiment by Tanford

and Kawahara (1966), most of the remaining properties of
aldolase to be discussed are consistent with the previously

accepted three subunit model for rabbit muscle aldolase.
II. Physical Properties 2£_Rabbit Muscle Aldolase
A. Native Enzyme

The physical properties of native rabbit muscle
aldolase have been measured in several laboratories. Gralen
(1939) demonstrated that rabbit muscle myogen A had a sedi-
mentation coefficient of 8.08. Taylor and Lowry (1956) first
measured the weight-average molecular weight of the native
enzyme using the Svedberg equation (see Methods). They
obtained a molecular weight of 149,000 using a sedimentation
coefficient of Sgo,w = 7.358 and a diffusion coefficient of
D20,w = 4.63 x 10‘7 cmZ/sec. However, adiabatic cooling of
the rotor was not taken into consideration; with this correc-
tion, the sedimentation coefficient is considerably increased
(Kawahara and Tanford, 1966). Higher values for the sedimen-
tation coefficient of the native enzyme have been obtained by
Stellwagen and Schachman (1962) and by Deal 22 2;. (1963) who

0

found values of 320 w = 7.98 and 7.88, reSpectively. Tanford
O

and Kawahara (1966) estimated a value of s30 w = 8.08, using
0
a sedimentation coefficient at a single concentration

together with the concentration dependence calculated from

8
the regression formula of Stellwagen and Schachman (1962).
Using an average value of 7.98 for the sedimentation coeffi-
cient and the above diffusion coefficient, a value of M(S/D)
a 160,000 was obtained. Unfortunately, the value of the dif-
fusion coefficient has not been confirmed.

Stellwagen and Schachman (1962) measured the molecular
weight of the native enzyme by the sedimentation equilibrium
technique at 5 mg/ml and found an apparent weight-average
molecular weight of 142,000. Using the short column sedimen-
tation equilibrium technique, Lewis and Haas (1963) obtained
a weight-average molecular weight at three concentrations of
140,000, in good agreement with the previous value. Finally.
Kawahara and Tanford (1966), using the high Speed equili-
brium technique of thantis (1964), found an apparent weight-
average molecular weight of 158,000 at 2 mg/ml.

The intrinsic viscosity of native aldolase has been
determined by two laboratories (Stellwagen and Schachman,
1962; Hass, 1964) to be 4.0 cc/g. The intrinsic viscosity
of aldolase indicates that it has a very compact structure,
and thus is a fairly globular protein, not an elongated rod
or random coil. Using the formula of Scheraga and Manderkern
(1953) relating sedimentation coefficient, intrinsic viscos-
ity, and molecular weight, a molecular weight of 161,000 was
obtained for the native enzyme by Kawahara and Tanford (1966).
The a parameter used in the above formula had a value of
2.12 x 106: this value is commonly used for ellipsoidal

molecules with moderate axial ratios and for Spherical

9

molecules. The sedimentation coefficient used was the same

as that used in the previous M(S/D) calculation, i.e. 7.98.
Values of 0.742 cc/g (Taylor and Lowry, 1956) or

0.745 cc/g (Hass, 1964), both at 20°C, for the partial Speci-

fic volume have been reported for native aldolase.
B. Subunits

The enzyme has been dissociated by urea (Stellwagen
and Schachman, 1962), by guanidine HCl (Schachman and
Edelstein, 1966a,b), by sodium dodecyl sulfate (Schachman,
1960), by acid (Deal 33 51., 1963; Stellwagen and Schachman,
'1962; Westhead and Boyer, 1963), and by alkaline pH (Sine
and Hass, 1966) into subunits approximately one-third the
molecular weight of the native enzyme. Although the orig-
inal subunit molecular weight values of Schachman and
Edelstein were too low for the three subunit model, they
corrected these values upward where they did agree with the
three subunit model using the assumption that there was pre-
ferential interaction of water with protein. This correction
for the 3. 5, and 7 M guanidine HCl solutions changed the
reSpective apparent molecular weights from 46,600, 43,000.
and 38,300 to 50,000, 49,500, and 49,900.

Hade and Tanford (1967) have disputed their correction
and provided evidence that, instead of preferential of water
with protein, actually there was preferential interaction of
guanidine with protein. Thus, in contrast to the previous

correction where the molecular weight was increased, Hade and

10
Tanford's correction would decrease the molecular weight.
Thus the molecular weight of 46,600-38,300 should be cor-
rected downward to about 44,000-36,000.

Kawahara and Tanford (1966) also measured the molecu-
lar weight of the subunits in guanidine HCl, but included ‘
B-mercaptoethanol as well. By the sedimentation equilibrium
technique they obtained an average molecular weight for two
experiments of 41,000 using a partial Specific volume of
925 = 0.747 cc/g. Since this experiment was done at essen-
tially zero concentration, extrapolation to zero concentra-
tion was unnecessary.

Hass and Lewis (1963) reported the rapid formation of
six subunits when the native enzyme was made alkaline to
pH 12.5. However, they later retracted this with their con-
clusion that peptide bonds were being hydrolyzed at this pH
(Hass, 1967). This later report also contained evidence for
four subunits in aldolase. The results of Kawahara and
Tanford (1966) and Hass (1967) are in agreement with those
results reported here (see Chapter I-A in Results) which
were completed in early 1965.

The value reported for the intrinsic viscosity of the
subunits formed in acid is about 24.5 cc/g (Stellwagen and
Schachman, 1962) which is in agreement with the results to
be presented here (see Chapter II-B in Results). Kawahara
and Tanford (1966) found an intrinsic viscosity of 35.3 cc/g
for subunits produced in 6M guanidine HCl with 0.1M

B-mercaptoethanol. This suggests that the acid dissociated

11
subunits were not completely unfolded, whereas the enzyme
dissociated in guanidine were probably random coils. Tanford
and Kawahara also reported an extrapolated sedimentation
coefficient value for the subunits in this medium: Sgo.w =
1.88.

They used their viscosity and sedimentation data to
give an additional independent calculation of the molecular
weight. Again, to use the formula of Scheraga and Manderkern
(1953) to obtain the molecular weight, a 8 value characteris—
tic of that for random coils must be used. A plot of the
intrinsic viscosity vs. the molecular weight for various
enzymes dissociated into random coils with guanidine HCl and
B-mercaptoethanol has Shown that the intrinsic viscosity (n)
varies linearly with the molecular weight (M) according to
the formula: (n) = M0’68 (Tanford, 1966). The 8 value for
random coils used for these molecular weight determinations
was 2.5 x 106. Using this B value the previously mentioned
values for intrinsic viscosity and sedimentation coefficients,
and a partial Specific volume of 0.747 cc/g at 25°C, Kawahara
and Tanford (1966) calculated a molecular weight for aldolase
subunits to be 42,000.

III. chemical and Catalytic Prgierties 2_f_
Rabbit Muscle Aldolase

Aldolase is a relatively unique enzyme because one of
its substrates can be reduced stereoSpecifically onto its

active site by sodium borohydride (Grazi g§_gl.. 1962). This

12
method has been used to quantitatively estimate the number of
binding sites per mole of native enzyme. Using this technique,
the number of estimated binding Sites has risen from one to
three (Horecker §t_gl.. 1963; Lai gtflal.. 1964; Ginsburg and
Mehler, 1966) during the last several years. Investigations
using other techniques led to values from one to three bind-
ing Sites: (1) Westhead gtflgl. (1963) found one binding site
using ultracentrifugal or equilibrium dialysis methods; (2)
Castellino and Barker (1966) found three binding sites for

“C when the diphosphate was

D-arabinitol-i,5-diphOSphate-1-1
equilibrated with enzyme and run through a Bio-Gel P-6 column
or analyzed by the partition-cell ultracentrifugation tech-
nique: and (3) Ginsburg (1966) found three binding sites when
a mixture of radioactively labeled DHAP and native aldolase
was rapidly percolated through a G-50 (coarse) Sephadex
column.

Winstead 22.31. (1963) found a dissociation constant
for the FDP-aldolase complex of kd = 4 x 10'°M. An associa-
tion constant for DHAP of ka = 1.6 x 10'3M was also reported;
thus FDP is bound much more strongly than DHAP. The Km value
of 1.0-1.5 x 10'5 (Richards and Butter, 1961) is somewhat
higher than the dissociation constant for FDP; this is
SXpected since the Km also includes the catalytic rate con-
stant. Carboxypeptidase treatment of aldolase lowers the Km’
but does not lower the kd for DHAP or FDP (Dreschlser 23.51..
1959). These data taken together suggest that the catalytic

rate constant is decreased by the loss of the c-terminal

13
tyrosine; this data provides a mechanism whereby carboxypep-
tidase treatmentcfi'aldolase decreases the Vmax to 7% of that

of the native enzyme (Dreschlser gt_§;.. 1959).
A. C-Termina;.Analeis

Dreschlser g§_gl, (1959) reported that 93% of the
activity of native aldolase was lost upon treatment with
carboxypeptidase. When the digestion was conducted in H2018
and the liberated C-terminal amino acids were isolated and
characterized (Kowalsky and Boyer, 1960), approximately three
moles of C-terminal tyrosine were found per 149,000 g of pro-
tein. Tyrosine was measured by the Folin-Ciocalteu method or
the Goodwin and Morton U-V absorption method. Winstead and
Wold (1964) also calculated three C-terminal tyrosines from
experiments using carboxypeptidase A and B. In addition, by
the hydrazinolysis method they found 2.47 and 2.68 moles of
tyrosine per 149,000 g of enzyme.

Carboxypeptidase treatment does not alter the Spectro-
photometrically demonstrated DHAP-enzyme or FDP-enzyme com-
plex; this fact is in agreement with the observation that
C-terminal tyroSine is not involved in the binding, but is

necessary for the catalytic process of FDP cleavage.
B. NQTerminal.Analysis

Udenfriend and Velick (1951) first reported 1.9 and

2.3 moles of proline per 149,000 g of aldolase using 1131-

p-iodophenylsulfonyl chloride. Hass (1964) reported 3.98

14
moles of proline per 142,000 g of aldolase by the Edman pro-
cedure.

Winstead and Wold, in their attempt to explain the
discrepancy in the literature for the number of subunits (3
or 6 subunits) of aldolase,hypothesized that the N-terminal
amino acid might have been "masked" by other groups such as
an acetyl group. This would then eXplain why only 3 or 4
moles of proline (instead of 6) were detected per mole of
enzyme. Since at least one other glycolytic enzyme, enolase,
was known to possess an N-acetylated amino-terminal amino
acid (Winstead and Wold, 1964a), they decided to investigate
this possibility in aldolase. In their attempt, they found
no evidence for an N-acetylated amino-terminal amino acid
upon treatment with hog kidney acylase; they therefore con-
cluded that the previous data which indicated that aldolase

possessed three or four N-terminal prolines was correct.

C. Sulfhydryl Content

The number of sulfhydryl groups in aldolase has been
reported to be between 28 (Swenson and Boyer, 1957; Westhead
22.51.. 1963) and 29 (Benesch.gtmgl.. 1955), depending on
the method used. Swenson and Boyer used a Spectrophotometric
assay of sulfhydryl groups with p-mercuribenzoate, whereas
Benesch‘gt‘gl. (1955) measured -SH groups by amperometric Ag
titration. Approximately 10 -SH groups can be titrated with-
out loss of catalytic activity (Swenson and Boyer, 1957). It

appears then that at least 10 —SH groups are on or near the

15
surface of the enzyme. If 8M urea is added to unfold the
enzyme, more and more -SH groups are eXposed until finally.
all 28 -SH groups are available for titration. Stellwagen
and Schachman (1962) found 27.1 -SH groups in BM urea; this

value agrees with the previous two results.
IV. Isozymes g£.Aldolase

The isozyme studies yield information which is very
pertinent to subunit structure; in fact, the results from
the isozyme work strongly support a model in which rabbit
muscle aldolase contains four subunits, and it is hard to
reconcile this work with a three subunit model.

Penhoet gt'al. (1966) have reported three different
aldolases in mammalian tissues, each with its own distinc-
tive catalytic, electrophoretic, chromatographic, and
immunochemical properties. Acid dissociation and reassocia-
tion of any two of the parental aldolases yield five membered
sets by electrophoresis, with the two parental type aldolases
enclosing this set. Any purified hybrid can also reproduce
the five membered set upon reversible dissociation. In addi-
tion to producing hydrids artifically. in|zitrg, hybrids are
also found in tissues containing more than one parental type
aldolase. For example, adult brain and testes contain all
the A-C hybrids; on the other hand, the adult liver and kid-
ney contain all the A-B hybrids.

The sequential appearance of isozymes of aldolase dur-

ing embryogenesis will now be considered. In the early stages,

16
isozyme A is the only form synthesized. Then as organogene-
sis takes place, synthesis of isozyme C results in the forma-
tion of A-C hybrids, for example, when differentiation of the
ectoderm into brain tissue begins. CorreSpondingly, the syn-
thesis of isozyme B results in the formation of A-B hybrids
when differentiation of the endoderm into kidney and liver
occurs (Rutter gtflg;.. 1967).

The isozymes have also been characterized by the
ratios of their catalytic activity using fructose diphOSphate
(FDP) and fructose-i-phosphate (F-i-P) reSpectively, as sub-
strates. After treatment with carboxypeptidase, residual
activities and FDP/F-i-P ratios are almost equivalent. This
suggests that the function of the C-terminal tyrosine in the
catalytic process is different for each of the different
isozymes. However they do seem to have certain similarities
with respect to their catalytic Sites; all three isozymes are
reduced and inactivated by NaBHu in the presence of FDP, indi-
cating a common Schiff base intermediate.

The production of five isozyme hybrids from the paren-
tal types suggests that aldolase is composed of four subunits
(Rutter gt_§;.. 1967). In addition to this evidence, Butter
and coworkers have found that when they dissociated radio-
active aldolase A (isolated from mice grown on 3H leucine)
together with non-radioactive aldolase C (isolated from rab-
bits), reassociated again, and separated the resultant 5
hybrids by electrophoresis, they discovered the following

percentages of Specific radioactivities (0% representing the

17
C4 tetramer): 0%, 25%, 50%, 75% and 100%. These results,
of course, support, if not prove, the four subunit model for

aldolase.

METHODS

I. Preparation and Assay 2; Rabbit Muscle Aldolase

Aldolase was prepared by the method of Taylor $3.21.
(1948) as modified by Kowalsky and Boyer (1960). After at
least three recrystallizations the crystalline suSpension
was stored in 50% saturated ammonium sulfate at 4°C. Enzyme
concentrations were determined at 280 mu, using the extinc-
tion coefficient, Eg’éé = 0.91 (Baranowski and Niederland,
1949). (All preparations were homogeneous in the analytical
ultracentrifuge and those that were analyzed on polyacryl-
amide gel electrophoresis gave a single band.

.Aldolase was assayed by the procedure of Richards and
Butter (1961) as represented in the following coupled reac-

tion sequence:

 

 

 

 

DPNH DPN+
DHAP“:El 3* " > a-glycerol
(( P: a-glycerol phOSphate
'5' phOSphate
g dehydrogenase
Aldolase I
FDP 2"
p
U)
o
a
(D
‘3
bG-B-PJ g

The reaction was followed, Spectrophotometrically, by the

decrease in DPNH absorbance at 340 mu. The concentrations

18

19

of the various components in the 0.4 ml assay were the follow-
ing: 0.012M FDP (Sigma), 0.003M DPNH (P and L Laboratories),
0.02 mg/ml triose-p-isomerase-a-glycerol phoSphate dehydrogen-
ase mixture (Sigma), and 0.2M tris HCl (Sigma), pH 7.5. The
FDP used for most of the work was 75% pure, and where it was
thought that the impurity might interfere with the results,
such as in the inhibition studies, FDP (Sigma) of 98% purity
was used. (All aldolase preparations had a Specific activity

of 13.2-13.5 “moles FDP cleaved/min/mg of enzyme at 25°C and

PH 7050
II. Ultracentrifugal Analysis

All eXperiments were performed in a Spinco Model E
ultracentrifuge equipped with a phase plate as a schlieren
diaphragm. Photographic plates were measured with a Bausch

and Lomb microcomparator.
A. Sedimentation Velocity and Diffusion Experiments

Sedimentation velocity eXperimentS were run either
with 30 mm or 12 mm Single sector cells, at 50,740 rpm in an
An E rotor for the former cells and at 59,780 rpm in an An D
rotor for the latter cells. Sample volumes for the 30 mm
cells and 12 mm cells were 1.25 ml and 0.55 ml, respectively.
The longer columns had the advantage of greater sensitivity
and could be used for sample concentrations as low as 0.3
mg/ml. Sedimentation coefficients were calculated from the

following equation:

ss__1__1n_2___
(t-to)w2 r (to)

where rP is the peak maximum in the refractive index gradi-
ent curve of picture taken at time to (first picture to be
measured) or at time t (later pictures), w is the rotor
velocity in radians/sec and s is the sedimentation coeffi-
cient in Svedberg units (8).

The diffusion coefficient experiments were determined
with a double sector synthetic boundary cell (12 mm) at
4,059 rpm using an An D rotor. A solvent volume of 0.41 ml
was used in one sector, and a sample solution volume of 0.14

ml was used in the other sector. The diffusion coefficient

was calculated from the following equation (Schachman, 1957):

D s In

4560

where m is the slope of a line from a graph which plots
AreaZ/Height2 vs. time (minutes). The diffusion coefficients
were calculated manually using the above graph.

Sedimentation and diffusion coefficients were corrected
to 20°C and water (Schachman, 1957). Densities of the solvent
were measured at the same temperature as the centrifuge
studies with the aid of a hydrometer. Since sedimentation and
diffusion coefficient studies were usually performed in paral-
lel, the same temperature was used in both series of experi-
ments. The precise temperature was obtained from the RTIC
meter on the ultracentrifuge.

To calculate molecular weights from sedimentation

coefficients and diffusion coefficients, the M(S/D) formula

21
of Svedberg (1940) is used:

M=.§_§2__
D (1-39)

where R is the gas constant, 8.31 x 107 ergS/mole/degree, T
is the absolute temperature, 3 is the partial Specific volume,

and p is the density.

B. Sedimentation Equilibrium and Archibald
Molecular Weight Analysis

Standard 12 mm double sector cells were used for the
sedimentation equilibrium eXperiments, whereas both 12 mm and
30 mm double sector cells (the latter for greater sensitiv-
ity) were used in the Archibald molecular weight analysis.

A volume of 0.06 ml of protein sample was used in one sector,
while a solvent volume of 0.02 ml more than the combined
height of sample volume and flurocarbon oil (PC-43) was used
in the other sector. The flurocarbon oil (0.05 ml. ordinar-
ily) was used to produce column height separation when more
than one cell was run at a time.

For the subunit analysis, short column sedimentation
equilibrium techniques (Van Holde and Baldwin, 1958) were
used along with a three cell arrangement to shorten the exper-
imentation time. The time to reach equilibrium in the subunit
analysis was usually 36 hours. The weight-average molecular

weight was calculated from the following equation:

Mzapp RT cb - 0m 2

“ 2 2 2
1 - o I-
: w (1 vp) C rh rm 3

22
but 1 1 (Cm + Cb)

MzaPP M2 2

 

To obtain true weight-average molecular weight Mw, the quan-
tity, 1/Mzapp (the "apparent" signifies that the molecular
weight was measured at a finite concentration), must be
extrapolated to zero protein concentration (eXpressed as
(Cm + Cb)/2), and the reciprocal be taken. In the above
equations, M2 is the true molecular weight, B1 is the first
virial coefficient, rm and rb are the distances from the
center of rotation to the meniscus and the bottom of the
cell, reSpectively, and Cm and Cb are the concentrations at
the meniscus and the bottom of the cell, reSpectively. C°,
the original protein concentration, was determined from
area measurements of synthetic boundary eXperiments, Similar
to those described for the diffusion coefficient experiments.
The Z-average molecular weights, M2, were also calcu-
lated and these "apparent" values were obtained from the fol-
lowing equation:
_1_ (22) - 3.69.) i. am Mo) w... (Cb - cm

rb dr dr RT
b m

 

1‘m

where (dC/dr)b and (dC/dr)m are the concentration increments

at the bottom and the meniscus of the cell, reSpectively.

The apparent Mz values were also extrapolated to zero protein

concentration to obtain the true Z-average molecular weight.
.Although it usually takes at least 24 hours to reach

equilibrium (and longer when using guanidine HCl solvents) in

the sedimentation equilibrium technique, the Archibald tech-

23
nique can be used to calculate molecular weight very early in
the run (within 2 hours). Since the meniscus or the bottom
of the cell does not allow any solute pass through these
points, conditions for equilibrium are satisfied at all times
at these two points. The same equation, which applies to the
rest of the points of the cell at equilibrium, then applies
at all times for the meniscus and bottom of the cell: the
calculation of the molecular weight for the meniscus of the
cell is given below:

M RT (dc/drim

(1-Vp) wz rm Cm

These experimental parameters have been previously defined.
If the subscript (m) is changed to the subscript (b), this
equation could then be used to calculate the molecular
weight from data obtained from the bottom of the cell.

(Archibald and sedimentation equilibrium calculations
and the plotting of the data were carried out by a Control
Data Corporation 3600 computer with a fully tested program
which included a statistical analysis.

III. Viscosity

Viscosity was measured by means of a Cannon-Ubbelohde
capillary dilution viscometer. The temperature was maintained
at 7.85°C and regulated to within t 0.00200. The densities
of the solvents were determined with a hydrometer. The den-
sities of the protein solutions were calculated from the par-

tial Specific volume, protein concentrations, and solvent

24
densities as outlined by Schachman (1957). All protein and
solvent solutions were filtered through polyvinyl chloride
Metricel Type VM-6 Millipore filters (pore size of 0.45 0)
under air pressure.
The equation used to calculate the Specific viscosity
nsp was the following:

TI =—2-E—-1
SP pct:

where p' and p are the densities (cc/g) of the protein solu-
tion and the solvent solution, reSpectively, and t' and t are
the flow times (seconds) for the protein solution and solvent
solution, reSpectively. The reduced viscosity and the intrin-
sic viscosity are defined in the text.

