EHEMIQAI. AND PHYSICAL CHARACTERIZATION OF
EDP-GLUCOSE PYRGPHOSPHORYLASE
FROM GALF LIVER

Thesis for #213 Degree of Ph. D.
MESHIGAN STATE UNIVERSITY
SIEVEN LEVINE .
1959'

1.455935 LIB.pAPY

Michigan State
University

 

This is to certify that the

thesis entitled

CHEMICAL AND PHYSICAL CHARACTERIZATION OF
UDP-GLUCOSE PYROPHOSPHORYLASE
FROM CALF LIVER

presented by
STEVEN LEVINE

has been accepted towards fulfillment
of the requirements for

PhoDo degree in BIOCHE‘AIS TRY
2 ' /"
// é/ //¢éfl ,

Date S EFT Ell'fl

0-169

 

 

 

 

 

 

 

 

yew

I "* “‘1' if?“
' n» 'Mfi’gg‘hflfi) '7/ /;} /A’f{[ {‘J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ABSTRACT

CHEMICAL AND PHYSICAL CHARACTERIZATION OF
UDP-GLUCOSE PYROPHOSPHORYLASE
FROM CALP LIVER

By

Steven Levine

Uridine diphosphate glucose, which is formed from glucose-
l-phosphate and uridine triphosphate with the concomitant
liberation of inorganic pyrophosphate, is an extremely important
intermediate in carbohydrate metabolism. The crystallization of
uridine diphosphate glucose perphosphorylase, an enzyme that
catalyzes the above transformation, has afforded an opportunity
to study the relationship of its structure to its function. The
physical and chemical characterization of this protein is the -
purpose of this research.

The pyrophosphorylase was found to be a large polydisperse
enzyme composed of large discrete subunits. The optical properties,
renaturation tendency, chemical composition, and subunit structure

of the enzyme were also determined.

CHEMICAL AND PHYSICAL CHARACTERIZATION OF
UDP-GLUCOSE PYROPHOSPHORYLASE

PROM CALF LIVER

By

Steven Levine

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1969

4)

ACKNOWLEDGEMENTS

The author wishes to express his appreciation to Dr. R. G.
Hansen without whose guidance and inspiration this work could
not have been completed.

Thanks are also extended to Dr. Pat Oriel of Dow Chemical
Company for performing the optical rotatory dispersion experiments
and for his assistance in the interpretation of the data.
Acknowledgement is made to Dr. Bill Rutter for allowing us the
use of his laboratory to perform the electrofocusing experiments.
Acknowledgement is also made to Miss Doris Bauer for her aid in
the amino acid analysis. The author also wishes to thank Dr.

w. C. Deal for his help with the analytical ultracentrifuge
experiments. Special thanks are extended to Dr. w. A. Wood whose

interest in the author and his work is greatly appreciated.

ii

TABLE OF CONTENTS

Page
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . 1
LITERATURE REVIEW . . . . . . . . . . . . . . . . . . . . . 5
EXPERIMENTAL PROCEDURE . . . . . . . . . . . . . . . . . . . 10
Materials and Methods . . . . . . . . . . . . . . . . . lO
Reagents . . . . . . lO
Spectrophotometry. and Measurement of Enzyme
Activity . . . . . . . . . . . . . . . . . lO
Absorption Spectroscopy . . . . . 10
Determination of the Extinction Coefficient .
Carbohydrate Content . . . . . . . . . . . . . . . l2
Ultracentrifugation . . . . . . . . . . . . . . 12
Optical Rotatory Dispersion . . . . . . l5
Titration of Cysteine Residues with Ellman‘ s
Reagent (DTNB) . . . . . . . . . . . . . . 17
Titration of Cysteine Residues with p-
Mercuribenzoate (PMB) . . . . . . . . . . . . 18
Amino Acid Analysis . . . . . . . . . l9
Electrophoresis on Cellulose Acetate Strips . . . 22
Isoelectric Focusing . . . . . . . . . 22
Trypsin Digestion and Peptide Mapping . . . . . . 25
Cyanogen Bromide Cleavage . . . . . . . . . . 25
Subunit Reassociation . . . . . . . . . . . . . . 26
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Absorption Spectra . . . . . . . . . . . . . . . . . . 2T
Extinction Coefficient . . . . . . . . . . . . . . . . 27
Carbohydrate Content . . . . . . 27
Homogeneity and Molecular Weight of the Enzyme by
Sedimentation and Diffusion Techniques . . . . 30
Molecular Weight of the Native Enzyme by the Sedimen—
tation Equilibrium Technique . . . . . . 39
Homogeneity and Molecular Weight of the Subunits by
Sedimentation, Diffusion, and Sedimentation
Equilibrium Techniques . . . . . . . . . . . . . . 59
Optical Rotatory Dispersion . . . . . . 51
Titration of Cysteine Residues with Ellman' s Reagent . 6h
Titration of Cysteine Residues with p- -Mercuri-
benzoate . . . . . . . . . . 69
Amino Acid Composition and Partial Specific Volume . . 69

iii

iv

TABLE OF CONTENTS - Continued

Page

Subunit Electrophoresis on Cellulose Acetate
Strips . . . . . . . . . . . . . . . . . 75
Isoelectric Focusing of the Subunits . . . . . . . . . 73
Trypsin Digestion and Peptide Mapping . . . . . . . . . 79
Cyanogen Bromide Digestion . . . . . . . . . . . . . 80
Reassociation of the Subunits . . . . . . . . . . . . . 80
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . 87
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . 98

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . lOO

LIST OF TABLES
TABLE Page

I. Sedimentation Coefficients for UDP-glucose Pyrophos-
phorylase under Varying Conditions . . . . . . . . . . . . 58

II. Optical Rotatory Dispersion of UDP—glucose Pyrophos-
phorylase . . . . . . . . . . . . . . . . . . . . . . . . 65

III. Amino Acid Composition of UDP-glucose Pyrophosphorylase. . 72

IV. Calculation of the Partial Specific Volume from the
Amino Acid Analysis . . . . . . . . . . . . . . . . . . . Th

V. A Summary of the Fingerprints of Tryptic Digests of
UDP-glucose Pyrophosphorylase . . . . . . . . . . . . . . 85

VI. Some Optical Rotatory Dispersion Parameters of UDP-
glucose Perphosphorylase and a Group of Proteins . . . . 91

VII. Cotton Effects of L- -Polypeptides and UDP- glucose
Pyrophosphorylase . . . . . . . . . . . . . . 92

FIGURE

10.

ll.

12.

15.

IA.

LIST OF FIGURES

Ultraviolet absorption spectra of the pyrophosphory—
lase . . . . . . . . . . . . . . . . . . . . . . . .

Schlieren sedimentation velocity pattern of the
pyrophosphorylase . . . . . . . .

Relative sedimentation rates for the enzyme in the
presence and absence of reducing agent . .

Apparent sedimentation coefficients, diffusion coeffi-
cients, and weight-average molecular weights for the
principal component of the pyrophosphorylase .

Apparent weight-average, z-average, and limiting z-
average molecular weights for the enzyme . .

The relative sedimentation rates for the native protein
and the subunits . . . . . . . . . . . . . .

The relative rates of sedimentation of the subunits in
guanidine hydrochloride and urea . . . . . . . . . . . .

Apparent sedimentation and diffusion coefficients and
weight-average molecular weights of the subunits . . . .

The apparent weight-average and z-average molecular
weights of the subunits . . . . . . . . . .

Ultraviolet optical rotatory dispersion of the native
and denatured enzyme . . . . .

Visible and near ultraviolet optical rotatory dispersion
of the native and denatured enzyme . .

Drude plots in the visible and near ultraviolet regions
for the native and denatured enzyme .

Moffitt—Yang plots of the native enzyme in the 280-500
mu-spectral region . . . . . . . . . . . . . . . . . .

Moffitt-Yang plot of the denatured enzyme in the 280—
500 mu spectral region . . . . . . . . . . . . . . . . .

vi

Page

29

52

5h

56

Al

AB

1+5

A8

50

55

55

57

59

62

vii

LIST OF FIGURES - Continued

FIGURE

15.

I6.

170

18.

19.
20.

21.

Rate of reaction of the sulfhydryl groups of the native
and denatured enzyme with Ellman's reagent .

The pH dependence of the number of SH groups titrated
per mole of protein . . . . . . . . . . . . . .

Spectrophotometric titration of the sulfhydryl groups
of the enzyme with PMB . . . . . . . .

Electrophoresis of the subunits on cellulose acetate
strips .

Isoelectric foching of the subunits
Fingerprint of a trypsin digest of the enzyme
Polyacrylamide patterns of a reduced and S-carboxy-

methylated sample of the enzyme before and after
cleavage with cyanogen bromide .

Page

66

68

71

76
78
82

85

INTRODUCTION

The enzyme uridine diphosphate glucose perphosphorylase
(UTP:a-D-glucose-l-phosphate uridylyltransferase, E. C. 2.7.7.9)
catalyzes the biosynthesis of the important coenzyme uridine
diphosphate glucose (UDP-glucose) from glucose-l-phosphate
(Glc—l-P) and uridine triphosphate (UTP) with the concomitant

formation of inorganic pyrophosphate (PPi)'

F. - “6““ fl:-.\

Mg2+

UTP + Clo-LP: UDP-glucose + PPi (1)

The enzyme appears to be ubiquitous in nature (h-8, 15-17,
19-27) and has been crystallized from calf (28), lamb, goat,
sheep, rabbit (52), and human (50) livers. The enzyme also has
been obtained in a highly purified state, if not crystalline,
from rabbit muscle.

The calf liver protein is the subject of this research. In
the original work of Albrecht et al. (28), the pyrophosphorylase'
was estimated to account for over 0.5% of the extractable protein
of calf liver and to have a molecular weight of approximately
MO0,000. Polydispersity was noted, but contamination could not
be eliminated as its cause.

This thesis describes: (l) a more complete study on the

molecular weight of the native pyrophosphorylase and its subunits

and some information concerning the nature of the minor molecular
components; (2) the Optical properties, renaturation tendency,
chemical composition, and subunit structure of the crystalline

enzyme.

LITERATURE REVIEW

Uridine diphosphate glucose pyrophosphorylase belongs to
a general class of enzymes which catalyze the formation of
various nucleoside diphosphate sugars according to the following

reaction:

\

XTP + S—l-P ‘ XDPS + PPi (2)

The notation, X, can be various purine or pyrimidine nucleosides
and the notation, S, can be various sugars. The important
nucleoside diphosphate sugar, UDP-glucose, is biosynthesized by
this route. UDP—glucose, or as some have called it, "activated
glucose," is an exceedingly important compound because of its
glycosyl transferring ability and its role in the interconversion
of sugars and formation of hexoses, pentoses and uronic acids (l-h).
While studying the metabolic route for the conversion of

Egalactose-l-phosphate to glucose-l—P in bakers' yeast, Leloir
§§E_al, in l9h9—SO determined that a necessary thermostable cofactor
iJi the conversion was uridine diphosphate glucose (5, 6). Shortly
tiiereafter, in 1952—55, an enzyme was discovered in yeast by
1Qalckar and Munch-Petersen §£_al. (7, 8) that would catalyze the
Ffiflrophosphorolysis of UDP—glucose to form glucose-l-phosphate and
uI‘idine triphosphate. This type of pyrophosphorolytic cleavage,

although extremely important to the eventual understanding of the

C-h interconversion of hexoses, was not new. In 19h8-50, Kornberg
§£.§l- had demonstrated the pyrophosphorolytic cleavage of some
nucleoside diphosphate compounds (9-11).

Several years after its discovery, Munch-Petersen purified
the UDP-glucose perphosphorylase from yeast approximately 260-
fold with an overall yield of 15% (12). She noted that the
enzyme was unstable in dilute solutions, even in the presence of
a 5 mg/ml solution of bovine serum albumin, and that 60-75% of
the activity remained after three month's storage. The enzyme
exhibited a broad pH profile (pH 6.5-8.0) and the reaction was
stimulated by magnesium at an Optimal concentration of 2 mM.
In addition to the fact that NaF at a concentration of 0.05 M
and ethylenediaminetetraacetate (EDTA) at a concentration of
0.01 M had no inhibitory effect, the presence of cysteine also
had no affect on the reaction. Following some radioactive
exchange experiments, she proposed the possible existence of a
uridylated intermediate in the reaction sequence.

The decade or so after 1955 was a time of identification of
the pyrophosphorylase for UDP-glucose from many sources. After
the 1955 demonstration by Kalckar 2E.il- (15) of the existence

of the perphosphorylase in galactose adapted Saccharomyges

 

fragilis, Smith gt 21. (1h, 15) described its presence in nuclei
of guina pig liver and mammary glands, and Burma gt g1. (16) in
1956 noted its activity in sugar beet leaves.

