, . Janna...

uh a:1:.2.xll!u toll
3. ;l. 135.. 9.2.3.0

 

 

(1;):

flu..:h1::fi3§1n {aqua-navy! {2.13. 3... 5135.1. 5...: .t

1.11.1.3}.xlfificiklme—MWJMA:~JMKPIW vmWW Wufo’QnHIIfiHIwfiulr9lele

 

IHESIS

This is to certify that the

thesis entitled

. "Sedimentation Behavior of the Glucose
3 Dehydrogenase from Spores of Bacillus
Cereus"

presented by

John W. Kools

.i iigfi..._iu__

has been accepted towards fulfillment
of the requirements for

Ph.D. &qmein Microbiology and

Public Health

We!

Major professor /
Due bruar 7 1964

0-169

LIBRARY

Michigan State
University

 

 

ABSTRACT

SEDIMENTATION BEHAVIOR OF THE GLUCOSE DEHYDROGENASE
FROM SPORES OF BACILLUS CEREUS

 

by John W. Kools

Glucose dehydrogenase from germinated spores of

Bacillus cereus was used as a model for the study of the

 

heat resistance of bacterial endospores. It had previously .

been found that the heat resistance of this enzyme, in vitro,

was a function of the NaCl and hydrogen ion concentrations

of its environment. In solutions of 5 M NaCl. the enzyme

was as heat resistant as it was in the intact spores and

appeared to have dissociated into two or more molecules.

A study was therefore made of the relationship between

the state of aggregation of the protein and its heat resistance.
Gel filtration was used in an attempt to separate

the aggregated from the disaggregated enzyme. This was not

a successful approach although an excellent separation of

glucose dehydrogenase from contaminating proteins was obtained.

_ fractionation, heat denaturation

4

of contaminating proteins, and gel filtration with G—lOO

A combination of (NH4)ZSO

Sephadex through a 900 ml column resulted in a purification

of 90—100% and a yield of enzyme that approached 100%.

John W. Kools

The G—lOO Sephadex was suspended in and eluted with a buffer
system composed of 0.05 M imidazole chloride buffer pH 6.5
and containing 0.5 M NaCl. This concentration of NaCl gave
sufficient heat resistance to the enzyme so that the column
could be run at room temperature.

The two sedimentation techniques used in this study
employed linear sucrose gradients for the stabilization of
the boundary of the sedimenting protein. These gradients
were 0—3 per cent and 5-20 per cent respectively. The
location of the boundary after the centrifugation runs was
determined by sampling the tube and assaying the enzyme
activity of each sample.

No change in the sedimentation properties of the
glucose dehydrogenase was observed over a pH range of 6.5—8.0
in 0.05 M imidazole chloride buffer pH 6.5 at 20 C. The
sedimentation constant was 8.4 S.

The effect of the NaCl concentration was determined
in 0.05 M imidazole chloride buffer pH 6.5. Without any
NaCl present, the molecular weight of the enzyme was 37,000,
but it was partially dissociated into 22,000 molecular weight
subunits in 3 M NaCl at 20 C. From 3—5 M NaCl the sedimen—
tation constant decreased with a corresponding decrease in
the diffusion constant, indicating a shape change with no
change in the molecular weight.

The sedimentation studies were conducted at 20 C

while thermal inactivation studies had been conducted at

John W. Kools

elevated temperatures. Although the temperature of the
preparative centrifuge could be regulated only over a limited
range, sedimentation studies in 3 M NaCl at temperatures of
5, 20, and 27 C indicated an undissociated, partially dis~
sociated, and completely dissociated enzyme respectively.

On the other hand, sedimentation studies in 0.05 M imidazole
chloride buffer, indicated only an undissociated enzyme at

5 and 20 C. It would therefore appear that heating can cause
disaggregation of the enzyme and that this phenomenon occurs
readily as the concentration of NaCl increases. As a
consequence of these observations it would appear that the
property of heat resistance resides in the subunit of the

enzyme.