.A portion of the crystalline aldolase suSpension was
centrifuged and the pellet was dissolved in 0.001M EDTA, 0.05M
NaCl, and 0.01M citrate buffer, pH 5.45. The stock solution
was then dialyzed against this buffer for 48 hours at 4°C.
Solvents containing 0.001M EDTA, 0.05M NaCl, and 0.5M citrate
buffer at the appropriate pH were used to produce the desired
Species: native, intermediate, or subunits. To prepare the
final solution for viscosity measurement, 4.8 ml of enzyme
stock solution were diluted with 0.6 ml of the appropriate
solvent. To make the final dilution solvent, 4.8 ml of the
dialysate were diluted with 0.6 ml of the same 0.5M citrate
(Na) buffer. This dilution solvent was then used to dilute
the enzyme solution in the viscometer. Three concentrations

of native enzyme and two concentrations of acid dissociated

25
subunits were examined to check the SXperimental technique
and verify the published values. The viscosity of the
intermediate was determined as a function of time after acid

addition at two concentrations at pH 3.40.

IV. Bio-Gel.P-150 Column Preparation

Bio-Gel P-150 (50-150) was allowed to swell in 0.001M
EDTA, 0.05M NaCl, and 0.1M citrate (Na), pH 3.35. and poured
into a column (1.5 cm x 75 cm) equipped with a large funnel
to facilitate a single addition of gel. After about 10 cm
of gel had settled in the column, the elution buffer was
allowed to percolate through the system.

To determine the void volume, 0.1% dextran (Sephadex)
was layered on the column and fractions were taken until the
dextran peak appeared. The volume obtained in this interval

was taken as the void volume.
V. Sucrose Density Gradient Centrifggation

The procedure used for the sucrose density gradient
sedimentation velocity (SDSV) experiments was that of Martin
and Amos (1961). The SXperiments were performed with the aid
of a Beckman Model L Analytical Ultracentrifuge at a Speed of
40,000 rpm (except where noted) and -6°C. The linear gradi-
ent of 4.6_ml was 5% to 20% sucrose. This, together with an
applied sample volume of 0.1 ml, made a total volume of 4.7
ml. Ten drop fractions were collected per tube and the total
number of tubes were noted to the first decimal point. The

26

following procedure was used to convert fraction number (tube
number) to the distance traveled from the meniscus (cm):
(1) cm/tube = 3.62 cm/total number of tubes, (2) "corrected
number of tubes" = total number of tubes - 001 totalmnumber
of tubes) (3) (corrected tube number from meniscus) =
"corrected number of tubes" - (fraction-tube number of each
tube collected), (4) (corrected tube number from meniscus) x
(cm/tube) = (distance traveled from meniscus in cm).

Activities (units/ml) were then plotted against the
(distance traveled from meniscus in cm). Peak positions (in
cm traveled from meniscus) were then used to calculate the
sedimentation coefficient of the dimer Species using the fol-
lowing relationship:

Sdimer Distance traveled

stetramer Distance traveledtetramer

where Stetramer

dimer

 

 

was taken as 7.98, which is an average sedi-
mentation coefficient (830 W) determined by the conventional
0

moving boundary technique using the Model E ultracentrifuge.

RESULTS

I. The Approach to the Problem and Initial Studies

AS outlined in the introduction, the primary objec-
tive of this research was to achieve an understanding of
the mechanism of subunit association of rabbit muscle
aldolase: of Special interest was the role played by the
3.58 obligate intermediate in both the dissociation and
association processes of this enzyme. But before a mean-
ingful study on the mechanism could be undertaken, the
subunit structure of the native enzyme had to be defined
as well as possible. Although essentially all the available
evidence at that time was consistent with the three subunit
models for native aldolase, a report by Hass and Lewis (1963)
had provided evidence that aldolase dissociated rapidly into
6 subunits when a solution of native enzyme was made alkaline
to pH 12.5.

The first objective then of this investigator was to
repeat the subunit molecular weight analysis to solve this
problem. To effect complete dissociation, the best disso-
ciation solvent system known, guanidine HCl and B-mercapto—
ethanol, was used. 1

The second objective was to characterize the 3.58

intermediate. However, before the physical characterization

27

28
of? the intermediate could be conducted, the effect of pH had
tc> be investigated to find the optimum pH for 3.58 interme-
diate formation. Such a study constitutes the second major
section of this introductory chapter (I) and involves an
analysis of the sedimentation velocity patterns for native,
intermediate, and subunits in the narrow pH range of 3.0 to

“.0.

A. Determination of Rabbit Muscle Aldolase Subunit
‘ Molecular Weight Ag Guanidine HCl and §¥Mercaptoethanol

Preparation of the aldolase subunits for molecular
weight analysis by the sedimentation equilibrium technique
was as follows: a portion of the crystalline aldolase was
centrifuged, and the resulting pellet was dissolved in a
solution containing 5.6M guanidine HCl, 0.15M NaCl, 0.001M
EDTA, and 0.04M tris buffer at pH 7.5. This enzyme solution
(approximately 16 mg/ml) was then dialyzed for 48 hours at
4°C against the above solution. Following dialysis the
enzyme solution was diluted with dialysate to yield several
protein concentrations. Synthetic boundary and Short column
sedimentation equilibrium eXperiments (Van Holde and Baldwin,
1958) were performed with the use of the Spinco Model E
analytical ultracentrifuge.

Figure 1 shows the weight-average molecular weight of
aldolase subunits as a function of protein concentration.
There is a strong concentration dependence and the data is

somewhat scattered. Using the extrapolated value of 42,000

29

.meosaoa one use» com .maaspos pom .xmmma .sHseHsm
use oeaom sm>v mama sou A90 + avg use am”: now N\Ano + aov ms commoaawo who:
msoapenpSoosOo one .SoflpsapSSoSoo ohms op mpswaes amazooaoa Aaamzv owsaoble

passages ase Aaamzv ommaobdipsmaos pacemaam esp mo Soapmaoamapwm .H ostHm

30

:2\o§ 2058:28on
m: 2 m. o o m
_ q

 

 

 

— — — d

m 088..

mu:

E

w 83?.

R

m

U s

m 80 mm.

o

M s

m 088..

E

M

DI s

M CON 0— LOZ<IHmOhm<Umw<< $3.0
JUI mZ_D_Z<Dmv 20.0

 

 

 

thDmDm $430044.

31
(Figure 1) and assuming a native enzyme molecular weight
between 142,000 (Schachman and Stellwagen, 1962) and 150,000
(Taylor and Lowry, 1956), the number of subunits was calcu-
lated to be between 3.4 and 3.6. Thus these results were
not consistent with the six subunit model of aldolase, nor
were they consistent with the trimer model. Therefore, in
addition to the trimer model, the possibility of a tetramer
model had to be considered. It should be mentioned, however,
that the tetramer was not consistent with any of the exten-
sive chemical or physical data on aldolase in the literature
and was not seriously considered by this investigator until
the analysis of the 3.58 intermediate Showed an inconsistency
with the trimer model.

.After this work was complete, two other laboratories
(Schachman and Edelstein, 1966; Kawahara and Tanford, 1966)
published results of the molecular weight of aldolase sub-
units in guanidine HCl. Although their SXperimentS were of
a limited nature (3 molecular weight determinations by the
former investigators and 2 determinations by the latter),
both of their results were in good agreement with ours.

(See Literature Review for more information on their work.)

B. Determination p£_the pg Range for Formation p:

the 3.58 Aldolase Intermediate

For reasons given in the introductory paragraphs of
this chapter, sedimentation velocity SXperiments were per-

formed to determine the most favorable conditions of formation

32
and stability for the 3.58 intermediate.

Approximately 90 mg of enzyme crystals were centri-
fuged, and the resulting pellet was dissolved in 18 mls of a
solution containing 0.001M EDTA, 0.05M NaCl, and 0.01M
citrate (Na) buffer at pH 5.45. This sample was then dia-
lyzed for 36 hours. Nine parts of this dialyzed solution
were then acidified with one part of solutions at the appro-
priate pH containing 0.001M EDTA, 0.05M NaCl, and 0.5M
citrate (Na) buffer to yield solutions at the indicated pH
values (between pH 3.0 and pH 4.0). After a one hour incu-
bation at 4°C and 4 mg/ml, sedimentation velocity eXperi-
ments were performed for the different samples at 59,780 rpm,
using the Spinco Model E analytical ultracentrifuge.

The results in Figure 2 Show the effect of pH on the
dissociation of aldolase. The top line represents unfolded
native enzyme with an apparent sedimentation coefficient of
6.58. If the pH is lowered to pH 3.6, a transition from
unfolded native enzyme to the intermediate occurs, as Shown
by the upper dashed line. .At lower pH values, below pH 3.2,
a second transition from the intermediate to monomers occurs,
as shown by the second dashed line. Since these eXperiments
Show that the intermediate is formed between pH 3.1 and pH
3.6, let us examinethe sedimentation velocity patterns in
this pH range.

Sedimentation velocity profiles are depicted in Figure
3 at the indicated pH values. In Figure 3A the upper peak,

at pH 3.1, iS skewed towards monomers with some intermediate

33

.mHaspoe Swansea

you vamp com .moaooam peep go Rooa msHpSommaaoa modoaao eoHHam .wSSUSSm

mo moawoe on» on Hmsoapaoaoaa ma ma some as peemoaa moaooam some mo pSSoam
one .Homafip Amzv opmapao ammo.o use .Hosz zmo.c_.danm Saoo.o ampsosoaaoo
weasoaaom esp decampsoo encapsaom Hanan on» mo Seam .Ha\wa 0.: we: Soapsap

leoosoo adepoaa one .omsaooas mo chaposapm on» so ma no poomuo 028 .N themes

34

0d

IQ

 

mm Om Sm NM Om
_ . _ _ . _ _
$2020: I»
.. OIOIIO\O\O I m
\\\
1 0/ It 0\0 .. m
525128582. 13.0 o
1 . 1 s
.I \ II n
525:8 .
In \JOIIIO l 0
o

 

 

 

35

Figure 3. Sedimentation velocity patterns of
aldolase for the pH range between pH 3.1 and pH 3.5.
Sedimentation is from left to right. The times at
which pictures were taken after reaching Speed
(59,780 rpm) were 140 minutes for photograph 3A and
80 minutes for photograph 3B.

A. Upper curve: 4 mg/ml at pH 3.1

Lower curve: 4 mg/ml at pH 3.3

B. Upper curve: 9 mg/ml at pH 3.5
Lower curve: 6 mg/ml at pH 3.4

 

2.13 3.131

 

36

 

37
present on the leading edge. At pH 3.3, in the lower pat-
tern, the peak with a sedimentation rate of 3.18 is skewed
towards intermediate, with some monomers present on the
trailing edge. Thus the intermediate and subunit peaks are
not resolved completely; whether this result can be attrib-
uted to a small difference between sedimentation coefficients,
or to a rapid equilibrium between the intermediate Species
and subunits will be discussed later.

To obtain a working upper limit for protein concentra-
tion and pH for the production of the intermediate, eXperi-
ments of the kind described above were performed at higher
concentrations (9 mg/ml and 6 mg/ml). At pH 3.5 and 9 gm/ml,
two peaks were evident in the upper profile of Figure 3B;
the small leading native peak indicates about 90% dissocia-
tion of native enzyme into the intermediate. The lower pro-
file, at pH 3.4, and 6 mg/ml, Shows complete dissociation of
native enzyme into the intermediate. Thus complete resolu-
tion of the native and intermediate peaks is clearly visible.
This complete resolution ruled out a rapid equilibrium
between native enzyme and the intermediate Species or native
enzyme and monomers.

Thus pH 3.5 is too high at 9 mg/ml to give complete
dissociation into the intermediate species, whereas pH 3.3
at 4 mg/ml is too low and causes further dissociation into
monomers. Since the intermediate is essentially the only
Species present between pH 3.3 and pH 3.45, further physical
characterization of this intermediate was examined in this

pH range.

38
II. The Nature _o_f_ the 3.53 Aldolase Intermediate

Before the quantitative results on the intermediate
analysis are presented, it is desirable to consider the dif-
ferent dissociation mechanisms by which a tetramer or a
trimer molecule could give rise to a 3.58 intermediate
(Figure 4). A.measure of the asymmetry of each of the dif-
ferent intermediates considered was determined by calculat-
ing f/fo, the frictional ratio. This ratio can be calculated

from the following relationship:
o

11.32235.
0
f0 SeXp

where Sgax is the theoretically determined sedimentation

coefficient for perfect Spheres and 8° iS the SXperimen-

exp
tally determined sedimentation coefficient for the aldolase
intermediate, i.e. 3.58. The only factor which changes in
the above relationship is Sgax' and this changes according
to the molecular weight assumed for each of the different _
intermediates considered. The reader is referred to Figure
5 from which the theoretically determined 823x values were
obtained, since it shows the dependence of 823x on changes
in molecular weight. Thus, for example, to accomodate a
trimer or a tetramer intermediate, the intermediate would
have to be extremely unfolded (f/fo = 3.0); on the other
hand, in order for the intermediate to be a monomer, it

'would have to be very tightly folded (f/fo = 1.2). Simi-

larly, a dimer intermediate would have a conformation with

39

.maampoe nonpo Mom

one modpma owes» Samppo on some sasaaow on» you pxop mom .AN.H n om\mv mHoSo
usos cocaou aaasmap Ana use .Am.a u oa\av unease seasons: aacasaooos Ame .Ao.m n
om\mv maossapop Ho macadap eoeaouss haoaoapwo Adv on on empSHSoHSo one: mopdadoa
lumped peoaemmae map mo some you .hapSSShmm mo oneness s .th\av modems Hanoap
Icahn one .opmdeoaaopsH mm.m m on mean ebam canoe oafiocaoa Heads» 6 no Hosmhpop

8 nodes an amazesoos haoxaa pmoa on» was one: oepodaoe escapades one .: chewam

TETRAMER MODEL

TRIMER MODEL

INTERMEDIATE-3.55

INTERMEDIATE -3.5$

 

 

 

 

 

41

.AmHmQSp

.Q.nm n.2daoaaom .m.3 Soap Swampy manpsaopaa can ad eonmdansa dude Hangoaaaoaxm
Scam oodeSoHoo madness HmoaaaasS can mucomoaaoa mafia deemed on» one .mesadb
eopmHSono hHHsoHpoaooSp esp manomoaaoa mafia eaaom use .mSaopona asaenoaw use

monogam poomhea how magmas: amazooaos .mb menoaoamwooo soapmpSesHoom .m shaman

42

nIOH H amuamz médbumqom

 

 

          
     

can com com com ooa as on on ow 0H
cassava—09am
N
cannon

Sowosdusuoabfi

58883344 0

     

 

canoamosom m
unseemoaeagoe opssamosaouoohawuo
omuaosu w
omaahfiuuo O
omgomoaehnoe 03on doorman—Hm a lily
Shadow: Hecauouoofi. all; m
owes“: owmfifimmm 0 0H
unosamosm 0
ma
t enema:
om

“152885 3830 mos 82
71988 mos 2.280538 .85

OTHER 0:) NOIIVINEN 1033

7A

43
a f/fo between the two extremes (f/fo = 1.9). The last pos-
sibility, a rapid equilibrium, would be too complex to ana-
lyze in f/fo terms.

As mentioned previously, a rapid equilibrium between
native enzyme and dimers or native enzyme and monomers has
been already excluded (see Section B of Chapter I); however,
a rapid equilibrium between dimers and monomers must still
be considered as one of the possibilities for the 3.53
intermediate.

The physical characterization of the 3.58 intermedi-
ate was approached as follows: First, an analysis for the
possible existence of a rapid equilibrium was conducted
(Section A). Next, sedimentation and diffusion coefficients
were measured and molecular weight values calculated by the
M(S/D) technique, under different conditions of dimer prepa-
ration, pH, and temperature (Sections B and C). Then.
viscosity analysis of the 3.58 intermediate was conducted
(Section D); the results will be shown to support the
molecular weight analysis in the previous two sections and
exclude other possibilities which were described in the

introductory paragraphs (see also Figure u).

A. Analysis for a Rapid Eguilibrium by _a_._n_ Investigation
gi|the Concentration Dependence 2£.Sedimentation

Coefficient and Molecular Weight by the Archibald
gechnique

To test the 3.58 intermediate Species for the possible

44
existence of a rapid equilibrium, sedimentation coefficient
analysis (for a theoretical treatment of this method, see
Fugita (1962)) and molecular weight analysis using the
(Archibald technique were chosen. The study by Rao and
Kegeles (1958) of the rapid equilibrium of a-chymotrypsin
using the Archibald method has shown the potential useful-
ness of such an analysis.

An enzyme solution of approximately 12 mg/ml, pre-
viously dialyzed against a solution of 0.01M citrate (Na),
0.05M NaCl, and 0.001M EDTA, was acidified with acid solu-
tion, 0.5M citrate (Na), 0.05M NaCl, and 0.001M EDTA (nine
parts of the former enzyme solution to one part of latter
0.5M citrate solution) to yield a final pH of 3.45. This
solution was then dialyzed for 2# hours. The dialyzed
solution was then diluted to 10.h, 9.0, 8.0, 6.0, 4.0, 3.0,
2.0, and 1.2 mg/ml. Sedimentation and diffusion coefficients
were then performed simultaneously with the molecular weight
analysis by the Archibald technique. For the sedimentation
coefficient and Archibald molecular weight determinations,
the four lowest concentrations were run in 30 mm cells for
better sensitivity (2.5x): the remaining concentrations for
these determinations and all concentrations for the diffusion
coefficient determinations, however, were run in the more
commonly used 12 mm cells.

The sedimentation coefficient data is shown in Figure
6 (closed circles). For a rapid equilibrium,a line with a

negative slope at high concentrations and a positive slope

“5

.Amoaoado umaov 0:.m mm and Amoaoado ommoaov m:.m mm pm nodpmhu

nsoosoo sampoaa oaou op mpsmaoammmoo soapmpsoaaoom go soapmfioamapxm

.0 oaswam

1+6

323;: ZOEEZMGZOU 3,;sz

 

 

 

«a. 0.2 3 on 3
_ _ d _ _
l J, /m/
0 O 4
33 n ;\omm\.
.. o 9% In
. 3m 1.. $383.52. $382

 

 

iNBDleBOD NOIi‘V’lNBWIGBS

47

or decreased negative slope at low concentrations is eXpected;

but this was not found. Instead, the graph shows a linear

line with a negative slope which, upon extrapolation to zero
0

concentration, yields a sedimentation coefficient of 820 w =
O

3.803. This result then contradicts what one might eXpect

for a rapid equilibrium between different molecular weight
species such as between monomers and dimers; on the other
hand, it is indicative of a single molecular entity such as

monomers, dimers, trimers, or tetramers.
The results from the Archibald molecular weight analy-

sis could not be interpreted by the above analysis because

the values were so badly scattered. This scattering might

have been eXpected, however, since the sedimentation velocity

patterns exhibited substantial amounts of aggregate or undis—

sociated native enzyme. It was later discovered that the

enzyme is not completely dissociated at pH 3.45 and thus,

that pH 3.146 was too high.
Diffusion coefficients were also measured from eXperi-

ments run simultaneously with the Archibald analysis; the

results are presented in Figure 7 (closed circles). The

details of the diffusion coefficients are presented in a
later section (Section B), so a discussion of those results
W1ll be deferred until then.

Following the Archibald molecular weight analysis,
°°mrerntional short column sedimentation equilibrium eXperi-
mentg were performed on the intermediate using different

3°1u'tion column lengths, produced by using different volumes

48
of 0.02 mg, 0.06 ml, and 0.09 ml, respectively. A qualita-
tively analysis of the curves indicated the presence of
higher molecular weight material. That aggregate was present
was attested by the fact that increased slopes were observed
near the bottom of the cell. Furthermore, the slopes
increased with time over a four day period and thus never
attained final equilibrium: these results indicated that
aggreagte was continuously being formed over this time
period.

In an attempt to reduce aggregation, thought to be
caused by non-specific interaction of the intermediate, and
to try to obtain complete dissociation of native enzyme into
the intermediate, molecular weight analysis M(S/D) was con-
ducted with what was thought to be two improvements: (1)
the pH was lowered from pH 3.45 to pH 3.40 to effect complete
dissociation, and (2) the total eXperimentation time was
decreased from 5 days to 12 hours to eliminate to the fullest
extent possible, the effects of aggregation with time. Also,
a new method for intermediate formation was used as described

in the following section.

B. Sedimentation Coefficient, Diffusion Coefficient,

and Molecular Weight M(SZD) of the 3.58
Intermediate §£,p§,3.40

 

Aldolase crystals were spun down, and the pellet was
dissolved in 2 ml of a solution containing 0.001M EDTA,
0.05M NaCl, and 0.01M citrate (Na), pH 6.9. The enzyme

49
solution was then dialyzed against this pH 6.9 solution for
six hours to remove any residual ammonium sulfate present
(if this dialysis was not done, the enzyme precipitated out
of solution in the following acidification procedure). After
dialysis a 1 ml aliquot of this solution was acidified by
direct addition of 1 ml of a solution containing 0.001M EDTA,
0.05M NaCl, and 0.1M citrate (Na), pH 3.20 (this solution
was first diluted five-fold to further reduce the salt con-
centration during the dissociation procedure). After acidi-
fication, the protein solution was dialyzed for three hours
against the following buffer: 0.001M EDTA, 0.05M NaCl, and
0.1M citrate (Na), pH 3.40. The solution containing the
intermediate was then diluted with the dialyzate to yield
four concentrations: 10, 8, 5, and 2.5 mg/ml. Sedimenta-
tion and diffusion coefficients were immediately determined
at 7°C.

Sedimentation velocity patterns revealed only a small
amount of aggregate or undissociated native in the first two
samples (10 and 2.5 mg/ml): but more aggregate was observed
in the second set of samples (8 and 5 mg/ml), indicating
aggregate formation with time.

The extrapolated value for the sedimentation coeffi-

cient is shown to be s = 3.808 (Figure 6, open circles)

o
20,w
and agrees closely with previous results at pH 3.45--both
were determined at 7°C. Presumably, the reason the slope
was more negative in this eXperiment than before was the

lower ionic strength (u = 0.035 vs. 0.11).

50

The diffusion coefficient, extrapolated to zero pro-
tein concentration, is Dgo,w = 3.85 x 10"7 cmZ/sec., (Figure
7, open circles). This data also includes three points from
the previously mentioned experiments at pH 3.45 (Section A).
Again, the temperature was maintained at 7°C for both eXperi-
ments.

The molecular weight obtained from the above data is
M(S/D) = 91,000 t 4.000. This value is too high for dimers
of subunits of molecular weight, 40,000; a value between
80,000 and 86,000 was eXpected. Since the molecular weight
may have been raised by the presence of a small amount of
aggregate, this eXperiment was repeated at pH 3.35 and 2°C
in an attempt to see whether such a molecular weight was
obtained under conditions where the 3.53 intermediate might

be more stable.