In 1957 Neufeld e_t 21- (17) found the pyrophosphorylase in

mung bean seedlings and a host of other plant sources. After

partially purifying the enzyme, they noted the need for a divalent
metal and found that Mge+, Mn2+, and Co2+ would serve equally as
well. Further work on the characterization of the mung bean
enzyme followed the next year from Ginsburg's laboratory (18).
After purifying the enzyme 800-fold, he noted that the pH
optimum was about 8.0 and the Michaelis constants (Km) for UDP-
glucose and PPi were 1.1 x lO-uM and 2.5 x 10-hM respectively.
His purified protein was not very stable and lost 90% of its
activity even in the presence of mercaptoethanol after storage
for 2 weeks at -70C.

In that same year the pyrophosphorylase was found in the

plant Impatiens holstii (19) and in pea seeds (20). The purified

 

enzyme from pea seeds exhibited a broad pH range (pH 7.0—9.0) and
its activity was not affected by fluoride or orthophosphate anions,
arsenate at a concentration of 10 mm, p-chloromercuribenzoate (PMB)
at a concentration of 1 mm, or Hg2+ ions at a concentration of

n2+ (

0.1 mM. M lmM) was as effective a divalent metal as was

M92+ (5mM) and Co2+ and Ni2+ stimulated at a concentration of
2.5 mM. For this enzyme EDTA at a concentration of 0.01 M inhibited
the reaction by 90%.

Using rabbit muscle as their source, Villar-Palasi gt a1. (21)
purified the pyrophosphorylase l500-fold and noted that the
stability point (pH 9.8) was different from the point at which
the enzyme was most active (pH 7.5).

Shortly afterwards, in 1961, Basu 35 a1. (22) purified 50-
fold a human brain pyrophosphorylase that showed an absolute

requirement for Mge+. Although the enzyme was not affected by

reducing agents such as cysteine and glutathione, 0.0001 M PMB
caused the enzyme to lose 50% of its activity. It should be
noted that the pyrophosphorylase has been reported in other
human tissues (25-26).

More recently, in 1965, Kamogawa gt a1. (27) purified the
§-.2211 K-12 pyrophosphorylase 280-fold. They noted that as with
other pyrophosphorylases, the pH profile was broad (pH 7.5-9.0).
With Mg2+ optimal for activity, they determined the apparent
equilibrium constant to be 5.0 in the direction of UTP formation.
NaF, PMB, and UMP at 2 mM concentrations were not inhibitory.

In 1966 Albrecht 35 a1. (28), using calf liver as a source
of the enzyme, finally obtained a 500-fold purified and
crystalline UDP-glucose pyrophosphorylase. When the base and
sugar portion of the nucleoside diphosphate sugar was varied
from uracil or glucose, the rate of the reaction was only a
small percentage of that found for UDP-glucose. The protein
was apparently present in large quantities in calf liver, as it
represented more than 0.5% of the total extractable protein, and
the specific activity and turnover number were found to be 2&0
and 85,000 respectively. The molecular weight of the protein was
estimated to be h00,000 and all the substrates had Michaelis
constants of about 10-5M except UTP which was 2 x 10—hM. Although
Mg2+ was optimal at a concentration of 2 mM, C02+ and Mn2+ were
25% as effective at a comparable concentration. The enzyme
apparently did not require reducing agents for stabilization.

It is important to note the many factors that indicate that

UDP-glucose pyrophosphorylase is the major catalyst for UDP-

glucose biosynthesis; (l) the large amount of the enzyme in

calf liver, (2) the high rate specificity for UDP—glucose, (5)
the low Km for UTP and glucose-l-phosphate, (h) the high turn-
over number, and probably of less importance under physiological
conditions, (5) the favorable equilibrium constant (approximately

0.5 in the direction of UDP-glucose biosynthesis).

 

More recently, DeFazio (29), employing the crystalline calf F
liver enzyme and using new Spectrophotometric and chromatographic
techniques, determined very accurately that the apparent equili-
brium constant was 0.2 in the direction of UDP-glucose synthesis. é
Because the perphosphorylase has recently been crystallized ‘

from sources other than calf liver, it is now possible to compare
the proteins on the bases of their physical and chemical charact-
eristics, along with the comparison of their reaction mechanisms.
Extremely important is the recent purification and crystallization
by Knop (50) of the pyrophosphorylase from human liver. The 500-
fold purified and crystalline human liver pyrophosphorylase was
almost homogeneous as judged by sucrose density gradient ultra-
centrifugation,polyacrylamine gel electrophoresis, and sediment-
ation in the analytical ultracentrifuge. The pyrophosphorylase
exhibited the usual broad alkaline pH profile. Mg2+ was most
active at a concentration of 5 mM, but Co2+ could serve as the
divalent metal at only lh% of the rate with Mg2+. Also,the
human liver enzyme required the presence of a reducing agent to
protect against inactivation.

Since the original study of Villar-Palasi gt a1., Bass e£_al,

(51) have highly purified, if not crystallized, the pyrophosphorylase

from rabbit muscle. The enzyme exhibited one band on poly—
acrylamide gel columns, but two bands on sucrose density gradient
columns. The protein also required reducing agents to prevent
against inactivation.

Still in progress are the investigations of Gillett g£_al.
(52). They were able to crystallize the pyrophosphorylase from
lamb, goat, and rabbit liver. Preliminary experiments indicate
that the enzymes from lamb and goat are extremely similar, if not
identical,to the calf liver pyrophosphorylase. These conclusions
have come from similarities in gross crystal structure, electro—
phoretic results on polyacrylamide gel columns, and the apparent
lack of necessity for protection by reducing agents. On the other
hand, the rabbit liver perphosphorylase is a sulfhydryl requiring
perphosphorylases and its gross crystal structure and electro—
phoretic behavior are different from the calf, lamb, goat, and
human liver preparations.

Many patients suffering from the hereditary disorder known
as galactosemia are able to metabolize a small amount of galactose.
Due to the reports of Isselbacher gt a1. (55, 5h) and other authors
(17, 55, 56) relating to a separate UDP—galactose pyrophosphorylase
pathway in these galactosemics, Ting et 31. (57) and Knop (50)
have studied the question in relationship to the crystalline
UDP—glucose pyrophosphorylase. The results of both studies
support the conclusion that there is not a separate UDP-galactose
pyrophosphorylase. Their conclusidns were based on the fact that,

throughout the purification of the calf and the human liver

pyrophosphorylases, the UDP-galactose/UDP-glucose activity ratio

remained unchanged.

EXPERIMENTAL PROCEDURE

Materials and Methods

Reagents

All chemicals were purchased from commercial sources and
where critical, the sources are indicated. Crystalline UDP—
glucose pyrophosphorylase was prepared according to the

procedure of Albrecht et_al. (28) as modified by Gillett _§>gl. (58).
Spectrophotometry_and Measurementgpf Enzyme Activity

Spectrophotometry and the estimation of pyrophosphorylase
activity were determined as described by Albrecht EE.§1- (28).

Assay system 1 was routinely used.

Absorption Spectroscopy

 

A solution of approximately 1 mg/ml of pyrophosphorylase in
0.01 M tricine buffer (pH 8.56) was dialyzed against 1000 volumes
of the above buffer for 18-2h hours. Values for the actual
protein content were obtained using the molar extinction coefficient,
and were checked with the Lowry procedure (59). The spectra were
taken at 2h°C in a Cary 15 Spectrometer in 1.5 m1 quartz cells

with a 1 cm light path.
Determination of the Extinction Coefficient

Approximately 50 mg of a crystalline suspension of the

pyrophosphorylase, that was stored in 0.01 M tricine buffer (pH 8.5),

10

11

which was 20% in ammonium sulfate, was diluted to about 6 ml with
0.01 M sodium phosphate buffer (pH 7.0). The material was then
dialyzed against 1 liter of the 0.01 M phosphate buffer for 2h
hours. The slight amount of precipitate was removed by centri-
fugation and the solution was redialyzed for a total of 5 days
against changes amounting to 10 liters of fresh phosphate buffer.
Dry weight was determined after high-speed centrifugation which
resulted in a clear solution.

Metal stainless steel planchets were used as tares. The
planchets were heated in a drying oven at 1050C for 2h hours,
and then allowed to equilibrate to room temperature in a dessi-
cator over CaC12 for 1 1/2 hours. After the equilibration
period, the planchets were removed with forceps and weighed with
a semi-microanalytical balance. To insure a constant weight,
further heating, equilibration, and weighing were performed
at 5 and 10 hours after the initial weighing.

Five samples of the pyrophosphorylase and two of the outside
dialysis solution were pipetted into the planchets. All planchets
were then dried at 85°C for 6 hours and then at 1050C until a
constant weight was obtained.

Extinction readings were made on the sample at 280 mu, 278.5
mu and 260 mu. Dilutions of this solution were measured on
spectrophotometers which were independently standardized.

Cross-standardization with the Lowry method involved Bovine
Serum Albumin (BSA) as the comparison standard. The dry weight
value was also cross—standardized with the 280/260 method of

Warburg and Christian (ho).

12

Carbohydrate Content

 

The phenol-sulfuric acid method of Dubois gt _1. (Al) was
used to determine the carbohydrate content of the pyrophosphorylase.
Galactose was used as the internal carbohydrate standard and

ovalbumin was used to check the sensitivity of the method to

protein-bound carbohydrates.

Ultracentrifugation

 

All ultracentrifugation experiments were performed near 50C
in a Spinco Model B analytical ultracentrifuge equipped with
phase-plate schlieren optics. Schlieren patterns were read with
the aid of a Bausch and Lomb or Gaertner microcomparator. Sedimen-
tation velocity analyses were performed at 59,780 rpm and
diffusion coefficient analyses were performed at h,059 rpm.
Diffusion coefficients (D) were calculated by the height-to-area
analysis (A2) and were converted as were the sedimnetation
coefficients (S) to standard conditions of water at 200C. Double
sector synthetic boundary cells were used for the diffusion
coefficient analyses.

Sedimentation equilibrium analysis was done by the short
column technique of Van Holde and Baldwin (A5). The solution
column depth was routinely 1.7 mm (0.06 ml of protein solution).
The rotor velocities were 5,000 for the native enzyme and 20,h00
for the enzyme dissociated in guanidine hydrochloride. Sedimen-
tation equilibrium experiments were allowed to proceed for 2h
hours for the native and from 2h-56 hours for the dissociated enzyme.

Molecular weights, M(s/D), that were evaluated from sedi-

mentation and diffusion data were analysed by the formula of

l5

Svedberg (Ah):

M(s/D) = SRT
D(l-vp)
where R is the gas content, T is the absolute temperature, p is
the solvent density, and v is the partial specific volume.
The apparent weight-average molecular weights determined
from sedimentation equilibrium experiments were calculated from
the equation:

1 (l-vp) mg Co r 2 - r 2

 

M App RT c

where w is the angular velocity, Co (determined from a synthetic
boundary experiment) is the initial protein concentration, Cb and

Cm, and r and rm are the concentrations and radii at the bottom

b
and meniscus of the solution column, respectively.

The apparent z-average molecular weights were calculated
from the equation:

2
.21, ES ’ _E_ (:3 = MZ(l-Vp) w (Cb‘Cm)
r dr r dr RT
b b m m

 

where E: and _: are the concentration gradients at the
dr b dr m.

bottom and the meniscus of the solution column.
The intrinsic weight-average and z-average molecular weights
were obtained by extrapolation of the apparent molecular weights

to infinite dilution. Concentrations for extrapolation of the

1A

apparent molecular weights were evaluated as em + cb)/ 2 for the

weight-average molecular weights and cm + c for the z-average

b
molecular weights (A5).

Sedimentation, diffusion, and equilibrium calculations were
performed with statistical analysis on a CDC 5600 computer with
programs kindly supplied by Dr. W. C. Deal. The densities and
viscosities of solutions not containing guanidine hydrochloride
were calculated using reference tables (AA, A5). The relative
viscosity of solutions containing guanidine hydrochloride were
measured at temperatures near those of the centrifuge experiments
with a Cannon-Ostwald viscosimeter. Densities were measured
pycnometrically using water to calibrate the pycnometer at
identical temperatures to thase used for the guanidine hydro-
chloride solutions.

To prevent interference from products of mercaptoethanol
and guanidine hydrochloride (A6), freshly distilled mercapto-
ethanol was used and synthetic boundary experiments were
performed immediately after dialysis. Native samples were
dialyzed for 2A-A8 hours and the denatured enzyme was dialyzed
for 56 hours with rapid stirring.

The partial specific volume utilized in these studies was
calculated from the amino acid analysis (see amino acid analysis

section)§and found to be 0.7A ml/gm.

15
Optical Rotatory Dispersion (ORD)

ORD measurements were performed on a Cary Model 60 Spectro-
polarimeter and critical wavelength readings were checked on a
Durrum-Jasco ORD/uv—5 Optical Rotatory Dispersion Recorder. A11
readings were made under a nitrogen purge at room temperature
(Ca. 2AOC) for the Durrum-Jasco machine and in a thermostatically
controlled environment at 270C in the Cary 60.

For the Cary 60 studies, A5 mg of the pyrophosphorylase in
crystalline suspension was added to a pH 8.5 solution of 0.01 M
tris-HCl (Mann Ultra Pure) that was 0.10 M in KCl to make a
solution of about 1A mg/ml. The solution was then dialyzed for
several days against the above buffer system. Before rotatory
dispersion analysis, the solution was diluted and the protein
content was determined by the Lowry method and then converted
to a dry weight basis. To obtain readings on the native protein
in the 195-500 mu spectral range, the protein solutions were
varied from 0.26 - 1.50% in layers that were 0.1 mm to 1 cm.