SEDIMENTATION BEHAVIOR OF THE GLUCOSE DEHYDROGENASE

FROM SPORES OF BACILLUS CEREUS

 

BY

John W. Kools

A THESIS

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

DOCTOR OF PHILOSOPHY
Department of Microbiology and Public Health

1964

ACKNOWLEDGMENTS

I appreciate the guidance and interest shown to me
by Dr. Harold L. Sadoff throughout this research. I am
also grateful to Dr. Ralph N. Costilow for his generous

advice and to Mrs. Gary Duke for her technical assistance.

ii

TABLE OF CONTENTS

Page

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . l
HISTORICAL REVIEW . . . . . . . . . . . . . . . . . . . 3
EXPERIMENTAL METHODS . . . . . . . . . . . . . . . . . 12
Organism and Medium 12
Production of Spores 13
Germination of Spores l4

Glucose Dehydrogenase Assay 15

Rupture of Germinated Spores 15

Partial Purification of Glucose Dehydrogenase l6
Sedimentation Studies 17

RESULTS . . . . . . . . . . . . . . . . . . . . . . . . 25
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . 45
REFERENCES . . . . . . . . . . . . . . . . . . . . . . 47

iii

LIST OF TABLES

Table Page

1. Summary of the purification results for
glucose dehydrogenase from germinated

spores of g. cereus . 29

Appendix Tables

1. A summary of the data used for
calculating the sedimentation constants

of glucose dehydrogenase in 0.05 M imidazole
buffer and various hydrogen ion concen-

trations . 54

2. A summary of the data used for calculating
the sedimentation constants of glucose
dehydrogenase in 0.05 M imidazole
buffer pH 6.5 and various NaCl concen-

trations . . . . . . 55

3. Data used for the Martin and Ames calculation 59

iv

Figure

1. Apparatus for the preparation of linear
sucrose gradients

2. Centrifuge tube sampling device, similar to
that described by Martin and Ames (1961)

3. The purification of glucose dehydrogenase by
gel filtration . . . . . . . . . .

4. The effect of pH on the half-life of glucose
dehydrogenase at 55 C . . . . . . . . . .

5. The effect of the NaCl concentration on the
half-life of glucose dehydrogenase at
85 C . . . . . . . . . . . . . . . .

6. The results of a typical sedimentation
experiment by the method of Hogeboom
and Kuff (1954) . . . . . . . .

7. The effect of pH on the sedimentation
properties of glucose dehydrogenase

8. The effect of the NaCl concentration on
the sedimentation properties of glucose
dehydrogenase . . . . . . . . . . .

9. The effect of 3 M NaCl on the sedimentation

LIST OF FIGURES

properties of glucose dehydrogenase

Page

l9

19

27

3O

31

32

34

35

37

LIST OF APPENDIX

Figure . Page

1. A plot of the minimum and maximum

densities as found in the centrifuge

tubes vs. the NaCl concentrations . . . . . 52
2. A plot of the minimum and maximum viscosities

as found in the centrifuge tubes vs. the

NaCl concentrations . . . . . . . . . . . . 53
3. The densities at 20 C of 3 M NaCl solutions

dissolved in 0.05 M imidazole chloride
buffer pH 6.5 vs. various concentrations
of sucrose . . . . . . . . . . . . . . . . 56

4. The viscosities at 20 C of 3 M NaCl solutions
dissolved in 0.05 M imidazole chloride
buffer pH 6.5 vs. various concentrations
of sucrose . . . . . . . . . . . . . . . . 57

SAMPLE OF CALCULATION BY THE METHOD OF MARTIN
AND AMES (1961) . . . . . . . . . . . . . . . . 58

vi

INTRODUCTION

Members of the family Bacillaceae are characterized

 

by their ability to form dormant cells called endospores.
The endospores are extremely heat resistant in comparison

to vegetative cells of the same family and to other micro—
organisms in general. The mechanism by which these spores
are made stable has been studied with the primary approach
dealing with the cytological and chemical differences be—
tween the labile vegetative cells and the resistant spores.
Many theories have been derived from these studies to
explain heat resistance, such as a dipicolinic acid-calcium—
protein complex (Young, 1959) and a gelled core (Black and
Gerhardt, 1962); but there is no direct experimental evidence
to confirm or refute these hypotheses.