C. Sedimentation Coefficient, DiffusiOn Coefficient,

and Molecular Weight M(SéD) g; the 2.53
Intermediate at pH 2.35

To reduce the amount of aggregation to a minimum,
sedimentation coefficients and diffusion coefficients were
determined as soon as possible (1 hour) after acidification,
omitting the dialysis in these eXperiments. The buffering
capacity of the protein was calculated from the amino acid
composition. This effect was then compensated fgg'in the
acidification procedure--the resultant enzyme stock was pH

3.34, only 0.03 pH units less than the solvent.

51

.Ameao
ado s
emov
0:.m
ma
use
Amoaoaao c
omoao
V m:
.m me
no so
apes
p

:sooso

o s

. Hopes

a o
you 0
9 mes
; edoa
m
meoo sodeMM
do o
3»
mo :0
apwao
may
me
.u oasw
am

52

:2\o§ 20:3:szon 22>sz
3. 0.2 3 on 2
_ _ _

 

I

 

_ _

O O O

3&5 To. x 3m ":3 \\\...

o 9% :e

17.”

 

. 2% :e $385.52. $582

 

iNBIDHJBOD NOISFHJIO

53

The enzyme was prepared somewhat differently than
before, but the main features remained the same. Freshly
prepared enzyme was dialyzed for 36 hours. In the first
12-hour period, the enzyme was dialyzed against 0.001M
EDTA, 0.05M NaCl, 0.01M citrate (Na), pH 7.00, to remove
any residual ammonium sulfate. In the next 24-hour period
the enzyme was dialyzed against 0.001M EDTA, 0.05M NaCl,
and 0.003M citrate (Na), pH 7.00, with one solvent change
after 12 hours; this was to lower the citrate concentration
and buffering capacity of the enzyme solution. A test
dilution was then run to check for precipitation: none was
formed and so the entire sample was acidified 1.25-fold
(eight volumes of the former to two volumes of the follow-
ing pH 2.90 buffer: 0.001M EDTA, 0.05M NaCl, and 0.5M
citrate (Na), pH 2.90) to a final pH of 3.35. Sedimentation
coefficients and diffusion coefficients were determined
immediately after a one hour incubation at 0°C. Simultan-
eously, the molecular weight analysis by the Archibald
technique was begun.

Unfortunately, the results from the Archibald molecu-_
lar weight experiments were inconclusive because of the
uncertainty found in computing the area under the curve of
dC/dr vs. r in the schlieren patterns. Since the vertical
lines at the meniscii were very curved at high Speeds and
not coincident, the direct subtraction of the solution line
from the solvent line did not properly subtract out the
sedimentation of the solvent itself. This had to be

54
corrected by shifting the curves until they coincided. The
resulting uncertainty was as large as 16%. The uncorrected
values ranged from 87,000 to 130,000, whereas the corrected
values (see Table I) ranged from 96,000 to 116,000. Because
of this error, these molecular weight values should be
interpreted with caution.

Extrapolation of the sedimentation coefficient to zero
concentration gave a value of SgO,W = 3.453 (Figure 8); this
value (at 2°C and pH 3.35) is 0.353 units below the previous
values (at 7°C and pH values of 3.45 and 3.40). This differ-
ence in the sedimentation coefficient is probably indicative
of a more unfolded dimer at lower pH values and lower temper-
atures. .A lower pH is eXpected to unfold the molecule to a
greater extent and lower temperature is eXpected to unfold
the molecule by lowering the intramolecular hydrophobic
interactions. Also, the lower temperature would be eXpected
to reduce the collision frequency and thus, prevent aggrega-
tion. In fact, no aggregation was observed in the sedimenta-
tion velocity patterns in contrast to the previous eXperi-
ments.

The extrapolated value for the diffusion coefficient
was Dgo,w = 3.73 x 10""7 cmZ/sec (Figure 9), a value 0.12
units less than the previous value. This decrease is prac-
tically within eXperimental error, but it may reflect a more
unfolded dimer, in agreement with the sedimentation coeffi-

cient data.

Since no apparent aggregation was observed in the

55

.mHHmme

you pump mom .mm.m ma pm mumaooaampsa mmmaooam odomsa pannmh on» you Scandapnmo

:soo sampoam oaeN op mpflmaoammeoo noapmpsmaaoom 03p mo aoapmdoamhpxm

.m madman

56

:2\o§
o.

ZOE<EZmUZOU m2>NZm

m 0

V

m

 

 

mom Ia

q _

fulfil-J1,
m 2% "saw \i

mp<_om2~_mh2_

:

mm<4004<

 

 

lNBDHJEIOD NOIiVlNEIWICJBS

55

.uaaspee

you use we . . .
p m mm m ma pm mpmapeaampsd mmmaooam oHomsa panama esp How coapmapsoo

:200 s
Hoponm ones on mpfioaoammmoo noapmpsmaaoom on» mo aoapmfioamhpxm

.w easwam

56

v

:23:
o.

ZOE<EZmUZOU m<<>NZm

m o

v

m

 

 

new Ia

_ _

bI-Iulqlqlbl

u

_

l

 

m mum “saw \

mp<_om<<mmp2_

mm<._Oo._<

 

 

lNElDleEIOD NOIl‘v’lNBWIGBS

57

.nadepee

Mom pxmp mom .mm.m ma pm mpeaoeahmpsa emmaoodm eHomsa Hannah 039 How :oapmap

Izmosoo sdepoaa ones on museaoawmmoo soam5mmao on» no sodpmaoaeapwm

.@ 09.5me

:2\o§
E

ZOE<~EZmUZOU m<<>NZm
m o v N

 

 

mmm Id

_ _ _ _

JIf-lql»!

.Sm\~5 To— x mmm u so

mp<_om<<~_mhz_ mm<._0n_._<

3.2 \l

 

 

lNSIDHJBOD NOISFHJIO

59
sedimentation velocity patterns, these values for sedimenta-
tion coefficient and diffusion coefficient were considered
the most reliable. For this reason, these values will be
used in all discussions of the intermediate.

The molecular weight can be calculated from the sedi-
mentation coefficient and diffusion coefficient data, using
the published values for the partial Specific volume. The
published values for partial Specific volume were determined
at 200 and 5°, and there is some uncertainty as to the -
interpolation procedure to use to correct for temperature
dependence. Thus, the amount of correction upon temperature
change on the partial Specific volume is from 0.000365/degree
(Hunter, 1966) to 0.001/degree (Taylor and Lowry, 1956). If
the former is correct, a partial Specific volume of 0.735
cc/g should be used to compute the sedimentation coefficient;
this yields a value of ng.W = 3.453. But if the latter is

correct, a corrected partial Specific volume of 0.722 cc/g

is obtained; from this, a sedimentation coefficient of 33

= 3.263 is computed. Then by using these values, 3.263 agdw
3.453, and a partial Specific volume of 0.742 cc/g at 20°C
(Taylor and Lowry, 1956) in the Svedberg equation, the
respective values for the molecular weight are 83,000 and
86,000. A summary of the molecular weight results for the
intermediate at pH 3.35, and pH values at pH 3.40 and pH
3.45, are presented in Table I.

The molecular weight of the 3.53 intermediate of

83,000 to 86,000 is most consistent with a dimer molecule,

60

TABLE I

PHYSICAL PROPERTIES OF 3.53 ALDOLASE INTERMEDIATE

 

 

 

pH s20 D20 % PURITY 0F M(S/D) MOLECULAR WEIGHT
'W 'W INTERMEDIATE (ABCHIBALD)
3.40- 130,000-
3.45 3.803 3.85 70-85 . 91.000 200.000
3.35 3.45 3.73 85-95 83:000' 96:000-

86,000 116,000

 

61
Since these molecular weights are slightly more than twice
the molecular weight of aldolase subunits, assuming a value
of 40,000 (see Results in Chapter I, Section A).

To provide supporting evidence for the identification
of the intermediate as a dimer of aldolase subunits, further
physical information on the intermediate structure was
obtained from intrinsic viscosity measurements. It will be
also shown that other possibilities, which were described
in the introduction of this chapter, such as unfolded native

enzyme or folded subunits are excluded.

D. Determination 22 the Intrinsic Viscosity of the
§.§3 Intermediate gf'Aldolase and Comparison

with that for the Native and Subunits

Let us first give a few definitions to make clear the
nomenclature for different types of viscosities. Specific
viscosity n8 is defined by the following identity, nsp =
(n' - n)/n, Ehere n' is the viscosity measured for the
macromolecular solution and n is the viscosity for the pure
solvent. But as such, Specific viscosity is not too useful
for comparisons between different macromolecular solutions
since it is proportional to concentration. 0n the other

hand, reduced viscosity "red is independent of concentration

(for ideal solutions) and is given by the relationship:

ns
‘EB'= nred

However, since most macromolecular solutions are not ideal,

62
the reduced viscosity is usually extrapolated to zero con-
centration, and this extrapolated value, the most useful
quantity, is commonly called the intrinsic viscosity [fl].
Before presenting the eXperimental details and the
results, it is of interest to first predict what viscosity
one might SXpect for each of the different intermediates.
If, in fact, the intermediate is a distinct species, such
as extremely unfolded tetramers or trimers, unfolded dimers,
or folded subunits, intrinsic viscosity should be a very
sensitive tool to help distinguish between these possibil-
ities--independent of the previous molecular weight analysis.
For example, if the intermediate is folded monomers, the
intrinsic viscosity should be close to those values found
for globular proteins, i.e. about 4 cc/g. Alternatively,
unfolded tetramers or timers would have to have a much
higher intrinsic viscosity than subunits, i.e. higher than
24 cc/g. The intrinsic viscosity for an moderately unfolded
dimer, however, would be mid-way between 4 cc/g and 24 cc/g.
In the first eXperiment, kinetics of intermediate for-
mation were followed beginning immediately after addition of
acid. The final conditions were pH 3.4, 8.3 mg/ml, and 6°C.
(See Methods for preparation of dimers). After 3 hours the
reduced viscosity for the dimer intermediate reached a limit-
ing value of 14.5 cc/g, which was constant for at least 10
hours (Figure 10). In a similar eXperiment at a Slightly
lower concentration of 6.2 mg/ml, a reduced viSCosity of 14.0

cc/g was obtained. These data, together with the viscoSities

63

on pxep mom
9

gm
0 nova
ado m

ow He

no noHpmSH

oodpeh

omab o

.hpamo

.mHprmU

.oa enemas

64

= o. F e. m m _

 

 

 

u on we Ia
>58; 20 20.2.5.0“. 325 do but“:

 

 

 

65

determined for the native enzyme and subunits, are reported
in Figure 11 (closed circles); here it is seen that our
values for reduced viscosities for the native enzyme and
subunits extrapolate to the same intercepts and yield the
same intrinsic viscosities as those (open circles) reported
by Stellwagen and Schachman (1962). The intrinsic viscosity
of 14.0 cc/g for the intermediate is approximately half-way
between the value of 4 cc/g for the native enzyme and 24
cc/g for the subunits (Figure 11). As discussed in the
introduction of this section, the fact that the viscosity
for the intermediate is mid-way between 4 cc/g (folded
native) and 24 cc/g (unfolded subunits) suggests, indepen-
dently of the molecular weight analysis, that the intermedi-
ate is a moderately unfolded dimer.

An independent method for the measurement of asym-
metry of the dimer molecule involves calculating the B

parameter from the following equation (Scheraga, 1961):

5 _ 8 [Nil/3
1112;3 (I-Vp)

Thus, by using the experimentally determined values for the
intrinsic viscosity (14.0 cc/g), sedimentation coefficient
(3.453), and molecular weight (84,000) for the dimer molecule,
the 8 value from the above formula is calculated to be 2.28

to 2.45 x 106. This value results in an axial ratio of about
8 to 11 (see Table II on page 6 of Scheraga's book (1961))
assuming a prolate ellipsoid. The value of B for an oblate

ellipsoid is essentially independent of axial ratio in this

range.

.mflaapee

/o How vamp mom .mesHm> ma ompmoaosa on» no mamaosoa one .maeaao .mShuSo ebapm: Mom
6
coapeapsoosoo Sampoaa oaou op mpamoomab decades one mo Soapmaoaeapxm .HH easwam

6?

323$ 20.2323on 3,;sz

 

 

 

 

 

 

m_ o. m o v m
_ . _ l, _ . _ . _ . _ b
as re Ignaz/Em: l e I .
10—
am £13.25 I. ?\ 4
low
as Islamzozoz ll. 1? + U.
:3

>.:mOUm_> ZO ZOC<EZmUZOU ”.0 Humtm

 

 

3/ 'ds ll

(6/33)

68

The values for the axial ratios, together with the
pertinent physical data from this and other sections are
summarized in Table II.

In conclusion the viscosity of the 3.53 intermediate
is most consistent with a dimer. The following discussion
Shows how the physical data in Table II excludes other pos-
sibilities for the intermediate, such as tightly folded
subunits, or extremely unfolded native enzyme.

If we assume the intermediate is subunits with a
molecular weight of 40,000, these subunits must be tightly
folded-~a molecule of this molecular weight must be very
compact (f/fo = 1.3) to have a sedimentation coefficient of
3.53 (Figure 5). But for reasons given earlier the observed
value of 14.0 cc/g is not consistent with a folded globular
protein; furthermore, using the B value calculated from the
previous formula, an axial ratio of 400:1 is estimated. Thus
a contradiction results: it is obvious that the subunits
cannot be both very compact and extremely unfolded. There-
fore, the intermediate cannot be subunits, Since the symmetry
data cannot be reconciled using a molecular weight of about
40,000.

We next analyze the possibility that the intermediate
might be an unfolded polymeric (trimer or tetramer) enzyme.
The same data is used, with the molecular weight being
changed to that of trimers or tetramers, i.e. 142,000 to
160,000. For a molecule of this molecular weight to possess

a sedimentation coefficient as low as 3.53, it must be

69

TABLE II

PHYSICAI.PROPERTIES 0F ALDOLASE MONONERS,
DIMERS , AND TETRAMERS

 

 

 

 

 

SPECIES pH INTRINSIC MOLECULAR AXIALf
VISCOSITY wEICHT RATIO
8.
2. 2 . 38 300-46 600
MONOMER 9 3 5 cc/g 41:000?42:0000 13
d
DIMER 3.4 14.0 cc/g 83,000-86,000 8-11
' 142 0008
TETRAMER 6.9 4.0 cc/g ' b 6

160,000

 

a. Edelstein and Schachman, 1966.

b. Kawahara and Tanford, 1966; and thantis, 1968. This
appears to be the most probable value.

0. See Results, Chapter I-A.
d. See Results, Chapter II-B, C.
e. Stellwagen and Schachman, 1962.

f. Axial ratios are not to be taken literally--includes a
factor of hydration.

70
extremely asymmetric (f/fo = 3.0, see Figure 5). When the
intrinsic viscosity of 14.0 cc/g, a molecular weight of
150,000, and a sedimentation coefficient of 3.53 are com-
bined in the B formula, a value for B of 1.5 x 106 is cal-
culated. But theoretically the lowest possible 3 value is

2.12 x 106

which applies for perfect Spheres; this value
can only increase with increasing asymmetry. Therefore,
this 3 value (1.5 x 106) takes a value not allowed and
results in a cOntradiction: that this molecule of 150,000,
assumed to be very asymmetric, is required to be more
symmetric than a perfect Sphere. Hence the results from
the viscosity eXperiments demonstrate that the intermediate
is not folded monomers, or unfolded trimers, or unfolded
tetramers. But, in fact, is most consistent with a dimer
of aldolase subunits. These results from the viscosity
analysis, then, do indeed support the conclusions from the

molecular weight analysis: that the 3.53 aldolase inter-

mediate is a moderately unfolded dimer.

E. Analysis for Incomplete Dissociation and Non-specific

Aggregation p£.Dimers: Attempts pp Separate Dimers
from Higher Molecular Weight Material and Estimate

Dimer Purity

1. Effect 23 Salt and Temperature pp Homogeneity

p§.Dimer Preparations

Aldolase crystals were centrifuged and dissolved in

the following solution: 0.001M EDTA, 0.05M NaCl, and 0.003M

71

citrate (Na), pH 7.00. The enzyme solution was then dialyzed
against this buffer for 15 hours; the final enzyme concentra-
tion was 13.9 mg/ml. After dialysis the enzyme solution was
divided into equal fractions. Fraction one was not dialyzed
further; but fraction two was dialyzed for 5 more hours
against the above buffer solution, diluted five-fold with
water. Both solutions were then dissociated by acidifica-
tion to form dimers with a 1.25-fold dilution with 0.5M
citrate (Na), 0.001M EDTA, and 0.05M NaCl, pH 2.90. After
one hour, the solutions were then analyzed by sedimentation
velocity eXperiments.

The results (Figure 12A) show that the sample, which
was dialyzed the second time (fraction two) against the
0.01M NaCl, buffer solution to remove most of the salts, was
essentially free of higher molecular weight material; while
the other sample, dialyzed against the 0.05M NaCl, buffer
solution, had a substantial amount of material sedimenting
at a faster rate. Thus it seemed best for subsequent experi-
ments to remove almost all the salts before acid dissociation.

For the second part of this study, to test the effect
of higher temperatures on the stability of dimers, a sample
of dimers at 11.0 mg/ml from the above preparation was divided
into two parts and treated as follows: (1) Sample I was
incubated for 16 hours at 0°C and the relative distribution
of tetramers (and aggregate) and dimers was analyzed by con-
ventional moving boundary sedimentation velocity (Figure 12B).

(2) Sample II was subdivided into two additional parts. The

72

Figure 12. Sedimentation velocity patterns of dif-
ferent dimer preparations eXposed to various condi-
tions. Sedimentation is from left to right. The
times at which pictures were taken after reaching
Speed (59,780 rpm) were 60 minutes for photograph
12A and 12B and 25 minutes for photograph 3C. The
main peak in A, B, and C (upper curve only) was
shown in previous eXperiments to be the dimer inter-
mediate with a sedimentation coefficient of approxi-

mately 3.1 (4 mg/ml). See text for details.

73

 

 

  
 

 

0°C-16HRS.

 

Run at 22°C ‘

it ‘C

74
first part (II-0°) was incubated for 16 hours at 0°C and the
other part (II-22°) was incubated for 16 hours at 22°C. Then
both were analyzed by sedimentation velocity eXperiments to
determine the reSpective dimer-tetramer distributions. The
peaks in the upper curve (Figure 12C) indicate that some of
the dimer sample (left peak) is converted into tetramer or
aggregate (right peak) as a result of warming the sample
(II-0°) to 22°C to run the sedimentation velocity eXperiment
(compare to the control in Figure 12B). The lower curve in
Figure 120 Show the effect of incubation at 22°C instead of
0°C for 16 hours, and run at 22°C as in sample (II-0°):
essentially all the dimers had aggregated within this period
of incubation, and some of the dimers had dissociated into
monomers. It is not known whether dimer dissociation into
monomers precedes aggregation, or whether it succeeds aggre-
gation when the resulting dimer concentration is very low.
The upper peaks in Figure 12C indicate that aggregation

precedes dimer dissociation, i.e. the latter case.

2. Attempts pp.Separate Higher Molecular Weight

Material from Dimers

The main purpose of these eXperiments was to try to
separate the dimer from aggregate in order to be able to
measure the molecular weight of the 3.53 intermediate sample
by sedimentation equilibrium. Separation by sedimentation
velocity separation cells, sucrose density centrifugation,

Bio-Gel filtration, and Sephadex filtration was tried. It

75

should be emphasized that sedimentation velocity eXperiments
showed no evidence of any higher molecular weight material
in the dimer preparations. However, sedimentation equilib-
rium, a technique much more sensitive to small amounts of
heterogeneity, indicated some higher molecular weight
material to be present.

Both of the first methods, separation by sedimentation
velocity separation cells and separation by sucrose density
sedimentation velocity SXperimentS, were unsuccessful.

The Bio-Gel and Sephadex filtration eXperimentS, how-
ever, were successful in the attempt to separate the dimers
from higher molecular weight material.

The procedure for preparation of the protein sample
was as follows: Aldolase crystals were Spun down, dissolved
in 0.001M EDTA, 0.05M NaCl, and 0.003M citrate (Na) buffer
at pH 7.0, and dialyzed against this solvent for 24 hours.
The dialysate was diluted five-fold with water, and dialysis
continued for another 12 hours. Four parts of this enzyme
solution were then acidified with one part of a solution con-
taining 0.001M EDTA, 0.05M NaCl, and 0.5M citrate (Na), pH
3.35. A 0.5 ml sample of dimers at 25 mg/ml was applied to
a column (80 x 1.5 cm) containing Bio-Gel P-150, equilibrated
in the pH 3.35 citrate buffer at 4°C.

Five ml fractions were collected and the optical den-
sity was determined at 280 mu. The results are shown (closed
circles) in Figure 13. The void volume, determined with a 1%

dextran marker, was 55 ml and is equivalent to the first 11

76

.mHHmpmo

pom pxmp mom .Amosaa Soxoan mm samnm SH czoswv m.u ma pm noapomnm some

so Soapwbapomoa Hopmm mosaahopoo mes unabapom use AmosHH odHom we samaw SH
szozmv 18 0mm pm oosmnhompo an penaaaopoo mes soapmap:mosoo anponm .Ha\wa

mm one mm.m ma so oaasoh ohdaooao so do soapdspaae Hoonoam omaam .mH enemas

77

ACTIVITY (UNITS/ML)

ON!

OR!

 

 

mmmEDZ ZO:.U<~I
om mm ON . 2 o_ n

 

 

 

. _ u Li
“.0 ...;
z... a
/.O. Inc
Ono-o... o...o
. /. _
as... o 19.
. . A .0
/ o .G
0.... me. i M
/..... lam
\ o _
522:: .L
0
sum Io om
.535 mm H .uon .sz

 

 

35:22 32557325 do 20:55... 3002 omTa

78
fractions. After the protein determination was completed,
the fractions were diluted five-fold with 0.2M imidazole (Na)
pH 7.5. incubated for 2 hours at 0°C and then for an addi-
tional hour at 22°C to reactivate the inactive dimers and
native enzyme (open circles, Figure 13). The higher Speci-
fic activity, 9.3, was obtained with fraction 11: therefore,
this peak was essentially all native enzyme which was only
unfolded at pH 3.35 and not dissociated. Thus even at pH
3.35, high protein concentration (25 mg/ml) prevented com-
plete dissociation. The fractions were too low in protein
concentration to be analyzed further by ultracentrifugal
eXperiments. However, an attempt was made in the Sephadex
filtration eXperiments to concentrate the protein for further
analysis. The method of protein concentration and the results
will be described below.