The denatured pyrophosphorylase was prepared by diluting
the native enzyme solution with an 8M solution of guanidine
hydrochloride that was 0.01 M in tris-HCl (pH 8.50) and 0.01 M
in KCl to obtain a 6 M solution. The pyrophosphorylase was
allowed to remain in the denaturing solvent at room temperature
for about 2 hours prior to the rotatory dispersion measurements.
Readings were made on a 0.16% solution in cells that varied from
1 mm to 1 cm in layer thickness throughoUt the 220-500 mu spectral

range.

16

All observed rotations were corrected for the buffer blank

and converted to specific rotations ([a];) by use of the formula:

 

where aobs is the corrected observed reading of the protein
solution, 1 is the path length in decimeters, c is the concentration
in gm/100 m1, and T is the temperature in degrees centigrade.
Reduced mean residue rotations, ([m’];), were computed from the

specific rotations according to the formula:

, T ’ w 5 T

where MRW is the mean residue weight and is taken as 111 for the
pyrophosphorylase and n is the refractive index of the solvent
at wavelength A.

The solvent for the native enzyme was assumed to be water
at 20°C and the refractive index at various wavelengths was

evaluated according to the following Duclaux-Jeantet formula (A7):

n2 = 1.762550 - 0.0155998A + O°OO630957
2

 

(A - 0.0158800)
where the wavelength, A, is in microns.

The refractive index for the 6 M guanidine hydrochloride
solution was evaluated from the following basic Sellmeier
equation (A8, A9):

n2 = 1 + 0.995AA2/(A2 - 15067)

where A is measured in mu.

17

Visible rotatory dispersion was determined according to the

following one term Drude relationship:

where AC is the dispersion constant and is obtained from the slope

of a plot of [0c]TA2 vs. [a]T.
A A
The Moffitt a0 and b0 parameters were evaluated according to
the phenomenological equation of Moffitt and Yang (50):
2 A
[a]; 2 a0; 2 + 20502 2
(A ‘A0 ) (A “A0 )

or

)1

[a1T (12-102) = avoe + boxo
2 2
(A -Ao )

Using a A0 of 212 mu, 216 mu, and 220 mu for various spectral

ranges, a0 and b0 values were determined from the intercept and

T h/(A2—A02).

A (A2-Ao2) vs. A0

slope, respectively, of a plot of [a]
Titration of Cysteine Residues with Ellman's Reagent (DTNB)

All titrations with DTNB_[5,5:dithiobis (2-nitrobenzoic
acid)] were performed at 25°C in a total reaction volume of
0.5 ml. The liberated thionitrobenzoate anion was quantitated
at A12 mu by the extinction coefficient of 15.6 reported by
Ellman (51). For experiments involving pH dependent titrations,
the extinction coefficient of the thionitrobenzoate anion was
taken to be constant over the pH 6.0 to 10.0 range, since it

had been shown to vary little in this region (52). The enzyme

18

concentration was determined by its extinction coefficient or by
the method of Lowry. EDTA was included in all titrations at a
concentration of 20 mM. When guanidine hydrochloride or urea

was used as the denaturant, a freshly recrystallized sample or

a Mann Ultra Pure product was used. All titrations were initiated
by the addition of 10 pl of 10 M DTNB (Aldrich) in 0.1 M phosphate
buffer (pH 7.0). Extraneous reaction of DTNB with the solvent
system was corrected for by using a blank solution that contained

all the components except the protein.

Titration of Cysteine Residues with p—Mercuribenzoate (PMB)

 

The procedure used to determine the sulfhydryl content of
the pyrophosphorylase with PMB was based on the Spectrophotometric
method of Boyer (55) and Benesch 3: a1. (5A). The PMB (Sigma)
was purified according to Boyer's directions and standardized
immediately before use. The reported extinction coefficient
of 1.69 x 102+ (pH 7.0) at 255 mu was used.

All titrations and PMB standardizations were performed in
5.0 M urea because the native enzyme became turbid upon titration.
The pyrophosphorylase samples were dialyzed overnight against
1000 volumes of 0.05 M phosphate or pyrophosphate buffers. The
initial solution volume was 1.0 ml, and a reference cell that
contained all the components except the protein was used to
correct for the reaction of PMB with the buffer system. Ten ul
aliquots of the standardized PMB solutions were added to both
the reference and protein cells, and optical density increments

were recorded at 250 mu immediately after the addition. The

19

reaction was considered complete when further aliquots of PMB
failed to produce an absorbance increment. The Optical density
increments for the protein solution were corrected for dilution

by the PMB additions, and for the reference readings.

Amino Acid Analysis

 

The sample was prepared according to the procedure of Moore
and Stein (55). Five mg samples of salt-free and lyOphilized
pyrophosphorylase were placed in constricted heavy-walled pyrex
tubes (16 x 125 mm). The protein was then suspended in 1.0 ml
of 6N HCl and the tubes were frozen, degassed, and sealed with
the aid of a vacuum pump which reduced the pressure to below
50 microns of mercury. Hydrolysis was carried out at 1100 i 1°C
for the indicated times and the contents were evaporated twice
to dryness at AOOC in 10 ml flasks on a rotary evaporator. The
residues were readily taken up in 1.0 ml of water, and the amino
acid content of the hydrolysates were determined with a Technicon
Amino Acid Analyzer using a procedure based on the Piez and
Morris (56) modification of the Spackman, Stein, and Moore (57)
procedure. Each analysis was performed in about 2A hours with a
150 cm column loaded with type A chromobeads.

Tryptophan was estimated separately by two of the more common
methods: (1) the chemical p-dimethylaminobenzaldehyde (PDAB)
technique of Spies and Chambers (58, 59); (2) the spectrophoto-
metric determination of Goodwin and Morton (60). Tyrosine was

also estimated using the latter technique.

20

In the PDAB method, the L—tryptophan (General Biochemicals)
and PDAB (Sigma) were purified according to the directions of
Spies and Chambers. Standards and pyrophosphorylase samples,
estimated to contain 10—1507 of tryptOphan were analyzed.
Quantities of solution A (15.92 mg of L-tryptophan and 5A8 mg
PDAB dissolved in 116.0 ml of 19 N sulfuric acid)and solution B
(500 mg PDAB dissolved in 100 ml of 19 N sulfuric acid) were
mixed in 25 ml glass—stoppered erlenmeyer flasks to give a final
volume of 10.0 ml and the desired tryptOphan content. In the
case of the pyrophosphorylase, 10.0 ml of solution B were added
to the solid salt—free, lyOphilized protein. All flasks were
mixed and allowed to remain in darkness at room temperature for
12 hours. One tenth ml of 0.0A% sodium nitrite was added to each
flask,and after mixing and incubation for 50 minutes, the
optical densities of the solutions were read at 600 mu in a
Coleman Jr. Spectrophotometer. The tryptOphan content of the
pyrophosphorylase was determined from the standard curve.

In the Spectrophotometric determination of tryptOphan,
approximately 2.5 mg of the lyOphilized, salt—free pyrophosphorylase
was dissolved in 0.1 ml of 0.1 N NaOH and the exact concentration
of one-tenth and one-twentieth dilutions of the stock enzyme were
measured every 10 mu from 280 mu to 560 mu. Readings were also
taken at 29A mu. The 280 mu and 29A mu readings were corrected
for haze by extrapolation of the optical densities in the 520—
560 mu region. The moles of tryptOphan and tyrosine were determined

by the equations of Goodwin and Morton.

21

In addition to the DTNB and PMB titrations (see respective
sections), the cysteine content of the perphosphorylase was
evaluated by chromatographic analysis of the acid—hydrolyzed,
reduced and S-carboxymethylated (RSCM) enzyme. RSCM pyrophos-
phorylase were prepared according to the procedure of Crestfield,
Stein, and Moore (61). Fifteen to twenty-five mg of the salt-
free, lyOphilized pyrophosphorylase were placed in a 15 ml
screw-cap vial. Urea (5.61 g of a Mann Ultra Pure product) and
the following solutions made up in 02 free water were added:

0.50 ml of a 50 mg/ml solution of disodium EDTA, 5.0 m1 tris
buffer (5.25 g plus 9.0 ml 1.0 N HCl diluted to 50 ml with water),
and 0.1 m1 2-mercaptoethanol. The solution was made up to the

7.5 m1 mark with water and 8 M urea that was 0.2% in EDTA was used
to fill the vial to the 12.0 ml mark. A layer of nitrogen gas was
then placed over the top of the solution. After incubation for

A hours at room temperature, 0.268 g of iodoacetic acidl (Eastman)
in 1.0 ml of 1.0 N NaOH (adjusted to pH 8.5 with concentrated NaOH)
was added. Further manipulation was performed in the dark. After
15 minutes, the solution was dialyzed against either 0.01 M NH4HC03
or 50% acetic acid to eliminate the components of the reaction.
Both the solution in the case of the acetic acid dialysate and

the suspension in the case of the NH4HC03 dialysate were shell
frozen and lyOphilized. The acetic acid dialyzed sample tended

to form a glass.

 

lThe iodoacetic acid was recrystallized from a diethylether-

hexane solution and dried en vacuo over silica gel.

22

Electrophoresis on Cellulose Acetate Strips

 

A 7 mg/ml solution of the crystalline pyrophosphorylase
that was 0.05 M in sodium bicarbonate buffer (pH 11.0) and
6M in urea was dialyzed for 2A hours against the same buffer
system prior to electrophoresis. Cellulose acetate strips
(2.5 x 18 cm) were soaked for 1 hour in the outside dialysis
solution used above. Excess solution was blotted from the
strips with the aid of a piece of filter paper. Five ul
were streaked across the strip, without scratching the cellulose
acetate surface or spotting the edges of the strip. The still
moist, but thoroughly impregnated strip was subjected to
electrophoresis at a constant voltage (65-70 volts per strip)
in a Shandon Electrophoresis Apparatus for 5 hours at AOC. The
strips were then gently blotted with filter paper and fixed by
submersion for 10 minutes in 5% trichloroacetic acid (TCA). The
excess solution was removed by blotting and the strips were
stained for 10-15 minutes by submerging them in 0.2% Ponseau S
in 5% TCA. The excess dye was removed by repeated washings of the

strips in 5% acetic acid prior to drying.
Isoelectric Focusing

Isoelectric focusing was performed on an LKB 8101 electro-
focusing column with a 110 m1 capacity. The technique used was
described by Svensson (62) and Vesterberg and Svensson (65). A
step-wise 0-50% sucrose density gradient was used with an equal
mixture of pH 5-10 and pH 5—8 LKB carrier ampholytes which were

used at a final concentration of 1% (w/v). All solutions including

25

those of the cathode and anode were made 6 M in urea. To prevent
oxidation and reduction, sulfuric acid and ethylenediamine were
added to the anode and cathode solutions, respectively. Six mg
of the pyrophosphorylase in 0.5 ml were added to an intermediate
fraction of the less dense solution, and solid urea was added to
a concentration of 6 M. The analysis was performed for A8 hours
at ADC and final focusing was accomplished at 900 volts. The
column was then drained and 9 drop fractions were collected and

were analyzed at 280 mu for protein.

Trypsin Digestion and Peptide Mapping

 

Fingerprinting experiments on the perphosphorylase were
carried out by standard procedures (6A, 65). Trypsin digestion
was found to proceed best when the enzyme was heat denatured.
A 10 mg/ml solution of the pyrophosphorylase in 0.2 M NH4HC03
(pH 8.6) was dialyzed for 2A hours against the same buffer. The
solution was then transferred to a small glass vial, and the protein
was coagulated by heating the vial at 90°C for 6 minutes. The tube
was then cooled by emersion in an ice bath. The protein was
dispersed by use of a small magnetic stirring bar. A 2% protein
ratio of TPCK-trypsin (Worthington Biochemicals Corporation) was
added over a period of 20 hours. Trypsin was a freshly prepared
5 mg/ml solution in 0.001 N HCl. The clear and colorless solution
was centrifuged to eliminate a trace amount of suspended material
and the digest was then shell-frozen and lyOphilized.

The white fluffy digest proved to be insoluble in water, but

soluble in dilute ammonia water (1:15). Two to three mg of the

2A

digest was routinely used for each map and was applied in enough
ammonia water to make about a 50 mg/ml solution.

Chromatography was performed in the first direction on full
sheets of Whatman 5 MM chromatography paper (18.5 x 22.25 cm).
Discrete spots were observed when the solvent system was n-butanol:
pyridine: acetic acid: water (90:60:18:72). The chromatograms
were routinely developed in a chromatocab for 27 hours at room
temperature and then air-dried overnight. Electrophoresis was
performed in the second direction in a Gilson Model D Electro-
phorator. The chromatograms were exposed to 2100 volts for
2 hours in the pH 5.5 buffer system containing pyridine: acetic
acid: water (1:10:289), and the papers were again air-dried
overnight.