In an attempt to by-pass the difficulties that arose
when the mechanism(s) of heat resistance was studied in
intact spores, an enzyme extracted from the spores was used
as a model. The justification for this approach lies in the
fact that all of the enzymes within the spore which are
needed for its germination (and possibly outgrowth) must

necessarily be heat resistant (Kools, 1960).

Recently an enzyme, glucose dehydrogenase, was

isolated from the spores of Bacillus cereus, and its heat

 

resistance was found to be a function of the pH and NaCl
concentration of the suspending medium. The enzyme could
be rendered as stable lg vitro as it was in the intact
spore. There was also dissociation of the enzyme into
smaller molecular weight fragments in solutions of high
NaCl concentration (Bach, 1963).

The purpose of the study reported in this thesis
was an investigation of the relationship between the heat
resistance and the state of aggregation of glucose dehydro-
genase in solutions of various salt and hydrogen ion con—

centrations.

HI S TORICAL REVIEW

The heat resistance of bacterial endospores has
played an important role in the history of microbiology.
Much of the furor over the theory of spontaneous generation
might never have occurred if early investigators had
appreciated the extent of the resistance of spores. For
example, John Tyndall, in trying to disprove the theory of
spontaneous generation, found in some cases that more than
five and one-half hours at 100 C was necessary to sterilize
his media. From these experiments, he concluded that heat
resistant organisms had contaminated his media. Ferdinand
Cohn, in 1876, demonstrated that certain bacteria were able
to produce heat resistant endospores and that these could
be microscopically differentiated from vegetative cells
(Stanier, Doudoroff, and Adelberg, 1963).

A spore has a unique structure. The external wall
of the spore consists of an outer coat which is often seen
to be split into two layers. Inside this coat lies a thick
electron—transparent layer called the cortex. The inner
spore coat seems to develop from the interior portion of
the cortex and becomes the cell wall of the new vegetative

cell when germination occurs. Inside the inner spore wall

is a mass of cytoplasm called the core. The spores of

B. cereus are characteristically covered, in addition, by
a loose-fitting envelope called the exosporium (Hashimoto,
Black and Gerhardt, 1960).

Over a period of 70 years, many studies have been
made in an attempt to elucidate the mechanism of heat re—
sistance in bacterial endospores. Subsequently, a number
of theories have been proposed to explain this unusual
attribute. The majority of these theories were based on
the chemical and structural differences that existed, or
were thought to exist, between the spores and the vegetative
cells. Among these were included:

1. A non-permeable spore coat (Lewith, 1890).

2. A relatively low water content in the spores
(Davenport and Castle, 1895).

3. A high concentration of dipicolinic acid (DPA)

and calcium in the spores (Powell, 1953).

4. A relatively high thiol content in spores, i.e.,

cystine and cysteine (Vinter, 1961).

By virtue of the discovery of the extensive per—
meability of spores to water and solutes, the concept of
heat resistance due to a non-permeable spore coat could be
discounted (Murrell, 1961; Gerhardt and Black, 1961: and
Black and Gerhardt, 1962).

On the other hand, Black and Gerhardt (1962) pointed

out that the spore interior is relatively dry compared to the

cytoplasm of a vegetative cell but not to the extent envisioned
by Davenport and Castle (1895). They were able to determine
the weight of the interstitial water surrounding the spores
thereby allowing them to make a valid dry weight determination.
The dormant spores 0f.§- cereus contained 65% water, germinated
spores 73% water, and the vegetative cells 77% water by
weight. Murrell (1961), using density determinations of
dry spore material and an estimate of water in spore pellets
found that any anhydrous portion of the spore would account
for less than 10% of the total spore volume. This value
was confirmed by Black and Gerhardt (1962).

Another hypothesis concerning the water content
of spores had been set forth which predicted the existence
of a dry, central core (Lewith, 1890). The experiments
of Black and Gerhardt (1962) showed that about 90% of the
volume of this central core was permeable to water. Thus,
it would seem that water is distributed throughout the spore
but that the core has a relatively lower concentration of
water. The authors have proposed that:

the core of the dormant spore exists as an insoluble
and heat stable gel in which cross— linking between

macro— —molecu1es occurs through stable but reversible
bonds so as to form a high— polymer matrix with entrapped

free water.
They further suggested that this cross-linking was due to
the presence of disulfide bonds, and, in support of this
hypothesis, they mentioned the high concentration of

cystine—cysteine groups in spores (Vinter, 1961).