Following the separation by Sephadex filtration, tubes
from the dimer peak and from the aggregate or tetramer peak
were separately pooled and concentrated for subsequent sedi-
mentation velocity analysis. The concentration of protein
was accomplished by absorbing the water from the dialysis
bag containing the protein solution with G-25 Sephadex. Sub-
sequent sedimentation velocity patterns of the concentrated
dimer fraction revealed a molecule sedimenting at the same
rate as the dimer, whereas the tetramer-aggregate peak which
was also concentrated by G-25 Sephadex Showed the presence
of native enzyme and aggregate. Presumably, the aggregate

was formed after separation by gel filtration in the concen-

79
tration step and because of this, further physical analysis

of the dimer was not pursued.

Summarizing this chapter, the molecular weight of the
intermediate is most consistent with a dimer of aldolase sub-
units. The best conditions for the production of the dimer
are (1) very low salt (0.01M NaCl), (2) 0°-2°C, (3) pH 3.35,
and (4) final enzyme concentration between 1.0 and 10.0
mg/ml. Also, the dimer is more reactive at higher tempera-
tures and consequently aggregates. And if the native enzyme
is not essentially free of salt before dissociation, the
production of the intermediate is hampered; the product
aggregates and in fact, if too high a salt concentration is
present, the enzyme precipitates out of solution.

The sedimentation coefficient is probably not affected
very much by any aggregation present because of the "kinetic
purification" during the run. What is meant by this is that
the aggregate, which has a much higher sedimentation rate
will move down the cell faster and be separated completely
from the intermediate boundary. Consequently, it will not
interfere with the slower moving boundary of the intermedi-
ate. 0n the other hand, the diffusion coefficient is a
weight-average of the diffusion coefficients of all the com-
ponents in the solution. Thus the aggregate indicated by the
sedimentation equilibrium eXperiments would tend to yield a
diffusion coefficient too small compared to what it should
be; therefore, proper correction of the diffusion coeffi-

cient would increase it and consequently would decrease the

80
molecular weight, according to the Svedberg equation. An
estimation of this error would be helpful in defining the
accuracy of the molecular weight data. The following sec-
tion describes attempts to estimate the error by an empiri-

cal determination of the purity of the dimer preparation.

3. Ag Estimation pf the Purity p£_the Dimer Preparation

Peaks lg the Sedimentation Velocity Patterns

A stock solution of native enzyme at 7.4 mg/ml was
prepared in the usual manner (see Section C-l above) and
divided into two parts. Part I was diluted 1.25-fold with
the second stock solution dialysate, while part II was
acidified 1.25-fold with the pH 2.90 citric acid buffer used
to prepare dimers. After one hour both were analyzed by a
sedimentation velocity eXperiment, and the areas of each
peak were measured and compared. The ratio of dimers to
tetramers was about 0.85; thus a maximum value of 15% is the
estimate for undissociated tetramer-aggregate present. To
calculate a corrected diffusion coefficient for the dimer,
and thus the percentage error in this value, (and the molecu-
lar weight value), the following weight-average relationship
for diffusion coefficients was used:

DMixture = DAggregate (15%) + DDimer (85%) = 3.73 X 10'7 cmZ/Sec

where 3.73 x 10"7 cmZ/Sec is the SXperimental value for the
mixture. If one now assumes a diffusion coefficient, 3.0 x

10"7 cmZ/sec (a minimal value for a molecule with a molecular

81
weight of 300,000, unfolded to the same extent as dimers at
this pH), the corrected value for the diffusion coefficient
for the dimer Species is calculated to be 3.90 x 10"7 cmZ/sec.
This increase in the diffusion coefficient for the dimer
intermediate is approximately 4% and would decrease the
molecular weight of the dimer from the 83,000-86,000 range
to 80,000-83,000 range. A value in the range of 80,000 to
84,000 is exactly what is predicted for a dimer of aldolase
subunits with a molecular weight in the 40,000-42,000 range
(see Results in Chapter I).

82

III. Characterization g£_the Reactivation Reaction

From p§,3.3 Dimers

Since tetrameric aldolase can be selectively dissoci-
ated into dimers (pH 3.3) or into monomers (pH 2.8) (see
Results in Chapter II), two distinct intersubunit binding
sites must exist, one for monomer-monomer interaction and a
different one for dimer-dimer interaction. Selective disso-
ciation was eXpected from an isologous tetramerl (pseudo-
tetrahedral) in which two isologous dimers, each with an
axis of symmetry, are associated into the tetramer with a new
axis of symmetry formed (Changeus, 1964). But what does the
effect of breaking the native enzyme into dimers have on the
catalytic properties of the newly formed dimer, presumably
each with half of the four catalytic sites. Presumably, when
dimers are formed, new previously uneXposed hydrophobic areas
then become eXposed to the aqueous medium and may cause a
conformational change in the dimer to minimize the hydropho-
bic exposure to water; this could alter the topology at the
active site and render the dimer inactive. Alternatively,
there may be little conformational change upon dissociation

into dimers. Thus activity analysis of the dimer would be a

 

1Isolo ous Association: The domain of bonding involves
two identicaI BI

nding sets.

Even though rabbit muscle aldolase is now thought to
have two different subunits, d and d', the domain of bonding
must still be equivalent for both subunits since they Show no
apparent preferential interaction for each other in associa-
gion-dissociation studies (see isozyme section in Literature

eview .

83
sensitive technique to test the effect of dissociation on the
conformation of the active Site. An analysis for possible
catalytically active dimers, therefore, was the subject of
research to be described in the rest of this thesis: (1) In
this chapter (III), the immediate aim was to find conditions
which would produce complete reactivation of activity from
pH 3.3 dimers, free of any aggregation; and an analysis for
pH 3.3 dimer catalytic activity. (2) Chapter IV describes
an investigation of the factors which inhibit dimer associa-
tion at higher pH's under conditions in which the native
enzyme remains as a tetramer and still possesses catalytic
activity. .(3) Chapter V will be concerned with the final
analysis for active dimers at higher pH values (pH 5.5 and
pH 7.5).

It was possible to define the conditions for total
recovery without the necessity of a complete systematic
study. The key eXperiments which Showed the effects of
variables upon recovery are described below.

Since the method for producing dimers was the same
throughout the rest of this chapter-~but is unique for this
chapter--this procedure will be described first. Native
enzyme was dialyzed against 0.01M citrate (Na), pH 7.0, for
24 hours at 0°C. This stock solution was diluted two-fold
with 0.025M citrate (Na), pH 2.35, to final pH of 3.3 and
incubated for one hour at 0°C. Usually the final dimer con-
centration was between 1.0 and 6.5 mg/ml. Kinetics of inac-

tivation demonstrated that the reaction is essentially com-

84
plete in 35 minutes.

Since the eXperimental conditions for obtaining opti-
mum recovery of activity were developed in successive stages
from exPeriment to eXperiment, comparison between different
eXperiments will be limited. However, any comparisons that

can be made will be emphasized in the text.

A. Effect pf B-Mercaptoethanol pp Native Enzyme and

Reversal pf Catalytic Activity from.p§ 3.3 Dimers

The first variable to be tested for its effect upon
reversal recovery was B-mercaptoethanol. After dimers were
formed at a concentration of 6.50 mg/ml, they were diluted
500-fold at 0°C to a final enzyme concentration of 0.013
mg/ml with 0.3M Imidazole buffer, pH 7.9, containing the
appropriate concentration of B-mercaptoethanol. After incu-
bating 5 minutes at 0°C the solutions were brought to 20°C
for 15 minutes before assaying. For the control, native
protein solutions were diluted in the same manner except
the pH was never changed. The results (Figure 14) Show
that low concentrations of B-mercaptoethanol (0-0.15M)
stimulate recovery of activity but have relatively little
effect on the native enzyme. However, at higher concentra-
tions of B-mercaptoethanol (0.7M) both the native control
and reversal samples lost activity; presumably, this is an
unfolding effect of the solvent on the enzyme.

Another critical factor in obtaining total recovery

of activity upon reversal was pH. The effect of pH at

.maumpeo Mom axon mom .Ha\wa mao.o mo sodpmapseo
isoo oaause assay m use .ooom pm soHumnsOsd Chanda ma e an oozoaaom 000 pm Soapmn
MW Izosd mascas m .m.m ma who: muoapausoo Hdmhmbom .maeaau m.m ma Song hpdbapoo OH»

:mHmpmo mo Hemaobmh use mahuso e>apmc so HOSTSpoopamoaoalm mo poommm .fia ondwam

E2302 Loz<rmoz<ofi2um

 

 

 

 

 

j d a N\ — q — —
-2
Ommmm>mm .50....
m>:<z |.| 13
.8
as
o .................... o ......... o.....
.+ . ...... as

 

$52.0 mm<._Oo._< m0 L<mmm>mm

AHEAODBH %

87
various times of incubation is described in the following

section.

B. Effect p£_pH‘gijeassociation from.p§,§.3 Dimers

 

22 various Times prIncubation 23 0°C and 20°C

The procedure for testing the effect of pH was quite
similar to that in the B-mercaptOethanol eXperiments. After
preparing the dimer (6.5 mg/ml), aliquots of 10 ul were
diluted into 5.0 ml of 0.2M Imidazole buffer or 0.2M citrate
buffer at the appropriate pH and incubated in this reactiva-
tion medium for various times--the final concentration was
again 0.013 mg/ml.

Almost 80% recovery was obtained after incubation in
the pH 7.0 reassociation medium for 6 hours at 0°C, followed
by a 2 hour incubation at 20°C (Figure 15). The overall pH
profile was quite broad both for the 5 minute incubation and
the 6 hour incubation at 0°C. The one point at pH 5.5 may
be in error, but any downward correction would just increase
the broadness of the curve.

As Judged from the last two eXperiments, recoveries
are beSt in 0.1M B-mercaptoethanol, pH near neutrality, and
incubation in the reassociation medium for 6 hours at 0°C
followed by 2 hours at 20°C. These conditions--except the
6 hour incubation at 0°C which was changed to 1 hour at 0°C--

were used in the reversal procedure in the next eXperiment.

88

.uadwpoo hoe cwoc com .ooom osw ooo so soacss

Izosd mo mead» mSOdam> pm maesHo m.m ma mo Hempobon no ma ho poemmm

.mH shaman

89

EON/3:2. Sud ..I In II 35:5 ind
m N 0N m. o 0.0 mum ohm
_ _ _

\N

 

 

2.2 «Tees 2%: 2.2 ml oo
«...—loos 22.: z.I2m oo

 

O
O
000000
0 0
oo 0.
I
'0 O
0. 00
0
o O
00 ‘0
.0 ...

a: a I oou 22:
.3: o loo

 

oop

 

3.52.0 34309.4. “.0 ._<m~_m>m~_

AUSAODSU °/o

90
C. Effect 3; Dimer Concentration and Time g£_Incubation in,

the Dissociation Medium: Attainment g£_Conditions for

9gfi-100Z Recovery 21; Activity

In contrast to the previously described eXperiments,
in which reversal conditions were varied, the purpose of this
eXperiment was to investigate various conditions in the
dissociation medium. It was thought that dimers might be
aggregating with time and thus cause a decreased activity
recovery in the reversal studies; moreover, high enzyme con-
centrations would tend to increase this aggregation. Thus
an eXperiment was designed to test two effects: (1) dimer
concentration and (2) incubatiOn time in the dissociation
medium.

Enzyme solutions were acidified as before to form
dimers at a concentration of 6.5 mg/ml. This solution was
immediately diluted with dimer solvent to 1.3 mg/ml and 0.70
mg/ml; these three solutions were then incubated at 0°C for
1.0. 1.5. and 2.0 hours. After the incubation period the
three dimer solutions were diluted into the reassociation
medium of 0.3M Imidazole buffer. pH 7.9. and 0.1M B-mercapto-
ethanol at 0°C, to a final enzyme concentration of 0.013
mg/ml. After one hour at 0°C, the solutions were warmed to
20°C for two hours and assayed. The results reported in
Figure 16 show that at least six dimer samples had been
reactivated to 90-100% of the control. Furthermore, the
results show that the recovery of enzymatic activity was not

significantly altered by dimer concentration at incubation

91

.pxop on» on powmh .madmpoo honpo Mom .Ha\wa mao.o Mo noapwhpzoo
Isoo oahNCm Hanan a one .ooom pm coaprSoca H50: 03p a an dozoaaom 000 no soap

Iwnfiona H50: ono .HosmSpmopmmonalm 2H.o .m.m mg ohms mSOHpadsoo mehmboh one

 

 

.hpabdpos mo hamboooa mooaIfiom Mom mzoapaosoo mo puma:HMppd ”asauoa zodpmao

 

Iommdu on» ma soapGDSozH go mad» was GoapMHpcoozoo Hoaad mo poommm .wa onswam

 

 

 

 

 

 

 

 

 

 

                 

 

 

INCUBATION TIME (HRS)

EFFECTS OF DISSOCIATION CONDITIONS ON REVERSAU

2 g g2 D I“

Q a o B ‘3 ‘2- '3

5 ° " E

3 _

E

mum1111.!1.11.1.1!.IJ.J.I.I.uJJJJIJJIIIIIJ.11.11le1111:uIJIUII1111quu,
4 z:223252525253235255252535232323232323232523a525:332525235323335322252523:5252533232322232525:2:isisésésézszszszsgsgz:5:szszszzzszszszsza:2:222:22:2:3.-;: ...-
U- F

§

2 L iiiiiiiiiiiiiiiiiiiiiiiiiiéésizisisizisisisis§s§s§s§s§s§s§s§s§s§s§s§s§5§5§siege;eggs;s;a;a;55ggags;agggggggggggggggggggggggj __.

 

 

 

BEACON °/o

93
times at 1.5 and 2.0 hours.

Although.not shown in the graph, other studies have
shown that the % recovery of the sample which was incubated
for one hour at 6.5 mg/ml in the dissociation medium can be
increased from 55% to 98% by simply increasing the reversal
concentration from 0.013 mg/ml to 0.065 mg/ml. The activity
remained constant at 98% for at least 5 hours.

The Z recoveries of activity are all based on dupli-
cate controls (arbitrarily placed at 100%) which were put
through the same incubations and dilutions but not the change
in pH. Any change in activity due to instability in the
eXperimental samples should also occur in the control. The
controls had a Specific activity (umoles/min/mg) of 12.2 and
12.3, close to the reported value of 12.5.

In conclusion, the main purpose of these eXperiments
was to find conditions for complete recovery of enzyme activ-
ity upon reversal of dissociation into dimers. This eXperi-
ment accomplished this purpose by identifying conditions
where at least 6 samples were reversed 90-100% and 2 samples
were reversed 98%. Also, at higher concentrationsin the
reversal mixture (0.065 mg/ml vs. 0.013 mg/ml) three more
samples were reversed 98% (the 0.065 mg/ml was not graphi-
cally represented).

Now that the conditions for complete reversal of
catalytic activity had been found, the rest of the work
described in this chapter was concerned with the following

two problems: (1) the question of whether pH 3.3 dimers are

91+
active; the next three sections describe studies on the
effects of temperature and dimer concentration on the
kinetics of reactivation and half-live of reversal, and (2)
the question of whether the substrate, FDP, could inhibit

reactivation and/or dimer association.

D. Recovery as a Function of Time at Two Temperatures

Dimers (pH 3.3) were prepared as before and incubated
for 2 hours in the dissociation medium at an enzyme concen-
tration of 0.75 mg/ml. These dimers were then diluted 50-
fold to a final concentration of 0.013 mg/ml with 0.3M
Imidazole buffer, pH 7.9, and incubated at 0°C or 20°C.
Aliquots were taken at various times and tested for activity.
Activity recoveries at these two temperatures are shown in
Figure 17. Kinetics show that the initial rate of recovery
was very fast (“0% in 2 minutes) at 20°C, and the reaction
was essentially over in 20 minutes. When compared to the
native control, the recovery was approximately 60% after
one hour at 20°C. At 0°C, however, only a small amount of
activity was found. Later, it was shown that a 2“ hour
incubation at 0°C produced about 10% reactivation. Thus
reactivation was occurring at 0°C, although it was very
slow.

The increased reversal at the higher temperature was
probably not only due to increased collision frequency of
the diffusion-limited process of association but also due

to increased dimer interaction strength caused by stronger

95

now pump mom

.nodpmanoana Hacodpdeem

.moHSpsaoaamp 03» pm mad» go coapossw m mm mambooom

.ma ohzwam

96

mmh32=2

 

 

CNN 02 o: oo. Fioo 0v om
_ fl . _ 1 4 q A
o.o\
In (To
UoON \
3 In ...232 Sod n .u20u 3.9.55“

 

 

oop

 

 

._<m~_m>w~_ 5:25 20 mEmhmO ...UmuEm

O/O

AUBAODBU

97
hydrophobic bonding at 20°C over that at 0°C. If the reversal

process involves an association reaction. it should be con-
centration dependent; in particular. higher concentrations
should also increase the rate of reversal. EXperiments to
test for concentration dependence of the reassociation

process are described in the following section.
E. Kinetics 2£.Reversal Process §£.Two Concentrations

After dimers were prepared, they were diluted into the
reversal mixture, 0.3M Imidazole buffer, pH 7.9 and 0.1M B-
mercaptoethanol at 0°C to a final concentration of 2.17 mg/ml
and 0.013 mg/ml. Aliquots were removed at various times and
assayed.

As can be seen in Figure 18 where kinetics of reversal
are plotted for the two concentrations, the higher concentra-
tion indeed increases the rate of reversal, but only 25%
recovery of activity was obtained at this concentration.

This low recovery was probably due to non-Specific aggrega-
tion. since a white precipitate had formed after two or three
hours. At the lower concentration much less reversal was evi-
dent; by the time the higher concentration (2.17 mg/ml)
plateau of reversal was reached, only about 1% reversal had
occurred at the lower concentration (0.013 mg/ml). However,
after longer times (20 hours) the lower concentration sample
had reversed about 10%. The concentration dependence of
reactivation at 0°C is consistent with a model in which an

inactive dimer must associate before activity is obtained.

98

.madm
pop a
on
9N
op mom .m
soaps
space

zoo

menus
o 03
9 pm
mad
p we
so“
poms
m a mm m
Head
e m.
n me
do .
Hams
mbmh
0:9

ma enemas

99

 

 

45:02 90.0

 

\

.:2\0<< N flu

122$ .23
as 1.. .uoo "mzoEoon 2252:.

 

O O
N '—
AHBAODBH

O
0')

O
V

 

 

._<m~_m>m~_ ZO .UZOU «mi—D ”.0 .0373

Om

°/o

100
Thus the pH 3.3 dimer is probably inactive at 0°C.

But the activity assays show significant lag times;
perhaps the pH 3.3 dimer which is inactive at 0°C folds in
the assay at higher temperatures and becomes active. If the
pH 3.3 dimers must associate before reactivation occurs, the
effect of concentration on the half-time for this reaction
should reflect a second order reaction. The following eXper-
iment examines the effect of concentration on half-time of

reactivation in the assay.

F. Effect 23 Concentration 2E.Half-Time 9;,Reactivation
‘32 the Assay (16°C)

Dimers were produced at 3.25 mg/ml and incubated for
at least 1.5 hours at pH 3.3. Then 10 ul and 20 ul aliquots
were diluted into an assay to a final concentration of 0.09
mg/ml and 0.18 mg/ml at 16°C (16°C instead of the normal
assay temperature of 25°C to prevent precipitation which
‘?occurs at the higher temperature) and assayed at 16°C.
Relative Specific activities from typical assays are repro-
duced in Figure 19. The half-time for reversal for 0.18
mg/ml sample was 1.0 minutes; whereas the half-time for
0.09 mg/ml dimer sample was 2.2 minutes. If pH 3.3 dimers
were active upon folding in the assay, the half-time for
this unimolecular reaction at various dimer concentrations
should be constant; however, the results show that the half-
time was reduced by approximately two when the dimer concen-

tration was increased by a factor of two. This inverse

101

.mHHmme Hospo How pHmp 0mm .m.n mm Una oowa pm

momma exp 2« Homamboa mo omaalwams mo soapmnpnoonoo Hoaac mo poommm

.9 chew;

102

3.52.2

 

l

 

2 m o.v a

— a _ . u

w m

3.232 8.8 .z=z «a u “a III
:23: 2.3 .23 S u d

‘ 4 4

0

- ’
‘ ‘ ‘

W 0U
.2 n ._. ”$9.0 ozoBm

3 _._Q .oofl H><$< z. Swami”.

0°. V.
o o

AllAllDV 'DidS

N.

 

 

._<m~_m>m~_ ".0 mmZInSdi ZO.UZOU «mi—D ...O .53.:

103
relationship is consistent with a second order reaction which
follows the half-time equation, l/Co = kt%. From this rela-
tionship a second-order rate constant was calculated to be
k a 1.35 x 1o“M liters/mole-sec at 16°C. These results then
are consistent with a model in which dimers must associate

before catalytic activity is obtained.

G. Effect 2£_FDP gn_Beactivation.gf,p§.3.} Dimers

_i_n the Assay

In the course of previous studies it was observed that
when FDP was present in the assay before pH 3.3 dimers were
diluted into the assay, much lower activities were found than
if FDP was added last. Subsequent studies showed that if a
pH 3.3 dimer sample at 0.0125 mg/ml was added to the assay
before FDP was added, substantial recovery of activity was
obtained after a lag period of one to three minutes; but if
FDP was present before dimers were added, no recovery of
activity was obtained even after ten minutes. Moreover, if
higher concentrations (0.025 mg/ml) of dimers were used, the
FDP inhibition of reactivation was partially overcome. The
fact that the FDP inhibition of reactivation is enzyme con-
centration dependent suggests that FDP somehow binds to the
dimer and prevents a conformational change required for
dimer association. 9

Thus eXperiments were designed to test the effect of
various concentrations of FDP on initial activities of dimers

at a dimer concentration of 0.025 mg/ml in the assay. The

104
values were then compared to dimers incubated in the assay
without FDP under identical conditions. The effect of FDP
on the native enzyme was also determined to make sure that
the native enzyme was not inactivated by FDP. The concen-
tration of FDP which caused a 50% inhibition of activity
over the control was considered to be a measure of the
binding constant for FDP inhibition (Figure 20):

_ -3
KFDP Binding - 2 X 10 moles/liter

This is approximately 500 times higher than the FDP binding
constant (kd = 4 x 10-6M) for the native enzyme. Evidence
which will be presented later (Chapter IV, Section B-7)
supports this data by the demonstration that FDP does indeed
inhibit association of the pH 3.3 dimer. Furthermore, the
binding of FDP to the pH 3.3 dimer is probably through the
phOSphate groups of FDP since pyrophOSphate also inhibits
dimer association. It has been shown that two phOSphate
binding sites exist at each FDP binding site on the native
enzyme (see Literature Review); perhaps these binding sites
still persist in the pH 3.3 dimer but in a slightly differ-
ent conformation. When FDP binds to the pH 3.3 dimer, it
presumably freezes the molecule in the altered conformation

and prevents association.