The peptides were visualized by dipping the papers in 0.25%
ninhydrin (General Biochemicals) in acetone; the spots were
allowed to develop at room temperature for 5 hours, and then in
the dark overnight. Tryptophan-containing peptides were
visualized by dipping the ninhydrin—developed chromatograms in
Ehrlich's stain (66) which was prepared by freshly mixing 900 ml
acetone, 100 m1 concentrated HCl, and 10 g p-dimethylaminobenzal-
dehyde. The HCl bleached all the peptides and after several
minutes, the tryptOphan—containing spots developed a blue color.
The modified Sukaguchi reagent (67) was used to detect the
presence of arginine on untreated maps. The chromatOgrams were
dipped in 0.0125% d-napthol in absolute ethanol. After air

drying, they were sprayed lightly with a solution of 1.5 ml of

25

A-5% sodium hypochlorite and 25.5 ml of 10% NaOH. Arginine

peptides turned a light red color.

Cyanogen Bromide Cleavage

 

Cleavage of the pyrophosphorylase with cyanogen bromide
was performed by the procedure of Gross e; g1. (68) as modified
by Steers et__l. (69). Three to four mg of RSCM—pyrophosphorylase
in a glass-stOppered tube were dissolved in 2.0 m1 of 70% formic
acid. A 50-fold excess of cyanogen bromide (Eastman) over
methionine (taken to be about 70 residues per mole) was added.
The reaction was allowed to proceed at room temperature for
16-20 hours and then the digest was dried and the volatile
components were removed by lyophilization. A white, fluffy
digest was obtained.

The components of the digest were monitored by electro-
phoresis on 7.5% polyacrylamide gel columns with the pH 8.7
system described by Davis gt g1. (70). All gels were 5 M in
urea and the digest (200-500 ug) was added to the sample gel
in a 10 M urea solution, so that the final gel concentration
was 5 M. Solubilization was also possible in a 1:15 solution
of NH4OH, but better acrylamide patterns were obtained with the
urea system.

The subunits and the RSCM-enzyme were compared with the
cyanogen bromide digests using 5% polyacrylamide gel columns

that were 5 M in urea.

26

Subunit Reassociation

 

Dissociation and reconstitution experiments were performed
according to the procedure of Deal (71). The dissociation
medium contained 2.A5 gm of urea (8 M), 0.1 ml of mercaptoethanol,
0.50 ml of 2 M ammonium sulfate, 2.A0 m1 of 0.A7 M glycine buffer
(pH 9.5), and water to make 5.0 m1. A crystalline suspension of
the pyrophosphorylase (1.5 mg) was diluted to 1.0 ml with the
dissociation mixture and a control system was prepared by
dilution of the enzyme into a mixture containing everything
except the urea. The control tube was warmed at 55°C to dissolve
the crystals and both samples were allowed to remain for 2 hours
at 5-A°C.

Reassociation experiments were attempted by two different
procedures. The first method was the direct dilution of the
control and dissociated enzyme into the system wherein activity
was measured. A second procedure was the 1:100 dilution at 0°C
of the dissociated enzyme into a reassociation mixture which
contained 0.25 M imidazole buffer (pH 6.3), 0.2 M KCl, 0.03 M
glutathione, and either 0.0A M DPN, 0.01 M UDPG or no additional
component.

In the method that utilized the reassociation mixture, care
was taken to keep all micropipettes at 0°C, and all dilutions were
made slowly with stirring at that temperature to prevent precipit—
ation. After the dilution at 0°C, the tubes were transferred to a
water bath at 15-16°C, and aliquots were removed for activity

measurements for times up to 5 hours.

RESULTS

Absorption Spectra

 

Figure 1 shows the typical aromatic amino acid spectra of
the pyrophosphorylase. A maximum was observed at 278.5 mu and
a minimum at 255.0 mu. Peaks were also noted in the spectra at
258.5 mu, 265.0 mg, 268.0 mu, 285.5 Hm,and 292.0 mu. The

purified enzyme exhibited a 280/260 ratio of 1.6.

Extinction Coefficient

 

The protein samples were weighed with a precision of
10.025 mg, and the Optical density readings were obtained with a
precision of about 1%. It was found that in 0.01 M phosphate
buffer (pH 7.0) a 0.1% solution of the pyrophosphorylase had an
optical density of 0.719 in a 1 cm light path at 280 mu. This

Sat

corresponds to a molar extinction coefficient of 5.A0 x 10
this wavelength.

Protein estimated by dry weight was 95% of the values
determined by the Lowry method with BSA as a comparison standard.

Also, protein estimated by the 280/260 method of Warburg and

Christian was 21% lower than the actual dry weight.

Carbohydrate Content

 

The carbohydrate content, as determined by the phenol-

sulfuric acid method of Dubois 35 gl., was no more than 0.02%.

27

WC

Figure 1. Ultraviolet absorption spectra of
the pyrophosphorylase
The protein concentration was 0.80 mg/ml in
0.01 M tricine (pH 8.56). A Cary 15 Spectrometer
was used with a 1 cm light path at a temperature

of 2A°C.

29

1E .Ihwzmqm><>>
mum 9m no.0. man EN new new now new

_ fl _ _ _ _ T

 

 

 

 

 

ALISNEO ‘IVOlldO

5O

Homggeneity and Molecular Weight of the Enzyme by Sedimentation

and Diffusion Techniques

 

When two and three times recrystallized pyrophosphorylase
was subjected to ultracentrifugal analysis, it became obvious
that the pyrophosphorylase was polydisperse. With the major
sedimenting component, there was at least one more rapidly
sedimenting component (Figure 2). At concentrations (0.2-1%)
of protein and conditions used in these experiments, the minor
components amounted to about A% or less of the total protein.
Recrystallization of the enzyme up to six times has not altered
the relative amount or mobility of these minor components. At
times, two rapidly sedimenting components may be visualized. In
an experiment that yielded three components, the sedimentation
coefficients,Sgééi,were 15.22 for the principal component, 19.66 S
and 25.08 S for the two minor components. Using the sedimentation
coefficients and assuming spherical entities with a partial
specific volume of 0.7A ml/gm, approximate molecular weight ratios
of 1.0, 1.8 and 2.6 were obtained for the major component and the
two minor components, respectively. It thus appears that multimers
of the pyrophosphorylase molecule may exist. The results were
essentially identical regardless of whether mercaptoethanol at a
concentration of 0.1 M was included in the buffer system. The
comparison of the mobilities of the native enzyme in the presence
and absence of mercaptoethanol is shown in Figure 5. It thus
appears that a reducing agent itself would not cause dissociation

of the enzyme.

51

Figure 2. Schlieren sedimentation velocity
pattern of the perphosphorylase
The temperature was A°C and the enzyme was
5.2 mg/ml in 0.01 M tricine buffer (pH 8.5), 0.1 M
in mercaptoethanol, and 0.1 M in NaCl. The
diaphragm angle was 65° and sedimentation is from

left to right.

52

,b

.a
..;
I
ll
.1
.
z
1-
1w”

 

55

Figure 5. Relative sedimentation rates for the
enzyme in the presence and absence of
reducing agent

The upper pattern shows the enzyme in the
presence of 0.1 M mercaptoethanol, and the lower

pattern shows the enzyme in the absence of 0.1 M

mercaptoethanol. Both samples were 0.01 M in

tris-HCl (pH 8.5) and 0.1 M in NaCl.

5A

 

55

Figure A. Apparent sedimentation coefficients,
diffusion coefficients, and weight-average
molecular weights for the principal
component of the perphosphorylase

The enzyme was analyzed in 0.02 M triethanol-

amine buffer (pH 7.5), 0.1 M in NaCl, and 0.1 M in

mercaptoethanol.

56

3.ON
bzw_o_.n_n_woo zo_mDn_n:o
5 5. 5 5

 

 

0.0. x :1ng ".3830:

5

3

0.

 

 

 

hzwariwoo 20—h4h2w25wm

 

 

 

5. 4 3 2 2 .3 5
_ 1 a .e . _ _
l -
. .
.. m . i l -mu
. * M
2m . /
1 com 1 1 18%.
_.,o l. m N
T S l u l l a. 16nlv
e .. . a m
m. a m _. ..., m
T .. 2% I I m l4~
We : . LQW E
08% m4 on m
.. 0D“. 1 .. 12m
/ .
_ _ e n_ _ _
w M B W. 4 3 2
3. 3
Afiocumv sag
o0. _

57

Since the minor components represented a small fraction of
the total protein, molecular weight determinations were made
using the sedimentation and diffusion techniques. Sedimentation
coefficients have varied under all conditions by about 6% and
combination of sedimentation and diffusion data were therefore
determined on the same sample. Table 1 compares the sediment-
ation behavior of the pyrophosphorylase under various conditions
of buffer and pH, and different protein preparations. In some
of the earlier work, the sedimentation coefficients varied by
more than 6%, but it is believed that this simply represented an
inaccurately known protein concentration. Figure A shows the
extrapolation to infinite dilution of the apparent sedimentation
and diffusion coefficients. The principal boundary was used for

this analysis. Extrapolation to zero protein concentration

0

yielded a value for S20,w

-7

of 1A.A5 and a value for Dgo,w of 2.87 x
10 cm2/sec. Combination of the apparent sedimentation and
diffusion coefficients have yielded a molecular weight for the
principal component (M8(s/D)) of A72 x 105 gm/mole. Although

the sedimentation and diffusion coefficients varied in some

cases, extrapolation to infinite dilution under various conditions
of buffer, pH, and different protein preparation yielded values

in close agreement. For example, an experiment in 0.01 M tricine

(pH 8.5) that was 0.1 M in NaCl and 0.1 M in mercaptoethanol

of 1A.10 S, a D20 of 2.72 x 10'7 cm2/sec and a

. 0
yielded a 820 w ,w

)

M3(s/D) of A85,000 gm/mole.

58

HOGmLDOODOmonoEIN u mmzm

 

 

mm.ma 00.: A m.m m0 ”H002 z H.0 mocAEoHocozooanu 2 No.0
m0.mH 00.: s m.w m0 ”H002 z H.0 mooflosamazosflm z H0.0
0:.ma 00.: e m.w ma ”H002 z H.0 mocfloan z H0.0
em.na 0m.m e m.e ma mmmz 2 H.0 ”H002 z H.0 mocaeoaocoguoann z m0.0
mm.mH ne.m 0 m.w mm ”mm: 2 H.0 maomz z H.0 ”defloHED z H0.0
n0.mH 00.: m m.w mo mmmz z H.0 maooz z H.0 mocflofinu z H0.0
0H.mH 00.m : m.w ma mmmz z H.0 ”Hooz z H.0 mocaofinn z H0.0
:m.ma mm.m m m.w mo ”mm: 2 H.0 maooz z H.0 moeaoano z H0.0
Hm.ma we.m m m.w mm ”mm: 2 H.0 ”Hooz z H.0 mocfiofleo z H0.0
mm.ma 0m.m a m.e mm Mam: : H.0 mauoz z H.0 “Hom-mano z H0.0
00.ma 0m.m a m.e ma maomz z H.0 ”Hom-mfinp z H0.0
sqomm AHE\mEv COHpmpm mQOHuflocoo
Gwouonm udmpm

 

 

m:0flufiocoo mcfi>nm> noon:

omma>nonamozmop>m omoozamumms pom mucoe0flmmooo COHumu:oEHUom

H mqmflH

59

Molecular Weight of the Natiye Enzyme by the Sedimentation

Equilibrium Technique

 

Sedimentation equilibrium experiments yielded molecular
weight estimates that were less equivocal than the sedimentation
and diffusion techniques. These experiments are shown in figure
5. Discrepancies are noted not only between the weight-and z-
average molecular weights, but also between the limiting z—average
and the intrinsic z—average molecular weights. Due to some
scatter in the weight—average molecular weights, a value between
A76,000 and 570,000 was found. The limiting value of A76,000
is in good agreement with that obtained from the sedimentation
and diffusion techniques. Similarly, values of 606,000 and
1,052,000 were obtained for the limiting z-average and the intrinsic

z-average molecular weights.

Homogeneity and Molecular Weight of the Subunits by Sedimentation,

 

Diffusion, and Sedimentation Equilibrium Techniques

 

Figure 6 shows the relative rates of migration of the native
enzyme in the lower pattern and the enzyme denatured in 5 M urea
in the upper pattern. It appeared from this experiment that
subunits of the pyrophosphorylase exist. Figure 7 shows the
similarity in mobility of the denatured states of the pyrophos—
phorlase in urea and guanidine hydrochloride.

Sedimentation, diffusion and equilibrium studies were
performed in 6 M guanidine hydrochloride and 0.1 M mercaptoethanol

to characterize the pyrophosphorylase subunits. In this solvent

A0

Figure 5. Apparent weight-average, z—average and
limiting Znaverage molecular weights for
the pyrophosphorylase

Concentrations were evaluated as (Cm + Cb)/2 for

the weight-average and Cm + C for the limiting and

b
intrinsic z-average molecular weights. The
analysis was performed at 5°C at a speed of 5,000 rpm.

The buffer system contained 0.01 M tris-HCl (pH 8.5)

that was 0.1 M in NaCl and 0.1 M in MSH.

