REVERSIBLE DISSOCIATION AND CHARACTERIZATION
. OF RABBIT MUSCLE PYRUVATE KINASE

Thesis for the Degree of Ph. Dy
MICHIGAN STATE UNIVERSTIY

Marlene Steinmetz Kayne
19616

  
      

  

LIBRAR Y
Michigan State
University

Thu-.SIS

This is to certify that the

thesis entitled
Reversible Dissociation and Characterization

Of Rabbit Muscle Pyruvate Kinase
presented by

Marlene Steinmetz Kayne

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Biochemistry

7/ «(an (‘

Major professor

Date M 6‘

 

 

0—169

 

   

 

 

 

 

 

 

 

 

 

ABSTRACT

REVERSIBLE DISSOCIATION AND CHARACTERIZATION

OF RABBIT MUSCLE PYBUVATE KINASE
by Marlene Steinmetz Kayne

Rabbit muscle pyruvate kinase (EC 2.7.1.40) (mol wt
237,000) has been found to be a tetramer, each polypeptide
chain having a molecular weight of 9a. 57,000. Structural
studies as a function of urea concentration revealed an
intermediate in 1.5 fl urea having a 336?? of 7.3 S, and
another in 3.0 M urea with an 336?: of 3.6 S. Complete
dissociation of the enzyme occurred in 4‘fl urea (830?: =
1.7 3) indicating noncovalent subunit bonding. Optical
rotatory dispersion studies on pyruvate kinase parallel the
sedimentation coefficient analysis and showed bO values of
-1080 for enzyme with no urea, -930 in 1.5 fl urea, -86° for
enzyme in 3 fl_urea and 00 in 4‘fl urea. This indicates only
slight unfolding at urea concentrations up to 3 E and com-
plete unfolding and loss of structure in 4 fl_urea.

Characterization of the subunits in 7.4 fl urea yielded

values of s0 = 2.01 S, D0 = 3.46 X 10'7 cm2/sec, and
2 20,w

0,w
M; (s/D«) = 56,300. Detailed subunit molecular weight analy-
ses utilized sedimentation equilibrium in three separate dis-

sociating systems; namely. (1) 7.4 fl urea, (2) 7.4 fl_urea -

Marlene Steinmetz Kayne
0.12 E E-mercaptoethanol, and (3) 6.8 M guanidine hydrochlor-
ide — 0.12 E fi-mercaptoethanol. In the guanidine-mercapto-
ethanol system the subunits were quite stable; the M; and M:
were essentially the same, indicating that the subunits had
approximately equal molecular weights of 57,100. Random
aggregation of the subunits occurred in both urea systems.

Detailed studies of the parameters influencing the
reversal of the dissociation process led to the conditions
under which 56% of the enzyme activity was recovered. Opti-
mum renaturation was achieved by dilution of the guanidine
denatured enzyme to final reversal conditions of pH 8.5, 20°,
in 0.01 mm ADP, 0.1 fl Tris—HCl buffer, 0.165 E ammonium sul-
fate, 0.06 E D-mercaptoethanol, 1.0 mfl EDTA, and a final
enzyme concentration of 0.04 mg/ml.

The sedimentation coefficient of the native enzyme was
found to decrease to ca. 8.5 S when the enzyme was allowed to
remain for 1% weeks at a concentration of 80 mg/ml in 0.02 E
imidazole buffer (pH = 7.0) at 5°.

Amino acid analysis of the enzyme indicated that 50%
of the amino acid residues were hydrophobic; it also showed
37 half-cystine residues, only 4-6 of which lie exposed on
the enzyme surface as judged by reactivity with performic
acid. A partial specific volume of .741 cc/g was calculated

from the amino acid content of pyruvate kinase.

 

REVERSIBLE DISSOCIATICH AND CHARACTERIZATION

OF RABBIT LUSCLE PYRUVATE KIEASE

:y

harlene Steinmetz Kayne

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1966

 

 

 

237’ 4/3/77“
$7197

ACKNOWLEDGMENT

The author wishes to express her appreciation to
Dr. W. C. Deal, Jr. for his guidance and assistance
throughout the course of this study. The technical
assistance of Miss Doris Bauer is also greatly appre-
ciated. The support of a National Institutes of Health

predoctoral fellowship is gratefully acknowledged.

ii

 

iii

 

 

TABLE OF CONTENTS

INTRODUCTION . . . . . . . . . . . . . . . . . . .
LITERATURE REVIEW . . . . . . . . . . . . . . . .
General Properties of Pyruvate Kinase . . . .
Chemical Properties of Pyruvate Kinase . . .
Enzyme Specificity . . . . . . . . . . .

Metal Requirements . . . . . . . . . . .

IHhibi tOI'S o o o O o O O O a n I o o o e

Mechanistic Aspects of the Pyruvate Kinase
Reaction 0 I O I I I I I I I I I O O O I

Physical Properties of Rabbit Muscle Pyruvate
Kinase I O I O I I O I I I I I I I 0 O I

Control of Rabbit Muscle Pyruvate Kinase . .
METHODS . . . . . . . . . . . . . . . . . . . . .
Preparation of Pyruvate Kinase . . . . . . .
Ultracentrifugal Analysis . . . . . . . . . .
Optical Rotatory Dispersion Analysis . . . .
Amino Acid Analysis . . . . . . . . . . . . .
General Procedure . . . . . . . . . . .

Amino Acid Calculations . .

Preparation of Performic Acid Oxidized
Enzyme . .

Preparation of Reduced, Carboxymethylated

Protein 0 o O I O n o o e o 0 O I O o
DTNB Ti tration e e o a o I e t O I I a 0

Activity Assay . . . . . . . . . . . . . . .
RESULTS I I O C O I I I I O I O I l 0 I I I I I I
Approach to Problem . . . . . . . . . . . . .

iv

0\

10
11
11
11
14
15

15
16

17

18
19

19
21
21

TABLE OF CONTENTS (CONTINUED)

Characterization of Native Pyruvate Kinase . . .

Determination of the Sedimentation
Coefficient (3%0 w . . . .
Determination of he Diffusion
Coefficient (D20w ) .
Determination of thé WMolecular Weight
and Frictional Coefficient . . .
Alteration of Pyruvate Kinase Upon Aging
Amino Acid Analysis . . . . . . . . . . .
Sulfhydryl Determination . . . . .
Partial Specific Volume of Pyruvate
Kinase . . . . . . . . . . . . . . . . .

Characterization of the Dissociation Process . .

Dissociation as a Function of Urea Con-
centration by Sedimentation Coefficient
Analysis . . . . . . . . . . . . . . . .

Dissociation as a Function of Urea Con-
centration by Optical Rotatory
Dispersion Analysis . . . . . . . . .

Determination of the Sedimentation and
Diffusion Coefficient of Pyruvate
Kinase in 0.5 M Urea . . . . . . . .

Determination of the Sedimentation and
Diffusion Coefficient of Pyruvate

Kinase in 1. 5 M Urea . . . .
Estimation of the —Enzymatic Activity of
Pyruvate Kinase in 1. 7 M Urea . . . .

Studies of the Molecular Weight of the
Enzyme in 2. 5 and 3.0 M Urea . . . . . .

Characterization of the Subunits of Pyruvate
Kinase I I I I I I I I I I I I I I I I I I

Molecular Weight Analysis in Urea . . . . .
Molecular Weight Analysis in Urea and
-mercaptoethanol . . . . . . . .
Molecular Weight Analysis in Guanidine
Hydrochloride and M-mercaptoethan l . .
Determination of s30 w, D30 w, and Mw(s/D)
in 7. 4 M Urea . . . . . . . . . .

Reversible Dissociation of Pyruvate Kinase . . .

Effect of Dissociating Solvent on the
Renaturation of Pyruvate Kinase . . . .

 

35

42

50

51
6O
61

61
61
65
65
68
75

78

DISCUSSION

SUMMARY .

Effect of pH, Temperature, Ionic Strength,
and Protein Concentration on the
Renaturation of Pyruvate Kinase
Effect of Reducing Agents, Substrates, and
Metal Cofactors on the Renaturation of
Pyruvate Kinase

REFERENCES .

TABLE OF CONTENTS (CONTINUED)

I

vi

 

79

87

98
108
111

LIST OF FIGURES

Extrapolation to zero protein concentration of
the sedimentation coefficients of fresh and
aged enzyme . . . . . . . . . . . . . . . . . .

Extrapolation to zero protein concentration of
the diffusion coefficient of pyruvate kinase .

The effect of urea concentration on the struc-
ture of pyruvate kinase . . . . . . . . . . . .

Sedimentation patterns of pyruvate kinase in
1.5-1.7 M urea at zero time and after 2 and 3
days I I I I I I I I I I I I I I I I I I I I I

Sedimentation patterns of pyruvate kinase as a
function of urea concentration . . . . . . . .

Conformational analysis by optical rotatory
dispersion of pyruvate kinase in the presence
and absence of 1.5 M urea . . . . . . . . . . .

Conformational analysis by optical rotatory
dispersion of pyruvate kinase in the presence
of 3.0, 4.0, and 5.4 M urea . . . . . . . . . .

Extrapolation to zero protein concentration of
the sedimentation coefficients of pyruvate
kinase in 0.5 M_urea . . . . . . . . . . . . .

Extrapolation to zero protein concentration of
the diffusion coefficients of pyruvate kinase
in OI 5 ‘Ii-L urea I I I I I I I I I I I I I I I I I

Extrapolation to zero protein concentration of
the sedimentation coefficients of pyruvate
kinase in 1.5 M_urea . . . . . . . . . . . . .

Extrapolation to zero protein concentration of

the diffusion coefficients of pyruvate kinase
in 1 I 5 .IV—l urea I I I I I I I I I I I I I I I I I

vii

r——————————'—_—f

Figgge-

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

LIST OF FIGURES (CONTINUED)

Extrapolation to zero protein concentration of
the apparent weight-average and z-average
molecular weights of pyruvate kinase in 7.4 M
Llrea I I I I I I I I I I I I I I I I I I I I I I

Extrapolation to zero protein concentration of
the apparent weight-average and z-average
molecular weights of pyruvate kinase in 7.4 M
urea and 0.12 M M-mercaptoethanol . . . . . . .

Extrapolation to zero protein concentration of
the weight-average and z-average molecular
weights of pyruvate kinase in 6.8 M guanidine
hydrochloride and 0.12 M Memercaptoethanol . . .

Extrapolation to zero protein concentration of
the sedimentation coefficients of the subunits
of pyruvate kinase in 7.4 M urea . . . . . . . .

Extrapolation to zero protein concentration of
the diffusion coefficients of the subunits of
pyruvate kinase in 7.4 M urea . . . . . . . . .

Effect of pH on the renaturation of pyruvate
kinase I I I I I I I I I I I I I I I I I I I I I

Effect of temperature on the renaturation of
pyruvate kinase . . . . . . . . . . . . . . . .

Effect of ionic strength on the renaturation
of pyruvate kinase . . . . . . . . . . . . . . .

Effect of enzyme concentration on the renatura-
tion of pyruvate kinase . . . . . . . . . . . .

Effect of -mercaptoethanol on the renaturation
of pyruvate kinase . . . . . . . . . . . . . . .

Effect of ADP on the renaturation of pyruvate
kinase I I I I I I I I I I I I I I I I I I I I I

Effect of cations (Mg2+ and K+) on the renatura-
tion of pyruvate kinase . . . . . . . . . . . .

viii

64

67

7O

72

71

81

84

86

89

92

94

97

Tables

I.

II.

III.

IV.

LIST OF TABLES

Amino acid analysis of rabbit muscle pyruvate
kinase I I I I I I I I I I I I I I I I I I I I I
Half-cystine content of rabbit muscle pyruvate
kinase I I I I I I I I I I I I I I I I I I I I I
Calculation of the apparent specific volume of
pyruvate kinase from its amino acid composi-
tion I I I I I I I I I I I I I I I I I I I I I I
Properties of the intermediates of pyruvate
kinase dissociation . . . . . . . . . . . . . .
Physical properties of the pyruvate kinase sub-
mits I I I I I I I I I I I I i

a I I e a o I 0

ix

31

34

36

62

76

 

ABBREVIATIONS

O

z and L; (s/D), the superscript,

-.~ 0 O 0

For ”20,w' s20,w’ 1‘w' L

0’ indicates that the quantity has been extrapolated to zero
protein concentration; kw (s/D), weight-average molecular

weight determined from the Svedberg equation; r

'w' weight

average molecular weight; E z-average molecular weight; CO,

2'
initial protein concentration; (Cm + Cb)/2, average of the
protein concentration at the meniscus, and bottom of the
solution; rtic, rotor temperature indicator and control;
CRD, optical rotatory dispersion; mol wt, molecular weight;
PEP, phosphoenolpyruvate; ADP,adenosine diphosphate; ATP,
adenosine triphosphate; .AD, MADE, oxidized and reduced
nicotinamide adenine dinucleotide; MINE, 5,5'-dithiobis
(2-nitrobenzoic acid); PELB, p-hydroxymercubibenzoate; EDTA,
ethylenediaminetetraacetic acid; RMR, nuclear magnetic
resonance; DES, diethylstilbesterol; LDE, lactic dehydro-

genase.

ILTECDUCTIDh

Recent years have seen the accumulation of a consider-
able amount of evidence indicating the presence of several
polv.eptide chains in the structure of many enzymes, espe-
eially in those with nol wt e5,000 or creater. Although most
of the @lycolytie enzymes of rabbit skeletal muscle are large

- _ . '\ __ .L. 1 o
x would oe expecten to eonSis

, . ,/ 1;. ",, A 1 \n
wan nelde, 1962; Jeel et a ., 10o}; spell e on aid :enaehr i,
1062) when this stvly res undertaken. It see'ed clear that e

elucidation of the relationship between structure and function
of thee glycolytic enzynes.

Cur attention was drawn to :yruvate kinase because of
its larfie size nol wt 237,000 (earner, 1959)] and because

the enzyme was renorted to bind 2 moles of phosphoenolpyru-

vate/nole of enzyme (Bevnarl et al. 1961}. RecentIV, it has
./ I. ’ d

 

 

also been shown that 2 Ln ions are bound/mole of enzyme
(Lildvan and Cohn, 1965).

horanieki (1960) reported that pyruvate kinase disso-
ciated into suounits of rel wt 150,000 in 6 L urea and con-
eluded that the enzyme consisted of at least two polypeptide

chains. Aowever, it is part of our workinfi hypothesis that

1

most polypeptide eLains W111 have a nol wt less than 65, 000.
Thus, it seemed to us that the pyruvate hinase complex might
consist of subunits smaller than 150 .000 mol ?t which might

simply require more stringent conlitions to produce disso-

(1‘
(,3
<3
(3.!
L:
t J
H l
C‘-
U)
I

eiation into ultirm This prediction has been
found to be correct and the existence of smaller su eunits has

been documented ir this work.

.0 .1. .'.°., .N 1- V ~ 4.; , ~._ 1.. .r 1- v - ‘ 4..
Cl the natiue enzvne, dissociation intesreaiates, ultimate

:19

s a result of the inportant role the enzyme pyrt

vate kinase plays in the scheme of glycolySi is, its physical
and chemical properties have been actively investi'e ted for
almost thirty vears. ....ee\,._.,l*r two extensive review articles
(Czoh and Lucher, 1960; loyer, 1962) have been published on

the general pro erties of the enzyn—.

1

r . ‘2:— - 1.: .— '1': m .-- - 7.x ' ~
general iroperties of iyiuuate ninase
M1,, (3* 1 o 1 1 a' -.—-r- J-
1.1.18 V r. 2U" .".‘.‘\1 purl lL] Ill CL be

- f \ I v ".
loaaarn and neyerhof (1; 30) and ubs entlr isolated ans

Iegelein was unfortunatel, lost during florld war :1. Since

[—5

then partial ourif ieation o the enzyme has seen reported
several inv es ti'etor“ (Lorcoeng and Prieer, 1951; lachmar and

Boyer, 1953) while cryste line preparations have seen

The enz'me widely distributed in nature, nas been detectev

' . - - .1. .,.,. a. . '3 ‘_ -... - A
in animals, y~ast, holds ane sons clans (scyer, 1302).

series of reactions involving the oxidation 0 J—glyeeralde-

F!)

hyde- 3-phosp-ate and PJ2 labeled inorranic phosphate, it was

q

3

 

 

4
shown that P32 labeled PEP could be obtained by reversal of
the pyruvate kinase reaction (Lardy and Ziegler, 1945).
McQuate and Utter (1959) and hrimsky (1959) have reported an
equilibrium constant (rec) for the reaction of as: 1 x 1011

l/pM in the absence of h+ and at a pH of 7.4 to 8.0.