Dipicolonic acid comprises 4-15% of the dry weight
of spores, whereas none is found in vegetative cells. It
was shown that this compound must be present in order
for spores to be heat resistant (Halvorson and Howitt, 1961;
Church and Halvorson, 1959; Black, Hashimoto and Gerhardt,
1960). During spore germination the DPA is released into
the medium in the form of a calcium chelate with the sub-
sequent 1oss of heat resistance of the spores. Calcium was
also implicated in the heat resistance of Spores (Curran,
Brunstetter and Myers, 1943; and Black et al., 1960). Young
(1959) found that peptides from spores form a complex with
the calcium—DPA chelate and has, therefore, suggested
that a DPA—calcium-protein complex could exist. However,
no meaningful experiments leading to an understanding of
the mechanism of the heat resistance due to DPA have
been performed.

Spores are heat resistant and it follows that their
essential proteins are also heat resistant under conditions
prevailing in the spore interior. This forms the basis for
a different approach to the problem, that of studying
individual enzymes from the spore. Enzymes extracted from
spores have been found to be either heat stable or heat
labile, suggesting that some proteins are intrinsically
Stable, while others are stabilized by factors in their
environment. Some spore enzymes are particulate and heat stable

When first extracted but tend to lose this stability if they

become soluble. Powell and Hunter (1956) extracted a stable
adenosine ribosidase from spores of_§. cereus which was
capable of withstanding a temperature of 100 C for 1 hour.
They did not state whether the enzyme was particulate or
soluble. The adenosine deaminase from the same organism
lost its heat resistance upon disintegration of the spore.
Another enzyme from spores of B. cereus, alanine racemase,
has been found to be particulate and relatively stable when
extracted by means of brief sonic oscillation. The enzyme
loses heat resistance and becomes soluble when subjected

to extensive sonic oscillation (Stewart and Halvorson, 1953;
Stewart and Halvorson, 1954).

Bach (1963) has recently described an enzyme, glucose
dehydrogenase, from the spores of B. cereus which was labile
when extracted from the spore but which could be made
stable by the addition of sodium chloride to the enzyme
solution. His experiments indicated that this protein was
stabilized by salts normally present in spores. When the
sodium chloride was added to glucose dehydrogenase, the
enzyme disaggregated into two or more units.

In contrast to the above examples, two intrinsically
heat resistant, non-spore proteins have been studied. The
first of these, the flagella from thermophilic bacteria,
were much more stable than those from mesophilic bacteria
(Adye, Koffler, and Mallett, 1957). The heat resistance

of these proteins was maintained through extensive purification

and no stabilizing material was detected (Koffler, Mallett,
and Adye, 1957; and Adye et a1., 1957).

The alpha-amylase from Bacillus coagulans has been

 

crystallized without any evidence for the existence of a
stabilizing factor. When the organism was grown at 55 G,
the crystalline enzyme retained 88 to 90% of its activity
after being heated for 60 minutes at 90 C. However, when
the same organism was cultured at 35 C, only 6—10% of its
activity remained when it was subjected to the preceding
conditions (Campbell, 1954).

Many protein molecules exist as aggregates of peptides
which are referred to as subunites. Pedersen (1940) has
reviewed the research concerning the dissociation of
proteins in varying hydrogen ion concentrations and con-
cluded that "for a given protein, the molecular weight as
a rule has its maximum value within a given pH range around
its isoelectric point." This dissociation of a protein at
a given pH may also be influenced by the salt concentration
of the surrounding medium. For example, haemoglobin in
moderately concentrated salt solutions (0.5—1.0 M) was
dissociated into molecules having half the normal molecular
weight.

McFarlane (1935) found that the dissociation of a
protein can further be influenced by the presence of
another, different protein. He discerned that globulin

from horse serum dissociated in the presence of albumin.