A1

X 1H9|3M HV‘anS'IOW

..Ol

00.

0.00

Qmm

Qmm

o_
8

35:05: zo_._.<m._.zmozoo
0m 0. 0. s N. 0. 0 0 v N

 

 

 

 

 

 

 

 

 

    

_ _ _ a _ _ _
000.80.. . Ns.
r. G o _
7...? 111 a a
I )0!
TI.) (WHHMiLi... -a
T I , m
000.000....__.Ms_
T N2 G V
E...N
T s. . 000.0% - 000. 0% as. 1 a
32 o no
_ _ _ _ _ _ _ _ r _

 

 

 

A2

Figure 6. The relative sedimentation rates for the
native protein and the subunits

The pictures were taken 1 1/2 hours after
attaining 59:780 rpm and were spaced by 52-
minute intervals. The lower pattern shows the
native enzyme at A mg/ml in 0.02 M triethanolamine
buffer (pH 8.5) that was 0.1 M in NaCl. The upper
pattern shows the subunits at A mg/ml in 0.02 M
triethanolamine (pH 7.8) that was 0.01 M in NaCl and

5 M in urea. The temperature was 6.6OC.

A5

 

I.“ .
.9 v'J .
. . in .12]. . .a ”4.11.1.1! "nL-V

 

AA

Figure 7. The relative rates of sedimentation of
the subunits in guanidine hydrochloride
and urea

In both cases the enzyme was 5.5 mg/ml in

0.01 M NaCl. The upper pattern shows the enzyme

in 7 M urea (pH 8.1) while the lower pattern shows

the enzyme in 5 M guanidine hydrochloride (pH 7.6).

The rotor speed was 59,775 rpm and the temperature

was A.2°C.

 

A6

system, the sedimentation coefficient decreased markedly, but

the diffusion coefficient remained essentially the same as the
native enzyme. Figure 8 shows the extrapolation to infinite
dilution of the apparent sedimentation and diffusion coefficients
of the subunits. Using a partial specific volume which was the
same as the native enzyme and uncorrected for preferential

O -7
of 2.15 S and a D20 w of 2.90 x 10

J

O
solvent effects, a 820 w

J

0
and D20 w gave a

cme/sec. was obtained. Combination of SO
20,w ,

M8(s/D) of 69,100.

Low speed sedimentation equilibrium experiments yielded
a value for the molecular weight of the subunits in good agree-
ment with that obtained from the Svedberg relationship. Figure
9 shows the extrapolation to infinite dilution of the apparent
weight- and z-average molecular weights. The fact that both
extrapolated to 67,500 indicated that the subunits were similar,
if not identical, in mass.

Production of these subunits is accompanied by a major
change in shape of the protein, as judged by molecular weight
and frictional coefficient analysis. The sedimentation and
diffusion coefficients are lower than expected for a globular
protein of molecular weight 68,000. Furthermore, a frictional
ratio of 2.8 was found for the subunits as compared with 1.5
for the native enzyme. The subunits are, therefore, highly

unfolded.

A7

Figure 8. Apparent sedimentation and diffusion
coefficients and weight-average molecular
weights of the subunits

The dissociating solvent system was 0.01 M

tris-HCl (pH 8.5) that was 6 M in guanidine

hydrochloride, 0.1 M in MSH, and 0.1 M in NaCl.

 

 

 

ammo.

500302
0 -0. x Eon; «4
590.1180 20.021. . 0 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5 5 5. 5. u... w m
. 4 3 2
_ m_ J _ _ — A
.
. 10)
I . I I C IL
0 O. O M
e
s _ .nnuu W
. J 9
MW 7m ’" = N
.... .- .. 1 / 5n
.0 O I“ A
m “ ow m
_. .. M 1m.
,. a... n c
D O an
“ I I I20
I .
.
/. [III
_ u

 

 

 

 

 

 

 

O
0.5 —-
3
2
I

 

 

kzmafimmoo zoFdhzwgn—wm

 

 

A9

Figure 9. The apparent weight-average and 2-
average molecular weights of the subunits

The weight- and z-average molecular weights

were analyzed as described in figure 5 and the

dissociating system was the same as described in

figure 8. The analyses were carried out at 5°C

for 2A-56 hours at 20,A00 rpm. Various preparations

of the pyrophosphorylase are indicated by the

symbols -A- and —O-.

50

:.Ol x .I.H9I3M HV'II'IOB‘IOW

 

3 2 \ 02 V Zo_._.<m._.zmozoo

NN ON m_ m_ E N_ O_ m m v N
_ _ _ l _ _ _ _ _ _ _

 

  

000... .2 .i 1.
00m .. .2 0.5 N
w
0mm I m mm!
x
0.0N ne m
9
0.0N .. 009.0... .2 10
N0. 1 10

 

 

 

51

Optical Rotatory Dispersion

 

When native and denatured pyrophosphorylase were subjected
to optical rotatory dispersion analysis in the 190-550 mu
spectral region, the results shown in figure 10 were obtained.
Figure 11 shows an expanded plot of the specific rotation versus
wavelength in part of the visible and near ultraviolet regions.
In these regions, the specific rotation decreased in a regular
fashion for both the native and denatured states. Re-evaluation
of these data according to the graphic procedure of Yang and
Doty (72) is shown in figure 12. This procedure involved plotting

[QIXAg versus [a] The fact that these plots are linear for both

A'
the native and denatured states indicates that the one term
equation of Drude (75) is adequate to account for the visible
and near ultraviolet dispersion. The dispersion constant, AC,
which was obtained from the slope of the Yang and Doty plot was
found to be 2A6 mu and 216 mu for the native and denatured enzyme,
respectively.

Analysis of the rotatory dispersion of the native enzyme in
the 280-500 mu spectral range according to Moffitt and Yang is
shown in figure 15. A value of 212 mu for AC was adequate to
obtain linear Moffitt plots in the 550-600 mu spectral range (7A, 75).
Choosing a A0 of 212, 216, and 220 mu, the linearity of the
Moffitt plots was extended to 520, 510, and 285 mu, respectively.
Values for ac of ~271, -20A, and -200 deg—cm2/decimole and values

for b0 of -l25, -87, and -76 deg-cmZ/decimole were obtained when

AC was 212, 216, and 220 mu, respectively.

.__4::.!!_(-
. o ‘-

___. 4....—
o

 

52

Figure 10. Ultraviolet optical rotatory dispersion
of the native and denatured enzyme

The symbol -0- represents the native enzyme
and —0- represents the enzyme denatured in 6 M
guanidine hydrochloride. Both solutions were
0.01 M in tris—HCl (pH 8.50) and 0.10 M in KCl.
The native enzyme varied in protein concentration
from 0.521% to 0.260% in layer thicknesses of
0.02 to 0.001 decimeters. The denatured enzyme
was 0.165% in layer thicknesses that varied from
0.1 to 0.001 decimeters. The analysis was performed

at 2700.

55

 

I40 I 1

I2.0I-

I0.0 -

 

-6OI____L___L- .

 

 

I I j I l I I I
.1.
. 1 l I 1 I I 1
220 240 260 280 300 320

4-”; ml.

 

:"w

5A

Figure 11. Visible and near ultraviolet Optical
rotatory dispersion of the native and
denatured enzyme

The symbol -0- indicates the native enzyme
solution and -0~ indicates the enzyme denatured
in 6 M guanidine hydrochloride. The conditions
of the analysis are given in figure 10. The
native enzyme was 1.505% in a layer thickness

of 0.1 decimeter, and the denatured enzyme was

0.165% in the same layer thickness.

55

own

u-“ -... ,

...._—

Ill... -Cilflu‘i. lla‘ti .i‘i!

000

09..

as:
owe

o

n

 

_r

 

 

 

 

56

Figure 12. Drude plots in the visible and near
ultraviolet regions for the native
and denatured enzyme

The symbol -0- depicts the native enzyme and

-O- the enzyme denatured in 6 M guanidine hydro—

chloride. The conditions are given in figure 10.

57

-. inn?

8
5.6

- 5.4

‘52

~5.0

- 4.8

" 4.6

 

 

~5.

 

 

3.3-
3.I I’

 

Figure 15. Moffitt-Yang plots of the native
enzyme in the 280—500 mm spectral
region

See figure 10 for the experimental conditions.

59

O.

 

m h

.5. . 3.2.2
w m ¢

5N

 

 

— fi

:5 N_Nu£
1E 0_N "OKN \

as 0NNu2

 

d _ -

O

 

 

apl x (°;-z‘<)<[>°]-

60

The Moffitt plot of the data for the enzyme denatured in
6 M guanidine hydrochloride is shown in figure 1A. A A0 of
212 mu was utilized and the ac and b0 values in this solvent
system were -6A2 deg-cme/decimole and approximately 0 deg-cm2/
decimole, respectively.

A summary of the optical rotatory dispersion parameters of

 

the native and denatured enzyme is given in table II. There was I;
a blue shift of l mu of the cotton effect trough and a red shift I
of A mu of the cotton effect peak from the accepted values for

helical proteins and synthetic polypeptides. Also, the cotton I
effect extrema showed relatively low amplitudes when compared 5;

with values for helical polyglutamic acid ([m’] of 15,000-

255

16,0000 (76,77) and [m’] of 70,000-80,000O (76,78)).

198
Estimates of helicity have varied depending upon which
method of calculation was utilized. Using a value of 212 mu
and 2A6 mu for the denatured and native dispersion constants
and 212 mu and 25A mn for 0 and 50% helicity of reference poly-
glutamic acid (7A), a helical content of 21.A% was calculated.
Also, employing a be of —125 deg-cm2/decimole (A0 of 212 mu) for
the native perphosphorylase and -650 deg—ch/decimole for the
be of helical polyglutamic acid, a helical content of 20% was
calculated. The excellent agreement in the calculated helical
content from the above methods is not substantiated analyzing
the amplitudes of the cotton effect trough and peak (77). A
value of 10% or less was obtained from the amplitude of the
255 mu trough. Standard values for the peak regions are still

uncertain, but a helical content of about 16% was calculated

when a suggested amplitude was utilized.

61

Figure 1A. Moffitt-Yang plot of the denatured enzyme
in the 280-500 mm spectral region

See figure 10 for the experimental conditions.

62

 

 

 

0.0
. .
0e 11
e: 7
7v
0: .
wxz
0.... H
0..» I.
«m 1
«m

 

 

65

 

 

 

$7.1 - .asallr
oHosfioop\msoumop Lo muficsm
as 00m-mwN A0
m m
as 00m-0am as oomuoan An 0 s so
Hmohuom m ESEHXME
as 00muon A0 III as NON . 2
ESEHXmE
I U n . U
as A I 5:5: a: 5.1 a
50 em - An on
Hm>0mw090
mnmau Am Ill. 15 mmm K
as
00N- :0 osmmm- oNHmm- mmm~.s_
m.N:0. :0N- An oo 0
39H
oHsN- A0 .III as NMN s UK
0
as 0NN A0 as OHN as 0:N 2
as NHN as GHN An 02
as 00m
as N N o 0.0» - m.» - 0
H A O H O m OJNfi H
nonsum:om o>flpmz nouosmpmm oonsumcwm o>wumz souwsmumm
unfimmoz Honosoo

 

 

 

 

wmmaznonmmosmop>m omoozamumms wo coflmuommflm Apoumuom HOOHDQO

HH mqmae

6A

Titration of Cysteine Residues with Ellman's Reagent

 

The rate of reaction of the sulfhydryl groups of a several
month old preparation of the pyrophosphorylase at pH 7.5 is
illustrated in figure 15. At this pH, 9-11 SH groups of the
native enzyme reacted in less than 2 minutes. In 6 M guanidine
hydrochloride, lA-l6 SH groups immediately reacted.

In order to detect if the DTNB reactable SH groups of the

pyrophosphorylase were sensitive to pH, titrations were performed

 

in the pH 6.0 to 10.0 range. The titrations proceeded for 15

ti.” Jan-41;: . .
.

minutes before the final readings were taken. Figure 16 shows
the effect of pH on the number of SH groups that react readily
with DTNB. There was a gradual increase in the number of SH
groups with increasing pH, and the maximum value found was 16
moles of SH per mole of protein. Other preparations have shown
the same increasing trend in reactable groups, but the absolute
number has been lower than that indicated in figurd 16. In fact,
preparations that were a year old (stored in a crystalline state
in (NH4)2SO4) yielded from 1-15 readily reactable SH groups in
the native state and from 5-16 readily reactable groups in the
denatured state. The various preparations did not appear to
have significanély different sedimentation coefficients. At
present, 16 moles of SH per mole of protein represents the maximum
number found. After these are titrated in the native state, the
enzyme maintained 70% of the activity of the unreacted protein.
As expected, 0 SH groups per mole of protein was found when
the reduced and unreduced S—carboxymethylated pyrophosphorylase

was titrated with DTNB.

65

Figure 15. Rate of reaction of the sulfhydryl
groups of the native and denatured
enzyme with Ellman's reagent

Both reactions contained 0.1 M tris—HCl
buffer (pH 7.5), 20 mM EDTA, 0.2 mM DTNB, and

1.250 mg/ml of pyrophosphorylase. The denatured

enzyme solution was 6 M in guanidine hydrochloride.