Chemical Properties of Pyruvate Kinase
In the cell, the enzyme functions in the degradation
of glucose by catalyzihq the conversion of phosphoenolpyru-
vate (PEP) to pyruvate with the transfer of a "high energy"

phosphate bond (reaction 1).

 

”v-9- 1'“ o " I 1 LL-
(1) h + PEEJ + an? A~>Pvruvate + ATB'

1112+ Tf+
.3 , -

In addition to iTP, the enzyme will also phosphoryl—
ate fluoride ion and hydroxylamine. Thus it has been shown
to be identical with ”fluorokinase” (Tietz and Ochoa, 1953),
an enzyme which had been reported by Flavin at al. (1956;

1957) to catalyze reaction 2.

 

(2) ATP + fluoride ) ADP + fluorophosphate

The ability of the enzyme to act as an "hydroxylamine
kinase" by catalyzing reaction 3 was observe; by Kupieki and
Coon (1960).

. phosphorylated
(3) ATP + NHZOH )7 ADP +

 

hydroxylamine
C02 , ZN.

5

'c“
l 1
is
51)
U)
(D

Both the "fluorohinase" and "hydroxylanine

reactions are narhelly stimulated by bicarbonate. The

"' 2+ (‘1 n‘}" 1 ' ' .
cations, Ln or cc‘C , out not h72+, promote "hydroxylamine

\

, 2+ .
:i and coon, 1960) whereas v: , out

in

g3

86?

'
1"}.A2' - J... n - ‘ e n . -| - — ’0 VJ-
' ' ' ‘ (JV -.U. - "' ' "J ' " ' A 1':

.L

activity (Tietz and CcLoa, 195C). The nonovalent cation,

h , required in the n"r_”ote hinase reaction, has also been

Enzy:e Specificity

detailed investifiations have not as yet seen undertaken to
determine the specifici'v of the ensyne towerl PEP. However,
Rose (1960) has found that tLe enzyne speci- ically detriti~
ates pyruvate but noto<-hotobutyrate orcK—hetOjluterate.
The nucleotide speciiicitv of the enzyme is not as
pronOaned. In addition to are, the 5'- Mi ahosphates of
guanosine, inosine, uridine and c
act as phosphoryl acceptors but at relatively lower rates
(Plowman and hrall, 1905). It has furthe; been oeserved that
phosphate, pyroph sthate, and triv1o phate may replace ATP in

O O _L_ "N . /
the enolization of pyruvate (nose, 19o0).

/

v- '0 v?‘ '2. ‘ - . ‘- ‘n J'- v- . «~—
- , s lirst sLo«n to he required fol catalySis ‘y

Boyer t a . (1942; 1993) anl later confirmed by Lardy and

Ex
I.)
(D

3’
[.1
(D
A
H
\ 3
.{r
km
.3
5
O
O
O"
(I)
(D

i
«1,4
(D
0.:

an e.honccne1t in the exchanoe

J

3
Both the "fluorokinase" and "ivrrorvlaniwe rinase“

‘

cicarbonate. The

reactions are narkedly stimulated by
2-'—- 2-1. _ _ 2+ L

cations, Ln or Go ', out nOt a? , pron0ce " IArOtvla“ ine

kinase“ actitity (lupieki and Coon, 1960) whereas If , out

not Lu T, promotes ”fluorokinase” and pyruvate l:ina se

activitv (Tietz and CcLoa, 195T). Ehe Mondalen cation,

---,, fl L' 1
-inase rea0c1011, has also seen

fSiS of reactions 2 and 3 (Tietz

..)

zvne jpeci"

11“ L1 '1 I _ ‘ I .1 4", Q " .- n C ‘ 1 Q -_ y a .I- - v--I-‘-‘-.-\
sue to vflp ni-;icuicJ in preparind derivatives 01 ref,

4.1

determine bile specificity of tae enzyme toward EVE. However,
Rose (1960) has found tlat tLe enzyme specifically de ri bi-

ates oyr1vatc but notc£-hoto'1tr ate orcL—hetotlutarate.
Ihe nucleotide spe ci
pronounced. In addition to £12, the 5'-dinhosr hates of

juanOSine, ino i1 e, Ln Miline and c

C”)

act as phosphoryl acceptors but at relativelr lower rates

'~ AN/r’ __
(P lo an a11“.:.ra l, i9e5). u
phosphate, Ulroveocihcte, and trii1ospiate raw replace ATP in

the enolization of pyruvate (Rose, 1960).

Letal “egp1re1m1 s

.2]? + - - . J. v -. J. ‘ _ 0 j o _ J- . 1

a has firSt ShOwfl DO ee required 10f cataleis cv
Boyer et a1. (19%2; 19b3) and later confirmed by Lardy and

Ziefiler (19h5) who observed an “1-.ccne1c i_‘: the exchanée

6
of P32vATP and PEP in the presence of K+. It was subsequently
.+
found that either Eb+ or EHE+ was able to replace h as an

activator but to a lesser extent (Kachmar and Boyer, 1953).

Inhibitors
Ca2+

 

was found to inhibit the activation of the enzyme

+

by K+; further addition of increasing amounts of K partially

2+

reversed the Ca inhibition. Loreover, the monovalent

- ~— ~.+ v o
cations, na+ and Ll , were found to counteract the activating

capacity of 1+ (hachnar and Boyer, 1953).

,-

i complete inhibition of the enzyme by sulfhydryl com-

_/ A —( N l
pounds (6 X 10 O L FELL, 5 K 10 J : 13012, 2 r 10'3 L ethyl-

h.

naleimide, 0.2 L iodoacetate) was observed with the hydroxyl--
aminekjnasereaction but it Las not been determined whether
this is due to reaction with an active site sulfhydryl group
or to a secondary structural change accompanying the reaction
with a sulfhydrjl not at the active site (hupieki and Coon,

1960).

hechanistic Aspects of the Pyruvate Kinase Reaction
Studies, utilizing 018 exchange, have established that
pyruvate kinase, in common with other kinases, catalyzes trans-
fer of the phosphoryl group without the exchanqe of the phos-
phate oxygens with those of water or substrates (Harrison gt
a;., 1955). This indicates that the reaction proceeds either
by direct transfer of the phosphoryl group or by the formation

of a phosphoryl—enzyme intermediate. Three different experi-

mental approaches (Harrison et al., 1955; Bass gt al., 1961)

7

have been used to detect the existence of a possible phosphoryl-

enzyme. ho evidence to support the existence of such an inter-

:-
.L

mediate species was 'ound; therefore the catalytic reaction

seems to involve a direct transfer of the phosphoryl group from

.u-

the PE: donor to the acceptor molecule. it has

0

rurther been
demonstrated that the keto rather than the enol form of pyruvate
is the actual enzvne substrate since enoli tion of pvruvate was

° I. .0 1‘. 1"” ‘.“- ' , ‘1 r ' + ‘f / —; V
enhanced in the presence 01 arr, n$2+, and a (dose, 1900). by

H
(I)
Ci.
0

observing the dependence of the of enolization of pyruvate

ionic state of the

:3
ct
c1-
0, J
(1)

on pn, it was fuither concluded tL~

han that of the n'f, is the rate de

step for enolizaticn and net phosphorylation.

-* ..~‘M\

for the active site on the protein. This tOjether with evi-

.1.

1ainst the forxation of a pncsonorvi-enzvme intermediate,

U)

has been interpreted a indicatiné that the cata 13m tic site on the
enzvne c011sists of one binding site for PEP or pyruvate and
one for ADP or ATP. Furthermore, a competition exists between
PEP and ATP because of a connon locus on the er zy11e for the
transferrable phosphoryl firoup (deynard gt afl,, 1961).

d value of two binding si
has been calculc1tec7_ for both FEP (Re
h32+ (1 ild1ran and Cohn, 1965) i1idi cc”""” the piesence of two
catalytic sites.

when PEP was bound to metal activated and non-activated
enzyme, a perturbation of a tu,r osine and crvptoonsa res pec-

tively was observed (Layne and Suelter, 1965) while pyruvate

 

8
elicited a difference spectra only in the presence of activat-
ing cations (Suelter gt gl., 1966).
Suelter and helander (1963) observed a change in the

. - 2+ 1
environment of a tryptophan reSidue when either hn or ng+

was added to the enzyme while the addition of Ca2+ to the
enzyme did not produce such an effect (Lildvan and Cohn, 1965).
With the use of nnr techniques, hildvan and Cohn (1965) con-
cluded that the functional catalytic unit is a divalent cation-

1. . . .. i 1 2+
enzyme complex. Additional nnr studies of the ATE—hr -enzyme

complex have indicated that a L32+ bridge structure exists
between the enzyne and the X-phosphoryl group of ATP in the
active complex (Lildvan and Cohn, 1966). In contrast to these
observations, helchior (1965), by determining the velocity of
the pyruvate kinase reaction over a wide range of concentrations
of ADP, Lg2+, and L+, has concluded that Lg-ADP is the specific
substrate for the reaction and that no metal-enzyme complex is
formed.

Studies of the effect of h+ on the kinetics of the

+ and PEP

reaction were best explained by assuming that the h
do not act as a metal—substrate complex; rather there is inde-
pendent binding of K+ and PEP to the enzyme (Kachmar and Boyer,
1953). It was further postulated, by the same authors, that

the activating (3+, EH4+ and Rb+), slightly activating (Ra+)
and non-activating (li+) cations combined with the same site

on the enzyme, but due to differences in crystal radii between

the various groups, structures adjacent to the binding site

were displaced to a greater or lesser degree. The above

O
/

postulate is fortified by the observation of a difference

Q1

spectra upon the bind in5 Oi nonovalent cation to the ctiv

and non-activated enzyme (Suelte; et al., 1906). lbs electro—

phoretic patterns of pyruve te Kinase have also been found to
o 1 u - .+— 1
vary conSiderably in the presence of n (Sorger et al., 1965).

sical Properties of Rabbit huscle

SD
L)
Q
Ki)
0
\Q
C\
« 4
I |
i—L
O
I
\)
O
' I
N
\
m
(D
O
s
(D
C’)
’ O
(1)
C)
("t-
H.
.«l
x.
1..
1
D
o
‘ _)
(D
A
("I
s
d
81
$0
d.
(D
Q.
C1-
0
N
(D
F:
0

protein concentration and corrected to the standard conditions
of 200 and water. UsinS a ‘ectial specific volume of 0. 7U a
100 for the native protein in aqueous salt solution, he calcu-
lated a nol wt of 237,000 for the native ensvne.

Liri UGd studies indicated that the native enzyme, in

the presence of 6 t urea, dissociated into at least two poly~

peptide chains havin.: a nol wt of 150,000 (horanieki, 1960).
The dissociation process seemed to be reversible i 2.5 L

urea and irreversible in 6.0 ; urea.

The absorption spectre of the enzvne indicates a low

0).

content of t3 ro ins and tryptophan boyer, 1962). A isoelec-
trio point of 6.6 was determined by ranining the electro-
phoretic mobility of the enzyme at various ph's. An ertinction
coefficient (E) of 0.54 OD/cn/n5/nl has been calculated for

rabbit muscle pyruvate kinase (hayne, 1966).

I

10

Tanaka t al., (1965) have recently observed that the

 

mammalian liver extracts of pyruvate kinase exhibited four
electrophoretically migrating bands, all of which possessed
pyruvate kinase activity. One of the four bands was found to
be electrophoretically identical to pyruvate kinase isolated

from muscle, brain and heart tissue.

Control of Rabbit huscle
Pyruvate Kinase
Although liver pyruvate kinase has been shown to be
under dietary and hormonal control (Weber gt al., 1965; Krebs

and Easieston, 1965; Tanara et

K, J'-

_l., 1965), few 1actors with
the exception of substrates, have been found to regulate the
activity of the muscle enzyme. ATP, as mentioned earlier, has
been found to produce product inhibition of the enzyme
(Reynard gt gl., 1961). Diethylstilbesterol (DES), as well
as certain hormones, has been shown to inhibit the muscle
enzyme by as much as 60% when present at a concentration of
8 x 10'"5 t. DES altered the viscosity and electrophoretic
properties but not the molecular weight of the enzyme; inhibi-
tion by DES was partially reversed by the addition of AD? or
ATP. Investigation of red blood cell pyruvate kinase defi-
ciency in humans seems to indicate separate genetic control

) of the leucocyte and erythrocyte pyruvate kinase; the erythro—

l cyte deficiency results in Type A nonspherocytic hemolytic

anaemia (Koler, 1964).

LETHODS
Preparation of Pyruvate Kinase

Pyruvate kinase was prepared from frozen rabbit muscle
(Pel-Freeze Biologicals, Rogers, Ark.) according to the pro-
cedure of Tietz and Ochoa (1953). Following the second heat
step, the protein was precipitated with ammonium sulfate
between 40% and 60S saturation, resuspended in 0.001 g EDTA
and 0.02 g imidaZOle buffer, pH 7.0, and stored under these
conditions at a protein concentration of 50-30 mg/ml. Enzyme
concentrations were determined by measuring the optical
density at 80 mu using the relation: conc. = 0D 7 1/O.54

(Kayne, 1966).

Ultracentrifugal Analysis
Except for a few noted exceptions, the enzyme was
routinely prepared for ultracentrifugation as follows: stock
enzyme solutions at hiyh concentrations (50-80 mg/ml) were
diluted with various dissociating and reducing agents con-
' taining 0.15 L EaCl or £01, 0.0H g Tris-£01 buffer, and 0.001
; EDTA, pE 8.0 (measured at 230). The resulting stock enzyme
solution was dialyzed against these reagents for at least 3
days at 5°. For the experiments at different protein concen—
trations, samples were then diluted directly with the dialysate
to yield enzyme concentrations of 1.5 - 19.0 mg/ml.
11

 

12

Spinco analytical hodel E ultracentrifuges with phase-
plate Schlieren optics were used for these studies. The
short column sedimentation equilibrium techniques (Van Holde
and Baldwin, 1958) were complemented by new methods. Tech-
niques were devised to allow rulticell (2 or 3 cells) opera-
tion using the Schlieren optical system, and calculations were
developed for extrapolation to 0 protein concentration using
the data from a single experiment. Concentrations were
expressed as (Cm + Cb)/2 and Cm + Cb where (Cm + Cb)/2 =
average of the protein concentration at the meniscus and
bottom of the solution. The calculations, including statis-
tical analysis of data, were performed using fully tested
programs on a Control Data 3600 digital computer. A value
of 0.74 cc/g at 100 (Warner, 1953) was used for the partial
specific volume (6) using the temperature correction factor
of 0.001 cc/g/deg (Taylor and Lowry, 1956). The solution
column depth in the cells was 1.7 mm (0.06 ml of protein
sample) and all sedimentation equilibrium experiments were
carried out at 9a. 70 for at least 36 hr. Due to the appear—
ance of a low molecular weight species upon allowing the
enzyme to remain in the presence of 0.12 p fi-mercaptoethanol
(Eastman) in urea and guanidine for more than 24 hr, all
synthetic boundary experiments were performed immediately
after dialysis. A detailed, careful study of the production
and effect of this low molecular weight material showed that

the material was not derived from the enzyme nor did it

13

influence the results.1

Sedimentation velocity experiments were run at 59,780
rpm at pa. 7°; exact temperatures were obtained from an rtic
unit. The diffusion coefficient experiments were performed
in a double-sector synthetic boundary cell at 4,059 or 4,90?
rpm at a temperature of 70. Diffusion coefficients were cal-
culated using height-to~area analysis (Schachman, 1957) and
converted, as were the sedimentation coefficients, to stand-
ard conditions of 200 and water.