105

.Umdpm was mahuzo Hopmw

zaopmaooaaa doves we: mam ommo hopped on» 2H .Amoaoaao nonov mam mo commons

on» ad no Amoaoaao oomoaov mam mo monomoaa on» ad momma map opsd dopsado

mam: whoaad m.m mm Hmpms mocooom 0H ponampno moapabapod pcommhnmh mmHOHao one
.omwaooam chaps: How Coapmaspmm opmameSm Hmaaoa on» paomoamma moamndm oaaom

one .mam an msoapaezoo hemmw amend Hemamboa Meade mo soapanaan .om mazwam

106

E a 2 a o e u I.
_ _ _ _ _ _ _ i
...... 1
....... ..., _./
....... x u
mu 4 ..... "299 u
.UZOU an". :93 E ><mm< ..o.... _
Y. Z_ ._<m~_m>mm ”.0 202.212.
H ..9.
V 0 I ...... G. ................... :6.
n ................ ...\\O
M ..... Aw .....
c. o - ..............
m .0 «vxxi
S
S-

 

 

 

m<¢>NZw m>_hdv-\OIIII1I\

1 - sixes muod n .u20o .

 

NZw

V".
O

O
o‘
- Ail/\IDV 'DEIdS

I
9°.
C

 

 

do". >m ._<m~_m>m~_ «mi—D "—0 20.2212.

107
IV. ALStudy Preliminary tg_the.Analysis for Active Dimers
g£_Aldolase: Attainment 2£_Aldolase Dimers Under

Conditions Where Native Tetramers are Active

This study is a general investigation of the physical
parameters which affect dimer reactivation and association
at pH values between pH 3.35, where the inactive, unfolded
dimer is formed (see Results in Chapter III), and pH 7.5,
where the native tetramer is stable. It was thought that
the 3.58 dimer was inactive at pH 2&2; because the low pH
used caused it to be unfolded; thus it seemed that at a
higher pH the dimer might remain or become folded and have
activity. There are two obvious approaches for trying to
produce dimers with activity at pH values between pH 3.35
and pH 7.5: (1) try to produce active dimers directly from
tetramers by dissociation without unfolding, by varying
factors which aid dissociation, such as low protein concen-
trations, temperature changes, and small changes in pH: or
(2) produce unfolded dimers at pH 3.35 and then find condi-
tions where the unfolded dimer folds, but does not associate,
as the pH in increased towards neutrality (Figure 22). The
former approach seemed more reasonable since an inactive
Species, the pH 3.35 dimer, would not have to be a transient
intermediate from active tetramers to active dimers, so it

was attempted first.

108
A. T332 323 Active My Production by Dissociation gt;
Tetramers

The aim of this eXperiment was to investigate the
first possibility above: dissociating tetramers into dimers
with catalytic activity at higher pH values than in previous
methods. It seemed reasonable that this could be accomplished
by lower enzyme concentrations and higher temperatures.
Sucrose density centrifugation was the method of choice to
analyze this problem, activity measurements being the only
method sensitive enough to detect protein at these low con-
centrations.

Aldolase crystals were centrifuged, dissolved in 0.2M
citrate (Na) buffer at pH 5.0, and diluted with this solu-
tion to the following protein concentrations: 2.0 mg/ml,
0.2 mg/ml and 0.04’mg/m1. Then 10 ul aliquots were taken
from each of these solutions and applied to the top of
separate sucrose gradient solutions in 5 m1 plastic centri-
fuge tubes. (For the procedure for sucrose density gradient
preparation, refer to Methods.) After centrifuging for 20
hours (up to #0 hours in other eXperiments to be described
later), the gradient tubes were removed, holes were punched
in the bottom of the tubes, and 10 drop fractions were col-
lected and assayed for aldolase activity.

The results from a typical sucrose density sedimenta-
tion velocity (SDSV) run are shown in Figure 21. The con-
centration chosen was the intermediate concentration, 0.2

mg/ml. The finding of a single, fairly symmetrical peak for

109

.pxop

mom .mHampoo Hopsoaaaoaxo noanSM Mom .aomama a ma cow: was omssmmoadeoo
opmgamonaIMIoeaSoonamomaw scans ca msodpaosoo awaaaam Moons mpgosaaoaxo
Hmnpo op somaawaaoo hp doaaaaopoc mm: mamasapop we Moon on» mo scapmodw
IapCooa one .HS\wa N.o use now pm ommaouam chapss mo shopped >QO m ma mane

.mamaad obapos cpaa o.m mm vs mamasapop mo sodpmaoommad you pmoa .HN ohfiwam

110

:23 33.252 3053 22: 3252.

 

Qm ON C.—

EEéE .382 «d .ooo

 

moo.

05.

m5.

ONO.

 

 

an 1..

._.< m¢w<<<~=mh ”.0 ZO_._.<_UOmm_D m0“— 5w»

(,_Iw-I_UIw-salow'rl) AilAliDV

111
all concentrations demonstrated that tetrameric aldolase was
the only active Species present. Since the presence of pro-
tein was analyzed by activity assay, only the active Species
would have been detected: inactive dimers or subunits would
have gone undetected. However, we have an indirect check for
inactive enzyme: the percentage recovery of active tetramers
can be calculated for each concentration. The percentage
recovery for the highest concentration, 2.0 mg/ml, was 97%,
to be compared with 65% for 0.2 mg/ml and 60% for 0.0“ mg/ml.
It is not known whether the loss of activity recovery is a
result of inactive tetramers or inactive dimers; the greater
loss at lower protein concentrations is consistent with the
latter.

If dimers had been formed at these low concentrations,
they could have been unfolded, and thus inactive because of
the low temperatures used. Reasoning that perhaps a higher
temperature might avoid any such unfolding of a dimer, this
experiment was therefore repeated as before, with the excep-
tion of increasing the temperature to about 20°C. Again, the
results showed only a single symmetrical peak present at the
tetramer position. In agreement with the previous results,
lower percentage recoveries were obtained at lower concentra-
tions: 61% at 2.0 mg/ml compared to 22% at 0.04 mg/ml.

Since the recoveries were lower in this eXperiment at com-
parable enzyme concentrations, higher temperatures must have
.caused this inactivation.. As before, it was not known if

dimers were formed; if they were formed, they were inactive.

112
Therefore, the second approach was investigated
(Figure 22). First dimers were produced at pH 3.35: these
inactive dimers were then placed under various conditions
which inhibited association but which presumably might allow

selective folding of the inactive dimer to an active dimer--

if such was possible.

B. Test for en Active Dimer 21 Studying the Effect 9;

Various Conditions 22 Selectively Folding (Without
Association) the RH 2.§5 Unfolded Dimer g§_Higher

p§_Values

Since in this section the procedures for many of these
studies are Similar in outline, differing only in some details
at the various steps, a general flow-sheet will be used in the
legend of each figure with the Specific differences emphasized
in the text.

In the following series of eXperiments dimers were pro-
duced under previously tested conditions (see Chapter II) at
pH 3.35. When the dimer samples had to be diluted further at
pH 3.35, the following buffer was used for the dilution step:
0.001M EDTA, 0.05M NaCl, and 0.1M citrate (Na), pH 3.35.

Most eXperiments entailed, in addition to the above dilution
step, diluting the dimer solution five-fold to a higher pH
with 0.2M citrate (Na) and subsequently testing the effect

of variables such as dimer concentration, temperature, sucrose
concentration, substrate concentration, and incubation time on

the dimer-tetramer distribution. The relative dimer-tetramer

 

113

.11.!{III I. I II .(I.x:. II

.hpabapos oathSpsO Meade
ao Soapmaoommm Meade you weapwop use osHmb aoswd: a on mm on» madmawa men»

use :.m mm vs Meade m wsaoSana mo sosoaaam esp macaw Hoooa mane .NN madwdm

 

114

no In

 

a

m>CU<

m m>.._.U<

m>:u<z_
mm In

m>:.U<Z_

 

m>:.U<

115
distribution was analyzed by sucrose density gradient sedi-
mentation velocity (SDSV) experiments (followed by activity
analysis of the collected fractions). It was necessary,
however, to positively first identify the dimer as a dimer,
in order to be confident of the results.

Rabbit muscle glyceraldehyde-B-phoSphate dehydrogen-
ase was used as a marker. In the aldolase test sample two
peaks were usually present: the leading peak was identified
as the tetramer species (7.98), (sedimenting the same dis-
tance as native enzyme), whereas the slower moving peak was
identified as the dimer Species with a sedimentation coef-
ficient between 3.55 and 4.38 which is equal to or higher
than the experimentally determined values of 3.958 to 3.808.
The sedimentation coefficients for the dimer were determined
previously (see Results in Chapter II) by conventional mov-
ing boundary sedimentation velocity eXperimentS in aqueous
solutions using the analytical model B ultracentrifuge.

Since pH seemed to be the most influential factor in
reactivation and association, the first experiments were
designed to determine the minimum pH necessary to prevent
dimer association but still exhibit activity reversal. The
first pH range to be tested was pH 4.0 to pH 5.5, since the
tetramer has been shown to be stable and remain as a tetramer

at pH 5.0 (see Results in previous section at 2.0 mg/ml).

116
1. Effect _c_>_f_‘_ pg 23 the Reassociation _o_f_‘_ Dimers £9 Tetramers

After a 11 Minute Incubation

The design of this eXperiment was as follows: dimers
were diluted five-fold to a final concentration of 0.11 mg/ml
with pH 4.0, 5.0 and pH 5.5 citrate (Na) buffers, incubated
for 15 minutes, and analyzed by SDSV in sucrose gradients
also at the reSpective pH values. Patterns from the activity
analysis for pH 4.0, 5.0 and pH 5.5 samples are shown in the
left, middle, and right graphs, reSpectively, of Figure 23.
There was no activity in the pH 4.0 sample (see sample assay
in left insert); in fact, at this pH, only at very much
higher concentrations was reactivation in the assay possible.

The initial activities of the pH 5.0 sample were almost
zero but Showed a gradually increasing activity with time in
the assay. Typical activity assay profiles are given by the
inserts. Because of the pronounced lag times at pH 5.0 which
probably represent association (see insert in middle graph of
Figure 23 for a typical assay), final activities were plotted
for the pH 5.0 sample.

In contrast to the pH u.0 and pH 5.0 samples, the pH
5.5 fractions immediately gave activities and these were con-
stant (little or no lag times were observed; see insert in
right graph, Figure 23). This observation was very exciting
at the time, since it seemed to provide overwhelming evidence
for the capability of dimers to be active: however, later
discoveries tempered this observation (see Results in Chapter

V). At any rate, this fact suggested that the pH 5.5 system

117

Assess pswamv Asdosw oaoodzv finesse peoav

 

m.m me so >mom o.m me so >QO 0.: me so >mom
wooden m.m me spas season o.m me spas seamen o.e me see:
oaocnm oopsado ofloaum.oopsaao oaoaum oopsaao
_ a L

Amm.m.mev

mmmzHQ QmBDAHQ

a

wooden posse sods eaoeuoa oopsaeo

Amm.m may
mmmzHo mmeaooae

.mHadme

Hogpo you pawn mom .ACOapeHSoHoo mo connos aom muoSpoz oomv hao>apooamoa .mm.m
use mw.m on op dopeHSono macs m.m ma use o.m mm pm Meade on» How mpfioaoammooo
coapdpsoaaoom one .msodpomam m.m ma non ooppoaa one moapdbapom Heapdsd .Hobozos
”0.: we use o.m me How Umppoad was moapa>apos assam .hdmms esp ad soapsHoommwoa
pom mpmop ovaboaa heap “madam Some do mpaomsa on» ad mHHmodpsaosom dopdomoaaoa one
we some pm mammmm madamm .maobdpooamoa .msasaw pswaa use .oauuda .pmoa Ga macaw
one m.m mg was .o.m mm .o.: ma non mdmhamao >mom 0:» Scan mpHSmoa one .ma some

pm sodpMDSonH opssaa ma a amped soapwdoommm needs no mm Mo poohmo one .mm ohfiwam

:23 3922...). $053 .20.... 32529
...N 3 3 m... ...N 3 3 ...... S 3 3 ......

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11.44 0. a ...W «lulu-lulu. _ u. IllinuI-‘_

.\ z. \ / .\ v

. . .

. / .i.

/ \ l c \ I. .1 No. A

5.25:2. . (. II...

.. . I , 5.25 I I z. .m

n e

5.25 m.

.z..z .z.s. .z.s. . s

I... .114 I 8 .

.... w
o .o .o I

a .0 a u.

md In . o.m In of In 18. .
:2er .3 .II 22: 20:332. .22.. m. m._

 

 

 

2525.5: 0. $3.25 .0 2926035..

119

would probably be the best to analyze, even though a small
amount of association did take place, as shown by the small
tetramer peak present. Thus in answer to our earlier quesa
tion: dimer association could be completely inhibited by
lower pH values, but reactivation without significant lag
times was not possible at pH values of 5.0 or lower at this
concentration--although apparently possible at pH 5.5.

But only one concentration (0.11 mg/ml) was investi-
gated in this eXperiment. Therefore, the effect of concen-
tration on the reactivation process was next investigated.
To ascertain then if reactivation could be accomplished at
pH 4.0, higher concentrations were investigated. These
eXperiments along with the pH 5.5 concentration dependence
eXperiments--which were more important since the pH 5.5
dimer gave immediate activity in the assay--were designed to
test the effect of dimer concentration on reactivation and

dimer association.

2. Concentration Dependence g£_Reassociation 22.2§.4'0
and pH 5.5

a. Effect‘gg Dimer Concentration gngeassociation‘gE

EH “.0 After a $2 Minute Incubation

Dimers were diluted five-fold into a pH 4.0 citrate
(Na) buffer solution to a final concentration of 0.05. 0.25,
and 1.30 mg/ml, incubated for 15 minutes at 0°C, and ana-
lyzed by SDSV. The two lower concentration samples were

completely inactive and showed no reassociation in the assay,

120
even after 15 minute incubation (see Figure 24). Samples at
the highest concentration, 1.30 mg/ml, were also inactive
initially; but in contrast to those at the lower concentra-
tions, some reactivation occurred during the incubation
period in the assay. Since reactivation was concentration
dependent, this indicated that reassociation had also taken
place. For obvious reasons, final activities were plotted
in Figure 2h. Two conclusions are suggested from this data:
(1) association into tetramers had to occur before catalytic
activity was obtained at pH 4.0, and therefore dimers were
inactive at pH #.0, and (2) pH “.0 dimers which had been
incubated for several minutes in the assay at pH 7.5 and
25°C at 0.03 mg/ml were inactive. The possibility that the
dimer concentration had some influence on a folding reaction
is not excluded, although it seems unlikely. Thus, these
results apparently contradict the evidence for active dimers

in the previous section.

b. Effect 2§,Dimer Concentration gn,Reassociation

3.2 21;! §.§ After 2.5 Hours

In contrast to the previous eXperiments where dimer
concentrations had to be increased to very high levels (1.30
mg/ml) to reactivate the dimers and activation took place
only in the pH 7.5 assay, for eXperiments at pH 5.5, dimer
concentration had to be decreased to avoid association. This
decrease in dimer concentration (0.12 mg/ml to 0.03 mg/ml)

was used to accomodate a study of the following problems:

121

.samaw pnwda. .soanm mascaa. .sgmnw puma.

Hs\wa om.a no >QO Ha\ma n~.o pm >mom Ha\ws mo.o no >QO
o.: ma .mpaapao 2m.o spas eaocum>.c empsaaa
_ _ _
.Ha\wa m.o .mm.n mg. .Ha\wa mm.H .mm.m mg. .Ha\wa mm.o .mm.m ma.
mmman omsaqu mmmzHa omaquo mmman amaaqu
empsaao pmacsn Roads spas “mucus “made spas
poz vacuum sopsaao caocumm smpsaao

_ r . .
.Ha\ma m.o .mm.m mo.

mmmzHQ mmdqoqu
.maampmo How vamp mom .samaw pnwda mg» ad omppoaa ohms moapabapom Hosam
.maap Spas Ha\wa om.a no oaaamm Hoado on» Scam :oapmbdpomon oaom was @903» monam
.ao>oson ”momma on» ma noapmpsona opscaa 0H m ampmo coho moagaom coapdapnmonoo amzoa
03» map :« mpdbapoo on was whose .ooo.o: no can Mao: om a Honda aoapawom moon can
up was amado a mo nmmaw unwaa on» ma moon can no noapdoamapcmoa one .hfloaapommmma
.msamaw psmda sow .maeeaa .pcoa 2. zzosm ma .aa\ma om.a sum .mm.o .mo.o. mnoapaap
asoonoo amaao obapooamoa on» mo 0.: mm pm >mmm an mamhamzd .noapGDSoqa opsnaa ma a

Hopmm 0.: mm pm :oapmaoommd Hmaao no Goapdhpnmonoo Hoado mo pommmm one .im madwah

122

.20. 3022.2 3053 20$ 324.55

 

 

 

 

5 3 3 ......

._<<\0<< n 0.0

 

=.~ .... 3 ...... 5 ....— 3 m...
.IIIJINJL/ . Von-111%]
. \
. .. n
. .
$25.} 1 1
.282 om. .282 mad

0... In I 205::szon E25 “.0 SEE

 

Ii.

8.

S.

8.

2..

 

 

$35235... Oh 335:0 m0 ZO:<.UOmm<wm

(._Iw-I_U!w-salowd) Ail/\IDV

123
(1) demonstrate that dimer association into tetramers, a
bimolecular reaction, was indeed concentration dependent,
and (2) try to find at lower protein concentration condi—
tions which exhibit complete inhibition of dimer associa-
tion.

Dimers were produced at pH 5.5 at three different
protein concentrations, 0.12, 0.06, and 0.03 mg/ml, incu-
bated at pH 5.5 at 4°C for 3.5 hours, and analyzed by SDSV
eXperiments (see Figure 25) at -6°C.

The results clearly demonstrated that dimer associ-
ation was concentration dependent and that 0.03 mg/ml was
probably the lowest concentration which could be used in
later eXperiments and still retain good sensitivity. More-
over, since essentially no association took place in 3.5
hours at this concentration, and since immediate activity
was present in the dimer peak, this strongly suggested that
pH 5.5 dimers were active. Nevertheless, the possibility
that the activity might be due to tetramer formation from

dimers during the incubation had to be considered.

3. Effect 22.1ncubation Time ggyAssociation at,Various

Concentrations at pH 5.5

This eXperiment was a part of Z-b above. In this
part of the experiment, after the 3.5 hour incubation
reported in 2-b, the pH 5.5 dimers were incubated for an
additional #8 hours. The effect of incubation for 3.5

hours or for #8 hours at the three dimer concentrations

mason m: a mason m.m mason m: a mHSos m.m mason m: 0 maaos m.m

.scmaw unwan. Asmaaw oa00aa. .snanw puma.
Ha\wa no.0 00 >000 Haxwa 00.0 as >mom aa\wa 00.0 00 >000
. 0 .

m.m m0 .omoaozm Rm 000 A020 opoaudo 2N.o
spa: 0Hocum HHH 0 .HH .H mmmzHo nmequn mmmaooad 00000.0

.mm.m m0. .mm.m 000 .mm.m mav
HHH mmmsz amaqum HH mmmzHo omaqum H mmmzHo omquHo
0Ho0uom umpsadm 0000-00 0mpsadm 0Hocum 0mpsaam
.... _ _ .
1..

 

.mm.m m0 .wmccsp 905.0 30.: smpsaau 000: mmmzHo mwaqongd

.mm.m mo.
mmmsz mmdqonqd
.mambapoommoa .m:.m
000 .m«.: .mm.m mam om oaswam mo msamhw pswda cam .oadcaa .pmma on» ma Amado
030 How mosam> omoSp 0:0 .haobapomamon .ma.: 0:0 .mm.m .mm.m 090 mm moafiwam
mo magmaw unwaa 0:0 .oaooaa .pmoa on» Ga 90800 on» you mpnmaoamwooo soapdp
Ismadoom one .m.m ma pm noapwpsoza mason o.m: dam mason m.m Hopmm macadapop op

mamado mo noaumaoommmma 0:» so scandapzoosoo Hoado mo poomum .cm 0:0 mm moaawam

125

.200 030.202 #0500 20.: 00250.0
.... ......

.2 m... 3 m...
A —

3 m.—
4

ed
.

r... 3 ......

 

J 4

._<<\02 9.0

 

 

la. a
.0
O. 0.0

LPN... $23.5;

./ ..

 

\ ......

11

\

\

.. . . w .. .

3 $25 Na

1.3. 1.8.
._<<\O<< 00.0

 

0<<\0<< m0.0

0.50: 0.0 I 0.0 :0: 2050500200 325 “.0 but:

 

 

$32.32.: O._. 335:0 “.0 ZO_h<_UOmm<m~_

3:.

=3.

2:.

=3.

(l_lw-I_U!w-sa|owr|) Ail/\IDV

126

323 mDUm_Zm<< mmOmUDm <<O~E muz<hm5

 

 

.2 ....— 3 ...... ...N 3 .... m... .2 3 0.. ......
a q A . so _ 2 .— _o\-.l _ o _ _ _
’0 o\0lo\o\o are 10/ \ / o\
q k. . .>. ..
. .. / \
. $25 1 . q 1 / \ .\. 4
$25 .. +
. <
$25
. \ 0.2/DE:
> 1 + 1 1
$23.5. E2<EE
.2\02 0.0 .202 00.0 .282 00.0 1

 

 

$501 w... l 0.0 In. I ZOC<~¢ZmUZOU E25 ”.0 HUmnEm

 

 

$32.32.: O._. 335:0 ...O ZO_h<_UOmm<m~_

2:.

3:.

m5.

2:.

(l_Iw-1_U!w-sa|owd) ALIAIDV

127

(0.12, 0.06, and 0.03 mg/ml) is depicted in Figures 25 and
26. .As expected, longer incubation times promoted increased
association-~even at 0.03 mg/ml (Figure 26) some association
took place, in contrast to the results for the 3.5 hour
incubation where, at 0.03 mg/ml (Figure 25), no association
took place.