66

05.32.: 2. win.

0.1N.0_00¢N00_.¢_N_0_00¢N0
u—udfi d—qu-afiq

 

l

 

 

52>sz oumPEzmo

 

 

”1030:1089: Nzazsao

 

P
I

 

w2>Nzw

m>_._.<z

l

00

0.0

2

 

 

"I“ 3H7 'AJJSNBO "IVOIldO

67

Figure 16. The pH dependence of the number of SH

groups titrated per mole of protein
All titrations contained 0.05 M glycylglycine

(-O-) or 0.05 M glycine (-O-) buffer that was 20

mM in EDTA and 0.2 mM in DTNB, and the protein

concentrations were about 1.0 mg/ml in each

titration. Samples were dialyzed overnight against

the appropriate buffer prior to titration. The

symbol -A- represents the percent of activity

relative to an untitrated sample at that pH.

68

>._._>_._.0< ._<z_0_m0 hzwomwn—

0
0 0
m 8 w 4 w
__._T_n___a

 

 

 

 

 

I4—

_
m

MAO: \ mM.—.45.; Im

I0.0

8.0 9.0

7.0

69

Titration of Cysteine Residues with p-Mercuribenzoate

 

The Spectrophotometric titration of the sulfhydryl groups of
a several month old preparation of the pyrophosphorylase with
PMB is depicted in figure 17. The end point in this titration
corresponded to 8 sulfhydryl groups that were available for
mercaptide formation per mole of protein. Up to 12 sulfhydryl
groups per mole of enzyme have been reacted when the buffer was

0.05 M pyrophosphate (pH 7.5) instead of 0.05 M phosphate buffer.

Amino Acid Composition and Partial Specific Volume

 

The amino acid composition of the pyrophosphorylase is given
in table III. Cysteine was determined from a DTNB titration of
the denatured enzyme. Conflicting values for the analysis of
cysteine based on the S-carboxymethylation, PMB, and DTNB methods
necessitated the choice of one of these techniques. The DTNB
method, however, suffers from the fact that only cysteine and
not total half-cystine is determined and therefore the value
found must be taken to be a minimum. Moreover, chromatographic
analysis of the hydrolyzed, reduced and S-carboxymethylated
protein has indicated a higher value for the half-cystine content.
However, the assessment of half-cystine has been complicated by
the unexpected finding of 0 moles of S-carboxymethylcysteine for
the hydrolyzed, unreduced protein.

Twenty—one residues of tryptOphan per mole of pyrophosphorylase
is presently the best estimate. This was determined by the Goodwin

and Morton spectrophotometric method and was in relatively good

70

Figure 17. Titration of the sulfhydryl groups
of the enzyme with PMB
The PMB concentration was 7.52 x lO—uM and
the pyrophosphorylase was 8.68 x 10‘6M in 0.05 M
phosphate buffer (pH 7.0) that was 5 M in urea.
Additions of the PMB were made to 1.0 ml of the

protein solution.

71

 

I
4—.—
I
IZO

 

r
l
80

I
L
60
PMB, ul

 

 

L I I I I

III“ 092 'lNBWBHONI AllSNBCI 'IVOIldO

 

 

72

TABLE III

Amino Acid Composition of UDP-glucose Perphosphorylase

 

 

 

 

 

TimeEThydrolysis Average Residues
. . or
Amlno AC1d 20 hrs A2 hrs 72 hrs Extggpgizted E81ed
Mole ratioa
Asp . 1.060 1.080 1.08A 1.076 502
Thr . .A75 .A91 .AA5 .520 2A5
Ser . .616 .591 .A8A .680 518
Glu . .972 1.008 .961 “980 A58
Pro . .A52 .A91 .A55 .A59 21A
Gly . .615 .655 .610 .619 289
Ala . .56A .591 .565 .575 17A
Val , .668 .755 .687 .696 525
Met . .156 .166 .1A5 .156 75
Iso . .50A .555 .559 .525 2A5
Leu . 1.000 1.000 1.000 1.000 A67
Tyr . .257 .225 .259 .2A0 112
Phe . .589 .A08 .597 .598 186
Lys . . . .720 .708 .71A .71A 555
His . .19A .216 .191 .200 95
Arg . .575 .A00 .572 .582 178
Cys . .05Ab 16
Try . .0A5C 21

 

 

 

 

 

 

8Relative to Leucine taken as 1.000

Minimum value

denatured enzyme.

CDetermined by the Spectrophotometric technique (60).

dThe molecular weight was taken to be A80,000.

determined from a DTNB titration of the

 

75

agreement with the chemical method of Spies and Chambers which
yielded 26 residues of tryptOphan per mole of enzyme. The
spectrophotometric method was used because it afforded an indepen-
dent check on the amount of tyrosine obtained from the amino acid
analyzer. One hundred and sixteen residues of tyrosine per mole
of enzyme determined by the spectrophotometric method was in

good agreement with 112 residues determined by the amino acid
analyzer.

The partial specific volume of the pyrophosphorylase, calcu-
lated according to the method of McMeeken and Marshall (79) is
shown in table IV. The partial specific volumes of the amino
acids were obtained from Cohn and Edsall (80). Little difference
was noted in 9 when estimates of glutamine and asparagine were
used in the calculation. A value of 0.759 ml/gm was calculated

for the partial specific volume of the perphosphorylase.

Subunit Electrophoresis on Cellulose Acetate Strips

 

Because the native enzyme did not yield a discrete band under
all conditions, cellulose acetate electrophoresis was performed
in the presence of a dissociating concentration of urea. 'Figure
18 shows the migration toward the anode of the dissociated enzyme
at pH 11.0. In the denatured state the enzyme migrated as a

discrete band.

Isoelectric Focusing of the Subunits

 

In contrast to other experiments in 6 M urea, isoelectric
fractionation revealed that the pyrophosphorylase was hetero-

geneous. Figure 19 shows the electrofocusing optical density

IL
3?

Calculation of the Partial Specific Volume

From the Amino Acid Analysis

7A

TABLE IV

 

 

 

 

 

Amino Acid Regijues % by Weight igi:;:ic % by Volume
Residue Mole of Residue Residue of Residue
Asp 502 12.15 0.59 7.17
Thr 2A5 5.16 0.70 5.61
Ser 518 L.82 0.65 5.67
Glu A58 12.A0 0.66 8.18
Pro 21A A.58 0.76 5.55
Gly 289 5.A9 0.6A 2.25
Ala 17A 2.61 0.7A 1.95
Val 525 6.75 0.86 5.85
Met 75 2.00 0.75 1.50
Iso 2A5 5.85 0.90 5.25
Leu A67 11.10 0.90 9.99
Tyr 112 5.85 0.71 2.72
Phe 186 5 75 0.77 A.A1
Lys 555 8.97 0.82 7.56
His 95 2.69 0.67 1.80
Arg 178 5.8A 0.70 A 09
Cys 15 0.58 0.65 0.9A
Try 21 0.8A 0.7A 0.62
TOTAL --- EN. 99 --—- FWiV = 75.9

V : fwivi/‘fwi =

l

1" /
ongh ml/gm

 

aTaken from Cohn and Edsall (80).

_ u.- _- I;

w} z

 

75

Figure 18. Electrophoresis of the subunits on
cellulose acetate strips
Electrophoresis of the subunits (55 ug) was
performed for 5 hours at 70 volts in 0.05 M
bicarbonate buffer (pH 11.0) that was 5 M in urea.
The protein was stained with 0.2% Ponseau S in 5%

TCA. The temperature was A°C.

76

53:.

 

77

Figure 19. Isoelectric focusing of the subunits

Isoelectric focusing of 6 mg of the subunits
was performed in 6 M urea as described in the
Materials and Methods section. The line (-——5
indicates absorbance at 280 mu and the line (---)

indicates the pH gradient.

78

om.

00.

0S

mmmZDz 20:05“?—

ON.

00.

00

00

0¢

0N

 

 

_

 

_

_

_

 

 

 

 

’0' ' - "l

0 0 . 0' 0 0
) fiw 092 ALISNBO 'IVOIldO

Q
0

I

 

 

79

and pH profiles of the subunits. At least eight fractions could
be detected.

An attempt was made to analyze each fraction on the basis of
its tendency to reassociate to form active enzyme. The following
reassociation procedure was used. Fifty ul of each fraction was
diluted at 0°C to 1.0 ml with 0.01 glycylglycine buffer (pH 7.5)
which was 0.1 M in ammonium sulfate and 0.1 M in mercaptoethanol. I
The solution was then incubated for 1 1/2 hours at 16°C and
analyzed for the presence of pyrophosphorylase activity. Using

this procedure, only peaks 1, 2, 5, and A showed activity, and

 

the activity was approximately in the same ratio as the peak é
heights. Fraction 6 which was present in the largest amount was

devoid of activity. When this fraction was mixed with other

fractions in the reassociation experiments, no additional recovery

was noted. Further characterization of this fraction indicated

that protein was not present. At the present time, the nature of

this fraction is unknown.

Trypsin Digestion and Peptide Mapping

 

Although the length of time for trypsin digestion of the
heat denatured pyrophosphorylase apparently varied from preparation
to preparation, the peptide maps were very similar. Best results
were obtained with the procedure adopted, but no technique completely
eliminated all "core" material. The likelihood of chymotryptic
contamination was essentially ruled out since the results were
identical whether trypsin (Worthington) was an L-(l-tosylamido-

2-pheny1) ethyl chloromethyl ketone treated or untreated sample.

80

A fingerprint of 2.5 mg of the tryptic digest is shown in
figure 20. A summary of the peptide staining pattern of several
maps is given in table V. The total number of ninhydrin—positive
peptides varied from 52-62 and arginine-containing peptides
varied from lA—19, while only one tryptOphan—containing peptide

was seen.

Cyanogen Bromide Cleavage

 

r .(.-fl‘“‘ VD

The polyacrylamide pattern of a cyanogen bromide digest of
the RSCM—pyrophosphorylase is shown in figure 21. Ten to eleven
fragments could be detected from various preparations.

Gel electrophoresis of the subunits showed one diffuse band.
However, upon total S-carboxymethylation of the protein, a very
lightly staining second band was detected on polyacrylamide gel
columns. The mobility and amount of this fraction has varied;
but in all cases, it represented only a small percent of the
principal component, and the cyanogen bromide fragmentation
pattern appeared to be very similar. The origin of the minor
band is as yet unknown, but may be due to the carboxymethylation

procedure.

Reassociation of the Subunits

 

When the dissociated enzyme was diluted directly into the
assay cuvette, the activity increased with time from zero. This
fact was indicative of a time dependent formation of active from
inactive enzyme. Due to the low recovery of active enzyme by this

procedure, an alternate reassociation method was next used. When

 

81

Figure 20. Fingerprint of a trypsin digest of
the enzyme
The sample (2.5 mg) was applied to the paper
in dilute ammonia water and subjected to chromato-
graphy in the first direction and electrophoresis
in the second as described in the methods section.
Lightly staining ninhydrin-positive spots are

indicated by dashed lines.

82

 

 

ELECTROPHORESIS. 210W, 2|Iours, pfl 3.5

 

‘ CHROMATOGRAPH Y——-—

83

TABLE V

A Summary of the Tryptic Digests

UDP—glucose Pyrophosphorylase

of

 

 

Number of Peptides

 

 

Amino Acids Residues/molea b
Predicted Found
Lysine plus Argininec Sll 6h 52-62
. . d
Arginine 178 22 lh—l9
e f
Tryptophan 20-26 2-5 1

 

aThe molecular weight taken as h80,000.
bBased on 8—identical subunits.
cNinhydrin—positive peptides.
dSakaguchi-positive peptides (67).
eEhrlich-positive peptides (66).

f . . . .
See amino ac1d analy51s section.

 

_ll'l

 

8h

Figure 21. Polyacrylamide patterns of a reduced
and S-carboxymethylated sample of the
enzyme before and after cleavage with
cyanogen bromide

The left pattern represents 250 ug of the
sample cleaved with cyanogen bromide and the right
pattern represents 200 ug of an uncleaved sample.

The cleaved sample was developed for 1 hour on a

7.5% gel column while the uncleaved sample was

developed for 1 1/2 hours on a 5% gel column. Both

samples were developed at a constant amperage of

5 ma/tube and after staining for 1 hour with

napthol blue black (1% in 7% HoAc), they were

destained electrophoretically.

85

 

.fiaw
$3? . .
a

.. a... In? . .
nu. . .

1.1. ...,.

 

86

compared with the undissociated control, 55-75% of the perphos-
phorylase activity was recovered after three hours of incubation
in this reassociation mixture at 150C and pH 6.h. As expected,
addition of 0.0h M NAD+ had no effect on the recovery process.
Furthermore, 0.01 M UDP-glucose had no effect. However, when the
reassociation was performed at pH 5.5, there was less recovery of

active enzyme.