It has, in fact, been observed that the presence of even an
amino acid can cause disaggregation of a protein molecule.
For example, 0.001% thyroxine produces appreciable dis—
sociation of thyroglobulin (Lundgren, 1936). In contrast
to the effect of the addition of a protein or amino acids
to the solution, the dilution of the protein suspension
resulted in the appearance of several disaggregation products.
The same author also noted that urea favored the dissociation
of thyroglobulin.

Temperature also seems to have some role in protein
dissociation. Pedersen (1950) reported that gamma globulin
disaggregated when the temperature of the substance in the
ultracentrifuge was raised from 10 to 30 C.

Svedberg and Brohult (1938) showed that haemocyanin
at pH 6.5 was dissociated into half molecules by irradiation
with 'ultraviolet light of wave lengths of 280 millimicrons.
Brohult (1937) reported that the haemocyanin molecule
could be partially split by ultrasonic waves with a fre—
quency of 250,000 cycles per second. At pH 6.2, half mole—
cules were formed, and at a pH of 7.2 one—eighth molecules
were formed after sonication.

Alpha—amylases are metalloenzymes whose heat
stability can be greatly reduced by the removal of calcium

(Fischer and Stein, 1960). The alpha—amylase from Bacillus

 

subtilis is a dimer whose two parts are bound together by

 

a single molecule of zinc. The removal of the zinc was

10

readily accomplished by dialysis against the chelating
agent ethylenediaminetetraacetate (EDTA) with no loss of
its heat resistant properties. However, after prolonged
dialysis and under conditions that were not optimal for
the enzyme itself, the calcium was lost as well as the
resistance of the enzyme.

Deal, Rutter, and Van Holde (1963b) have described
the reversible dissociation of aldolase from rabbit muscle.
At pH values 2.9 and below the enzyme dissociated into three
subunits of approximately equal molecular weight. They
suggested that the native protein, a folded trimer, first became
an unfolded trimer, then dissociated into three unfolded
monomeric subunits.

Deal et al. (1963a) have studied enolase, alpha-
glycerophosphate dehydrogenase, fumerase, lactic dehydrogenase,
and glyceraldehyde-3-phosphate dehydrogenase in acidic
solutions. In all cases except the latter, structural
modifications were observed as indicated by changes in
sedimentation properties. The lactic dehydrogenase dis-
sociated into unfolded subunits.

Samejima and Shibata (1961) have reported the
dissociation of beef liver catalase with formamide and urea.
They found that the native catalase was dissociated into
one—third size subunits which could either dimerize into
two—third size subunits or dissociate again into one—sixth

size subunits. Since the catalase contains four iron

ll

molecules, the authors concluded that the iron was distri-
buted unevenly in the native catalase molecule. Further-
more, the catalase molecule dissociated into halves by
acid treatment while dissociating into one-third size
subunits by alkali treatment.

There are several environmental factors which can
contribute to the dissociation of any particular protein
molecule, although the degree to which any one of these
factors is effective will depend largely on the particular

protein involved.

EXPERIMENTAL METHODS

Organism and Medium

 

The organism used throughout this study was a strain
of B. cereus which had been obtained from the University of
Illinois culture collection in 1955. It was maintained at
4 C in the sporulated form on the surface of nutrient agar
slants (Difco Laboratories). Modified G medium (Stewart
and Halvorson 1953; and Hashimoto et a1., 1960) was used in
growing large quantities of spores and consisted of the

following materials per liter of solution:

Glucose 4.0 g
(NH4)ZSO4 4.0 9
Yeast Extract 2.0 g
K2HPO4 1.0 g
MgSO4 0.8 g
CaCl2 0.1 g
MnSO4-H20 0.1 g
ZnSO4 0.01 g
CuSO4°5HZO 0.01 g
FeSO4-7H20 0.001 g

In addition, Dow Corning Antifoam AF emulsion (Dow

Corning Corporation) was used as needed. The initial pH

12

13
of the medium was 6.6 and needed no further adjustment.