Densities of the solvents were measured with a hydro-
meter. Viscosities of the solvents were interpolated from a
graph based on data from the International Critical Tables
(1929), while viscosity contributions of buffers and salts
were estimated from other refereLce tables (Svedberg and
Peterson, 1940). Frictional coefficients (fO/f) were calcu-

lated from D20 , values, using equations described by Tanford
9

(1961).
Urea (baker‘s analyzed reagent) used in the experiments

was recrystallized from 95$ ethanol. The crystals were dried

o . .
in an oven at 40 to remove reSidual ethanol. Guanidine

 

1The following considerations provide a partial basis
for this conclusion: (1) The effect is associated with mer-
captoethanol; dialyzed protein solutions containing urea or
guanidine also did not show the low-molecular weight species
after aging for several days, whereas those with mercapto-
ethanol did; (2) the low molecular weight material is not
derived from the protein; after several days axing of the
mercaptoethanol-containina urea or guanidine solutions of
protein described in (1), synthetic boundary analysis revealed
that the area under the curve had increased significantly,
with a concomitant appearance of the low molecular weight
species.

 

14
hydrochloride was prepared from guanidine carbonate (Eastman)
according to the method of Anson (1941). The dissociating
solvents involving urea and guanidine were always prepared
just prior to use. The dialysis casing was always boiled and

then stored in 0.001 p EDTA until used.

Optical Botatory Dispersion Analysis

The optical rotatory dispersion (OED) measurements
utilized a Jasco 03D/uv-5 Optical Rotatory Dispersion Recorder
equipped with standard CED cells. For analysis, aliquots of
aged2 enzyme solution were diluted to a concentration of pa.
0.45 mg/ml with one of the following solutions: (1) no urea,
(2) 1.5 g urea, (3) 3.0 ; urea, (4) 4.0 L urea, and (5) 5.4 L
urea. All solutions contained 0.04 L Tris—H01 buffer (pH 3.0),
0.15 L KCl, and 0.001 g EDTA. The enzyme solutions were kept
cold prior to testing. After rapid equilibration of the solu-
tions to room temperature, the solutions were injected into
5 mm standard CED cells for the measurements.

The parameter, be, was determined from the optical rota~
tion measurements in the 290 to 240 mp wavelength range. The
bO value (hoffitt and Yang, 1956) was obtained from the slope

of a plot of Ex] ()2 — XE ) XE )E//()2 -)E ), where Ed is the

 

2Pyruvate kinase analyzed within 24 hrs after isolation
has the following properties: 520 w = 10.07 3, D20 w = 3.95 x
10‘7 cmZ/sec, 1-w (s/D) = 246,000. This is in agreement with
Warner (1958) and represents the values :enerally assumed by
workers in this field. But analyses performed 10 days after
initial isolation indicate that the enzyme undergoes a change
in structure during the first 10 days of standing in the pure
state. The sedimentation coefficient of this "aged" enzyme is
g estimated as s30 W c .56 The use of the word "aged” simply
refers to any sample of purified pyruvate kinase which is over
. 10 days old.

 

15
xthe specific rotation at the wavelength, and is a constant,
.in this case chosen to be 220 my (Jirgensons, 1965). The

variation in refractive index of the solutions was neglected

(I.
l
i

as was the residual molecular weight corrections since these

were found not to affect the b0 value.
Amino Acid Analysis

General Procedure
Preparation of the enzyme for amino acid analysis was

performed according to the method of Crestfield g3 al. (1963).
Several 5 mg samples of enzyme, which had been desalted on a
Sephadex G-25 column and lyophilized to dryness, were weighed
directly into clean constricted Pyrex test tubes. To this was
then added 3 ml of a 1:1 mixture of concentrated HCl (Baker,
analytical grade) and deionized distilled water. The tubes
were evacuated to pg. 50 u, carefully frozen and thawed to
remove dissolved oxygen, and then sealed and hydrolyzed at
110°. Following hydrolysis, the solutions were transferred
to another test tube, and evaporated to dryness on a flash
rotary evaporator. The dried hydrolysates were then placed
in a freezer until analyzed. For the analysis, which was
performed according to the method of Loore and Stein (1954),

: the dried hydrolysates were diluted with 0.2 L citrate buffer,
pH 2.2. The amino acid analysis was carried out on a Spinco
Model 1203 Amino Acid Analyzer. Sufficient samples were pre-

) pared in this manner to allow duplicate samples to be hydro-

lyzed for 24, 48, 72, and 96 hr. All the amino acids, with

16
the exception of cysteine and tryptcphan, were recovered after
hydrolysis. A pyruvate kinase tyrosine/tryptcphan ratio of
2.45, determined by Kayne (1966) according to the spectro—
photometric method of Bencze and Schnid (195?), was used to

determine the number of destroyed tryptcphans.

Amino Acid Qalculations

The amount of each component amino acid in the sample
was determined by measuring the area of the corresponding
peak by the height—width method; this entailed measurement of
the peak height and multiplying the value by the half-height
width. The values were then converted to uncles by dividing
the area by the calculated constant (area/uncle) of that
particular amino acid. A new set of constants (CHM) was
determined for each ninhydrin solution used in the analyses.
Constants for cysteic acid and S-carboxymethylcysteine were
determined by averaging the constants of all the acidic amino
acids, with the exception of proline, and multiplying byo.97
and0.8€ respectively.

To calculate the umolar ratio of each amino acid to a
reference amino acid, four ordno acids were chosen as stand-
ards because of their stability to hydrolysis, i;g., lysine,
leucine, arginine and aspartate. Of these four standards,
aspartate was found to be the most stable and was therefore
chosen as a base upon which the relative amounts of the remain-
ing amino acids were determined at the various times of hydro-
lysis. In order to determine the relative amount of each

amino acid actually present in the enzyme, the values calculated

 

 

 

 

17

at the various times of hydrolysis were plotted as a function

of time. For those amino acids which showed a progressive
decrease with time, the true value was obtained by extrapola-
tion of the data to zero time. If the quantity of the amino
acid (iLg. histidine, valine, tyrosine, phenylalanine, and
isoleucine) gradually rose to a peak value, the limiting maxi-
mal value was taken to be the correct value. The extrapolated
values were then used to calculate the amino acid content of
the protein based on an enzyme molecular weight of 246,000.
The number of molecules of aspartate per native pyruvate
kinase molecule was adjusted until the molecular weight cal-
culated from the amino acid analysis was as near as possible
to the figure of 246,000. Total nitroren in the samples was
not determined. The molecular weirhts of the amino acids used
in the determination were correctel for the loss of water in

forming the peptide bond.

Eggparation of Performic Acid Oxidized Enzyme

To analyze the number of cysteine residues in the native
enzyme, performic acid oxidation of the enzym was carried out
prior to hydrolysis, according to the procedure of hoore (1962).
Performic acid was prepared by allowing a mixture of 1 ml of
30% H202 (Fischer) and 9 m1 of 38% formic acid (Baker) to
stand for 1 hr at 230. A 5 mg sample, prepared by the general
procedure was weighed directly into a constricted tube to which
was added 2 ml of performic acid solution. After standing for
4 hr at 00, 0.3 ml of 48S HBr (Baker) was added to remove

residual performic acid. The solution was then evaporated to

18
dryness on a rotary evaporator and 3 ml of 6 3 H01 added. The

hydrolysis of the protein was then performed as indicated in

the amino acid general procedure section.

firsparation of Reduced, Carboxynethylated Protein

Since the results of performic acid oxidation were
unexpectedly low, the carboxymethylation technique was also
used to determine the total sulfhydryl content of pyruvate
kinase.- Protein was reduced and carboxymethylated according
to the procedure of Crestfield et al. (1963). Guanidine

hydrochloride (4.3 a) was weighed directly into a 12 ml test

".J

('1‘

tube and to this was hen added 0.30 ml of 0.1 1 EDTA, 3.0 ml

of Tris buffer, pH 9.6 (5.23 c of Tris and 9 ml of 1.0 E HCl
diluted to 30 ml with E20) and 0.1 ml of p-mercaptoethanol.
Deionized distilled water and 15 mg of stock enzyme were then
added to give a volume of 7.5 ml.. A solution of 7 L guanidine
hydrochloride and 0.001 L EDTA was added to give a final
volume of 12 ml.. In order to exclude oxygen from the system,
nitrogen was bubbled through the solution before addition of
enzyme, and passed over tie solution after the addition of
enzyme. After Q hrs at room temperature, the solution was
transferred to a larger tube. To this was added a solution
of 0.263 g of iodcacetic acid (Eastman) in 1.0 ml of 1.0 E
NaOH which had been adjusted to pH 3.5 with concentrated haCH.
Nitrogen was again passed over the solution and the test tube
covered to exclude light. The reaction was allowed to continue
for 15 minutes whereupon the solution was dialyzed for 3 hrs

at 5° against frequent changes of 1000 mls of 50% acetic acid

 

 

(Baker). Subsequenth, Q ml aliquots of the dialyzed solution
were tra lS sfe red to constricted tubes and the acetic acid
removed on a rotary evaporator. d 31. -aliquot of 6 g. HCl was
added and the amino acid analysis carried out in the usual

manner 0

i“it‘ll-3- Titration

the

(’1‘
(D
"3’
,3.
:5
(D
P
U‘
C(‘i

-;-1‘. ‘ .3 ,; _ _ . p :‘ . _ f a '.
ine number of cySteine residue was ce

Six 5.0 n? stiflec of 'csalted lyophilized protein

were weiqhed directlv into t,st tubes to wLich were then added

2 ml of eithe: u.1 L Tris-I31 buffer (pH 7.0) or 6 L rres and
0.02 1 Iris-I”l buffs: (pt ”.0). Subs eqmc ntly, 7 ul of TEXE

Solution (39.? us DTII in 10 ml of 0.1 ; Tris—501 buffer. pL

7.0) was added to 0.5 ml aliquots of the protein solutions

"iocl1enicals) was used as

3
P
,C“
c
p
E
i
O
,1
(D
(D-
i )
(Z)
1—5
m
}.J

Reduced
a standard for the determination of the extinction coefficien“
(€) of Drug; a value of 6 = 9931 was oatained. Standard sluta-
tione solution was prepared by eddinm 1 ml of a glutatione
solution (0.027 mf/io nl) to 7.0 ml deionized distilled water

.0 n "7 F‘-

and 2. 0 :11 of 0.1 L Eris—301 bui:er, p; ..0. 10 this was then

added 0.02 ml of DT.I solution.

The catalytic activity of pvruvate kinase was routinely
measured using a sp ectrophotometrio assay; in certain experi-

ments, such as those inrolvin: an assay in the presence of

 

 

20

  
 
   
  
  
  
  
  
  
  
  
  
  
  
  
  
   
 
  
  
  
  

urea, a titration assay was used.

In the former method, the pyruvate kinase reactior was
coupled to the lactic dehydrogenase reaction and the decrease
in absorbancy at 340 mp was followed (Lubowitz and Ott, 1944).
Specific activity units were expressed as: umoles BABE con-
verted/min/mg protein. The final concentration of reactants

in the 0.4 ml assay media were as follows:10.0 2g LgClZ (Baker
Analyzed Reagent), 0.1 E 101 (Lerck), 2.0 mi ADP, sodium salt
(Sigma, from equine muscle), 0.1 g imidaZOle (Eastman) buffer,
pH 7.5, 6.3 EL LA‘H (Siana, from yeast), 14 PT lDH (Sigma, from
rabbit muscle) and 14 Hi PEP (Sijna). ADP solutions were
adjusted to pH 7.5 with tetramethylammonium hydroxide. Solu-
tions of PE?, NADH, and LDH were prepared fresh prior to each
series of experiments. A Beoknan DU Spectrophotometer attached
to a Gilford Lultiple Sample Absorbance Recorder was used; it
was set to a sensitivity of 1.0 CD units full scale and a

chart speed of 1 inch per minute. The temperature in the
cuvette was controlled by circulating water from a constant
temperature bath through thermospacers inthe cuvette compart—
ment.

For the assays using the titration technique, a Radio—
meter TTT-i/SER -SDU1/TTA31 automatic recordin: titrator was
used, followin: the assay procedure of Kayne and Suelter (1965).
Reagents used in the assay were obtained from the same sources

as mentioned above.

 

RESULTS

The Approach to the Problem

To determine the subunit structure of an enzyme, a
reliable molecular weight value for the native enzyme must
be known. Because only a limited study of native pyruvate
kinase had been made, it was desirable during the course of
this study to reexamine the physical properties of the native
enzyme, eSpecially its molecular weight. Due to the rather
rapid changes occuring in enzyme structure upon standing;2
(to be discussed later). molecular weight calculations had
to be based on sedimentation and diffusion coefficient data
rather than on sedimentation equilibrium measurements. In
addition, the amino acid composition of the enzyme was also
examined.

For the subunit studies, the enzyme was subjected to
a variety of dissociating agents to determine what conditions
were necessary to achieve complete dissociation into stable
subunits. Sedimentation coefficient analysis was used in
preliminary tests for the dissociation, but since this quan-
tity is a function of both size and shape, the effective con-
ditions had to be further tested using unequivocal determina-

tions of the molecular weight.

21

 

 

22

After clear proof that the dissociation was obtained
under a given set of conditions, series of molecular weight
analyses using sedimentation equilibrium techniques were
performed at several (8 - 15) different protein concentrations
to allow an extrapolation to zero protein concentration and
determination of a precise, unequivocal subunit molecular
weight.

In another phase of the research, attempts were made
to define the properties of several structural intermediates
which were found to exist at urea concentrations of 0.5 -
3.0 M; the lack of stability of these various intermediates
made unequivocal molecular weight assignments impossible.
Nevertheless, the factual results of this phase of the
research are of intrinsic interest and are therefore presented.

After these aspects of the study were completed, the
next stage of the research was to determine the conditions
necessary to reassemble the polypeptide chain into its native
structure upon the removal of the dissociating agent and to
define as precisely as possible the influence of the more

important variables upon the reversal process.
Characterization of Native Pyruvate Kinase

Determination of the Sedimentation Coefficient (sgo’w)

 

Freshly prepared enzyme was dialyzed for 12 hrs at 20
mg/ml against standard buffer solution (see Methods). The
enzyme was then diluted with the dialysate to concentrations
of 10, 8, 6, 4, 3, and 2 mg/ml and the sedimentation coeffi-

cients at the various protein concentrations were immediately2

 

23
determined (Figure 1).
The extrapolated value for the sedimentation coeffi-
cient of the native enzyme was found to be ng,W = 10.07 S
which is in excellent agreement with the previously reported

value of 10.04 S (Warner, 1958).

Determination of the Diffusion Coefficient (D3O W)

 

In order to calculate the mol wt of the native enzyme,
the diffusion coefficient of the enzyme was evaluated.
Freshly isolated enzyme2 was dialyzed for 12ium3at 10 mg/ml
against standard buffer solution. The enzyme was then diluted
with the dialysate to 9.3, 5.6, 4.0, and 2.0 mg/ml and the
experiments at the various concentrations completed within
twelve hours. The value of Dgo,w = 3.95 x 10'7 cmZ/sec

(Figure 2) found by extrapolation to zero protein concentra-
tion is in excellent agreement with the value of D066% =
,w

3.96 x 10‘7 cm2/sec (Warner, 1958).

Determination of Molecular Weight and Frictional Coefficient

O .—
20,w _

3.95 x 10'"7 cmZ/sec, the molecular weight of native pyruvate

Combining the values of ego W = 10.07 s and D
f

kinase was calculated to be 246,000, using a partial specific
volume (‘7)3 of 0.75 cc/g at 200. From the diffusion coeffi-

cient, the value for the frictional coefficient (fO/f) of the

 

3This is based on a value of v = 0.74 cc/g (Warner
1958) at 10°, which yields a value of v = 0.75 cc/g at 206
when a small correction (Taylor and Lowry, 1956) for the
temperature dependence of v is made. Warner's experiment
involved one experimental determination using a protein con-
centration of 10 mg/ml using a density gradient technique.
From our amino acid analysis we have calculated a value of
0.741 for the native enzyme.