 

DISCUSSION

UDP-glucose pyrophosphorylase from calf liver is a relatively

large enzyme. Unequivocal molecular weight determinations are

I I “q

complicated by this fact and also by the presence of small
quantities of more rapidly sedimenting components. Preliminary

analysis of these more rapidly sedimenting components in the

 

ultracentrifuge indicated that the pyrophosphorylase might be a i
polymerizing system in which the rate of association is slow.

This hypothesis has received substantiation by Levine £3 £1. (81).

They observed that the same distribution of material that was

seen in the ultracentrifuge could also be seen on polyacrylamide

gel columns and that each component contained pyrophosphorylase

activity. Furthermore, the components which were separable by

sucrose density gradient ultracentrifugation could be re—equilibrated

to yield all the components of the original system.

Hydrodynamic studies of the principal component in the
analytical ultracentrifuge yielded a molecular weight of about
h80,000 as determined by sedimentation and diffusion techniques.

The results were identical whether or not mercaptoethanol was
included in the solvent system. This fact indicated that the
mechanism of association is presumably not by disulfide formation.

The use of the low speed sedimentation equilibrium technique
for the determination of the molecular weight of the pyrophosphory-

lase presented serious problems in obtaining unequivocal results.

87

88

One limitation of the technique was the low speed (3,000 rpm) at
which the ultracentrifuge had to be used. In addition to this
limitation, the low speed technique is abnormally sensitive to
small amounts of high molecular weight material; and since
molecular weight averages are determined, serious reservations
must be placed in the analysis under this conditions.

Assuming that the higher molecular weight material is a

dimer of the h80,000 molecular weight enzyme and that it represents

‘-" A9353—‘m

 

about h% by weight of the total material present, it may be

.—

calculated that the weight-average and the z-average molecular

 

-‘ D

weights should be h95,600 and 507,900, respectively. The
calculated z—average molecular weight is substantially lower than
was observed and presumably indicated an exaggeration of the
molecular weight due to a species outside the limitation of the
low speed technique.

It must therefore be concluded that, due to the "purification"
of material during the sedimentation experiments, the sedimentation
and diffusion techniques are more representative of the true
molecular weight of the perphosphorylase molecule. Similarly,
the average molecular weights as determine from sedimentation
equilibrium must be taken as a maximum, but presumably indicated
a molecular weight range for the enzyme.

Subunits of the perphosphorylase were produced in 5 M urea
or 6 M guanidine hydrochloride. The highly unfolded subunits
which were produced in 6 M guanidine hydrochloride appeared to

be homogeneous in size and to have a molecular weight of about

89

68,000. Confidence may be placed in the subunit analysis due to
the good agreement between the sedimentation-diffusion and
sedimentation equilibrium methods. Dissociation was obtained in
the absence of reducing agents, again indicating no interchain
disulfide bonds. Using h80,000 for the native molecular weight
and 68,000 for the subunit molecular weight, seven subunits per
mole were calculated.

Electron micrographs have revealed that the pyrophosphorylase
is tetrameric (81). The subparticle molecular weight was calculated
to be 55,000, a figure which, when combined with the native mole—
cular weight would indicate an octomer model.

In addition to circumstances which have precluded the calcula-
tion of an unequivocal molecular weight of the native enzyme, there
are also factors which contribute to the uncertainty in estimating
absOlutely the molecular weight of the subunits. Corrections for
the partial specific volume of proteins (6) in high concentrations
of guanidine hydrochloride are still debatable (82-87). Some
sources have indicated that v should be lowered by 0.01 - 0.02
ml/gm under these conditions. Furthermore, others have indicated
that 5 should also be lowered somewhat for experiments conducted,
as these were, at low temperatures (an, 88-90). Both of these
corrections if applied would tend to make the bouyancy correction
factor (l-vp) higher and the molecular weight correspondingly
lower. If the value of 0.72 ml/gm were used for the partial
specific volume in the subunit analysis at low temperatures, a
molecular weight of about 60,000 would be obtained and this would

be in excellent agreement with an octomer model. In view of the

“'1' ” 6“

90

present hydrodynamic data and the subparticle structure that has
been revealed from the electron micrographs, the conclusion is
supported that the pyrophosphorylase must be octomeric in nature
and that the subunits are similar, if not identical, in mass.
Optical rotatory dispersion measurements were obtained in
the 190-500 ml region for the native enzyme and in the 220—500 mu
region for the denatured enzyme. The more levorotatory behavior F
of the denatured state was expected on the basis of the helix to
coil transitions of other systems (7h). The dispersion constant

of 2A6 mu for the native enzyme is in the 230-280 mu range which .

 

has been found for most globular proteins studied to date. In

Why. .

addition to this, the dispersion constant for the denatured
protein of 216 mu is in the 210—250 mu range which has been found
for many denatured proteins and is only slightly higher than
212 mu, a value that has been accepted for randomly coiled synthetic
polypeptides.

The optical rotatory dispersion data for the pyrophosphorylase
closely resemble those of a group of proteins studied by
Jirgensons (91), and the comparison is shown in table VI. All the
proteins appeared to have low negative b0 values (A0 of 220 mu)
and similar cotton effect characteristics. Beta structure was
suspected in some cases.

Table VII summarizes some of the characteristics of various
secondary structures (92). The possibility of B-structure is
further supported by the blue shift of the cotton effect trough

and the red shift of the cotton effect maximum. However, the

91

 

 

 

.mHOEHowU\mEoummU we mpflcso

.Aamv mcomcwmpfid Eopm :mxmu mosam>n

.10 0mm 00 0K0
mom 00m.HH 00:.0 oomsn we mmmasnogamozaopsm 00000H0-mm:
mom 000.ma 00w.m 00m.m 0m ncflmbopaoosao 0H0<as
How 00m.ma 00m.n 00a.m mm 00H00>0u05>:o-0
mom 000.mm 000.0 000.0 am 00000on 00000Ho
mom 000m.sa 0000.: 0000.: use nwmmmaozconausxomm

AiEv ESEHXmE
um . fiGH mmmhogu mnmfiogn won: cflwboum
EDEHsz

 

 

mafimbopm MO @5090 m paw mmmaxpocamozdop>m

wmoosamumma mo mpoqumpmm coampmmmflm >90pmuom Hmofipmo meow

H> mammH

92

TABLE VII

Cotton Effects of L-Polypeptides and
UDP-glucose Pyrophosphorylase

 

 

 

 

 

 

 

 

Confirmation
or trough, mu crossover, mu peak, mu T
Protein
’ b
232-253 ~22h
0 Helixa 198-199 é
182-18h vv190 i
D
229-230 ~22O
B-forma 205
'vl90 rvl96
258 (small) 228 (small)
Coila 198
eon-205 189
Poly-L-prolinea 210 205 19h
UDP—glucose
Perphosphorylase 252 222 202

 

aValues taken from Yang (92).

bWith a shoulder near 215 mu.

95

212 mu shoulder presumably indicates the presence of some helical
regions. Thus, the present hypothesis is that the pyrophosphorylase
may contain a-helical, random coil, and B-regions. Substantiating
evidence from infrared spectroscopy would be informative.

In view of the possible presence of structures other than
the a-helix and random coil, semiquantitative estimates of helicity
must be viewed with some reservation. However, all facts indicated 05
that the enzyme is a protein of relatively low helical content. ‘-1
Discrepancies in the calculated helical content were noted using 8

different methods of calculation and this fact might be a i

 

further indication of the presence of structures other than the a ‘
d—helix and random coil.

The denatured perphosphorylase exhibited the typical optical
rotatory dispersion spectrum of randomly coiled polypeptides and
proteins. Furthermore, a ho of 0 (A0 of 212 mu) is interpreted
to indicated the lack of helical regions.

Titration of the sulfhydryl groups of the pyrophosphorylase

with DTNB and PMB has resulted in some interesting findings. The

 

rapidity of the reaction of the cysteine residues of the native

protein at a given pH with DTNB indicated that some of these

groups were quite available to solvent, and presumably they are

on the surface of the protein or in a cleft. The gradual increase

in the number of titratable SH groups with this reagent suggested !
a pH-dependent conformational change that resulted in an opening

of the protein structure at high pH to make more SH groups

available for titration. The magnitude of this conformational

change must be small because little change was noted in the

9h

sedimentation coefficient in this pH range. The observed difference

in the absolute number of groups titrated with different prepar-
ations might indicate that some were more highly oxidized than
others. There appeared to be a decrease in the number of
titratable SH groups with increasing age of the preparation.
However, the maximum number of 16 moles of SH per mole of protein
was found for some preparations that were a year old. The fact
that 70% of the control activity was maintained after 16 cysteine
groups were titrated with DTNB indicates that the sulfhydryl
groups are presumably not essential for activity.

The value for the cysteine content which was determined on

a limited number of preparations was lower when PMB was used as

the titrant than when DTNB was used. Two residues per subunit were

found with DTNB while 1.0 to 1.5 residues per subunit was found
with PMB. The reason for the lower value with PMB is as yet
uncertain, but may reflect the possible oxidation problem that
has been mentioned for the DTNB titrations or perhaps neighboring
group influence on the reactability of the various sulfhydryl
groups. It is clear that more preparations need to be titrated
with PMB.

Nothing abnormal was noted in the amino acid composition of

the pyrophosphorylase. As expected from other proteins, glutamic

and aspartic acids were present in the largest amount and cysteine

and tryptOphan, followed by methionine, were present in the lowest

amounts. There was no detectable carbohydrate material bound to
the purified protein. The calculated partial specific volume
of 0.739 ml/gm is also to be expected for an average globular

protein.

If;
I

 

a
.i’

 

95

Since the analytical ultracentrifuge data indicated that
the subunits of the enzyme were similar in mass, if not identical,
experiments were performed to assess the further likelihood that
they are chemically identical. Electrophoresis on cellulose
acetate strips indicated that the subunits were similar, if not
identical, in charge. Furthermore, polyacrylamide gel electro-
phoresis, a method which is dependent on both charge and mass, s
indicated that the subunits were similar in these respects. The
unequivocal interpretation of a single species suffers from the

fact that polyacrylamide patterns of the intact protein which

V533;

was in the native state, denatured, or derivatized always yielded
material at the origin and in the sample gel.

Fingerprinting experiments were also performed to assess the
likelihood of subunit homology. From the known specificity of
trypsin for lysine and arginine residues and from the number of
these residues present, one would expect 512 peptides if there
were no repeating sequences in the primary structure. Fifty-two
to sixty-two ninhydrin Spots were observed. Thus, one-eighth to
one-tenth the number of ninhydrin spots that would have been
expected if there were no repeating sequence was observed. This
result is consistent with eight to ten homologous subunits and is
in good agreement with the prediction of eight subunits by other
methods. The number of observable arginine-containing spots
varied somewhat from preparation to preparation, but the higher
values are consistent with nine to ten homologous subunits. A
lower than expected number of tryptOphan-containing peptides was

observed. Several explanations may account for this. First, the

96

two methods used for tryptophan analysis have been shown to be
unreliable in some instances (95). Other explanations might be
that the two or three residues of tryptophan were located on
the same peptide or free tryptophan, or a very small peptide
containing tryptophan, occurred because of an unusual sequence
and was not detected. Still a further possibility might be that
not all the tryptophan-containing peptides gave a positive test
with Ehrlich's reagent. Due to the fact that a lower than
expected number of tryptophan spots was observed and the number
of arginine-containing spots varied, the conclusion of eight

to ten homologous subunits muSt be viewed cautiOusly. However,
the results do not negate this possibility.

Cyanogen bromide cleavage yielded lO-ll bands which were
observed on polyacrylamide gel columns. From the number of
fragments, the specificity of cyanogen bromide for methionine
peptides, and the number of these residues per mole, eight
extremely homologous, if not identical, subunits were again
postulated. For this conclusion, complete cleavage of the
methionine bonds and complete reduction and S—carboxymethylation
of all disulfide linkages was assumed. This fact has been
demonstrated by Steers 3; £1. for B—galactosidase (69) using the
same cleavage system. When monitored on polyacrylamide gel
columns, the presence of a second faint band for the RSCM-
perphosphorylase was found. In light of the single diffuse
band found for the underivatized subunits, this is not yet

understood.

—-‘- ~fl0'r40nur.

 

y

 

97

Electrofocusing experiments, on the other hand, indicated the
enzyme was heterogeneous in 6 M urea. At least eight fractions
could be detected. These results are compatible with: (1) the
presence of differently charged subunits which were present in
varying amounts, (2) nonequivalent binding of small ions to a
single identical subunit, or (5) modification of the protein
due possibly to proteolytic digestion during fractionation. It
appeared from the electrofocusing experiments that at least
four fractions have the competence to form active pyrophosphorylase
molecules with approximately equal specific activities. These
preliminary results do not in themselves rule out the possibility
of chemically identical subunits.

Reassociation experiments demonstrated that active enzyme
could be reconstituted from a sample denatured in 8 M urea.

There was a strong pH dependence on the reassociation process,

and presumably the higher recovery at more alkaline pH can be
correlated with the alkaline pH stability range of the protein (28).
The high recovery in these preliminary experiments was encouraging
because of the prospects for further optimizing the yield of

active enzyme. Furthermore, the high recovery of the perphos—
phorylase activity demonstrated an important biochemical principle,
which is that the information necessary for the proper tertiary
structure of some proteins resides in the linear sequence of

their component amino acids.