But if the association is occurring continuously with
time, then why doesn't association take place during the 20
hour centrifugation period, and cause one broad peak? The
following analysis gives one possible explanation. If the
amount of tetramerization at 0.03 mg/ml, for a 20 hour
incubation period, is estimated by interpolation from the
48 hour incubation data, about 15% of the total dimers would
have associated. If this activity were then distributed in
the gradient position between the dimer and tetramer peaks--
such a distribution process would be eXpected to occur during
the 20 hour centrifugation period--the activity due to this
association would be too low to be detectable at any one
point in the gradient. Thus two sharp peaks would still
emerge and it would appear as if association had not
occurred during the centrifugation period. Thus, whether
this explanation is the correct one is not known; however,
it is consistent with the data.

Finally, the major conclusions from these eXperiments
as to the best combination of factors are the following:
(1) pH 5.5, (2) initial dimer concentrations of 0.03 mg/ml,

and (3) short incubation times (less than four hours),

128
eSpecially since enzyme from the dimer peak was active
immediately upon assaying. There was another variable, how-
ever, which might have been eXpected to influence the rate
of dimer association: that variable was sucrose. Sucrose
solutions are very viscous, eSpecially at low temperatures,
and could effectively inhibit the diffusion-limited bimo-

lecular reaction.

4. Effect‘gg Sucrose Concentration 9g.Dimer.Association
23.213322-

To test the effect of sucrose, dimers were prepared
in the usual manner (see legend of Figures 27 and 28 for
details) at pH 3.35 and diluted to pH 5.5. Dimers at pH
5.5 were then incubated with or without 4% sucrose for
various times at 4°C and 0.11 mg/ml and analyzed by SDSV.
The samples without sucrose were incubated for 10 minutes,
1 hour, and 3 hours (depicted in left, middle, and right
graphs, reSpectively, of Figure 27); whereas the samples
2122.2Z sucrose were incubated for 10 minutes, 3.5 hours,
and 16 hours (depicted in the left, middle, and right graphs,
reSpectively, of Figure 28). The conclusion from this eXperi-
ment was that sucrose does indeed inhibit dimer association
into tetramers. The results show that samples containing
4% sucrose had to be left for longer times to allow the
same amount of association found for the control samples
without sucrose (Figures 27 and 28). For example, the

dimer-tetramer distribution favored tetramers (40%-60%) in

 

129

.maampm0 you 0x00 000 .0000 soapsad0 m.m

m0 Hanan Gd psomoam was omonoSm Rm Mo on nonpam was» coaumooxo 0:» Spas Asamaw
pmoav mm 000 mm mmaswam you 0oms 0039 mm 080m 0:» was coapdammoam aoaa0 m.m

m0 you 09:0mooam A.so«pmcaaaopm0 unoaoammooo coapdpcoaa0mm H0800 03» How 0mm:
soapwsoo on» How m003poz oomv .haobapomamoa .mm.m 0cm .mo.: .mm.: and 0N oaswam
no mammaw unwaa 0:0 .ma00aa .000H 0:» ca 90800 on» you mucodoammooo Goapdpnma
Ia0om one .haobapomamma .mm.m 0:0 .mw.m .mm.: 0:0 mm madman no mammaw pswaa 0:0
.oH00aa .pmma 0:» ca 90800 on» pom mpsoaoammooo soapmpsoaa0mm 0:9 .mm madman no
.hambdpoogmoa .mnawaw unwaa 0nd .0H00aa .pmma 0:9 00 :Bosm ma mason 0H 0:0 .mHSos

 

 

m.m .mpznaa ca 900 codpGQSoSH Hopgm omohodm.&fl Spas muoapsnaapma0 Hoamapoplaoaa0

mo mammamnm >m0m .mm oaswdm go .mambapooamoa .mnawaw pnwaa 0:0 .0H00Ha .pmoa

 

 

SH QKOSm ma maso: m 000 .HSoQ H .mopszaa 0H How Soapwnsoca Hmpmm omoaoSm paogpaz

mcoapdpdapma0 amamapopnaoaa0 mo mamhamsm >m0m .moaap 000909000“ mfioaawb Hmpmw

m.m m0 pd codpwaoommw amaa0 no noapwapmmosoo 0moaozm mo poommm .wm 0:0 um mohswam

130

:28 mDUm_Zm<< mmOmUDm <<O~E mUZSm—o

 

 

 

 

...N ....— 3 0...
JAN/V). 40.1

....

E235: . 1

 

 

$25

$0.32: 0—

 

 

.: m. 0.. ...... 2 .0... 3 ......
...IIIT q . .N a)”. a.“ 0 od.
/0 0’0 \ 4 \ / o\
o\ / l \o o \
. .. .... C ..
0 ¢ + (o
/ K mw<<5 l mm<<<.~_.—m._. d. l
< $25
$240.50. .1 1
3501 m ~50: _
.2\02 ...0 1.0.0 :01 0000000 02

 

S.

3.

 

 

3.32.5.5» 00 33.2.0 “.0 ZO_._.<_UOmm<m~_

(l_Iw-I_U!w-sa|owd) AilAliDV

131

.20. 030.202 0.0053 20$ 00250.0
N

 

 

 

 

 

 

 

 

 

 

a 0.. .... ...... ...N .... .... ...... 310...... ......
q u .q a d, 11 W
W .>. .. ... <.\.. /. .. 1.).)
. :2 . .
. . . 1 /.\./. \. 1 $23.5: .
. ._ $25 $2<$:.<
1 . . .
. $25 .
. .
. 1 I .
. $25
$25.5:
9.30.1.0— mmDOI 0.0 mthZ_<< 0.
.232 :0 I 0.0 IQI 0000020 00..

 

 

3.524233. 0.— mmmszo ".0 ZO:<.UOmm<m~.

N...

2..

(,_lw-,_U!w-se|owd) Ail/\IDV

132
the sample without sucrose after a 3 hour incubation; whereas
in the sample‘with sucrose after a slightly longer time of
incubation (3.5 hours), the dimer-tetramer distribution still
favored dimers (60%-40%).

Although association was inhibited by sucrose, it was
eXpected that higher protein concentrations would overcome
the sucrose inhibition and yield rapid association. In addi-
tion to testing this possibility, it was also of interest to
test the ability of sucrose concentrations higher than 4% to
inhibit dimer association. Anticipating the effects to be
obtained, the time of incubation was changed, in the eXperi-
ments to follow (reSpectively), to obtain a similar dimer-
tetramer distribution in the higher dimer concentration and
in the higher sucrose concentration samples.

When pH 5.5 dimers were incubated for only 10 minutes
at 0.55 mg/ml (instead of 0.11 mg/ml), approximately 60% of
the area was found in the tetramer peak (Figure 29, left
graph); but in the previous eXperiment (Figure 28) conducted
at 0.11 mg/ml, it had required a much longer time (4.5 hours)
to obtain the same amount of tetramer. Both samples con-
tained 4% sucrose. When the 0.55 mg/ml sample (for details
of procedure, see legend of Figure 29) was allowed to react
for 4.5 hours in 4% sucrose, about 85% reversal was obtained
by activity analysis; almost complete association had taken
place by 17 hours (Figure 29, middle graph).

The third eXperiment in this series was to gain fur-

ther information concerning the strength of sucrose inhibi-

 

133

.mHaepeo nenpo you pxep eem .omad .mpnedoammeoo nodpepneaaoem mo nod»

IeHnoHeo non moonpez eem .hflebapomnmea .ma.: one .ma.: .mm.: one enmehw pnwaw
one .eHooaa .pmea on» n« neado en» pom maneaoammeoo noapepneaaoem one .pnemenm
me: emononm Rea pen» ma mnnenw puma one eHooHn on» nd emonp one nneaw unwaa en»
na oepoaaeo manaem on» neezpen eonenemmao hano esp one .Hn\ma Ha.o mo oeepwna
Ha\wa mm.o we: noapenpneonoo Rondo en» pen» pneoHe emoaonm mi pe pneaanenxe
mnoabonn on» na oemn penp me mnaenw eHooHa one puma enp nu oopoaneo meamaem

Mom eaem on» mes ennoooong one .m.m mm pe noapeaoomme Amado no ARQHV nod»

lenpneonoo emononm nonwan one noapenpneonoo naeponm Henwdn no poemmm .mm ehnwdm

134

 

:23 26an $053 EOE 37.255

 

 

 

 

 

lee In I n<<oz 3.0 I

=.~ m; =._ ...... =.~ m“.— 3 wt: 3 m..— =.— m...
J 4 _ a as. n . J . q u
,. / \ ...,.\_. x .\
/ . \. ./ x. .\
. . .. a $25 1 . .\
/< / .> .\
/ o l . _ .1 o\ +
4 + . E25
» 525 e P H m »
E2452 5 A <* 1 $25an
ea: 3. $25.5: 2:5 852.2 2
$053... 902 $068 oev

 

 

33.2.3.me Oh $3.25 “.0 20:<_U0mm<m~_

(,_IW-I_U!w-sa|owd) Ail/\IDV

135
tion of association. Dimer samples (0.55 mg/ml) at pH 5.5
were incubated for 4.5 hours in the presence of 16% sucrose
and analyzed by SDSV. The conclusion was that 16% sucrose
inhibited dimer association more than 4% sucrose.

The results (Figure 29, right graph) show that the
dimer-tetramer distribution favored dimers (40%-60%) after
the 4.5 hour incubation at 16% sucrose. In comparison, in
the previous eXperiment (Figure 29, left graph) at the same
dimer concentration but at 4% sucrose, the dimer association
rate was much faster: only 5 minutes were required to obtain
a similar distribution; after 4.5 hours, the reaction was
85% complete. Since the enzyme concentration was diluted
four-fold in this experiment, the activity scale for Figure
29, left graph, was consequently increased four-fold so that
all three graphs in Figure 29 could be compared. The reason
for the four-fold dilution of the enzyme was to accomodate a
dilution of the 16% sucrose solution to 4%, so that the
enzyme sample incubated in 16% sucrose could be layered on
the 5%-20% sucrose gradient without disturbing it. There are
two reasons why this method should not affect the results:
(1) the dilution was accomplished immediately before the SDSV
analysis, and (2) previous studies at 0.14 mg/ml had demon-
strated that a 5 minute incubation did not produce any asso-
ciation in 4% sucrose (see Figure 27, left graph).

Sucrose was also found to be able to inhibit non-
specific aggregation as well as the Specific association into

tetramers. Thus, when a sample of pH 3.35 dimers was diluted

136

into 0.2M imidazole buffer at pH 7.9 to a final enzyme con-
centration of 1.3 mg/ml with no sucrose, immediate aggrega-
tion occurred, as indicated by a white precipitate formed;
but when 20% sucrose was present, no precipitation was evi-
dent. Also the percent recoveries of activity increased
from 45% with no sucrose, to 70% with 20% sucrose.

In conclusion, these eXperiments show that sucrose
did indeed strongly inhibit association, as eXpected, but
the inhibition was still not complete. As stated in the
introduction of this chapter (IV), for a study of active
dimers, it was desirable to be able to use eXperimental con-
ditions (1) where a sample of dimers exhibited activity upon
assaying-~which has been achieved with the pH 5.5 dimer sys-
tem--but (2) where dimer association was completely inhibited.
Any additional inhibition of association would aid in achiev-
ing the second objective; for this reason, all reversal solu-
tions were made 4% in sucrose for all subsequent eXperiments.

One final variable which had not been investigated was
temperature. Since in previous eXperiments the reaction mix-
tures were incubated at 4°C and analyzed by SDSV at -6°C, it
was of interest to test incubation temperatures of 0°C and
-6°C. The results from these eXperiments could then be com-
pared with the 4°C eXperiments to see if the lower tempera-

tures would more effectively inhibit dimer association.

137

5. Effect 2£_Incubation at 0°C and -6°C for Various Times

922—20-0 _sL_m mlmgm

Since the inhibitory effect of lower temperatures
(0°C and -6°C vs. 4°C) on dimer association was of primary
interest in this experiment, and since association was
measured by rate of sedimentation in SDSV eXperiments, no
Special precautions for temperature regulation were neces-
sary once the dimer and tetramer peaks were separated, i.e.
after SDSV analysis. However, the following Special precau-
tions were needed and taken for that part of the eXperiment
preceding the SDSV separation, to make sure that the temper-
ature of the rotor, gradients, and enzyme solutions did not
rise above 0°C or -6°C: (1) Rotors were cooled to 0°C and
-6°C by incubation at -20°C for 12 minutes and 20 minutes,
reSpectively. (2) All gradients were cooled at 4°C for 3
hours, followed by (a) incubation of three gradients at —20°c
for 7 minutes to lower the temperature to 0°C, and by (b)
incubation of the other three gradients at ~20°C for 12
minutes to lower the temperature to -6°C. (3) The pH 5.5
dimer samples were maintained at 0°C and -6°C in salt-ice
baths. That these procedures accomplished their objective
was verified by using a thermocouple probe to measure the
temperature in an earlier trial run.

A sample of pH 3.35 dimers was prepared (see legend
of Figures 30 and 31 for details), diluted five-fold to pH
5.5 at 0°C and -6°C to a final dimer concentration of 0.03

mg/ml, incubated for 3.5, 14, and 24 hours, and analyzed by

138

.maaepeo Henunnm non use» en» op Homem A.mpneaodhheoo

nodpepneaaoem nenao en» mo noapeanoaeo enp Mom moonpez eemv .haebdpoemmea
.mu.m one .mm.m .mm.m one am meanwam nd menaep emonp one .haebapoeamen

.me.m one .mo.m .ma.m eye on enswnm we manage used“ one .eaooaa .pnea men

n« neado on» no mpneaoammeoo noapepneaaoem ena .Hm ehnwam nu nsonm one cow:
pe noapennona mo poemme enp meenenz .om ehnwam na ownedmeo one coo pe mnodp
Iennona man no poewmo ene .menap noapennond oepeodona enp we maeaeapep opna

noapeaoomme Rondo no oool one 000 pe noapepnona ho poewmm .Hm one on meanwam

139

223 33.52 3053 20$ 3245.5

 

 

=.~ 5 3 m... =.~ m; 3 m5 =.~ 3 3 ......
null/1 . ...?“ i N 13.1 e ... 1!. 1.4.... e
. six. I. \o /) x. / .\ or o\
a /.\. a ./ \ l .2
$2155: fl F $23.5: .... 1 \
$25 < (
$25 .. A
$25
PE em PE 3 2; md

 

 

" m2; ZO_.F<mDUZ_
Ma Sexes SS.

 

 

 

 

33549.5... Oh mmmcfio “.0 ZO_h<_UOmm<m~_

v
3
u
S... M
I.—
IA
as. .xn;
w
.0:
8
ms. 4
m.
u_
:8. .l
M
_

140

223 mDUmLZmE mmOmUDm 20¢“. mUZ<hmE

 

3 m... 3 2. 5 3 3 ...... 5 m... 3 a...
j .0). a Co. 1.1V???” :1 q dunner. o Mull
0/0\ /0 0 fl / O / O\
n / .\ ./ \ .z \ .
$25.5: .< 1 $23.5: 4. 1 ..... 1 8:
32.0 $25 52.0
1 1 1 as.
1 1 1 m8.
3; em 3;. 3 PE Wm
52: 20.2262. 1 as.
com... 1.5352.

 

 

 

 

 

 

 

mmm<<<mhmh 0... 332.0 “.0 ZO:<_UOmm<m~_

(l_Iw-._U!w-se|owd) Ail/\IDV

141
SDSV. The results (Figures 30 and 31) Show that only 3-6%
of the dimers had associated by 20 hours; thus almost com-
plete inhibition of association was found at both 0°C and
-6°C, using a protein concentration of 0.03 mg/ml.

Since lower temperatures strongly inhibited associa-
tion, an eXperiment was designed to test the effect of incu-
bation at higher temperatures (12°C) on dimer association
into tetramers. Although somewhat unlikely, it was hoped
that increased temperatures might aid dimer folding and thus

produce active dimers, without significant association.

6. Effect 23 Incubation g£_g Higher Temperature (12°C)

on Dimer Association

Dimers were diluted five-fold into the 4% sucrose
solutions at pH 5.5 and incubated for various times at 0°C
and 12°C at a final concentration of 0.05 mg/ml. The con-
trol was incubated for 3.5 hours at 0°C while the two
eXperimental samples were incubated for (a) 3.0 hours at
0°C, followed by 0.5 hours at 12°C (sample #1), or for (b)
10 minutes at 0°C, followed by 3.5 hours at 12°C (sample #2).
The control and eXperimental samples, 1 and 2, were then
analyzed by SDSV. The activity patterns obtained for the
control and experimental samples, 1 and 2, are depicted in
the left, middle, and right graphs, reSpectively, of Figure
32. The results demonstrate that increased temperature
(12°C vs. 0°C) stimulates the rate of dimer association into
tetramers and significantly, the stimulation is time-depen-
dent.

142

.pxep en» op Hemen

.madepeo ohoa Mom A.menHe> m enp no noapeanoaeo en» Hon moonpez oemv .maebop
noeneen .mm.m one .ma.m .mm.m ene mnnenw pnwnn one .eHoonn .pnea esp nn nendo
en» mo mpneaodmmeoo noapepneadoem .Aooma no ooov mennpenenaep oepeoaond 0:»
we manon m.m no end» Hepop e 909 m.m mm pe oepennona eyes wheaan .maeaehpep

opna noapeaoomme neado no “oomav eanenenaep Seaman e no poehwm .Nm onnwdm

143

23 33.52 3053 20$ $62525

 

 

 

A: m; 3 m... A: m. A... m... =.~ 3 A... a...
o A not} 41/ . on A q 1f. at“ _\&l
,. \1 A x . «A . xx ./ \
A .4 (AAA .\
A \ $25 1 < .< /. 1
A A . $23.5: <
1 $2<EE A A 1
/._
+ I. .1
$255:
.3 we den. .5 modes ...; no .er
.se 2 .600 .2 on .600 1

 

 

 

60m. 2 205382. 5 Belle

 

 

mmw<<<~2mh O._. 3325 ”.0 ZO:<_UOmm<m~_

E.

No.

8.

3.

(l_Iw-I_U!w-sa|owd) MIAILDV

144
Presumably, the dimers are already somewhat folded
and the increased temperature increases the collision fre-
quency of the diffusion-limited association reaction.
"Hydrophobic bonding" is also stronger'at 12°C than at 0°C

and thus may help stimulate dimer association.

7. Effect g£_Substrate gn.Dimer Association §§_p§,5.0

 

After a‘fl Hour Incubation

 

Even though dimers at pH 5.0 were not catalytically
active (see eXperimental results in Section B-i in this
chapter), the possibility existed that they might still
bind the substrate, i.e. FDP. (If the dimers did bind sub-
strate, the substrate might affect dimers in two ways: (1)
The substrate might stabilize a folded dimer structure;
this would allow dimer activity to be tested. (2) The sub-
strate might change the conformation of the dimers in such
a way as to stimulate dimer association.

In general, substrates, when bound to their native
enzymes frequently tighten their structures and stabilize
them against denaturation and dissociation. If the sub-
strate could bind to its (postulated) subpolymer binding
Site and induce a conformational change to a more folded
state, it conceivably could shift the subpolymer-polymer
equilibrium towards association. The other possibility is
that FDP could induce an active dimer formation. The sub-
strate could stabilize the dimer structure in a conforma-

tion which was catalytically active while the physical

145
forces such as pH, low temperatures, low protein concentra-
tion could be used to inhibit dimer association. Examples
of enzymes whose substrates stabilize them against unfold-
ing and denaturation are phOSphofructokinase, pyruvate
kinase, and carboxypeptidase.

The purpose of the following eXperiment was to test
whether such an effect of substrate might occur with rabbit
muscle aldolase dimers at pH 5.0. It was also of interest
to see whether the substrate might have effect on the
aldolase dimers when FDP was present in the sucrose gradient
solutions during centrifugation.

Dimers were prepared, diluted 5-fold into pH 5.0
citrate buffer to a final concentration of 0.08 mg/ml, incu-
bated for 4 hours in the presence and absence of 0.05M FDP.
and analyzed by SDSV. The activity pattern for the test
sample which was both incubated in FDP and run in a gradient
containing FDP is shown in the left graph of Figure 33. The
activity pattern for this test sample Should be compared to
that for the control, which was incubated without FDP and run
in an FDP-free gradient (see right graph of Figure 33 for the
control pattern). For the effect of incubating with FDP and
running without FDP in the gradient, refer to the middle
graph in this figure.

A comparison of the test sample and control sample
Shows that FDP did not stimulate association or dimer reac-
tivation as originally postulated; in fact if anything, FDP

inhibited association. Other SDSV SXperiments Showed that

146

.maaepeo pom pNep

op nemem ,A.menHe> m en» mo noapeanoaeo en» now moonpez eemA .mm.m one .mm.m
.mm.m one ennenm pnwnn one .eHooHn .pnea enp nn nenao enp no epneaodnneoo node
lepneaaoem .nnehw pmeH on» n« oepoameo pneaaheawe en» mo pneaoenw 0:» nd ano
pnemeam we: mam .nmehw pnwda en» na manaem en» no pnemenm mez mam on meenenz
.noopepnond anon : en» wnanno eHQaem eahnno npdz pnemenm mes mam .mnmenw
eaoonn one pnea en» nn oepOdneo mannee en» nH .o.m mn pe nodpennend noon

3 e Hepme wheaenpep opna noapeaoomme heado no epenpmpnm mo poemum .mm ehnwdm

147

23 36252 $053 20$ 32525
A3 3 A... 3A A: 3 A: ....AA A: ....A A... 3A

 

 

’ A .4 H? q .../H \o—. J-[ A it. a a o
u/ 03/ eoszeo o ’03.}.
< . A A. . A .A.
A /. .\ 1 $25. / \ 1 ./ .\.1 1s.
$2<$E. ... m: .1. $25.5 .<.
< A A
1 » $25 1 18.
$25 A , $25
1 1A 1 8.
#5536 5.. l1 555.05.. I 553.0 5.. 28.
e2>Nze 5.. II e2>~zem51 28. e2>Nze 1.5.. 28.
15.

 

 

 

whamhmmnm It; ZOC<mDUZA .mI V 4. m0 hUwHIm

 

$352233 0.. $3520 “.0 20:<_U0mm<m~_

(l_Iw-I_U!w-sa|owd) Ail/\IDV

148
pyrophoSphate also inhibited dimer association (these eXperi-
ments will not be reported) and suggested that the FDP inhi-
bition was probably through the phosphate binding site on the
dimer which remains partially intact after dissociation.
These results are consistent with the FDP inhibition of dimer

reactivation and dimer association found in Chapter III-G.