24

AmHHMme Mom pawn oomv .Qoapflmsmap wnflmm mHSp moms pommmm SOHPMMpSmoCoo
m wo apaaanammoa map ummp 0p doawfimoo mpsoaaamaxo mo woaamm oHHSp m mpcom
Images omHm pH .om .mm pm a: mu Mom dome p53 .m>opm mm oHQEMm mahwnm meow
oSp pfiommamoa on mmaoaflo oHHom 6:9 .oowhamsm haopmaomssfl QmSp 6:6 Meghan
damosmpw pmsfimwm M: NH 909 owNszHo mm: Scans mahmsm mo soapmamgmaa Smoky
m pgmmogaoa on mmHOHHo moao one .oahnso comm one Sneak go memHOflwmooo

N
soapmpsmsaoom 6:9 “0 Soapoapsmoaoo :Hmpoaa osoN op Soapmaogmapxm .H ohsmfim

25

Figure 1

{

3.255 20.252828

 

 

o. m m an N o
q _ 4‘ _ _ mw
. -3.o~
m mm m - om - s
Immmm O
omo< O
- m

(6(5le

-m

0
(M‘OZSMNElIOIddI-JOO Nouvmawnoas

 

m8? . 20%|“) -

n — b

 

.aa\wa m use .: .m.m .m.m .OH op mummaamae spas empsaae we:

oaaNsm .o.m me .aaom a Hoo.o use .Hog a ma.o .ueeesp Homumfiee a :o.o pmgammm

/O
2

as NH Mom dmNaHme was mammsm omhmmoam hHSmoam .mmmcax opm>59hm mo psmflo

uammmoo Soamsemflo 6:» mo soapohpsmosoo Sampoam OHmN op Soapmaogoapxm .N madem

 

27

Figure 2

3.28% 20.252328

 

we me No «m m.
— — — — _
Tommdzowo. x 8» " ammo
AHYIAUTiw nu (JIIAUlliw (IAVJIri

 

 

m2>sz m>_._.<z

p

F

—

 

 

<r N) (V

(“30) 1N3I0l:1:1300 NOISfldle

ID

28
native enzyme was calculated to be 1.32. Both molecular
weight and frictional coefficient are in good agreement with
those previously reported. Although pyruvate kinase analyzed
immediately after preparation has a mol wt of 240,00q,data to
be discussed later indicate that structural changes occur in

the enzyme upon aging.2

Alteration of Pyruvate Kinase Upon Aging2

In our survey of the literature, prior to initiation
of experiments, we noted a reported sedimentation coeffi-
cient of 320,w = 8.6 S at a protein concentration of 2.5
mg/ml (Tompkins and Yielding, 1962). It was evident that
this was too low to correspond to the freshly prepared2
enzyme, which should have a minimum sedimentation coefficient
of $20,w = 9.3 S at concentrations below 6 mg/ml, based on
the concentration dependence shown in Figure 1. The dis-
crepancy in these values is far outside experimental error
which is usually about 2%.

Because of the desirability of knowing the structure
of the native enzyme as accurately as possible, it seemed
necessary to investigate this observation further. One
plausible explanation for the observed behavior was that the
enzyme might undergo structural changes with time after iso-
lation.

Therefore, a study was initiated to compare the
extrapolated sedimentation coefficients (Sgo,w) obtained 24
hrs after isolation with Hume obtained 3% days after isola-

tion. Figure 1 shows the sedimentation coefficient data

29
determined from this comparison. It is seen that a struc-
tural change had definitely occurred with time, since the
extrapolated sedimentation coefficient of the 24 hr material
was Sgo,w = 10.07 S while that of the 3% day "aged"2 enzyme
had decreased to Sgo,w = 8.53 S.

It was also necessary to consider the possibility
that enzyme concentration may be an important factor in the
observed structural alteration. The experiment described
in Figure 1 were designed to also allow us to investigate
this possibility. The lower line in Figure 1 represents two
experiments, in which the sequence of dilution and "aging"
were reversed. The following scheme represents the proce-

dure in the experiment:

 

 

 

 

CONCENTRATED
FRESH (20 mg/m%) >
ENZYME Aged 3% days
DILUTION DILUTION
BEFORE AFTER
AGING DILUTED AGING
\V
‘r (10 mg/ml or less) \ RUN
Aged 3% days ’

Enzyme which was aged 3% days while still concentrated
(20 mg/ml) and diluted just before running gave identical
results to enzyme which was diluted first, aged 3% days in
the diluted State, and then run. This finding indicates that
the structural alterations occuring in the enzyme are not

dependent upon enzyme concentration.

‘\-

 
 
 
 

 

 

 

30

Amino Acid Analysis

The next step in the characterization of the native
enzyme was an investigation of its amino acid composition.
Table I shows the ratio of each amino acid to aspartic acid
calculated at a given time of acid hydrolysis (for procedure
and calculations see Methods). The first column of numbers
from the left, representing the number of molecules of each
amino acid per native pyruvate kinase molecule (mol wt
246,000), was obtained by multiplying the ratio values, given
in the second column from the left, by an arbitrarily chosen
aspartate plus asparagine residue value of 201.5; calcula-
tions based on this value gave a total amino acid residue
weight of 246,000. In computing the amino acid composition
(Table I) a cysteine/aspartate ratio of 0.185 was used; this

value was determined in separate experiments described later.

Sulfhydryl Determination

Because cysteine was completely destroyed during the
acid hydrolysis step of the general amino acid analysis, it
was necessary to use other methods to determine the half-
cystine content. In the next procedure, pyruvate kinase was
oxidized with performic acid (Moore, 1963) to yield cysteic
acid, a derivative which is completely stable to hydrolysis.
Duplicate samples of performic acid oxidized enzyme yielded
an average cysteic acid/aspartate ratio of 0.0199 after 24
hr hydrolysis. Based on this value, the half-cystine content

of the enzyme was calculated to be 4 residues/mol wt 246,000.

FE

 

31

Table I. Amino acid analysis of rabbit muscle pyruvate kinase.

 

amino acid

number of

extrapolation to zero time ratiosa

 

 

 

 

residue residues/ residues/aspartate residue
moleculee time of hydrolysis (HoursTfii
0 24 48 72 96
Aspartic acida 141.4 1 1 1 1 1
Asparagineb 60.1 - - - - -
Glutamic acid 160.5 1.095 1.047 1.094 1.201 1.097
Glutamineb 60.1 - - - - -
Lysine 160.6 0.797 0.792 0.783 0.789 0.756
Histidine 61.5 0.305 0.219 0.249 0.307 0.328
Arginine 131.0 0.650 0.647 0.627 0.631 0.615
Threonine 114.4 0.568 0.490 0.516 0.485 0.453
Serine 123.9 0.615 0.552 0.515 0.441 0.400
Proline 90.5 0.449 0.448 0.405 0.415 0.448
Glycine 174.1 0.864 0.817 0.887 0.839 0.800
Alanine 239.8 1.190 1.171 1.201 1.175 1.183
Half—cystine 37.2 0.185C 0.192 0.178 - —
Valine 191.8 0.952 0 876 0.946 0.943 0.943
Methionine 69.5 0.345 0.363 0.325 0.358 0.332
Isoleucine 150.1 0.745 0.668 0.736 0.729 0.727
Leucine 175.3 0.870 0.818 0.829 0.813 0.809
Tyrosine 33.2 0.187 0.187 0.166 0.196 0.191
Phenylalanine 67.3 0.334 0.334 0.278 0.353 0.342
Tryptophan 13.8 0.069d - - - -

 

aAn arbitrary value of 1 was given to aspartic acid

residues in order to calculate the ratios of the

remaining amino acids.

bTo take into account the 120 amide groups per molecule,
it was assumed that these were 9a. equally distributed
between the glutamic and aspartic residues.

CValue was not extrapolated, rather it represents an

average obtained from the S/carboxymethylcysteine

analysis.

dValue calculated from a tyrosine/tryptcphan ratio of

2.45.

eAssuming 246,000 g enzyme/mole; this column obtained by
multiplying the zero time column ratio by 201.5, the

total number of aspartate plus asparagine residues per
molecule of native enzyme.

32

This seemed to be an unusually small number of sulfhy-
dryl group for such a large enzyme molecule. One possible
explanation of the data was that the oxidation of the native
enzyme had been incomplete, due to the inaccessibility of
sulfhydryl groups buried in the interior of the molecule. If
this were the case, denaturation of the enzyme prior to reac-
tion with sulfhydryl reagents would tend to expose such hidden
sulfhydryl groups.

To test this possibility, native and urea-denatured
enzyme was reacted with the sulfhydryl reagent 5,5'-dithiobis
(2-nitrobenzoic acid) (DTNB) according to the method of Diez
§§_al. (1964), as described in Methods. Immediately upon the
addition of DTNB to the urea denatured enzyme, the optical
density (GB) at 412 mp was continuously recorded and subse-
quently compared to that of controls containing DTNB or enzyme
alone. After an observed large immediate increase in 0D upon
the addition of DTNB, the 0D did not change significantly over
a period of 20 min. An average cysteine content of 24.7 was
calculated for the urea denatured enzyme based on an enzyme
molecular weight of 246,000.

Upon the addition of DTNB to native enzyme, a small
increase in CD was observed immediately, followed by a slow
continuous change in optical density over a period of 30
minutes. Examination of the test sample revealed that the
protein was slowly precipitating. This suggested the exist-
ence of two classes of sulfhydryl groups, i.e., groups

reacting very rapidly and groups reacting rather slowly. The

33

increment in 0D produced immediately corresponded to a value
5.8 sulfhydryl groups titrated per molecule of native
pyruvate kinase (mol wt 246,000).

It was clear that the values of 25 half-cystines
obtained from the DTNB titration of urea denatured enzyme
disagree with the value of 4 obtained from the performic acid
analysis while the value of 5.8 half-cystine molecules obtained
with non-denatured enzyme agrees rather well with this value.
Presumably the fast reacting groups are on the surface of the
enzyme and the slow reacting groups are buried somewhat in the
interior of the molecule.

In order to determine a more reliable value for the
amount of total cysteines in the pyruvate kinase molecule,
an amino acid analysis was performed on enzyme reduced with
0.1 fl B-mercaptoethanol in 7 M_urea and reacted with iodoace-
tic acid to form the S-carboxymethyl derivatives of cysteine
(see Methods). A ratio of S-carboxymethylcysteine/aspartate
of 0.185 was calculated from the amino acid data; the value
corresponded to 37.2 half-cystine residues per 246,000 g of
enzyme (Table I). This quantitative value agrees with the
range (30-40) previously reported by Mildvan and Leigh
(1964). This value was used for the amino acid composition
calculation (Table I). A summary of the results obtained
from the various methods of sulfhydryl analyses is given in

Table II.

34

Table II. Half-cystine content of rabbit muscle pyruvate

 

 

 

 

kinase.
Enzyme Structure Molecules of half-cystine per
molecule of native pyruvate kinase
Method of
Determination
of Half-cystine Native Denatured
DTNB Titration 5.8 24.7
PHMB Titrationa - 30 - 40
Performic Acid Oxidation 4.0 -
b
Carboxymethylation - 37.2

 

aThe value was obtained by Mildvan and Leigh (1964).

bThis is the most reliable value.

Partial Specific Volume of Pyruvate Kinase

Although the estimation of the partial specific volume
of a protein from its amino acid analysis is not theoretically
rigorous, the empirical calculations have been found to agree
quite closely with experimentally measured values for several
proteins (McMeeken and Marshall, 1952). This method (Cohn
and Edsall, 1943) was therefore used to calculate the partial
specific volume of pyruvate kinase. The specific volume (VP)

Wa o - — -
3 calculated from the equation Vp -2V1Wi/zwi where W,L is

35

the percent by weight of the 134 amino acid residue and V1 is

the specific volume of this residue. The molecular weights
of the amino acid residues were calculated from the molecular
weight values given in the Handbook of Chemistry and Physics
(41§£ Ed, 1958). The corresponding specific volumes of the
amino acid residues were taken from the work of Cohn and
Edsall (1943) with the exception of half-cystine for which
the value calculated by McMeeken g§_al. (1949) was used.
These calculations (Table III) yield a value of 0.741 cc/g
for the partial specific volume of rabbit muscle pyruvate
kinase. This value is in good agreement with an apparent
partial specific volume of 0.74 cc/g determined at a protein
concentration of 10 mg/ml at 100 by density gradient tech-

niques (Warner, 1958).

Characterization of the Dissociation Process

Dissociation as a Function of Urea Concentration by Sedimentation
Coefficient Analysis

A detailed analysis of the structural transitions of the
enzyme as a function of urea concentration was also performed.
For this study, a dialyzed stock native enzyme solution was
diluted with concentrated urea solutions to various final urea
concentrations ranging from 0.5 M to 6.0 N. .All solutions con-

tained 0.15 M_KCl, 0.04 M Tris-HCl buffer, 0.001 M EDTA, and

36
Table III. Calculation of the Apparent Specific Volume of

Pyruvate Kinase from Its Amino Acid Composition

 

 

Amino acid Number of W1 7 Wivi
residue residues/ (% by weight (spec1fic (% by
molecule of residue) volume of volume of

residue) residue)
Aspartic acid 141.4 6.61 0.59 3.900
Asparagine 60.1 2.79 0.60 1.674
Glutamic acid 160.5 8.42 0.66 5.557
Glutamine 60.1 3.14 0.67 2.104
Lysine 160.6 8.37 0.82 6.863
Histidine 61.5 3.43 0.67 2.298
Arginine 131.0 8.32 0.70 5.824
Threonine 114.4 4.70 0.70 3.290
Serine 123.9 4.39 0.63 2.766
Proline 90.5 3.57 0.76 2.713
Glycine 174.1 4.04 0.64 2.586
Alanine 239.8 6.93 0.74 5.128
Half cystine 37.2 1.56 0.63 0.983
Valine 191.8 7.73 0.86 6.648
Methionine 69.5 3.71 0.75 2.783
Isoleucine 150.1 6.91 0.90 6.219
Leucine 175.3 8.07 0.90 7.263
Tyrosine 33.2 2.21 0.71 1.569
Phenylalanine 67.3 4.03 0.77 3.103
Tryptophan ALE: _1_-_0_i M _0_._7_zg
Total = 2256.1 2111:9958 iwivi=74.04i

ilini/Ewi = Vp = 0.7406 cc./g.

37
had a pH of 8.0 at 25°. The final enzyme concentration was
6 mg/ml for each sample. Except for the special cases men-
tioned in the legend for Figure 3, these were then dialyzed
in the cold 1s. the appropriate urea solvent for ga. 12 hrs.
Sedimentation velocity experiments on these samples at vari-
ous urea concentrations were immediately conducted. Figure
3 shows that marked structural changes in the enzyme first
occurred in 1.5-1.7 M urea yielding Intermediate II. Here a
single sedimenting peak was observed, with a sedimentation
coefficient of sgéég = 7.3 S, in contrast to a value of 8.9
S in 1 M_urea (Intermediate I). Standing several hours in
1.6-1.7 M urea yielded partial conversion of the 7.3 S species
to a 3.6 S product (Intermediate III). Thus, in 1.6-1.7 M
urea the equilibrium clearly allowed the 3.6 S species,
although the kinetics were so slow that only the 7.3 S
material was observed initially. This illustrates the general
slowness of attainment of the equilibrium distribution of
products which was observed in the 1.5 to 2.5 M range of urea
concentration. The Schlieren patterns showing the shape of
the curves and the distribution of products for this system
are shown in Figure 4. Photographic Plate A illustrates the
homogeneous peak observed immediately upon dilution of the
enzyme into 1.5 or 1.7 M urea while Plate B shows the two
peaks observed when enzyme was incubated in 1.6 M (lower) and
1.7 M (upper) urea at 36 and 72 hours respectively. In 3 M
urea, a single peak (Intermediate III) was formed which

0.6%

possessed an 820,w of 3.6 S, representing the same product

.mopooam HdHSOprdm pdSp md psomoaa :pdeHa do SOHpOdaw dSp op HdQOppaoaoHQ
mp SOHMoa dos: a m.Nlm.H map SH woadawm end mofioppo oSp mo wsfipdSm Ho doamop

0:9 .pxdp oSp Sp ponflaommp md popdmppmmpsfi odd as NH Mom pdnhadpp dams moadadm

38

adSpo HH< .hs Nu amped 25a md2.A UV dmas a ©.H pd pdSp oflpss .ddas opsp Cowpsapo
Moped hampdppmssp SSH dams dons a m.H pad m.H pd moaasdm .ddas a m.a pad w.H
.m.H pd mpsoapaddxo omosp pom .QOHpodop esp do moppospx esp dpdSHdpo op addao Sp
.pomaadpp pos dds dsaNso ope .omdo zodo SH HE\ms m mds Soapdapsdosoo SpopOHa use

.omdspx dpdbsaaa mo mapposspm oSp So :odeppSooQoo dos: mo poopmo one .m dpswpm

39

Figure 3

>._._m<...0_2 4mm:
N. m n v m N _

O

 

 

. _ . 4‘ _ A, a _ (A _ _ _ _ _ .