 

 

SUMMARY

Uridine diphosphate glucose pyrophosphorylase which was
recrystallized as many as six times exhibited molecular poly- F
dispersity. Analysis in the ultracentrifuge pointed to the ;.3
presence of polymers. The molecular weight of the principal i

component was estimated from sedimentation and diffusion

 

analysis and found to be about h80,000. This molecular weight a )

was roughly substantiated by low speed sedimentation equilibrium

analysis. I
In the solvent system containing 6 M guanidine hydrochloride
and 0.1 M mercaptoethanol, pyrophosphorylase subunits were
produced that had a molecular weight of about 68,000 with a value
of 0.7h ml/gm for 6 (60,000 with a G of 0.72 ml/gm).
The pyrophosphorylase exhibited a typical aromatic amino
acid spectrum, and the extinction coefficient was determined with
concomitant dry-weighing experiments. Optical rotatory dispersion 1
studies indicated a low helical content with the possibility of I
some B-structure regions.
Chemical studies demonstrated a typical protein amino acid
content withou:bound carbohydrate material. The total number
of cysteine groups which would readily react with Ellman's
reagent was pH dependent. Although the number of SH groups per

mole appeared to depend on the preparation chosen and possibly

its age, a maximum of about 2 per subunit was obtained. Other

99

methods have indicated the possibility of a larger number for the
total half—cystine analysis.

Polyacrylamide and cellulose acetate electrOphoresis, and
cyanogen bromide digestions were consistent with subunit homology,
if not identity. Certain aspects of the fingerprinting experiments
also supported this hypothesis, but due to the variance in the
arginine-containing spots and the observation of a lower than I”?
expected number of tryptophan-containing spots, the results
are less conclusive. Preliminary electrofocusing experiments

indicated the presence of different electrOphoretic species

 

under denaturing conditions. a
The subunits produced in urea appeared to be catalytically
inactive, but active pyrophosphorylase was able to be reconstituted

with a high overall yield.

 

 

10.

ll.

l2.

15.

1h.

15.

l6.

l7.

18.

REFERENCES

Newfeld, E. F., and Hassid, W. Z., Advan. Carbohydrate Chem.,
1Q, 509 (1965).

Cabib, E., Ann. Rev. Biochem., 52, 55h (1965).

 

 

Ginsburg, V., Advan. Enzymol., 26, 55 (196A).

 

Neufeld, E. F., and Ginsburg, V., Ann. Rev. Biochem., 55,
601 (1965).

 

Caputto, R., Leloir, L. F., Trucco, R. E., Cardini, C. E.,
and Paladini, A. C., J. Biol. Chem., 179, h97 (19h9).

 

Caputto, R., Leloir, L. F., Cardini, C.-E., and Paladini, A. C.,
J. Biol. Chem., 18A, 555 (1950).

 

Kalckar, H. M., and Cutolo, E., IIe Congr. Intern. de Biochemie,
Paris, 260 (1952).

 

Munch-Petersen, A., Kalckar, H. M., Cutolo E., and Smith,
E. E. B., Nature, 172, 1056 (1955).

 

Kornberg, A., J. Biol. Chem., 176, 1u75 (19u8).

 

Kornberg, A., J. Biol. Chem., 182, 779 (1950).

 

Schrecker, A. W., and Kornberg, A., J. Biol. Chem., 182, 795
1950 .

 

Munch-Peterson, A., Acta Chem. Scand., 2, 1525 (1955).

 

Kalckar, H. M., Braganca, B., and Munch-Peterson, A., Nature,
172, 1058 (1953). ""

Smith, E. E. B., and Mills, G. T., Biochim. Biophys. Acta, 15,
586 (195A).

Smith, E. E. B., and Mills, G. T., Biochim. Biophys. Acta, 18,
152 (1955).

 

 

 

Burma, D. P. and Mortimer, D. C., Arch. Biochem. and Biophys.,
62, 16 (1956).

Neufeld, E. F., Ginsburg, V., Putnam, E. W., Fanshier D., and
Hassid, W. Z., Arch. Biochem. Biophys., 62, 602 (1957).

 

Ginsburg, V., J. Biol. Chem., 252, 55 (1958).

 

 

19.
20.

21.

22.

25.

2h.

25.

26.

27.
28.

29.

50.

51.
52.
35.
5h.
55-

36.

37.

58.
59.

101

Ganguli, N. c., J. Biol. Chem., 252, 557 (1958).

Turner, D. H., and Turner, J. F., Biochem. J., 62, hh8 (1958).

 

Villar-Palasi, C., and Larner, J., Arch. Biochem. Biophys.,
86, 61 (1960).

 

Basu, D. K., and Bachhawat, B. K., J. Neurochem., 1, 17h
(1961).

 

Anderson, E. P., Kalckar, H. M., Kurahashi, K., and Isselbacher,
K. J., J. Lab. Clin. Med., 50, h69 (1957). -

 

Seits, I. F., Luganova, I. S., and Vladimivova, A. D.,
Biokhimiya, 28, 295 (1965).

 

Blinov, M. N., Vopr. Med. Khim., 19, 165 (196M).

Hal rin, K. M., and Ohkawara, A., J. Invest. Dermatol., £6,
A5 1966).

Kamogawa, A., and Kurahashi, K., J. Biochem., 51, 758 (1965).

 

Albrecht, G. J., Bass, S. T., Seifert, L. L., and Hansen, R. G.,
J. Biol. Chem., 2&1, 2968 (1966)

 

DeFazio, K., Master of Science thesis, Michigan State University,
1968.

Knop, J., Master of Science thesis, Michigan State University,
1969.

Bass, S. T., and Hansen, R. G., in preparation.
Gillett, T. A., and Hansen, R. G., in preparation.
Isselbacher, K. J., J. Biol. Chem., 252, h29 (1958).

Isselbacher, K. J., Science, 127, 28 (1958).

 

Gitzelman, R., personal communication to R. G. Hansen, Sept.
1968.

Abraham, H. D., and Howell, R. R., J. Biol. Chem., 2hh, 5h5
(1969).

Ting, W. K., and Hansen, R. G., Proc. Soc. Exptl. Biol. Med.,
127, 960 (1968).

 

Gillett, T. A., Levine, S. L., and Hansen, R. G., in preparation.

Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.,
J. Biol. Chem., 195, 265 (1951).

 

 

 

MO.

hl.

#2.
hi.

uh.
MS.
M6.
AT.
AB.

M9.

50.

51.
52.

55.
5h.

55-

56.

57.

58.
59.
60.

102

Warburg, 0., and Christian, W., Biochem. Z., 510, 58h (l9hl).

 

Dubois, K. A., Gilles, K. A., Hamilton, J. K., Rebers, P. A.,
and Smith, F., Anal. Chem., 26, 550 (1956).

Schachman, H. K., Methods Enzymol., E, 2 (1957).

 

Van Holde K. E., and Baldwin, R. L., J. Phys. Chem., 6g,
75h (1958).

 

Svedberg, T., and Pedersen, K. 0., The Ultracentrifuge,
London, Oxford University, Clarendon Press, l9h0.

 

Kawahara, K., and Tanford, C., J. Biol. Chem., 2hl, 5288 gig
(1966). g

 

Steinmetz, M. A., and Deal, W. C., Jr., Biochemistry, 6,
1599 (1966).

 

 

Dorsey, N. E., Prgerties of Ordinary Water-Substance, y
Reinhold, New York, 280 (19h0).

Hooker, T. M., Jr., Doctor of Philosophy thesis, Duke
University, 1966.

Tanford, C., Kawahara, K., Lapanje, S., Hooker, T. M., Jr.,
Zarlengo, M. H., Salahuddin, A., Aune, K., and Takagi, T.,
J. Am. Chem. 300., 89, 5025 (1967).

 

Moffitt, W., and Yang, J. T., Proc. Natl. Acad. Sci. U. S.,
.53. 596 (1956).

Ellman, G. L., Arch. Biochem. Biophys., 6g, 70 (1959).

Kemp, R. G., and Forest, P. B., Biochemistry, 1, 2596 (1968).

 

Boyer, P. D., J. Am. Chem. 809,, 16, A551 (l95h).

Benesch, R., and Benesch, R. A., Methods Biochem. Anal., 16,
60 (1962).

 

Moore, 8., and Stein, W. H., in Methods in Enzymology,
Colowick, S. P., and Kaplan, N. 0., editors, Academic Press
Inc., New York, 819 (1965).

Piez, K. A., and Morris, L., Anal. Biochem., 1, 187 (1960).

 

Spackman, D. H., Stein, W. H., and Moore, 8., Anal. Chem.,
29. 1190 (1958).

 

Spies, J. R., and Chambers, D. C., Anal. Chem., 66, 50 (l9h8).

 

Spies, J. R., and Chambers, D. C., Anal. Chem., 16, l2h9 (l9h9).

 

Goodwin, T. W., and Morton, R. A., Biochem. J., MO. 628 (19h6).

 

61.

62.
65.

6h.

653

66.
67.

68.
69.

70.
71.
72.
75.

7h.

75-

76.

77.
78.

79-
80.

81.

105

Crestfield, A. M., Moore, 8., and Stein, W. H., J. Biol.
Chem., 258, 622 (1965).

 

Svensson, H., Arch. Biochem. Biophys., Suppl. 1, 152 (1962).

 

Vesterberg, 0., and Svensson, H., Acta Chem. Scand., 26, 820

 

(1966).

Ingram, V. M., Biochim. Biophys. Acta, 28, 559 (1958).

 

Katz, A. M., Dreyer, W. J., and Anfinsen, C. B., J. Biol.
Chem., 25A, 2897 (1959).

 

Smith, I. Nature, 171, A5 (1955).

 

Bennett, J. C., in Methods of Enzymology, Colowick, S. P., and
Kaplan, N. O., editors, Academic Press Inc., New York,550 (1968).

 

Gross, E” and Witkop, B., J. Biol. Chem., 257, 1856 (1962).

 

Steers, E., Jr., Craven, G. R., Anfinsen, C. B., and Bethune,

J. L., J. Biol. Chem., 2h0, 2h78 (1965).

 

Davis, B. J., Ann. N. Y., Acad. Sci., 121, hon (196A).

 

Deal, W. C., Jr., Biochemistry, in press.

 

Yang, J. T., and Doty, P., J. Am. Chem. Soc., 19, 761 (1957).

 

Drude, P., Lehrbuch der Optik, 2nd ed., S. Hirzel Verlag,
Leipzig, 1906T7

 

Urnesj P., and Doty, P., Advances in Protein Chem., 16, h01
(1961 .

 

Blout, E. R., in Optical Rotatory Dispersion, Djerassi, C.,
editor, McGraw-Hill, New York, 258 (1960).

 

Sogami, M., Leonard, W. L., Jr., and Foster, J. F., Arch.
Biochem. Biophys., 100, 206 (1965).

 

Yang, J. T., and McCabe, W. J., Biopolymers, 2, 209 (1965).

 

Blout, F. R., Schmier, I., and Simmons, N. S., J. Am. Chem.

 

Soc., §1: 5195 (1962).

McMeekin, T. L., and Marshall, K., Science, 116, 1&2 (1952).

 

Cohn, E. J., and Edsall, J. T., eds., Proteins, Amino Acids

 

and Peptides, Reinhold, New York, 570 (l9h5).

 

Levine, S., Gillett, T. A., Hageman, E., and Hansen, R. G.,
J. Biol. Chem., in press.

 

Jn‘i‘. ... ...-4

 

F“-

b

 

 

 

82.

85.
8h.

85.

86.
87.

88.
89.

90.

91.

92.

95.

10h

Kielley, W. W., and Harington, W. F., Biochim. Biophys. Acta.,
3;, u01 (1960).

 

Noelken, M. E., Federation Proc., 86, 777 (1967).

 

Marler, E., Nelson, C. A., and Tanford, C., Biochemistry, 8,
279 (196A).

 

Hade,)E. P. K., and Tanford, C., J. Am. Chem. Soc., 89, 505A
1967 .

 

Castellino, F. J., and Barker, R., Biochemistry, Z, 2207 (1968).

 

Edelstein, S. J., and Schachman, H. K., Federation Proc., 82,
12u8 (1966).

Hunter, M. J., J. Phys. Chem., 19, 5285 (1966).

 

Taylor, J. F., and Lowry, C., Biochim. Biophys. Acta., 88, 109
1956 .

 

Reithel, F. J., and Sakora, J. D., J. Phys. Chem., 61, 2h97
1965 .

JirgenSons, B., J. Biol. Chem., 210, 106A (1965).

 

 

Yang, J. T., in Poly-d-Amino Acids, Vol. I, Fasman, G. D.,
and Timasheff, S. N., editors, Marcel Dekker, Inc., New York,
Chapter 6 (1967).

 

Light, A., and Smith,E. L., in The Proteins, Vol. 1, Ed. 2,
Neurath, H., editor, Academic Press, New York, 1 (1965).

 

 

-.. L.”
T‘L-‘M—‘ru