149
V. Analysis for Active Dimers

In the previous chapter it was shown that aldolase.
dimers could be produced at pH 5.5 and, 2£Eg£.§2§l'gnglygi§,
exhibited immediate activity upon assaying the dimer peak.
Significantly, the activity assays from the SDSV fractions
were linear with time, showing no lag times and thus, there
apparently was no reactivation in the assay. In this chap-
ter we exPlore further the question of whether the dimers
are catalytically active. This involves detailed analysis
of the kinetics of the reversal process of pH 5.5 dimers,
got subjected 19.9. §p§1 analysis, (Section B--first part) and
further eXperiments related to the possibility that dimers
might be inactive--with the "apparent" dimer activity being
due to formation of heretofore undetected active tetramers
at some point in the analysis. In particular, we analyzed
the two possibilities that the apparent dimer activity might
be due to tetramer formation either (1) during SDSV analysis
or immediately after SDSV analysis and before the assay pro-
cedure (Section A), or (2) in the assay (Section B--second
part). The two possibilities were both tested directly by
SDSV analysis for tetramers in the products in the two cases--
in the first case, by SDSV "rerun" of the dimer fraction
from the initial SDSV, and in the second case, by SDSV
analysis of samples taken directly from the assay.

In Section B (first part), kinetics of activity rever-
sal from pH 5.5 dimers (0°C), not subjected to SDSV analysis,

were analyzed to see if this technique would answer the

150
question of whether dimers were active or not. The pH 5.5
dimers, not subjected to SDSV analysis, were not active
initially in contrast to the pH 5.5 dimers which were sub-
jected to SDSV separation. Therefore, the immediate ques-
tion was whether the reactivation of the inactive (initially)
pH 5.5 dimer was first-order or second-order with respect to
dimer concentration. A first-order reversal process from pH
5.5 dimers would be consistent with active dimers (although
other possibilities exist, see Section B of this chapter);
whereas a second-order process would require a dimer associ-
ation step before activity reversal and would therefore, be
inconsistent with active dimers.

But as will be shown, the results from these two
sections were quite perplexingly contradictory. These des—
criptive words understate the dilemma of this complicated
research problem. Fortunately, Section C resolved this
"apparent" dilemma and provides a unifying touch to this
chapter--and for that matter, to chapter III as well. How-
ever, as will be shown, there still remains a discrepancy
between the results with pH 5.5 dimers which have been sub-
jected to SDSV analysis and those which have not; the former

seem to be active but the latter do not.

151
A. Test for Dimer Association After the SDSV Analysis

but Before the Assay Procedure

0
1. Effect 21; goat—SDSV Incubation £2.12 -6°c, 4°C, and 12 0
9g Activit : Ag Indirect Test

This eXperiment was designed to test for activation--
and thus association (assuming that this correlation can be
drawn)--after the SDSV run. It was thought that the elution
temperature of 4°C might allow dimer association after the
run and might be reSponsible for the immediate activity
found in the dimer peak: thus it seemed that going to lower
temperatures would avoid this problem if it were occurring.
We thus decided to attempt to conduct an eXperiment at -6°C,
taking Special care to insure that at no time did the tem-
perature of the solutions rise above this. Thus Special
precautions were used in this procedure to insure that at
all stages of the experiment, the solutions were kept at
-6°C. Procedures outlined before for the previous -6°C
experiment (see Section B-5 of Chapter IV) were used again
here for the first part of the experiment through the SDSV
run. However, an additional feature in this experiment was
the maintenance of a temperature of -6°C for the gradient
tubes and fractions collected following the SDSV run. Cool-
ing coils with circulating ethylene glycol-water solvent
were used to maintain the plastic centrifuge gradient tubes
at -6°C during the drop collection procedure, and fractions

were collected in glass tubes equilibrated at -6°C using a

152
salt-ice bath.

A sample of pH 3.35 dimers was diluted five-fold to
pH 5.5 to a final concentration of 0.04 mg/ml. incubated for
3.5 hours at -6°C, and aliquots applied to three gradients
for subsequent SDSV analysis.

After SDSV runs, the control gradient was eluted at
-6°C and assayed; its activity pattern is represented in the
left graph of Figure 34. Another gradient was incubated at
4°C for 30 minutes, eluted at 4°C and assayed immediately.
The activity pattern obtained is represented in the middle
graph of Figure 34. This sample was then assayed again
after two incubations, an additional 4.5 hour incubation at
4°C, followed by a 0.5 hour incubation at 12°C: the result-
ing activity pattern is represented in the right graph of
Figure 34. The results Show that little or no reactivation
occurred upon incubation at higher temperatures (of. 4°C
and 12°C to -6°C); but there was, however, some preferential
reactivation (two-fold) of the tetramer peak with incubation
at the higher temperature. This data suggested that Since
there was no increase in activity in the dimer peak with
time of incubation, no association of dimers had occurred
after the SDSV analysis; however, the possible complication
of simultaneous activation and dimer inactivation at the dif-
ferent incubation conditions due to instability cannot be
excluded. Moreover, the validity of our analysis depends
upon the assumption that the material in the dimer peak is

dimers alone (not active tetramers) in the ~6°C control.

153

.mHaepmo Hon pxmp mew .nodpebdpoema Amado

no mmnnpenmnamp ompeoaona mnp pe noapennona >mamlpmom mo pommmm .nm madman

.ll‘llllnll iii

154

.28 33.52 3053 20$ 3245.5

 

 

 

 

=.~ As... A... 2 A: A..— 3 3A 3 3 3 2A.
1|1|J.o(/ a A A o o Anna-mo 1..
.<.. A .<. ..A. .A. ..
. A A. 1 A . A. 1 /./1\
A $23.5: A
$235: A . .
A . 1 .A A. 1 /./.A .A
. A
A _ .AA
.. .. .... 1
$25 A A
.558 ooNA $25 $25
.....o .....uv .528 don ....50. 0%!
”20:45:62. >QO 50.. ..o 53$ 1

 

 

 

 

 

 

 

l

 

 

 

mmm<<<mhmh 0... 332.0 “.0 ZO:<_UOmm<m~_

3:.

18:.

«S.

as.

(,_Iw-IAU1w-sa|owd) Ail/\liDV

155

Unless unequivocal proof of the correctness of this assump-
tion can be provided, a shadow of doubt will be cast on this
conclusion. This direct proof will be presented in the next

section.
2. Direct Test for Possible Tetramers‘gn the Dimer Peak

To test for the presence of tetramers in the dimer
peak from SDSV analysis, an aliquot from the dimer peak frac-
tion was "rerun" in another SDSV eXperiment. The Special
precautions used in the previous section were again used here
to insure temperature regulation at -6°C.

To accomplish this "double" SDSV eXperiment, the
initial enzyme concentrations in the first SDSV eXperiment
had to be higher, 0.6 mg/ml, so that the second SDSV eXperi-
ment would yield enzyme concentration high enough for assay
detection. In addition, the first SDSV centrifugation time
was extended to 40 hours to insure complete separation of
the dimer and tetramer peak; however, the "rerun" experi-
ments were centrifuged for the normal 24 hour period.

The "original" peaks from the first sucrose density
eXperiment are shown in the left graph of Figure 35. .Ali-
quots from these "original" dimer and tetramer peak frac-
tions were applied to two new sucrose gradients and run
again. The results from the "dimer rerun" and "tetramer
rerun" experiments are shown in the middle and right graphs,
reSpectively, of Figure 35.

As Shown in the middle graph, there is no tetramer

156

.maoepmo Mom pxmp mmm A.mmnHe> m mo noapeflnoaeo How moonpmz mmmA .hambop

nomnmmn .mw.m one ma.: mne mnneam mHooan one nmma mnp no enema nmnao mnp mo mpnmao

sawmmoo noopepnmaaomw one .noapenenme >mmm :Henawanoz mnp nmpme mpnmadsmnxm >mnm

snnnmn Hmaenpmps one :nnnma nmaoos an Memn hmaao no hmaeapmp you pmme .mm mHanm

157

:23 mDUm_Zm<< mmOmUDm <<O~E mUZ<hmAO
A...

A: m..— AA.A 3A
11.11;]
A

.AA. 1
AA. 1

525::

2:5: 525::

 

AS...

A: A...A

3a.

 

A:

 

 

 

.A .
A .. A. A . A ,
5A.... 1Am8.A «AMA... 1
523:3
2:5: 523 8.5.— ._<z_u_~3 11

 

 

 

 

A: A...—
A

A: 3..
— 1

. \

1.

 

 

¥<mm «mszo Z. ¢w<<<~=mh 10". ...mm»

3.

S.

E.

8.

(AIw-Auw-salowd) Ail/\IDV

158
contamination in the dimer peak; thus dimers remain as dimers
after SDSV analysis and do not associate before the assay
procedure. If the activity in the dimer peak had been due
to tetramer contamination, a tetramer peak similar to the
activity pattern of the "tetramer rerun" eXperiment would
have been found in the "dimer rerun" SXperiment. This then
shows that the sensitivity of the assay would have been
sufficient to detect tetramers, had they been present.

Other eXperiment, identical in procedure with the
exception that the dimer concentration was reduced to % and
i the concentration used in this SXperiment, produced results
identical to those presented above.

The conclusion then from this section is that the
activity found in the dimer peak is not due to tetramer con-
tamination formed before the assay procedure. Thus the

linear activity rates from the dimer peak after SDSV must
'be due to either (1) active dimers or (2) association in the
assay. The following section deals with the eXperiments

which were designed to test for association in the assay.
B. Test for Dimer Association in the Assay

Since most ofnthe following eXperiments involve a
kinetic analysis, it is best to discuss the various mechan-
isms before presenting the eXperimental details. Listed
below are the four most likely reaction mechanisms followed
by the appropriate exPlanations. The open symbols represent

subunits of an inactive enzyme and the closed symbols repre-

159

sent subunits of an active enzyme. The k's are the reaction

lst

rate constants, a k st symbol being used for a order

1
reaction and kznd for a second-order reaction.

 

 

 

 

 

 

 

 

klst
1' 00 > O.
k nd
II. CI) 2 > a
k st k nd
III. CD 1 > .0 2 _> “
kznd klst ' '
co . 88 a» a

The first mechanism postulates that dimers are active
and that a folding process is necessary to regain activity:
it predicts that the reaction rate for producing active
enzyme (filled circles) from inactive enzyme (open circles)
depends upon the dimer concentration to the first power. In
addition, the half-time Should be independent of concentra-
tion, as expected for a first-order process, such as a fold-
ing reaction.

The second mechanism postulates that dimers are
inactive, so that association is necessary to regain activ-
ity. The rate of association in such a process Should be
concentration-dependent and therefore, the rate of activity
recovery should be concentration-dependent, and in particular,
it should be proportional to C2. Thus the kinetics of
activity recovery would be expected to be second-order in

this mechanism. In addition, the half-time should depend

160
concentration to the first power.

The third mechanism is identical to mechanism I, with
the additional feature that active dimers associate into
active tetramers. A kinetic analysis would not distinguish
between these two mechanisms (I and III); however, a physi-
cal analysis of the dimer-tetramer distribution in the assay
would make this distinction (see Section B-2 below).

The fourth mechanism postulates that dimers are inac-
tive and must associate into inactive tetramers before a
final folding reaction produces active tetramers. The rate
of association in such a process should be concentration-
dependent, but the rate of activity reversal should be inde-
pendent of concentration. Since the overall rate of activ-
ity recovery will be measured, the kinetics will reflect the
one step in the reaction which is rate-limiting. Thus at
higher concentration, the kinetics of activity recovery might
be SXpected to be first-order; but at very low concentrations,
where the association reaction must become rate-limiting,
the kinetics of activity recovery would be eXpected to be
second—order. In particular, the initial rate of recovery
Should be proportional to CE. Likewise, the half-times
should be independent of concentration at high concentra- .
tions; whereas at low concentrations, they should depend
upon concentration to the first power.

Thus we have three tests which should help us to dis-
tinguish between the different possibilities. If the unfolded

pH 5.5 dimers can be active with only a folding process

161
required, then the kinetics of activity recovery should be
first-order and the half-times Should be independent of
initial concentration of unfolded dimers (mechanisms I and
III). 0n the other hand, if the unfolded pH 5.5 dimers
must associate to be active, then the kinetics of activity
recovery should be second-order and the half-times should
depend on the concentration to the first power (mechanism
II, and over a wider concentration range, mechanism IV).
As stated previously, a distinction between I and III can-
not be made by a kinetic analysis; only a physical analy-
sis of the dimer-tetramer composition in the assay will
distinguish between these two possibilities.

The first eXperiment in this section was a kinetic
analysis of activity recovery; this experiment Should dis-
tinguish between mechanisms I (III) and mechanism II. The
second eXperiment was designed to distinguish between mech-

anism I and mechanism III by a physical analysis.

1. Kinetics 2; Activity Reversal from.p§_§.§ Dimers (0°C)
Not Subjected 22_SDSV Analysis: Determination g£_the

Rate Constant and the Order 22 Activity Recovery from

2E 5.5 Dimers at 25°C and pg z.§

In contrast to all the previous eXperiments in
Chapters IV and V, pH 5.5 dimers were not separated by SDSV
before the activity analysis. Instead, the following pro-
cedure was used: Various concentrations of pH 3.35 dimers

were diluted five-fold in the usual manner to pH 5.5 at 0°C.

162
Using chilled micropipettes, aliquots were immediately
removed from the pH 5.5 stock solutions (before any associa-
tion could take place, i.e. 0-10 seconds) and diluted into
the pH 7.5 assay at 25°C (mixing immediately): the cuvettes
were then incubated for various times before adding substrate
to determine catalytic activity. The concentration range
used in the assay in this study was between 0.0032 mg/ml and
0.032 mg/ml.

Before turning to the results, the problem of inacti-
vation should be discussed. A significant amount of inacti-
vation was found after the initial reactivation period of 5
to 10 minutes. Because of this inactivation, final activ-
ities (after 30-60 minutes) were unreliable and therefore,
half-times could not be determined at each concentration.

It was evident that the problem was more acute at lower con-
centrations. However, since the inactivation did not affect
the initial reactivation rates, the other method of determin-
ing reactivation order could be studied. This method, to be
described below, is the kinetic analysis of the dependence

of the initial rates of reactivation upon initial dimer con-
centrations.

Specific activities were determined from the initial
optical densities changes. Reactivation curves were then
determined by plotting specific activity changes as a func-
tion of time for each concentration studied. Initial rates
of reactivation (dC/dT)t=o were then plotted against initial

dimer concentrations (CO) raised to the first or second

163

power for first-order and second-order kinetic reactions,
reSpectively. These data are shown in Figures 36 and 37,
reSpectively.

When (dC/dT)t=o was plotted against CE for a second-
order reaction, a curved line was obtained2 (Figure 37).
On the other hand, when (dC/dT)t=0 was plotted against CD
for a first-order reaction, a straight line was obtained

(Figure 36). From the slope of this line, a first—order

rate constant was calculated to be klst A 2.4 x 10'3/sec.
From these results, it follows that in the concentration
range studied the rate-limiting step is a first-order
reaction. Therefore, the production of an active enzyme
from pH 5.5 dimers is a folding reaction, and not an asso-
ciation reaction. Mechanism II (see introduction to this
section) must be excluded since the rate-limiting step in
this mechanism is a second-order reaction. The other
three mechanisms (I. III, and IV) are consistent with
these results. Mechanisms I and IV are included in Figure
36 (as both mechanisms, I and II. preclude an active dimer,
only one of them was included in the graph).

Since the first-order rate constant found above does

not distinguish between mechanism I in which only dimers

 

The "pseudo" second-order rate constant at low con-
centration, i.e. for the linear portion of this curvilinear
line, was calculated to be kznd = 6.4 x 10LP liters/mole-
sec at 25°C and pH 7.5. This can be compared with the
second-order rate constant fialculated in Chapter III,
Section F: kznd = 1.4 x 10 liters/mole-sec at 16°C and
pH 7.5.

 

 

164

.3233 son use» eem .Aoomm one mg. mg.

m
emme mnp nd Hemnmbma amndo m.m we no moapmnax Hmonoipmnam you pmma .mm mhnwom

165

Om

 

MWo. x 70. 55502 $25

  

.05 >72 x on n c. .I. Eon /v

.I :v

oun .o 0.. I w...

“iglmwflm .u .
aqm ..

 

2

o l

 

 

l
U)
('Das-aailvsanom 0101 x °=‘(iP/DP)

8.52.0. $9.0 ... «on. E:

166

.eanepeo non use» eem .Aoomm one m.a me.

memme mnp no Hemnmbmn amaao m.m mg no moopmnax Hmonolonoomm Hon pema

.mm manwam

 

167

03

 

voA x NAouA £5502 $25
8.. oo. 8 oo oe on
_

.Ummlwaozxmmg: #9 x 10 H v. N meAm 35:23 .IY

 

V

..1 Isl
cutout... o

l

l
N

I
or)

 

 

mUZhZCA «momOpcu «0”. 5w.—

(‘335-213111/5310w) mm x°='(1P/DP)

168
are present and mechanisms I and IV in which both dimers and
tetramers are present. the eXperiments in the next section
were designed to determine whether tetramer formation accom-
panied the reactivation by directly measuring the dimer-
tetramer distribution in the assay after various times of

incubation.

2. Direct Test for Possible Dimer.Association in the

Assay: SDSV Analysis After incubation 22 Dimers

in the Assay

The purpose of this eXperiment was to test for pos-
sible-dimer association in the assay after various periods
of incubation. To accomplish this, it was necessary to
determine the relative dimer-tetramer distributions in the
assay by the SDSV method of analysis.

Dimers at pH 5.5 were diluted into assay mixtures,
complete except for FDP, at 0.016 mg/ml, 25°C and pH 7.5.
and incubated for 0, 1, and 5 minutes, respectively, before
beginning the assay by addition of FDP. The initial activ-
ities in the insert of Figure 38 show a reactivation due to
incubation in the assay. To determine the dimer-tetramer
distribution in the assay at the various incubation times,
0.1 m1 aliquots were removed within about 15 seconds from
the reSpective assay samples after activity analysis, cooled
to 0°C to "freeze" the dimer-tetramer distribution, and
layered on a pH 5.5 sucrose gradient. Subsequent SDSV

analysis for the assay incubation times of 0, 1. and 5

169

 

 

.madmpoo How pump mom .ammmm on» as mmaap ompmoadsd map wo some

new soapmbdpommn mo undoam on» mSoSm pammnd one .hdwmd an» op whoadc m.m ma

madcom scams mead» muoaamb pm hammm 2d soapdmomaoo ho mdmhamsw >QO

.mm eeswdm

 

170

:23 33.sz $053. EOE 324:5

=.~ m..— 3 3.

...: 2 ...: 2 2 .3 3 3
A

 

 

 

 

 

O u 4 A CHOII
/0 NM \ on V
.o \ o o/ro\0190/o. + o o D
i. l _ m . u
. 1 1 325:: 1 S: M
/./. :23 1
$23 A
o .z=z . )
_ 1 522:: . 1 n g. 1 :5 .m
l i K m.
a 52:. 1 1 K Nm 18a. 4.
52:9 0K m. m.
L m A U_
L25. ...
2:2 n 2.2 P 20.23%. 02 W.

 

 

 

 

 

 

 

 

 

><mm< 2. 9.52.0

mo 20:532.. fir? m_m:<z< .._m> .omm

171
minutes are shown in the left, middle, and right graphs,
reSpectively. of Figure 38. Since the amount of tetramer
in the 5 minute sample is greater than that in the 1 minute
sample, these results show that dimer association does take
place in the assay, and that the amount of association is
proportional to the incubation time.

The question which now arises is, "Why does the
amount of activity in the dimer peak also increase with time
in the assay?" This raised the possibility that the dimers
might become catalytically active, as a result of folding of
dimers in the assay, i.e. mechanism III. However, an alter-
nate possibility was that the apparent activation associated
with the dimer peak was an artifact caused by (1) using
assay micropipettes which were not chilled or by (2) the
method of analyzing the test samples. That is, the 5 minute
sample was analyzed first, followed by the 2 and 1 minute
samples, respectively. During the interval between assays,
the material might have become inactivated. This would have
caused a reduction of peak height in the left graph over that
of the right graph. Further support for this "inactivation"
explanation was furnished by two other eXperiments where
special precautions were taken to insure that the assay
micropipettes were chilled (-20°C) and where the order of
assaying the samples was reversed, i.e. tubes one, two, and
then three. ~Under these circumstances, there was no increase-
in the activity in the dimer peak, in direct contradiction

to the previous results. However, the activity in the tetramer

172
peak did increase as before. Thus the apparent increase in
the activity in the dimer peak as shown in Figure 38 appears
not to be real, but rather due to inactivation and the order
of assaying the samples.

These results are consistent with mechanism IV in
which the only active Species is tetramers, but the data do
not unequivocally exclude mechanism III in which both active
dimers and active tetramers are formed. That is, both asso-
ciation and folding could be occurring, each of which could
yield activity--Figure 38 supports this contention but the
inactivation explanation contradicts this. Therefore both
mechanisms will be retained for further examination. On the
other hand, since we have indeed shown that an active
tetramer can form in the assay, mechanism I_i§ excluded, as
this mechanism does not allow the formation of an active
tetramer in the assay. Mechanism,I;_hg§ already been excluded
in previous section.

In conclusion then, only two mechanisms remain: (a)
mechanism III, which postulates that the formation of an
active dimer occurs first, followed by the formation of an
active tetramer, and (b) mechanism IV, which postulates that
the formation of an inactive tetramer occurs first, followed
by the formation of an active tetramer (no active dimer
intermediate). Therefore, final proof of mechanism III or
IV rests on diSproving one of these two mechanisms. This
again will have to be done by a kinetic analysis, and this

can most conveniently be done directly in the assay as in the

 

 

 

 

173
previous kinetic analysis.

To exclude the active dimer mechanism (III), it will
have to be shown (a) that the concentration-dependence of
the activity recovery changes from first-order (found in the
previous section) to second-order (or at least, becomes
greater than first-order) or (b) that the half-time of the
activity recovery vary with dimer concentration. But since
further quantitative analysis for a change in half-time at
still lower concentrations would be clouded by the inacti-
vation process under the present assay conditions, a study
was undertaken with the eXpress intent of eliminating the
inactivation problem. The successful completion of this
study then made it possible to conduct an analysis of half-

times at various initial dimer concentrations.
C. Proof that Dimers are not Catalytically Active

1. Stimulation of Reactivation in.the Assay with Bovine
Serum Albumin andgor figMercaptoethanol to Aid Later

Kinetic Analysis at Lower Concentrations

The minimum requirements fora.meaningfu1 kinetic
analysis of reactivation from inactive dimers in the assay
are (1) no association for at least 5 minutes at pH 5.5 and
(2) no inactivation in the assay for at least 30 minutes.
To eliminate any association at pH 5.5 before the kinetic
analysis, the temperature was controlled to -6°C. In line
with the second requirement above, the eXperiments in this

section were designed to stabilize the dimers against the

174
inactivation.