 

I
I

n? b r 0, TI. mtzamnm

I

an 34.82222. III. qumlblm
a

p
v
'

¢
o
o
c
o

a
H 34.8282. I|.. PIP-f.

H 34.82252. ll VI...

_ _ _ _ _ _ p _ _ _ _ _ _ _ _

 

.9

 

(22238) 1113033300 N0I1Vi~3wms

Figure 4.

40

Sedimentation patterns of pyruvate kinase in 1.5-1.7

M urea at zero time and after 2

A.

upper:
lower:

upper:

lower:

enzyme
enzyme
enzyme
50

enzyme

at pa.

in
in

in

in

1.5
1.7
1.7

1.6

and 3 days.
M urea at zero time
M urea at zero time

M urea dialyzed for 36 hr at 93.

M urea left standing for 72 hr

Figure 4

41

 

42
formed in increasing amounts in 1.5-2.5 M urea; this product
in 3 M urea converted to the slower sedimenting species upon

aging. In 4 M urea the maximum structural change possible

0.6% _
20,w -

1.7 S) observed at this concentration closely approximated

had occurred since the sedimentation coefficient (s

that observed at all higher concentrations of urea (Figure 3),
including 7.4 M urea (Figure 15). Typical schlieren pat-
terns observed for enzyme in 0.5-6.0 M urea are shown in

Figure 5.

0.6%
20,w

both of the size and shape of each enzyme species, the

Since the s values presented above are a function
decreases in sgé’f, described in Figure 3, may be a result
of dissociation or unfolding. To distinguish between these
two possibilities, another technique was needed which would
show changes in only one of these characteristics. The ana-
lytical tool chosen for this purpose was optical rotatory
dispersion, which for protein molecules, is primarily a

function of the degree of folding.

Missociation as a Function of Urea Concentration by Optical
Botatory Dispersion Analysis

To obtain a measure of the changes occurring in shape
(i.§. degree of unfolding) as a function of urea concentration,
the optical rotatory dispersion (OED) properties of the enzyme
were studied. In conjunction with the sedimentation analysis
previously described, such a study would allow classification
of the structural alterations into two groups: folding and

mass changes.

Figure 5.

43

Sedimentation patterns

function of urea concentration.

A.

upper:
lower:
upper:
lower:
upper:
lower:
upper:

lower:

enzyme

enzyme

_enzyme

enzyme
enzyme
enzyme
enzyme

enzyme

in

in
in
in

in

0.5 _

3.0
1.0
2.0
2.0
4.0
6.0

M
E
M

M

of pyruvate kinase as a

urea
urea
urea
urea

urea

_ urea
_ urea

5.0 -

urea

44

Figure 5

 

45
For the studies, the optical rotatory dispersion of the
enzyme was measured at several urea concentrations (1.5-5.4 M)
and the data plotted according to the method of Moffitt and
Yang (1956). The parameter bO was calculated from the slope
of the line obtained from such a plot. Since the value for.

b is known to decrease from -6300 to 00 ( )0 = 212 mu) with

o
decreasing 4,-helix content (Yang and Doty, 1957), a change
in the b0 value of the protein would be indicative of a struc-
tural alteration. The bO value calculated for rabbit muscle
pyruvate kinase in deionized distilled water at 230 was
reported to be -1200 (3.0 = 220 mu) (Jirgensons, 1965).
Although our studies were not performed on enzyme just
prepared,2 the results should probably apply for the native
system as well as the aged2 system. The bo value for "native"

2 enzyme was found to be -1080; for enzyme in 1.5, 3.0,

aged
4.0 and 5.4 M urea, bO values of -93°, -860, 00 and 00 respec-
tively were found (Figures 6 and 7). The discrepancy in the
b0 value for native enzyme reported above (-1080) and that
reported previously (~120) (Jirgensons, 1965) was probably

due to differences in solvents; it may also have been due to
the use of two different enzyme forms2 by the two investiga-
tions. Since the b0 values of pyruvate kinase in the urea
concentration range 0-3 M changed only slightly, it was
qualitatively concluded that the degree of unfolding of the
various species was approximately the same, i.e., no pro-

nounced structural change occurred. The marked decrease in

bo to 00 in 4.0 M urea indicated the enzyme had been completely

46

ddhd a m.a CH oahngm "m sadao

.ddas o: Sp damwsm u< smdso

.omm .8 ad .208 pwv 9m mm :88 2 80.0 .82 a. 3.0 .883 Saundra a 8.0

“pmspdpsoo mmadsdm Spom .dth z m.H mo doaowpd pad mosomoam exp Sp omdQHM

mpdpsazd mo gonaoampp aaopdpop Hdoppao an mamaHde HdQOdeaaopsoo

.w dhfimdm

47

Figure 6

 

 

 

 

 

 

 

<._.om _2 .00.
6x29. <._.om_2_oo.
_qum_m._.ZmO. flux—20..
4mm: .2 m._ _o InmE... .2 m0.
wN NN m. ¢_ 0. m wN NN m. S O.

.o. x a. 0.2 we}...

O¢

.dchs on a :.m pad on

48

.mmhs

E

z

0.: Ga mfihwsm “m sgdao

o.m mp dammsm ”4 :Qdao

.omm .8 as .208 as; ed mm 498 a. 80.6 .8: 2. 3.0 {885 82-3.5 a 8.0

upwspdpsoo owad moadsdm one .doh: E :.m pad .o.: .o.m mo oosommam oSp SH omdspx

opdppphm do 20pmhddmpp hsopdpoa adeppao an mpmhfidfid HdQOdeaaomsoo .m osswpm

119

Figure 7

 

OOMOD 00ml "OD

 

 

 

<Pow .2 _OO. 4.5m _2 _00. «from .2 _OO.

flux-29. _oxég. _Uv._20_.

_o I m_m._. .2 no. .0 I 2m... .2 mo. _0 I mE... _2 mo.
4mm: _2 9&0 a 4mm: _2 O...» 0 4mm: .2 O.m

 

 

mN mm w. .v. 0. m 0N NN m. .V. 0.
F0. x w o~< $90.5

50
unfolded; the data indicated the absence of detectable d.-helix
and,presumably, of any ordered structure.

When related to the decreases in bO values at the same
urea concentrations, the observed decrease in the sedimentation
coefficient of the enzyme with increasing urea concentrations
suggests that molecular weight changes are occurring in the
enzyme in the 1-3 M urea range but that the gross structure
of the enzyme is not markedly changed until a urea concentra-
tion of 4 M or greater is attained; here complete unfolding is
presumed to occur.

Studies were next undertaken to determine the molecular
weight of the enzyme intermediates present in 0.5, 1.5, and
3.0 M_urea. Since the half-life of some of the intermediates
was very short, molecular weights were determined from sedi-
mentation and diffusion coefficients rather than by sedimen-
tation equilibrium techniques. To ensure a precise determina-
tion of the sedimentation and diffusion coefficients, experiments
were performed at several protein concentrations and extrapolated

1K) zero protein concentration.

Determination of the Sedimentation (s 0,w) and Diffusion

 

(D20 w) Coefficient of Pyruvate Kinase in 0.5 M Urea
’

 

Enzyme in buffer was diluted with a stock urea solution
to a final concentration of 10 mg/ml in 0.5 M urea, 0.04 M
Tris-HCl buffer, 0.15 M KCl and 0.001 M EDTA (pH = 8.0 at 23°).
The protein solution was then dialyzed against the same solution
for 9a. 12 hrs at 5°. After appropriate dilution of the enzyme

solution with the dialysate, the sedimentation and/or diffusion

51

experiments were conducted as soon as possible; the experi-
ments were always finished on the same day they were begun.

The experiments described above were conducted twice
to ensure the validity of the results; the 820,w values were
found to be in excellent agreement, but the D20,w values
exhibited a significant variation (Figure 9). Based on a
value of Sgo,w equal to 8.41 S (Figure 8) and a tentative
value of D20,w equal to 4.4 x 10"7 cmZ/sec (Figure 9), the
molecular weight of the enzyme in 0.5 M urea was calculated
to be 184,000. It must be emphasized that this is only an

approximate value which may be in error or may represent the

average of a mixture of products.

0

20.W) and Difquion

Determination of the Sedimentation (s

(D20,w) Coefficients of Pyruvate Kinase in 1.5 M Urea

 

Enzyme in buffer was diluted with a stock urea solution
to a final concentration of 10 mg/ml in 1.5 M_urea, 0.04 M
Tris-HCl buffer, 0.15 .111 KCl and 0.001 M. EDTA (pH = 8.0 at 23°).
The protein solution was then dialyzed against the same solu—
tion for 4 hrs at 9a. 50. Sedimentation and diffusion coeffi-
cients were either immediately determined or the enzyme solu-
tion was allowed to remain at a concentration of 10 mg/ml for
72 hrs at 50 before the molecular weight analyses were performed.

The variability of the sedimentation and diffusion data
indicated that the system was quite unstable (Figures 10 and 11).
The experiments presented in Figure 10, in addition to the other
experiments conducted in the same manner, showed a variability

0.6%

in 820,w from 6.6 S to 8.0 S. Also the strong concentration

.omm aw o.w n ma .spom a

52

poo.o .p02 a mp.o .socess pomunpsp a 20.0 "oospwpsoo owpw sopwaw psoepow tsp

.ddpd op QOpppppd SH .dmps a m.o Sp mmdspx dpd>5phm po As.ONmV mpsopoppmooo

QOppdpsoEppom 63p mo Ceppdppsdozoo :pdpopg opmN op QOppdponpme .m opsmpm

53

Figure 8

 

 

 

 

 

l T l l
— -s.>
O
" -00
q
LIJ
if,
h- ‘0
2 “i
U)
o’ ‘2
q-
- O (15 "Q’
n
3
is“
— C (D “N
‘ fol/l
1
* O
O m 00 I" LO

(szsuwalcli-iaoo NOIlViNEJWIOZ-IS

CONCENTRATION (MG/ML)

54

.omm pw o.m so ma

“spam 2 Hoo.o ass p02 a mp.o .soecss pomuwpse.p no.0 oospwpsoo oops psmepow

03p .dmps op Sepppppd SH .ddpa a_m.o Sp dmdpr opdbppam mo As.ommv mpSmpo

ipppdoo QOprmppp mSp go QOppdproozoo onpopm opdN op QOpdeOdeme

.m opsmpm

55

Figure 9

3.28.): 20.252828

 

 

 

 

No. N .m No I» .N
. _ _ _ _
. .. 3.0m
_uUmm INZUFIO_ XNV .V .. OnQO
O O O Ololfl
O
0
4mm: 2nd

b -

 

 

<l' N) (\I

(”30) 1513005300 NOISHdAIG

IO

56

.Aomm pd o.m so may spam

2 poo.o osw .poa a mp.o .sosess pomnmpsa a no.0 oospspsoo onpw soonah psoepow one
.woas op soaspood sp .om aid as Asa as we nosed one A. .08 upmapwpo as a soaps
zpmpdppmaap pdpdwppmm>sp mdz dammzo 0&9 .dmps.a m.p Sp mmdspx mpdpapmd po A3.0va

mpzopOpppmoo QOppdpgoappmm oSp p0 QOppdppsdosoo opmN op SOppdpoadpme .Op mppwpm

57

Figure 10

3.28.): 20.252828

m

 

 

- p

4mm: in;

 

-

 

 

O
(Mme) .LNEJIOIddf-JOO NOIlVlNBWlOI—IS

58

.Aomm pd o.m co adv spam
a poo.o oss poa_a mp.o .smcass pomuwpse a 20.0 oosswpsoo owpw smpwan psmepow one
.dmps op SOpppppd mp .As .0 .ov mQOppproo mo pmm dadm oSp Hoods ppd .opda opms
mSOpdepapmpmp pdSpppppzp dopze .ddps_m m.p :p omdspx mpdbppmm p0 As.ommv pcopo

Ippmooo QOpmSpppp osp mo SOppdppgmoQoo Spopopm opoN 0p QOpdeOdeme .pp opsmpm

59

Figure 11

3.20.2. 20.252828

0..

 

_.m MN in mam
p p p

d

. .. 3.0m
70mm #2090. Ch I» .. 0.0a

 

—

 

r h P F

 

 

 

V ['0 N

(Mme) 1513003300 Nousnssm

ID

60
dependence exhibited by the samples analyzed immediately (open
and closed circles - Figure 10) decreased considerably after
the samples were incubated for 72 hrs (hatched circles -
Figure 10). The reasons for this behavior are not known. The
diffusion coefficients were observed to vary considerably,
especially at lower protein concentrations; in some instances,
D20,w values ranging from 1.5 to 6.6 x 10'"7 cmZ/sec were
observed. The possible molecular weight range indicated by
these data (Figures 10 and 11) is 120,000-190,000; an unequiv-
ocal determination of the molecular weight of this species

must await further work.

Estimation of the Enzymatic Actiyity of the Pygpvate Kinase
in i;7 M Urea

In order to test for enzymatic activity of the enzyme
in 1.7 M urea, the enzyme was first incubated for 93. 1 hr in
1.7 M urea at a protein concentration of 0.1 mg/ml. Then,
aliquots of the enzyme solution were added to 1 ml of assay
media containing 1.7 M urea, 0.008 M MgClZ, 0.002 M ADP,

0.001 M PEP and 0.1 M KCl (pH 7.5). The reaction was followed
on a recording titrator (see Methods).

Activities in the range of 50-90% that of the native
enzyme were determined, based on the initial rate of activity.
IFrom these results it was concluded that the form of the
enzyme present in 1.7 M urea was catalytically active. The
Ikbssibility that dilution of this species to assay concentra-
tixbns and conditions caused it to be converted to a catalyti-

Oailly active state seems extremely unlikely.

 

61

Studies of the Molecular Weight of the Enzymeggn 2.5 and 3.0
M Urea

Several attempts were made to determine the molecular
weight of the enzyme in 2.5 and 3.0 M urea (sgbéé = 3.6 S).
In all cases tested, the presence of a contaminating species
(sgb?§ = 7.6 S) made accurate determinations of the sedimen-
tation and diffusion coefficients impossible.

A summary of some of the physical properties of the

various intermediates studied is given in Table IV.

Characterization of the Subunits of Pyruvate Kinase
The sedimentation equilibrim technique was used to
provide the most precise determination of molecular weight
values of the subunits. The apparent molecular weights were
extrapolated to 0 protein concentration to eliminate the
effects of protein interactions upon the calculated molecu-
lar weights. All phases of the experiments were conducted

at pg, 50.

Molecular Weight Analysis in Urea

Initial attempts to dissociate the enzyme into small
subunits involved a series of short column sedimentation
equilibrium experiments utilizing 7.4 M urea as the disso-
ciating solvent. The results shown in Figure 12 extrapolated
to values of 60,600 and 75,700 for the weight-average (M3)
and z-average (Mg) molecular weights, respectively. This
result proved that rabbit skeletal muscle pyruvate kinase
consisted of subunits smaller than 150,000 mol wt, and prob-

ably contained 4 subunits of about 60,600 mol wt. However,

62

.mpSmSppmaxd do moppom 03p pmdop pd Sp opppospopddp mp0: mmSpd> mmmSBo

.ploommao N...o.p dad mppSDor

.SOOS apppppdppop de appppnpoSpopdop de ompdp dds moSpd> po odep
oSp “apSo mmpdappmm stop apmb md poppdwop on ppSOSm mmmmSpSdea Sp moSpd> oSBd

 

s o00: a n n u m 0.m a 0.0-0.0
000.00p 0.0 0 s.0
op 0p op
+ one- Amm.av 000.0ma 0.: 0 s.a an 0.0-0.00 m_m.a
+ u 20N.p. 2000.00a0 as.a0 om a.0 on 0.0 .m 0.0

 

   
      

.oSoo
dds:

appappos o o 00 2.0m 2.0m s.0m dependaosm
appdaaNSm S p\ m Ammvz S 0S 00 m depmaSm

 

 

d.S0ppdpoommpp mmdpr opd>SpaQ go mdpdppmapopSp 0Sp po mdpppmaopm .>H mande

63

.SQS 0p:.Mp pd pSo popppdo mds SOppdeppppSmo SSppppppSSm .00

6.0080 ads 830083 2.0 .omm p0 0.0 mes as one. 0:0 .0000 a... 80.0 .802 a 3.0
.Smpppp pomumppe a 20.0 poSpdpSoo ompd Sopmmm deppom oSp .dmpS op SOpppppd

SH .admz Sop A90 SUV de dam: pop N\Apo SUV md popdSpdpm opms mSOppdpdeoSoo
.doSS a 3.0 Sp omdpr mpdpSpad p0 mpSwpms pdpSompoS mwdpopdlm de Admmzv mwdpopd

npSwpds pSdean dSp po SOppdppSmoSoo Spmpopa oSmN op SOppdpoadprm .Np opawpm

60

Figure 12

APPARENT MOLECULAR WEIGHT

35(9): ZO_._.<m._.zmozoo

ON

0. N. m .V
_

 

0.00.00 ,

OOOdN

comm.

CORN.

 

OOHmh u

 

N
0.62

 

000.00 "obs.

4mm: .2 ¢N

L) _ h _

 

 

W
gOIX T

65
the discrepancy in weight- and z-average molecular weights
needed clarification since this could have been caused by
either of two possibilities; namely, (1) aggregation, or
(2) subunits of different molecular weights. Furthermore,
the possibility of even smaller subunits, covalently linked

by disulfide bonds, needed to be tested.

Molecular Weight Analysis in Urea and M-mercaptoethanol

 

The dual need for a dissociation solvent which would
prevent aggregation and also break disulfide bonds led us to
study the effect of mercaptoethanol in the urea dissociation
system. For this purpose, our next dissociation solvent
possessed the disulfide bond-breaking ability of 0.12 M
M-mercaptoethanol in addition to the noncovalent bond-breaking
capabilities of 7.4 M urea. As seen in Figure 13, extrapola-
tion of the apparent molecular weights determined at the
various enzyme concentrations yielded values of 54,300 for
M8 and 55,600 for M2. The subunit molecular weight results
in 7.4 M urea and 0.12 M M—mercaptoethanol were approximately
the same as those obtained in 7.4 M urea alone, indicating

the absence of inter-subunit disulfide bonds.

Molecuiar Weight Analysis in Guanidine Hydrochloride and

M-mercaptoethanol

 

Examination of the molecular weight values obtained in
the urea-mercaptoethanol solvent indicated some inherent
instability, as manifested by the slight scattering of the

molecular weight values (Figure 13). In order to test a

66

.SSS 0p:.Mp pd pSo pmpppdo dds SOppdeppSpSmo asppppppavm

.os .00 was osspwsdasms 0:0 .2000 000 0.0 was ms was new .epom_a p00.0
.pOdz a mp.0 .pmppSp pomumppe a 30.0 pmSpdpSoo ompd Smpmam pSobpom oSp .poSdSpe
nopswosossm 0:0 was: op sopppoow sp .asms pop 200 000 osw 00”: S00 0\Ano 000
md pdpdSpdpd mpms mSOppdppSooSoo .poSdSmepadopmaLm_m Np.0 de dmps_m 2.0

Sp omdSpS opd>5pad mo mpSwpms pdpSoopoS Adawzv mmdpopdlw 0Sd Amdmzv mwdpmpd

npSwpds pSdeamd mSp do SOppdppSmoSoo Spmpopm opoN op SOpdeOdeme .Mp opSmpm

67

APPARENT MOLECULAR WEIGHT

Figure 13

35:05: zo...<m...zwozoo

 

 

N. O. m m .V N
_ _ d _ A _
I .

. a
I. I
. I N . I 3 I

i 000 00 .. 0.02 000 v0 .. 08....
I N: . I
I 32 0 II.

I

I ..OZ<I._.MO...n.<omms. .2 N_.O .4mm: .2 Th

L _ . b _ _

 

 

 

68
system in which the subunits might be more stable, and to
further ensure that the native enzyme had been completely
dissociated into its ultimate subunits, the enzyme was
analyzed in a solution of 6.8 M guanidine-HCl and 0.12 M
firmercaptoethanol. As shown in Figure 14, the molecular
weight values obtained under these conditions extrapolate
to a value of 57,100 for both w; and Mg.

To provide a direct test for stability, several
selected concentrations were subjected to molecular weight
analysis after dialysis and again after remaining in the
cold for 1 or 2 weeks. In experiments on five such samples,
the reproducibility of the results was quite remarkable. In
comparing the molecular weights from fresh and aged subunit
samples, three of the samples had molecular weight differ—
ences of less than 1% while the other two had differences of

less than 2%.

Determination of S20,w' D20,w' and Mg, and M%(s/D) in 7.4 M

 

lass

To further characterize the subunits, diffusion and
sedimentation velocity experiments were performed in 7.4 M
urea. The enzyme was prepared for study following the same
procedure used in the sedimentation equilibrium experiments.
Extrapolation of the 520,w values obtained at various protein
concentrations yielded a value for 520,w of 2.01 S (Figure
15). In a similar fashion, the plot (Figure 16) of the
D20,w values as a function of protein concentration yielded

a value for D20 w of 3.46 x 10"7 cm2/sec. Calculation of a

69

.sss 0ps.0m so 0:0

pmpppdo dds SOpdeSppppSmo SSppppppdwm .om 2mm mds oSSpdpmmaop 0S9 .AomN pdv

0.0 was as was 0:0 .spsm_a p00.0 .p002.S 0H.0 .socpsp.pomuwpsa 2.00.0 oospwp
ISoo ompd adpmpw pSoppom oSp .poSdSpdodeopmSIm de opppopSooppmS mSppdesm

op SOpppppd SH .ddmz pop .90 So. de Q 2 Sop N\h£o So. md popdSpdpm opds

s
dd
mSOppdppSmoSoo .poSdSpmopddopoaImia Np.0 de mpppopSoopphS oSppdeSw.fl 0.0
Sp mmdSpS 0pd>Spaa po mpSmpos pdpSompoS Adams. omdpodeN de Admmzv omdpmbd

IpSMpms pSopded mSp mo SOppdSpSooSoo Spmpopa open op SOppdpoadpme .dp oSSwpm

70

APPARENT MOLECULAR WEIGHT

Figure 14

000.00

OOOdN

OOHO.

OOm.N_

AI..2\0.2V ZO....<m....szZOO
.VN ON m. N. m .V O

 

 

—I

p p p _ _

~20
3
2o .. N

.04
Our.

  
  

 

o o.

. 00- 50 "0002

  

   

o 10

oz<IhmOPd<Omm2 .2 N..O .wz.n._z<30 .2 m6:

0 . _ _ _ _

 

 

71

.EQH owm.mm pa pso umHHHmo mmz :oHpmMSmthsoo .om . 0
was 35330.33 9; ..omm £3 o.m mm: mg 2.» Em .«Em a 80.0 .Hom a 3.0
.Mmgmsn Homnmfipa a 30.0 UmcampSOQ omam ampmzm psm>aom mSp .mmHS op Soapaddm
SH .mmhs m :.u SH mmmcflx mpm>sghm mo mpHQSQSm mSp mo A3.ommv mpgmflofimmmoo

godpMpSmaawmm asp mo Sofipmhpcmosoo campoag 09mm on SoHpmHogmeNm .ma mhswam

72

 

 

 

 

 

I l l T I ‘l I l I
— -+
q
M
— a: .q
3
2
U)
.— q- _ -
" O.
N
ll
L. 3 ._
‘30"
ON
(I)
L ..
l 1 I I l 1 l 1
t0. 0. 0. O. to

(\l ‘ N
(“38) .LNBIOI

0
:HBOO NOIiViNBWICIBS

l2

IO

CONCENTRATION (MG/ ML)

73

.EQH woo: pm p50 umahhmo mm; sowpmm50

Iahpsoo .05 .mm was mpSpmHmmamp mse .Aomm pm. o.w mmz mm ms» 62m ¢Eam a Hoo.o
.HOM a mH.o .Homysn Homlmfiha a 10.0 UmQHMpSoo omam Empmhm pswbaom mSp .Mmhd

op godpfidvd SH .mmhs a 3.5 2H mmdsflx mpm>SHhm mo mpflQSQSm ms» mo “3.0NQV mpnmflo

uahwmoo seamswwfld mSp mo Soapaapcmomoo campopa OHmN op soapmaommhpxm .oa mhswfim

74

Figure 16

Nm.

35:05: zo....<m...zm0200

0N

 

 

0.m. m0. NM. No
_ _

‘ a

. 13.0N
_-omm-~zo #9 x 9%. can

4mm: .2 .VN

_ '— [F _

_

 

 

r0 N —

(“030) iNBIOIdJEOG NOISflddIO

q-

 

75

molecular weight using these values gave a value for M§(s°/D°)
of 56,300. This value is in excellent agreement with those
obtained by the sedimentation equilibrium techniques and
allows greater confidence to be placed in the previous results.
The extrapolated values for the subunit sedimentation
and diffusion coefficients are much lower than would be expected
for a globular protein with mol wt 57,200, suggesting that the
subunits are extensively unfolded and/or asymmetric in shape.
Specifically, the frictional ratio calculated for the subunits
is 2.4, whereas that of the native enzyme is 1.3; this indicates
a drastic change in structure, although a small portion of the
change is probably due to differences in solvation of the native
enzyme and the subunits.
A summary of the molecular weights of the subunits,
determinaiin.various solutions and by different methods, is

given in Table V.

Reversible Dissociation of Pyruvate Kinase

Upon achieving dissociation of pyruvate kinase into its
constituent polypeptide chains, it was of interest to determine
what conditions would be necessary to reform the enzyme upon
removal of the dissociating agent. Since the stabilization of
the subunits was a prerequisite for renaturation, parameters
affecting dissociation, as well as renaturation, needed to be
considered. It was expected that such a study would yield

valuable information about the requirements needed for i vivo

 

folding of the native enzyme after its polypeptide chains had

been synthesized. Therefore, a systematic study was undertaken

noom Mao buoH mum muwc:

 

 

a n

cowumwouwwm mo omamomn swan hanmnoua w
oo~.sm ocaasm Hocmnumoummoumanm
z NH.O + odwUHame ®.m
oom.sm oom.mm Hoamsumouamuuma-n
a i 2 $6 + m3: 2 i1.
scrum 39m oomdm «08.3. 08.8 8.3 2 :9
3.8 . a N Egaom

on 3 owm va w: o2 l w:
mum<Hm¢>

 

 

 

 

 

 

 

mHHZDmDm MW€ZHM MH¢>DMMm mmH m0 mMHHmmmomm H<0Hmwmm :wmdfia

 

 

 

 

77
of the parameters in dissociation and reversal procedures
which would be expected to possibly affect reformation of
the native enzyme. For the reversal process, the following
variables which generally influence protein structure and
function were considered: (1) environmental conditions;
including temperature, pH, ionic strength, and enzyme con-
centration, and (2) chemical factors, including reducing
agents and the substrates and cofactors of the pyruvate
kinase reaction.

Since the activity of the native enzyme governs the
percent recovery calculated for renatured enzyme, it was
necessary to evaluate the activity of the enzyme after incu-
bation in the dissociation solvent minus the dissociating
agent and/or the reversal solvent. The control enzyme was
taken through the identical dissociation and reversal pro-
cedures (including time of exposure to a given condition)
with the single exception that the dissociation solvent
contained its usual composition except for the absence of
the dissociating agent.

In order to accurately determine the percent recovery
of the enzyme in various reversal media, the activity of
stock native enzyme used in the experiments was determined
with each set of experiments. The general procedure for
dissociation was as follows: stock enzyme solution was
diluted with freshly prepared and chilled stock dissociating
solvent to a final protein concentration of 4 mg/ml at 00,
and allowed to remain there for 1 hr. Final dissociation

conditions consisted of: 7.0 fl guanidine-HCl, 0.12 E

 

 

 

 

 

 

 

 

 

 

78
{B-mercaptoethanol, 0.04 y; Tris-H01 buffer (pH 8.0 at 23°),
and 0.001 fl_EDTA. The general procedure for reversal was
then carried out in the following manner; the dissociated
enzyme was diluted one hundred fold into a O0 renaturing
solution with a X pipette, mixed by gentle swirling, and
brought to reversal temperature. After 1 hr the specific
activity of the enzyme was determined at 250 using a
coupled assay, as described in the methods section. Tests
of each reversal condition were performed separately in
triplicate, using 3 individual dissociation and reversal
solvents.

The data indicate that recovery of 56% of the native
enzyme is reproducibly obtained under optimal reversal con-
ditions of 0.04 mg/ml enzyme, 0.165 M ammonium sulfate,

0.1 mfl ADP, 0.1 fl Tris-HCl buffer, 0.06 M B-mercaptoethanol,
pH 8.5, at 200 (initially the reversal solvent was at 00) In the
following experiments, these optimum conditions were

employed for all variables save that under investigation.

Effect of Dissociating Solvent on the Renaturation of Pyruvate
Kinase

In initial studies on reversal, urea rather than
guanidine was used for the dissociating agent for reasons of
convenience and economy. While reversal of urea-dissociated
enzyme did occur, recovery of activity was only about 20%,
even after extensive studies in which numerous variables were

optimized.

 

79

Since the previously described molecular weight studies
indicated that the enzyme subunits probably aggregated in urea
or urea-mercaptoethanol, it seemed reasonable that random
aggregation in urea might be a major problem and that the
guanidine-mercaptoethanol system might best avoid this. Com-
parison studies revealed that reversal of guanidine dissoci-
ated enzyme yielded approximately 1% times the per cent
recovery of that found for the urea dissociated enzyme and
so 7 fl guanidine was used as the dissociation solvent for all

the reversal studies presented here.

Effect of_pH, Temperature, Ionic Strength and Protein Concen—
tration on the Renaturation of Pyruvate Kinase

To test the effect of pH on the reversal process, the
pH of the reversal media was varied from pH 5.0-8.75 with
0.1 fi_cacodylate, imidazole, tris and glycine buffers.

Enzyme was dissociated as described above and then diluted
into the various reversal solutions at O0 and brought to
20°. Assays performed after 1 hr in the reversal solution
showed that maximum enzyme activity was achieved in the pH
range of 8.0-8.5 (Figure 17). This pH range is considerably
:higher than the eXpected pH of the cell.‘

To study the effect of temperature on enzyme renatura-
tion, the following procedure was used: (1) stock enzyme was
added to chilled dissociating solvent and allowed to remain
there for 1 hr, (2) the dissociated enzyme was then added to
O0 reversal solvent, and (3) the enzyme-reversal samples were

then incubated for one hour at various temperatures ranging

 

 

.osawso

co 2&5 3.0 use Ema a 80.0 .Hosmfimopgmowmafi in). mod .3938 asagoaam a

O
8

mma.o .m0¢ mm H.o umsflmpcoo omam waves Hmmho>mh mSB .msmmmsn osaohaw Ho wasp
.oaowocflsfl .mpmaaooomo m a.o wsflms .mm.w op m.m Bong ooflhwb was «Home Homho>oh

onu no mm egg .omwSax mpd>5~mg go GOHpmHSpmsma So mg mo poogmm .ma ohdwflm

 

 

 

Figure 17

81

% RECOVERY

 

 

 

 

 

:6 I; —‘ —.
|\ ID :2 93 0
l I I l I [T
LLJ
_(f) __
<[
Z
_x _
LU _
('r—
<[
_> _
I)
(I
_>— \\\ __
O. \\
_E‘ q
of <
r-LL _
O 6
__J _
<[
(D C
p Q: _
Lu
>
—LLJ .—
(I
llll I
o o
co 8 v 53.

All/\llDV OlleBdS

l O

 

 

82
from 00 to 370. The maximum enzyme activity recovery was
found to occur in the sample incubated at 230(F1'g.18). However,
recovery was only slightly lower at values up to and includ-
ing 370. Of particular interest is the finding that at 0°,
no noticable activity was recovered. It is not known whether
the lack of recovery of activity after incubation at 00 is
because a) protein is unable to refold, b) incomplete refold-
ing of the enzyme has occurred, or c) incorrect refolding
of the enzyme has occurred.

To observe ionic strength effects, the concentration
of ammonium sulfate in the renaturation media was varied to
produce ionic strengths ranging from 0 to 0.8. As illus-
trated in Figure 19, no detectable activity is recovered
when the ionic strength of pa, 0. A sharp increase in the
recovered activity was observed at 0.1 ionic strength while
maximum activity is achieved at ionic strengths of 0.3 or
greater. In relation to the conclusion above, findings that
a threshold level of 2a. 0.3 ionic strength is necessary for
in 11339 formation of active enzyme is of particular interest
since the intra-cellular ionic strength is approximately 0.19.