A sample of pH 3.35 dimers was diluted five-fold to
pH 5.5 and incubated in a -600 salt-ice bath. Within 1
minute, 10 pl aliquots were removed with chilled micro-
pipettes (-20°C) and quickly added to the assay cuvettes
with mixing. After a 30 minute incubation at 25°C in the
assay cuvette, which contained the appropriate concentra-
tions of BSA and/or 0.1M B-mercaptoethanol, FDP was added
to determine the activity. The final dimer concentration
in the assay was either 0.025 mg/ml or 0.001 mg/ml. Four
different concentrations of BSA were tested: 0.1, 0.5, 1.0,
and 5.0 mg/ml.

The results shown in Figure 39 demonstrated that
although BSA was effective alone in stimulating reversal at
both enzyme concentrations, 0.1M B-mercaptoethanol was far
superior in its ability to stabilize the dimers in the
reversal process against inactivation. Moreover, when
B-mercaptoethanol was present, the BSA effect was less
pronounced. The combination of both B-mercaptoethanol and
BSA, however, did produce greater stabilization than either
one alone. Therefore, in the following kinetic analysis

both were used in the assay.

175

 

.maampcu
Mom use» mom .hmmmm can ca macapmapsoosoo Hmaao assay on» ones smmhw on»
mo macshoo songs can :a :303m Aaa\m1 H one Ha\m1 mmv ascendapScosoo swede

one .QodmeSOCa mpssaa on a Hopmm momma an» :« ammacbcn Meade no Ammzimv

aosm£pmopmmohoalm SH.o use A¢mmv :Hasnam aspen osabon mo poowmm .mm madman

 

3283 2.222 22mm, mz_>om
S no ..o 0.0 es 0.. no

:52. 52<1$OE<0$2$ I y/Z
loz<rmoE<ufizlm 25 see

><mm< Z_ ._<m~_m>mm 52.0 20 Im<<1m 024. 4.2 ".0 HUMEHE

_E\ml mm

 

IIIAII“\\_I —\III"\'LI(‘

177
2. Kinetic Analysis gg'Beversal ig.§hg_駧§y_gygy.150-Fold
21233 Concentration @2252 (0.00016 mgéml_§g_9;9§j_mggml):
Demonstratigg The; M Association i332 Tetramers,

Followed'by a Tetramer Conformational Change,
Precedes Catalytic Activity

Now that we had obtained conditions in which the pH
5.5 dimer was stable in the assay, it was possible to under-
take quantitative experiments on the kinetics of the activ-
ity recovery upon reversal.

The same procedure was used in this eXperiment as in
the previous eXperiment except a third enzyme concentration
(0.00016 mg/ml) was added, and dimers were incubated for
various times in the assay instead of the constant 30 minute
incubation period used before. Both 0.1M B-mercaptoethanol
and BSA (1 mg/ml) were present in the assay.

The kinetics in Figure 40 show that the half-times
increase from 2.5 minutes to 5 minutes when the dimer con-
centration is lowered from 0.001 mg/ml to 0.00016 mg/ml.
Furthermore, when the eXperiment was repeated with only 0.1M
B-mercaptoethanol in the assay, essentially the same results
were obtained.

Therefore Mechanism.£l;, which postulates an active
dimer, _i_s_ 112.2 consistent with this data. The correct
mechanism (IV) and the only remaining mechanism for the con-
version of inactive pH 5.5 aldolase dimers into catalytically
active enzyme is that aldolase dimers must first associate

into tetramers and then go through a conformational change

178

com

.naacpoe soc use»

cQOHPN>dPmeH MO wmbdfilMHG—A SO QOHPQHDQGOCOO .Hmadfi .HO Pomkflm

 

.0: ostHm

 

 

 

179

><mw< 2_ 3.52.2

 

 

s s s :E s 2 .
_ _ _ J\ _ — u _

ES: 3. .ii. .55 3

 

Ex: _ 111.. .55 Z

 

 

 

 

 

 

 

m”._._|u_._<I ZO ZO_._.<~_._.ZmUZOU mm<<5 ".0 ...UmnIm

"IVSHBASEJ %

180
before catalytic activity appears. Thus under all conditions
studied in this research, aldolase dimers have been shown to
be catalytically inactive; only the tetramer possesses cata-

lytic activity.

DISCUSSION

As stated in the introduction: "The 3.58 intermediate
is of Special interest for its significance to protein bio- .1]
synthesis since the intermediate may also occur in the process E
of subunit association in living cells." But before a mean-

ingful role for the intermediate in the subunit association

\ x .
‘ w .
Vlfl.b-'

could be presented, the subunit structure of native aldolase,

a problem not unequivocally resolved, had to be reexamined.
Subunit Structure 2£_Rabbit Muscle Aldolase

When aldolase was dissociated into subunits with a
0.1M B-mercaptoethanol, 5.6M guanidine RCl solution, the
subunits were found to have a weight-average molecular weight
of 42,000. This is in agreement with the results subsequently-
published in the literature (see Literature Review). These
results, together with the recently published molecular
weight of 160,000 for the native enzyme (Kawahara and Tanford,
1966), strongly suggested that aldolase was composed of four
subunits (further confirming evidence is presented in the
Isozyme section of the Literature Review). In addition, the
characterization of the 3.53 intermediate as a dimer of
aldolase subunits, which is described below, provides further

support for the four subunit models of rabbit muscle aldolase.

181

182
Characterization 2;,the 2.58 Aldolase Intermediate

as a Dimer of; Subunits

The extrapolated sedimentation coefficient for the

dimer was found to be s30 w = 3.803 at pH 3.40 or pH 3.45
9
at 7°C and 830 w = 3.uSS at pH 3.35 and 2°C. The f/fo value
O

which was taken as a measure of the asymmetry of a molecule
was calculated to be f/fO = 1.9 for the dimer: this can be
compared to the more compact structure, f/fo = 1.3, for the
native enzyme, or to the more unfolded structure, f/fo =
2.3, for acid dissociated subunits.

The extrapolated diffusion coefficient for the dimer
was found to be Dgo,w = 3.85 x 10"7 cmZ/sec at pH 3.40 or
pH 3.05 at 7°C and Dgo,w = 3.73 x 10'? cmZ/sec at pH 3.35
and 2°C.

The molecular weight of the 3.58 intermediate of
aldolase was calculated from the M(S/D) technique (using
extrapolated S and D values) to be 83,000-86,000, approxi-
mating a dimer of aldolase subunits (h2,000). The 3.58
intermediate was produced and was stable under the follow-
ing conditions: pH 3.35, o—2°c, and low salt (0.01M NaCl)
during the dissociation. Some higher molecular weight aggre-
gate was found as shown by sedimentation equilibrium in
agreement with 108,000 limiting value (Deal ggugl.. 1963).

An estimate of the inaccuracy of the diffusion coef-
ficient due to aggregation was made by an empirical deter-
mination of the amount of aggregate present in the inter-

mediate preparations. It was shown that the impurity would

183
lead to a diffusion coefficient too low and the maximum error
was 4%. With this correction, the diffusion coefficient was
increased from 3.73 x 10'7 cmZ/sec to 3.90 x 10"7 cmZ/sec.
Using the Svedberg equation for M(S/D), this correction
decreased, in turn, the molecular weight from 83,000-86,000
to 80,000-83,000. Significantly, these values are most con-
sistent with a dimer of aldolase subunits (40,000 to 02,000).

That the intermediate is indeed a dimer is strongly
supported by the measurements of another hydrodynamic para-
meter, the intrinsic viscosity. Upon formation of the inter-
mediate, the apparent (6.2 or 8.3 mg/ml) intrinsic viscosity
of the native aldolase molecule (4 cc/g), shows a rapid
initial increase to 9 cc/g within 10 minutes: then it slowly
attains a limiting value of 14.5 cc/g in 3 hours, and remains
constant for at least 10 hours. These data suggest that the
intermediate is: (1) formed rapidly, (2) stable, (3) some-
what unfolded, and (u) a dimer of aldolase subunits. In
addition, the viscosity data was shown to exclude the only
other possibilities for the intermediate: (1) tetramers,

(2) trimers, and (3) monomers.

Thus, in contrast to the previously-held trimer model,
the mechanism of subunit association in living cells for the
tetramer model of aldolase is somewhat anticlimactic,
eSpecially with the physical characterization of the inter-
mediate as a dimer. That is, the most likely mechanism of
subunit association into tetrameric aldolase is a simply two

consecutive bimolecular association reaction: (1) subunit

 

 

 

184

association into dimers, followed by another bimolecular
reaction, (2) dimer association into tetramers. This then
concludes the primary objective of this thesis: to char-
acterize the 3.58 intermediate of rabbit muscle aldolase.

The discovery that the 3.58 intermediate was a dimer,
together with the fact that aldolase contained at least 3
catalytic sites (see Literature Review) led to the second ““3
major aim of this thesis: "to answer the question of A
whether aldolase dimers could possess catalytic activity

independent of their 'partner dimer' in the native tetramer."

1'4 ‘

But before the question of active dimers could be “I
examined, a study was conducted to find the proper conditions
for complete reversibility of activity from pH 3.3 dimers.
The basis for such a study was to eliminate, as much as
possible, any non-Specific aggregation that might interfere

with the analysis for active dimers.

Activity Reversal from the 2;; 2.2 Dimers: Attainment

23 Conditions for Complete Reversibility

 

Studies on the reversibility of dissociation to pro-
duce the pH 3.3 dimers have shown that 100% recovery of
activity can be obtained under the following conditions:
(1) pH 7.9, (2) 0.1M B-mercaptoethanol, (3) 1 hour incuba-
tion at 0°C followed by a 2 hour incubation at 20°C, and
(4) reversal concentrations between 0.013 mg/ml and 0.065
mg/ml.

The rate of activity reversal from pH 3.3 dimers was

185
found to be temperature-dependent and concentration-depen-
dent. The rate of activity recovery in the assay at 16°C
was found to be second-order with reSpect to dimer concen-
tration (0.09-0.18 mg/ml). The second-order rate constant
was calculated to be k = 1.35 x 10“ liters/mole-sec. These
results suggested that the p§.3&2.dimgy was inactive and
had to associate into tetramers before catalytic activity 1 11

was regained.
FDP Inhibition 23 Reversal

In the course of these studies, it was discovered
that when FDP was present in the assay before pH 3.3 dimers
were diluted into the assay, much lower activities were
found than if FDP was added last. Since it was previously
shown that the pH 3.3 dimer was inactive, the observation
that FDP inhibited activity reversal coupled with the
observation that this inhibition was strongly dependent
upon dimer concentration suggested that the FDP was prevent-
ing dimer association. Perhaps the binding site on the
native enzyme for FDP had remained partially intact after
dissociation into dimers. This hypothesis was supported by
a later eXperiment (Chapter IV) where it was shown by SDSV
analysis at pH 5.0 that FDP also inhibited dimer association.
The subsequent observation that pyrophOSphate could substi-
tute for FDP in this inhibition supported the above conten-
tion that the inhibition by FDP was due to binding of the

effector to the phosphateabinding site on the dimer, perhaps

186
"freezing" the dimer in a conformation which prevented a
necessary folding reaction which preceded association. 0n
the other hand, it may have interfered with the association
reaction directly. More detailed studies on the effect of
various concentrations of FDP showed that, at 0.17 mM FDP
concentration, FDP would no longer inhibit reversal. At
50% inhibition, a dissociation constant for FDP dissocia-
tion from the dimer was calculated to be 4 x 10'3M FDP, and
can be compared to published FDP dissociation constant for
the native enzyme, kd = A x 10'6M FDP. In conclusion to
this section, conditions for complete reversibility were
found as there were few problems with non-Specific aggrega-
tion. The pH 3.3 dimer was inactive and therefore had to
associate before catalytic activity was regained. And
finally, FDP was found to inhibit dimer association into
active tetramers. These results should be kept in mind,
especially the inactive pH 3.3 dimer results, when discuss-
ing the studies in the next section which are preliminary

to the analysis for active dimers at higher pH values.

The Study Preliminary‘gg the Analysis for Active Dimers
2£_Aldolase: Attainment QQDAldolase Dimers Under

Conditions Where Native Tetramers are Active

The approach to finding an active dimer was two-fold:
(1) The first approach was by direct dissociation of the
active tetramer into a possible active dimer without passing

through an inactive dimer intermediate. (2) The second

187
approach was to form dimers at pH 3.35 and then raise the
pH to find conditions which would allow the selective fold-
ing of the pH 3.35 dimer without association into tetramers.
The method used to physically separate the dimer and
tetramers was sucrose density gradient sedimentation
velocity (SDSV) eXperiments.

In the first approach no evidence for an active dimer
was found at pH 5.0 at 7°C or 20°C by direct dissociation of
tetramers. Therefore, the second approach was investigated.
The results showed that when the pH of pH 3.35 dimers was
raised to pH 4.0 or pH 5.0, association was inhibited; but
reactivation in the assay yielded essentially zero initial
rates so that dimers at theSe pH values must have been
inactive too. In contrast to these results, the pH 5.5
dimers which were incubated and separated by SDSV immediately
gave activity and were constant with little or no log times.
Subsequent studies on the concentration-dependence of dimer
association at pH 5.5 showed that only about 3% dimer asso-
ciation had occurred in a 20 hour period at -6°c and 0.03
mg/ml. However, incubation at higher temperatures (0°C,
4°C, and 12°C) stimulated dimer association. In other
studies, it was shown that increased concentrations of
sucrose (4%, 16% and 20%) could more effectively inhibit
dimer association, when compared to lower sucrose concentra-
tions. In conclusion to this chapter (IV), pH 5.5 dimers
subjected to SDSV analysis exhibited immediate activity upon

assaying with essentially no lag times; therefore, it was

188
under these conditions (except the SDSV analysis was ommitted)

that the final analysis for active dimers was investigated.

Analysis for Active Dimers
(Not Subjected tg_SDSV’Analysis)

The last chapter (V) of this thesis is devoted entirely
to proving whether the pH 5.5 dimer at -6°C was active when
assayed at pH 7.5 and 25°C. To examine this question, a
kinetic analysis of activity reversal, together with the
physical analysis of the dimer-tetramer distribution in the
assay, from -6°C, pH 5.5 dimers (not subjected to SDSV
analysis) was undertaken. The only way to "prove" anything
by kinetics is to disprove all other likely reaction mech-
anisms--the remaining mechanism, of course, being consistent
with all the known physical and kinetic data.

At high concentrations in the assay, data was obtained
which was consistent with a first-order reaction, but not a
second-order reaction. This excluded mechanism II which
postulates that inactive dimers must associate directly into
active tetramers with no intermediates.

A physical analysis of the dimer-tetramer distribution
in the assay at various times during the reversal process
showed increasing tetramer formation; this excluded the mech-
anism which postulates that inactive dimers must fold into
active dimers with no subsequent association into active
tetramers.

Thus only two mechanisms remained at this time in the

_.._:

189
kinetic analysis: (1) inactive dimers folding to active
dimers which subsequently associated into active tetramers
(mechanism III), or (2) inactive dimers associated into
inactive tetramers which subsequently folded to form active
tetramers (mechanism IV). Both mechanisms were consistent
with the previous two results that (a) a first-order reaction
preceded a molecule with activity (active dimers in mechanism
III above or active tetramers in mechanism IV above), and
(b) tetramers were formed in the assay (tetramers were formed
in both mechanisms.

To exclude mechanism III, the half-times for reversal
should have increased at lower concentrations. 0n the other
hand, it would be much more difficult to exclude mechanism
IV. If it was found that the half-times did not change at
lower concentrations, reversal would begin to show changes
in half—times. Fortunately for the analysis of this probe
lem (but unfortunate for the active dimer model) the half-
times at lower concentration showed an increase. These
results then excluded the active dimer model that postulated
that inactive dimers folded in the assay to form active
dimers which subsequently associated into active tetramers.

The only remaining model was the following:

 

m kznd ’ % kJ st u ‘

INACTIVE INACTIVE ACTIVE

At high concentrations ---- ist order; k1 _ 2.4 x 10'3/sec

At low concentrations ---- 2nd order: k 6.4 x 10ul/m-sec

2 -

*‘Pl “mama; ’L"'.'.'~ '_.._. ‘

”‘13:.“

190
k
1
—=k
C 2
at 3 ul/ml or 0.003 mg/ml, the rates and half-lives for a

first-order and for a second-order reaction are equal.

Vic-its

10.

11.

12.

13.
14.

15.
16.

BIBLIOGRAPHY

Baranowski, T., and Niederland, T. R. (1949) g. Biol.
Chem. g 180, 543.

Benesch, R. E.. Lardy H. A. and Benesch, R. (1955)
J. Biol. Chem., 21 , 663.

Castellino, F. J. and Barker, R. (1966) Biochem.
Bioetho BeS. Commune. :22. 1820

Chan, W.. Morse, D. E. and Horecker, B. L. (1967)
Proc. Natl. Acad. Sci. U.S., 51, 1013.

Deal, W. C., Rutter, W. J., Massey, V. and Van Holde,
K. E. (1963) Biochem. Biophys. Res. Commun.. 12, 49.

Deal, W. 0.. Rutter, W. J. and Van Holde, K. E. (1963)
Biochemistry, a, 246.

Donovan, J. W. (1964) Biochemistry, 3, 67.

Dreschsler, E. R.. Boyer, P. D. and Kowalsky, A. G.
(1959) g. Biol. Chem., 234, 2627.

Edelstein, S. J. and Schachman, H. K. (1966) Fed. Proc.,
22. .12, ""' ""'""

Fujita, H. E. (1962) "Mathematical Theory of Sedimenta-
tion Analysis," Academic Press, New York.

Ginsburg A. (1966) Archives g§_Biochem. and Biophys.
112, 445. ” ' ' ' ””” ' '”' '

Ginsgurg, A. and Mehler, A. H. (1966) Biochemistry, 5.
2 23.

Gralen, N. (1939) Biochem. g,. 22, 1343.

Hade, E. P. K. and Tanford, C. (1967) g. Am. Chem. Soc.,
Q2. 5034.

Hass, L. F. (1964) Biochemistry, 2, 535.
Hasséég. F. and Lewis, M. S. (1963) Biochemistry, a,
1 o

191

“‘2...

192

BIBLIOGRAPHY--Continued

17.

18.

19.
20.

21.

22.

23.

24.

25.
26.

27.

28.

29,

30.

31.
32.
33.

34.

Hill, R. L., and Kanarek, L. (1964) Brookhaven Symposia
12 Biology, $1, 80.

Holleman, W. H. (1966) A Thesis, Michigan State Univer-
Sityo

Hunter, M. J. (1966) g, Phy . Chem., 20. 3285.

Kawahara, K. and Tanford, C. (1966) Biochemistry. 5.
1578.

Kobashi, K.. Lai, C. Y. and Horecker, B. L. (1966)
Arch. Biochem. Biophys., 112, 437.

Kowalsky, A. G. and Boyer, P. D. (1960) g, Biol. Chem.,

£22, 604.

Mehler, A. H., and Bloom, B. (1963) g, Biol. Chem.,
228. 1050

Morse, D., Lai, C. T. and Horecker, B. L. (1965)
Biochem. Biophys. Res. Commun.. 18, 679.

Nygaard, A. P. (1956) Acta Chem. Scand., 12, 397.

Penhoet, E., Kochman, M.. Valentine, E., and Rutter,
W. J. (1967) Biochemistry, 6, 2940.

Penhoet, E., Rajkumar, T. and Rutter, W. J. (1966)
Proc. Natl. Acad. Sci. U.S., 56, 1275.

Rajkumar, T. V.. Penhoet, E., Rutter, W. J. (1966)
Fed. Proc.. £5, 523.

Rao, M. S. N., and Kegeles, G. (1958) g, Am. Chem. Soc.,
$29,. 5721. 5724.

Richards, 0. C. and Rutter, W. J. (1961a) J. Biol. Chem.,
226. 3177-

Ibid. (1961b) 3185.

 

Rutter, W. J. (1964) Federation Proc., £2, 1248.

Rutter, W. J.. Hunsley, J. R., Grores, W. E., Calder, J.,
Rajkumar, T. V., and Woodfer, B. M. (1966) Methods
in EnzxmolOEY. 2;. 479-

Rutter, W. J., Woodfin, B. M. and Blostein, R. E. (1963)
Acta Chem. Scand.. 1 , Suppl. 1, $226.

193

BIBLIOGRAPHY--Continued

35.
36.

37.

38.

39-

40.

41.

42.

43.

L111.

45.

46.

47.

48.

49.

50.

51.
52.

Schachman, H. K. (1957) Methods in EnzymoloSY. £2, 32.

Schachman, H. K. (1960) Brookhaven Symposia ig_Biology,
12o 9-

Schachman, H. K. and Edelstein, S. J. (1966) Biochem-
istry, 2, 2681.

Scheraga, H. A. (1961) Protein Structure (AP) (N.Y.)
Po 5.

Scheraga, H. A., and Manderkern, L. (1953) g, Am. Chem.
.5990 o 220 1790

Sine, H. E., and Hass, L. F. (1966) Abstract no. 279c,

presented at the 152nd National meeting of the
American Chemical Society, New York.

Swenson, A. D., and Boyer, P. D. (1957) g. Am. Chem.
800.. 22, 217uo

Spolter, P. D., Adelman, R. C. and Weinhouse, S. (1965)
J. Biol. Chem., 240, 1327.

Stellwageg, E. and Schachman, H. K. (1962) Biochemistry,
.L. 105 .

Svedberg, T. and Pedersen, K. 0. (1940) "Ultracentrafuge,"
Clarendon Press, Oxford.

Taylor. J. F., Green, A. A., and Cori, G. T. (1948)
g. Biol. Chem., 123, 591.

Taylor, J. F.. and Lowry, C. (1956) Biochim. Biophys.
Acta, 39, 109. ‘

Udenfriend, S. and Velick, S. F.. (1951) g. Biol. Chem.,
120. 733-

Van Holde, K. E. and Baldwin, R. L. (1958) g. Phy .
Chem., 63, 734.

Westhead, E. W., Butler, L. and Boyer, P. D. (1963)
Biochemistry, a, 927.

Winitegd, J. A. and Wold, F. (1964) J. Biol. Chem., 239,
21 .

Winstead, J. A. and Wold, F. (1964) Biochemistry, 2. 791.
thantis, D. A. (1964) Biochemistry, 2, 297.