The concentration of enzyme in the reversal media was
expected to be very important in renaturation, since at low
concentrations the subunits might not recombine while high
concentrations might enhance random aggregation and thus
lower the percent recovery. To test the effect of enzyme
concentration on reversal, stock enzyme was diluted to 40

mg/ml with denaturing solvent, the final composition of which

 

 

3
8

 

.oshwso Ha\wa

:06 use «Em m 80.0 .Bmfiafi agnoaam a mode .Hommfimopgmowoa-m a mo.o
(:2 a do .26 so mo: 8.25 Homumfle m do 88328 Sees $333
ose .omm 0p 00 .mm ach0 omHHw> mm: Mecca mehoboa exp mo ohdpmhogaop

oSB .omwQHM opdbSth mo SOprHSprmh no oHSpmpoQEmp wo pommmm .wH ohswam

 

 

84

°/o RECOVERY

Figure 18

00 On

 

 

 

 

0 _ _ _
_.m_ I
_.mm I
.N01 100
: mm<Z_v. m...<>3w.>n.._>..m “.0 44mmm>mm 1
N.®N _ . . _ . . _ . . . . 0m

 

All/\LLOV OlleBdS

 

 

 

650qu dime do one flow a Hoo.o .Hogmfioooomowosnm a moo Mn? slam do

.Ao.m 0o mgv 000059 Homlmfiae m H.o oosampsoo ome swoos Hamsmbmg exp .opwsm

85

IHSm asflcosEm op Soflpaooc SH .wfiooa Hmm0o>m0 ogp 0o noflpwspgoosoo opoSQHSm

asfisoaaw esp wsflampaw an .w.o 00 0 .mm 8000 omfihm> was Spmsohpm 0020“ one

.omcsflx opwbsghm 0o SOHpMHSpmSmM esp go summoapm oHQoa 0o poo00m .mH ohsmam

 

86

Figure 19

°/o RECOVERY

v.0 N

m0?

m..®

...m

V
EOZMEm 929 8%.: 20

 

 

 

N. 0.. m. o. v.
. . . _ . . . _
I O )
O C
I mm<z§ m0<>3mi 2.x do nqmmmim I
_ . . . . . _ _ . _ _

 

 

 

00

00

All/\llOV OlleEdS

 

 

87
was 3.5 M_guanidine—HCl, 0.06 E gamercaptoethanol, 0.02 E
Tris-HCl buffer (pH 8.0), 0.01 M_imidazole buffer (pH 7.5),
0.07 E KCl and 0.001 M EDTA (although the final concentra-
tion of guanidine was relatively low, the enzyme was com-
pletely unfolded since at an enzyme concentration of 0.002
mg/ml no preceptable amount of activity was detected at 0.1
0D full scale). After 1 hr incubation, varying aliquots of
the enzyme solution were then transferred to tubes contain-
ing 7.0 fl guanidine-H01, 0.12 E fi-mercaptoethanol, 0.04 E
Tris-HCl buffer (pH 8.0), 0.15 M_KC1 and 0.001 E EDTA. This
was done to change the protein concentrations in the disso-
ciation solvent rather than in the reversal solvent so that
equal aliquots of the dissociation solution could then be
transferred to the reversal solvent system. This dilution
procedure yielded reversal solutions containing various
enzyme concentrations, ranging from 0.002 to 0.4 mg/ml. The
data obtained (Figure 20) indicated that enzyme concentra-

tions in the range of 0.03-0.09 mg/ml were optimal.

Effect of Beducing_Agents, Substrates and Metal Cofactors on
the Renaturation of Eyruvate Kinase

Since all the reversal experiments were conducted under
aerobic conditions, the possibility existed that random forma-
tion of disulfide bonds might be hindering specific refolding
of the enzyme. Tests were therefore conducted to appraise
the effect of the reducing agents, firmercaptoethanol and
glutathione, on the reversal process. The concentration of

reducing agent in the reversal solvent system was varied from

 

 

 

.4080 a Hoo.o see .Hogeeoe

 

nopoeoeeaum a mo.o .epesoHSm seasoaae a moa.o ..ms.w 00 may Heoosn Homumfipe

8
8

a H.o .mmo mm H.o oesaepeoo omae eaoea Hemwe>ee e50 .Ha\ma H.o op mmoo.o
a000 om00e> we: efiooa Hem0®>m0 ozp SH SOHpenpSmosoo ofihuflo.o£9 .mmwsfix

ope>5000 0o QoHpeMSpegos exp So Cofipehpsmozoo mahnso 0o @0000m .om ohsmam

 

 

89

Figure 20 °/. RECOVERY

:EBE zGCEzmozoo msfizm

 

v.0. I

man I

mwmi

 

0m. 0V9. 00. 00. v0.
_

N0.
_ «A _ _ _ _ _ _ _

—

 

/0
OOIIKDOO
O 3 00
OO 00

mm<z.v. m...<>3m.>n. .3. .m “.0 4<mmm>wm

_ _ C. _ L _ _ _ _ _ _ _

 

 

 

NKN

an

0m

0.»

00

00

All/\llOV OlleBdS

 

 

90
0 to 0.15 E. In the absence of firmercaptoethanol, 27% of
the control enzymatic activity was recovered while in the
presence of 0.06 E mercaptoethanol, 54% of the control
enzyme activity was recovered (Figure 21). Tests with
glutathione showed the same concentration maximum (0.06 E)
and the same total recovery of enzymatic activity as
observed with_§-mercaptoethanol.

These results indicated that in the absence (or at
low concentrations) of reducing agents, non-specific disul-
fide bond formation may be hindering enzyme refolding while
at high concentrations (0.15 E), the reducing agents are in
some way interferring with enzyme renaturation.

The possibility exists that under in_yiyg conditions,
a newly synthesized polypeptide chain may be influenced in
achieving its correct conformation by the presence of one or
all of its substrates, which are presumably readily avail-
able in the cytoplasm. To test this possibility under
12.21222 conditions, the substrates of the pyruvate kinase
reaction, i;§,, ADP and PEP, were incorporated (both together
and individually) into the reversal solvent at assay concen-
trations in the presence of 0.15 E KCl and 0.008 g MgClZ.
The percent recovery of activity was not found to be signifi-
cantly increased in the presence of PEP, but was definitely
enhanced when ADP was present (see Figure 22). The investi-
gations showed that at an ADP concentration of 1 x 10"5 fl,
maximum activity was achieved; increasing the ADP concentra-

tion above this level did not produce any further increase of

 

1
Q/

 

.omm we: manpehomaop SoomeSoQo Hom0o>mh osa

.easene oaks :o.o one Eon a Hoo.o .Bennda annnoane a modo .36 00 mo:
.8003 Honumfine a do .030 we do oenflepnoo omae enoen Hemnepen 80 .a mdo
on o 5000 ooaymb we: aflooa Hmm0o>o0 oSp SH Scopwhpsoosoo HosMSpooerohoalm one

omen? 39503 00 noopeaspegma 9.3 no Hosefioopgeoaoanm 0o 0800M .do. 92th

92

Figure 21

RECOVERY

°/o

m..

>....m<:.0.>. ...OZ<IFM.0...n.<0mm:2 I m.

0.. N_. mo. 00. m0.
.

 

0

0.1

va

 

_ . . _ . _ . .

O/Q/

mm<2_v. m....<>3m>n. ..>. .m. “.0 I.<mm.m>m.m
. r . . . . . . . _

.

 

 

V0

0N

Ow

All/\llOV OlleBdS

 

 

 

.thNQo

fines .36 one some 0 Hood . 893.05 ananonne a modo .HonenpeopneonesIm

3
O/

a mo.o .Am.m 0o mmv 000059 Homlmflhe a H.o oosoepsoo omae Modes H0m00>00
0:9 .mm 0.00 00 o 8000 ooaheb we: eaoos Hem00>o0 030 so 004 00 coopehpsoo

Isoo one .mmesfix ope>5000 00 CofipeHSPesom 030 So mm< 0o 00000m .mm opswom

 

 

94

RECOVERY

°/o

Figure 22

(D

(\l
N)

00
“-

V0

0.0.

0.0

322E. 20...<m...200200 n54
0.. om. o_. no. .0.

 

 

.

 

. . . . 4 o 0 . _ .

mm<Z_v. m...<>3m>n. 2.x “.0 I_<mmm_>m.m

 

I.

 

0?

00

00

All/\llOV OlleEdS

 

 

95
activity (Figure 22). The fact that an enhancement in renatur-
ation was observed at ADP concentrations as low as 1 x 10.5 fl,
strongly suggests that ADP may play an important part in the

i vivo formation of pyruvate kinase tertiary structure.

 

Tests were also conducted to determine the effect of

2+ .
and K+, on enzyme renaturation.

the activating cations, Mg
MgClZ and KCl were added to the reversal solvent to yield final
concentrations of 0.01 gfl to 0.1 E. In order to maintain a

constant ionic strength, the ammonium sulfate concentration

was adjusted appropriately to compensate for the cation concen-
tration change. Figure 23 shows the results of varying cation
concentrations on the reversal of pyruvate kinase. No signifi-

cant increase in catalytic activity was observed with either

cation.

 

 

.000N00 00\00

no.o one .0000 0. Hoo.o 09030000080000 0 moo .000 000 do .30 0o 03

000000 Homlm00B a 0.0 000000000 0000 00000 00000000 003 .m.o 0000 0000000

6
0/

00000000 00000 00 00000000 00 00 cm .000000000 00: 0000000000000 000000 000

N

00 000000000 003 0000000000000 00000050 00000000 009 .000 00 0002 a 0 00 o

0000 000000 003 00000 00000000 000 00 000000 00 0000000000000 009 .000000

000>0000 00 000000000000 000 00 0+0 000 mwzv 0000000 00 00000m .mm 000000

+

 

97

Figure 23

°/o RECOVERY

000. 00. 0.0. o._ _.o _o. 0
O _ _ _ _ _ _ _ _ _ _ _ 0
r _o x 0
£002.
m.m_ :
mém:
mle
I wm<2§ m0<>3m>a 2.0. “_O 04mmm>mm
06w _ _ _ _ _ _ _ _ _ _ _

:\§0: 29qu HZmuwflaou ZOEQ

 

 

 

 

 

ON

om

om

All/\ILOV OlleBdS

 

 

DISCUSSION

Native Pyruvate Kinase

Freshly2

isolated enzyme was found to have a mol

wt of 246,000 which is in good agreement with the value of
237,000 previously reported (Warner, 1958). A considerable
amount of information regarding the bonding forces oper-
ating in the native enzyme has been obtained using several
different experimental procedures. The amino acid compo-
sition of pyruvate kinase was found to consist of a large
number of hydrophobic groups ( over 50%) indicating that
the enzyme was hydrophobic in nature. The inability of
performic acid (a highly polar solvent) to unfold the enzyme
and oxidize all the sulfhydryl groups in the enzyme is

also indicative of the hydrophobic character of the enzyme.
The observation that only 4-6 half-cystine residues, out of
a total of 36 per mole of native enzyme, are accessible to
attack by performic acid strongly indicates that these
groups are situated on the exterior of the enzyme molecule.
The optical rotatory dispersion properties of aged2 enzyme
indicate a relatively lowo<-helix content in the protein;
the b0 value of -1080 atJAO = 220 mu is considerably lower
than that for most glycolytic enzymes (bo = 1500-2000)
(Jirgensons, 1965). Since the tertiary structure of the

enzyme could be easily disrupted by urea (a denaturing

98

 

 

99
agent thought to break only noncovalent bonds), it was
concluded that the individual subunits are bound together

by noncovalent bonds.

Subunits

The excellent agreement between the extrapolated
values of Na, Mg and M% (s/D) obtained in the guanidine-
mercaptoethanol and urea solvents respectively, allows con-
siderable confidence to be placed in a subunit molecular
weight of 57,000. The excellent agreement between the
weight- and z-average molecular weights obtained in the
guanidine-mercaptoethanol solvent system indicates that
the subunits have equal molecular weights. That the sub-
units exist in a highly unfolded form under the minimum
dissociation conditions (4 m_urea) is indicated by the
sedimentation coefficient (sgb§% = 1.9) and the optical
rotatory dispersion properties of the subunits (bO = 0°)
in this solvent.

Considering (l) the value of 240,000 for the mol-
ecular weight of the native enzyme, (2) the value of 57,000
for the molecular weight of the subunits, and (3) that
the subunits are of equal size, it is concluded that rabbit
muscle pyruvate kinase is composed of four subunits.

Whethersmaller polypeptide chains of about 15000-
30,000 mol wt exist within the 57,000 mol wt subunits of
pyruvate kinase is open to question. However, the strength
of the guanidine-mercaptoethanol solvent as a dissociating

medium is generally assumed to be adequate to achieve

 

 

100
complete dissociation into ultimate polypeptide chains.
Implicit in this hypothesis is the assumption that the only
covalent link possible between polypeptide chains is the
disulfide bond. Since the molecular weights obtained in
urea in the absence of disulfide bond-breaking reagents are
essentially identical with those obtained in the guanidine-
mercaptoethanol system, we conclude that there are no disul-
fide bonds holding the subunits together. Thus, weak non—
covalent attractive forces must maintain the native enzyme
in its subunit complex. Therefore, it appears that if sub-
units smaller than 57,000 exist, the attractive forces or
bonds maintaining the 57,000 mol wt subunit structure must
be of such a nature that they either have not been detected
before, or are thought not to be significant in protein
bonding.

Our hypothesis that most ultimate polypeptide chains
will have a molecular weight less than 65,000 is supported
by the pyruvate kinase data. Furthermore, enzyme struc-
tures which had been thought to contradict this hypothesis
have now been found to support it. The subunit molecular
weight of Escherichia coli fi-galactosidase is now thought to
be 50,000 (Steers 213. _a__l_., 1965; Weber, K., at al., 1963)
while that of rabbit muscle myosin has been reported to be
46,000 (Dreizen gt §l°t 1966). The general concept of
maximum size of polypeptide chains cannot be classified
as an inviolable rule because apparent exceptions to it

still exist; however, it does appear to hold true for most

 

101

protein whose structures have been intensively investigated.

Aging Process

This appears to be the first report of the rapid and
pronounced change which occurs in the sedimentation coeffic-
ient of rabbit muscle pyruvate kinase within the first 10 days
after isolation. Although the observed change in sedimem-
tation coefficient is quite significant ( 10.04 s to 8.5 3)
previous investigators (see Literature Review) have not
reported any variation in enzyme characteristics with aging.
That this change occurs has been shown by numerous exper-
iments in addition to those presented in Figure 1.

That the transition from "fresh" native enzyme (s30.W
= 8.6 S) probably involves a molecular weight change is indi-
cated by two different kinds of indirect information.

First, the 14% difference in sedimentation coefficients is
near the maximum that could be expected for an unfolding
change. Secondly, if this change in sedimentation coef-
ficient were due to an unfolding, the resulting unfolded
enzyme should have a more pronounced dependence of the sed-
imentation coefficient upon concentration; this is not ob-
served as seen in Figure 1. In fact, the concentration
dependence of the aged species is less than that of the
"fresh" native enzyme.

The change occuring in the enzyme upon aging2 may be
the result of a number of possible factors including a)
the loss of a stabilizing group from the enzyme, §;g;,

co-factors or metals, b) destabilization of the native

 

 

102

form of the enzyme due to the manner in which it is stored
in zitrg, and c) the enzyme when isolated is an artifact,
produced during isolation. That the first possibility may
be true is supported by the findings that freshly prepared
enzyme dialyzed 12 hrs was converted in 3% days to the

aged species, while a stock enzyme solution (not dialyzed)
required 9a. l0 days for the same conversion. If there were
a "stabilization factor" which was slowly released from the

enzyme, dialysis of the enzyme might accelerate its removal.
Pyruvate Kinase Intermediates in Urea

Analysis of the sedimentation coefficient of pyruvate
kinase at various urea concentrations have shown that the
dissociation procedes via a series of intermediate struc-
tural forms rather than by a single dissociation step (see
Figure 3). The discussion which follows presents evidence
that the transition from the native to the completely un-
folded form of the enzyme proceeds via three discrete steps
(designated as Intermediates I,II and III).

Since in two series of experiments, the sedimentation

coefficients of Intermediate I (0.5 fl urea) were in good
0

20,w
stable and uniform. The variability of the diffusion coef-

agreement (s = 8.5 8), this species seems to be quite
ficient data (Figure 9) under the same conditions, may be
explained by assuming the presence of an aggregated species
in addition to the intermediate. Under the conditions of

high speed used to determine sedimentation coefficients,

 

 

103

such aggregated species would move more rapidly than the
main component and therefore would not interfer in the ex-
periment.

Intermediate I may well be the same form of the enzyme
found after aging2. Not only are the values of the sedimen-
tation coefficients the same but the values are both repro-
ducible and the sedimentation coefficients of both show very
little concentration dependence. If these two species are
the same, subjection of the enzyme to 0.5 M_urea should pro-
vide a convenient system with which to study the "aged"
species,since it rapidly and reproducibly yields the con—
version of the native species to the aged species.

In discussing Intermediates II and III, it is con-

venient to refer to the sedimentation coefficient obtained at

0.6%
20,w

0
protein concentration (320 W) because of the variability of

6 mg/ml (s ) rather than to those values obtained at zero

the extrapolated values. Using this frame of reference,

016%

Intermediate II and III will be referred to as having 820 w =
9

7.3 S and 3.6 S respectively (see Figure 3).

We have already provided arguments that Intermedi-
ate I represents a discrete and reproducible state of the
enzyme. Evidence indicating that Intermediates II and III
are also discrete ( although not necessarily stable) is
obtained from the data present in Figure 3. This shows
that in a number of different experiments, at different urea
concentration, only these two species were found, i;§,, there

is no continuous variation of sedimentation coefficients

 

 

104

since no values between the two extremes of 7.3 S and 3.6 S
were found. Thus the sedimentation coefficient values are
characteristic of two discrete states.

Comparison of the dissociation properties of pyru-
vate kinase with those of other rabbit muscle glycolytic
enzymes reveals several similarities. The 3.6 S species
produced in 2-3 fl urea is strikingly similar to the 3.6 S
intermediate of aldolase found at pH 3.4 (Deal and Van
Holde, 1962; Deal §£.al., 1963). The high degree of unfold-
ing of the subunits of pyruvate kinase is also reminiscent
of that of the aldolase subunits, whose molecular weight,
sedimentation coefficient, and frictional ratio are only
slightly lower than those of the pyruvate kinase subunits.

A third similarity is that the requirement for a minimum dis-
sociation condition of 4 fl_urea is the same as that for

aldolase (Stellwagen and Schachman, 1962).
Renaturation

The in_vi££g renaturation of fully dissociated pyru-
vate kinase was achieved by the optimizing of certain chem-
ical and physical properties. Examination of the % recovery
of activity obtained in the presence of two dissociating
solvents, i;§., urea and guanidine hydrochloride, showed
that the greatest amount of activity recovered was obtained
when guanidine was used. This is presumably due to the stab-

ility of the subunits in the guanidine solution. The renat-

uration solvent, in which 56% of enzymatic activity was

 

 

105

recovered routinely, consisted of: 0.1 mfl ADP, 0.165 M
ammonium sulfate, 0.1 E Tris-HCl buffer, 0.06 E fi-mercapto-
ethanol, and 0.001 E EDTA, pH 8.5, at 230. The reversal pro-
cedure consisted of diluting the dissociated enzyme to a
final concentration of 0.04 mg/ml in this chilled (00) re-
versal solvent, mixing, and bringing the temperature of the
solution to 230. The substrate of the pyruvate kinase re-
action, PEP, and the activating cations, Mg2+ and K+, were
found to have no effect upon renaturation of pyruvate kinase.
However, the other substrate of the reaction, ADP, was

found to have a marked effect upon renaturation at a con-
centration of 10.5 m or greater. The relatively high pH

and ionic strength required for renaturation, may have some
particular significance in the in_yizg formation of the
enzyme's tertiary structure. Although it can be concluded
from these reversal studies that the enzyme has indeed re-
verted to an active form, it does not necessarily follow
that the complete original native structure has been regain-

ed. However, this does seem unlikely.
Partial Specific Volume

An accurate knowledge of the partial specific volume
(V) of the protein, as it exists in the analysis solution,
is essential for valid molecular weight determinations.
Two possibly complicating factors which directly affect
the selection and use of a correct V are: l) the occurence

of preferential interactions, in which the V'of a protein

   

 

106

in a particular solvent is not that of the pure protein,
but rather that of a protein-solute complex, and 2) the
possibility of a change in the V of the folded protein upon
unfolding, such as occurs upon denaturation in a dissociat-
ing solvent.

The problem of preferential interaction arises if
one of the components of a solvent system, in which the
enzyme is dissolved, is preferentially bound to the enzyme.
The specific volume of this enzyme-solute complex may be
quite different from that of pure enzyme alone. When the
enzyme binds all members of the solvent equally, the bound
material does not affect the specific volume of the enzyme
as far as transport properties are concerned.

The possibility does exist that pyruvate kinase ex—
hibits such preferential binding but limitations of time
have not allowed an evaluation of it by analyses which
have recently been reported (Casassa and Eisenberg, 1964;
Schachman and Edelstein, 1966). Evidence against the ex-
istence of preferential binding is obtained from the obser-
vation that the same subunit molecular weights are obtained
in urea dissociating mixtures as were found in the guani-
dine dissociating mixtures, despite the great differences
in the polarity and density of these two solutions.

It can be argued on a qualitative basis that since
for pyruvate kinase (see Results, p. 35) and other enzymes
( McMeeken and Marshall, 1962) the values for the partial

Specific volume calculated from the amino acid composition

 

107

agree well with those determined directly, the partial.
specific volume of the unfOlded protein should be the same

as that for folded protein. That is, the calculation of

7 from the amino acid composition should yield a value
corresponding to the unfolded protein. Since the value for

7 calculated from the amino acid analysis of pyruvate.kinase
in this work agreed with that measured directly (Warner, 1958)
it seemed safe to use this value of F’for both the native

and the fully dissociated enzyme.

Studies of the affect of urea and guanidine sol-
vents on the V of proteins has yielded conflicting results.
Kielly and Harrington (1960) and Marler §t_al. (1964) have
observed that the V'decreases by about 0.01 cc/g in 6—7
M guanidine solutions compared to that for native protein
in dilute aqueous solution. In contrast, Heithel and
Sakura ( 1963) have reported little change in V with dis-
sociating solvents for a number of proteins. It should be
noted that both the factors mentioned in the opening para-

graph may be operating in the experiments using dissociating

media.

 

 

SUMMARY

(1) Freshly prepared enzyme was found to have the
following properties: (a) sgo’w = 10.07 s, (b) ogo’w = 3.96
x 10”7 cmZ/sec, (c) fO/f = 1.32, (d) M; (s/D) = 246,000,

(e) 5': 0.7406 cc/g. This data confirms the findings of
Warner (1958).

(2) Freshly prepared enzyme was found to undergo
structural changes when stored in 13339, The sedimentation
coefficient of freshly prepared enzyme was observed to de-
crease from 830 W = 10.07 s to 5:0,w = 8.5 s.

(3) The data indicate that the native enzyme is
composed of four polypeptide chains which are bound together
noncovalently.

(4) The subunits were found to have the following
characteristics:

a. In 7.4 fl urea, the enzyme was found to have

the following properties: 830 w = 2.01 S,
0

Do
20,w

56,3000 and fO/f a 2.4

= 3.46 x 10'"7 cmZ/sec, M3 (s/D) =

b. In 6.8 fl guanidine hydrochloride and 0.12 H
firmercaptoethanol, the M3 and M2 was 57,100.
(5) In 4.0 and 5.4 fl_urea, optical rotatory dispersion
measurements indicated a bO value of 00.
(6) A total of 3? half-cystines ( determined as S-

Carboxymethylcysteine) are present in the native enzyme;

108

   

 

109

only 4-6 half-cystine residues reacted with performic
acid and thus are presumed to be on the enzyme's surface.

(7) The amino acid composition of pyruvate kinase
has been determined.

(8) Loss of activity upon dissociation of the enzyme
into individual polypeptide chains is reversible; upon dilu-
tion of the dissociating agent, the active enzyme is reform-
ed and presumably the native structure is obtained.

a. A recovery of 56% activity is obtained
routinely.

b. The optimal conditions for reassociation
were as follows: 0.01 mfl ADP, 0.1 fl Tris-
HCl buffer, 0.165 m ammoniumsulfate, 0.06 E
E-mercaptoethanol, pH 8.0, at 230, enzyme
concentration of 0.04 mg/ml.

c. PEP and the activating cations, Mg2+ and
H+, were found not to affect renaturation
of the enzyme.

(9) The dissociation of pyruvate kinase into unfold-

ed subunits has been found to occur in three discrete steps.
o.6%
20,w

iate species formed in 0.5 E, 1.5 fl and 3.0 M urea are 8.4 S,

The sedimentation coefficients (8 ) of the three intermed-
7.3 S, and 3.6 S respectively.
(10) The data presented above provide a broad basis

upon which other extensive experiments may be devised:

110

Investigation of the possibility that one
or more of the 4-6 half-cystine residues
exposed on the enzyme surface may be dir-
ectly or indirectly involved in catalysis.
Investigation of the structural identity
of the subunits.

Extensive investigation of all parameters
involved in the dissociation process.

Further examination of the aging transition.

 

REFERENCES

Anson, M. L. (1941), J. Gen. Physiol., 24, 399.

Boyer, . D., Lardy, H. A., and Phillips, P. H. (1942),

Biol. Chem., 146, 673.

D., Lardy, H. A., and Phillips, P. H. (1943).

P

J

Boyer, P.

J. Biol. Chem., 149, 529.
P

Boyer, . D. (1962), The Enzymes, 6, 95.

Buoher, Th., and Pfleiderer, G. (1955), Methods in
Engymology, 1, 435.

Casassa, E. F. and Eisenberg, H. (1964), Adv. in Prot.
Chem., lg, 287.

Cohn, E. J. and Edsall, J. T. (1943), Proteins,_Amino Acids
and Peptides as Ions and Dipolar Ions, Reinhold Publ.
Co., New York, N. Y.

Crestfield, A. H., Moore, 8., and Stein, W. J. (1963),
J. Biol. Chem., 238, 622.

Czok, H., and Bucher, Th. (1960), Adv. in Prot. Chem.,
15; 315.

Deal, w. 0., Jr., and Van Holde, K. E. (1963), Fed. Proc.,
Zia 254.

Deal, W. C., Jr., Rutter, W. J., and Van Holde, K. E. (1963),
Biochemistry, g, 246.

Diez, M. J. F., Osuga, D. T., and Feeney, R. E. (1964),
Arch. Biochem. Bipphys., 102, 449.

Dreizen, P., Hartshorn, D. J., and Stracher, A. (1966),
Jo BiOLO Chemo ’ 214‘]. 3 “1+3 0

Flavin, H., Castro—Mendoza, H., and Ochoa, S. (1957),
J. Biol. Chem., 229, 981.

Flavin, H., Castro-Mendoza, H., and Ochoa, S. (1956),
Biochim. Biophys. Acta, 29, 591.

111

 

 

112

Gamble, J. L. (1954), Chemical Anatom Ph $1010 and
Extracellular Fluid, 6E5 ed., Harvard University Press,
Cambridge, Mass.

Harrison, W. H., Boyer, P. D., and Falcone, A. B. (1955),
J. Biol. Chem., 215, 303.

Hass, L. F., Boyer, P. D., and Falcone, A. B. (1961), J.
Biol. Chem., 226, 2284.

Hill, R. L., and Kanarek, L. (1964), Brookhaven Symp. in
Biol., 1], 80.

International Critical Tables (1929), 5, 22.

Jirgensons, B. (1965), J. Biol. Chem., 24 , 1064.

Kachmar, J. G., and Boyer, P. D. (1953), J. Biol. Chem.,
200, 6690

Kayne, F. J., and Suelter, C. H. (1965), J. Am. Chem. 800.,
i2. 897-

Kayne, F. J. (1966), A Thesis, Michigan State University.

Kielly, W. W., and Harrington, W. F. (1960), Biochim.
Biophys. Acta, 41, 401.

Kimberg, D. V., and Yielding, K. L. (1962a), Fed. Proc.,
21. 237.

Kimberg, D. V., and Yielding, K. L. (1962b), J. Biol. Chem.,
221. 3233.

Koler, H. D., Bigley, R. H., Jones, R. T., Higas, D. A.,
Vanbellinghen, P., and Thompson, P. (1964), Cold
Spring Harbor Symp., g2, 213.

Kornberg, A., and Pricer, W. E., Jr. (1951), J. Biol. Chem.,
193, 481.

Krebs, H. A., and Eggelston, L. V. (1965). Biochem. J., 32.
3c.

Krimsky, I. (1959), J. Biol. Chem., 234, 232.
Kubowitz, F., and Ott, P. (1944), Biochem. Z., 312, 193.

Kupieki, 54 P., and Coon, M. J. (1960), J. Biol. Chem., 225.
19 .

Lazflyg H. A., and Ziegler, J. A. (1945), J. Biol. Chem.,
122. 343.

 

113

Levinthal, C., Signer, E. R., and Fetherrolf, K. (1962),
Federation Proc., 48, 1230.

Lohmanné K., and Meyerhof, 0. (1934), Biochem. Z., 223,
0.

McMeekin, T. L., Groves, N. L., and Hipp, N. J. (1949),
J. Am. Chem. Soc., 1;, 3298.

McNeekinfi T. L., and Marshall, K. (1952), Science, 1 6,
1 2.

NoQuate, J. T., and Utter, N. F. (1959), J. Biol. Chem.,
234, 2151.

Marler, E., Nelson, C. A., and Tanford, C. (1964), Biochem-
istry, 3, 279.

Melchior, J. B. (1965), Biochemistry, g, 1518.

Neyerhof, 0., 0h1meyer, P., Gentner, W., and Naier-Leibnitz,
H. (1938), Biochem. Z., 298, 396.

iildvan, A. S., and Leigh, R. A. (1964), Biochim. Biophys.
éQEEJ QB: 393.

Mildvan, A. S., and Cohn, M. (1965), J. Biol. Chem., 24 ,
238.

Mildvan, A. 8., and Cohn, M. (1966), J. Biol. Chem., 24 ,
1178.

Moffitt, W., and Yang, J. T. (1956), Proc. Natl. Acad. Sci.
U- 8.,‘92, 596-

Moore, 8., and Stein, W. J. (1954), J. Biol. Chem., 2 1,
893.

Moore, S. (1963), J. Biol. Chem., 238, 235.
Horawiecki, A. (1960), Biochim. Biophys. Acta, 44, 604.

Plowman, K. H., and Krall, A. R. (1965), Biochemistry, 4,
2809.

Reithel, F. J., and Sakura, J. D. (1963), J. Phys. Chem.,
62, 2497.

Reynard, A. H., Hass, L. F., Jacobsen, D. D., and Boyer, P. D.
(1961), J. Biol. Chem., 236, 2277.

Rose, I. A. (1960), J. Biol. Chem., 235, 1170.

 

114
Schachman, H. K., and Edelstein, S. J. (1966), Biochemistry,
5, 2681.
Schachman, H. K. (1957), Methods in Engymology. 5,32.

Stellwagen, E., and Schachman, H. K. (1962), Biochemistry,
;, 1056.

Svedberg, T., and Pedersen, K. O. (1940), The Ultracentri-
fuge, New York, N. Y., Oxford University Press,
Appendix III.

Tanaka, T., Harano, Y., Morimura, H., and Mori, R. (1965).
Biochem. Biophys. Res. Comm., 21, 55.

Tanford, C. (1961), Physical Chemistry of Macromolecules,
New York, N. Y., John Wiley and Sons, Inc., p. 356.

Taylor, J. F., and Lowry, c. (1956), Biochim. Biophys. Acta,
29, 109.

Tietz, A., and Ochoa, S. (1958), Arch. Biochem. Biophys.,
1&9 LL77-

Van Holge, K. E., and Baldwin, R. L. (1957). J, Phys. ChQE-.
_Zn 734-

Warner, R. C. (1958), Arch. Biochem. Biophys., 8. 494.

Weber, 0., Stamm, N. B., and Fischer, E. A. (1965). §gign22.

1.12. 65.

Weber, K., Sund, H., and K. Wallenfels. (1963). Biochem. 2..
228. 714.

Yang, J. T., and Doty, P. (1957), J. Am. Chem. Soc., 29, 761.

Young, D. H., Harrington, W. F., and Kielly, W. W. (1962),
J. Biol. Chem., 232, 3116.