Production of Spores

Cultures of B. cereus were grown in a 100 liter
fermentor (Stainless and Steel Products Co.), maintained at
30 C. The contents were stirred at 225 rev/min and
aerated at 100 liters/min. In order to obtain sufficient
inoculum for the 100 liter fermentor, the following proce-
dure was used. The spores from a storage slant were trans-
ferred to a fresh nutrient agar slant which was allowed to
incubate for 8 hours. The growth on the slant was trans—
ferred to 50 ml of medium contained in.a 500 ml dimpled
flask and shaken for 2 hours at 30 C. Successive transfers-
were then made into 4 and then 18 similar flasks using a
20 volume per cent inoculum and incubating with shaking for
2 hours. All of the contents of the 18 flasks were used to
inoculate three New Brunswick 3 liter fermentors (New
Brunswick Scientific Co.)° The contents of the small
fermentors were stirred at 450 rev/min and aerated at 5
liters/min for 2 hours. The resulting 9 liters of B. cereus
culture were used to inoculate the 100 liter fermentor.

The culture had sporulated and the sporangia had
begun to lyse after 22 hours. This lytic process was
hastened by raising the temperature in the fermentor to

45 c, decreasing the rate of stirring to 100 rev/min and

14

discontinuing the air flow. After 1 hour under these condi-
tions, the spores were free of their sporangia and most of
the vegetative cells had lysed. The fermentor tank was
rapidly cooled and the spores were harvested in a Sharples
Super Centrifuge, model A 5-12 (The Sharples Corporation).
The spores were washed once in distilled water and stored

at -15 C until needed.

Germination of Spores

The yield of glucose dehydrogenase could be greatly
increased by breaking germinated spores rather than dormant
spores. The germination procedure used was that described
by Bach (1963).

A typical germination procedure consisted of heating
400 g (wet wt).of spores in 1 liter of distilled water for
1 hour at 65 C, followed by their suspension in a 3% yeast
extract solution (Difco Laboratories). The final germination
medium contained 400 g of heat shocked spores suspended in
8 liters of 3% yeast extract. The spore suspension was
incubated at 30 C for 1 hour. The extent of germination was
ascertained by observing the loss of refractility of spores
by means of phase contrast microscopy. The germinated spores
were washed once with distilled H O and suspended in 1200 ml

2
of 0.1 M phosphate buffer pH 6.8.

15

Glucose Dehydrogenase Assay

 

Glucose dehydrogenase is a nicotinamide adenine
dinucleotide (NAD) specific enzyme and the reduced form of
NAD (NADH) can be assayed spectrophotometrically at a wave-
length of 340 millimicrons. A Beckman spectrophotometer,
model DU (Beckman Instruments, Inc.), equipped with a
Model SR Sargent recorder (E. H. Sargent and Co.) and a
Ledland log converter (Ledland Instruments) was used to
measure the NADH produced in a 1 m1 cuvette containing 800
micromoles of tris buffer pH 8.0 (trishydroxymethylaminomethane),
100 micromoles of glucose, 2 micromoles of NAD and sufficient
enzyme preparation and distilled water to make a final volume
of 1.0 ml.

Since the activity of glucose dehydrogenase is very
temperature dependent, the temperature of the sample chamber
of the spectrophotometer was maintained at 25 C by the use
of thermospacers and a constant temperature bath.

One unit of enzyme was defined as that amount which
caused an optical absorbance change of 0.001 per minute

under the above conditions.

Rupture of Germinated Spores

 

An Eppenbach Colloid Mill (Gifford-Wood Co.) was
used to effect the rupture of germinated spores (O'Conner,
Doi and Halvorson, 1960). The instrument contained 1200 ml

of the spore suspension and 740 g of 110 pavement marking

l6

beads (Minnesota Mining and Manufacturing Co.). The mill
was run at 10,000 rev/min with a gap setting of 0.03 inches
and the temperature was maintained at 10-15 C by circulating
ethanol at -20 C through the cooling coils of the mill.

The extent of breakage was observed by phase contrast

microscopy.

Partial Purification of Glucose Dehydrogenase

 

The cell brei from the colloid mill was clarified
by centrifugation at 14,600 x g for 1 hour in a Type SS—4
Servall enclosed centrifuge (Ivan Sorvall, Inc.). The
supernatant liquid was made 60% saturated with solid ammonium
sulfate and the precipitate that developed was removed from
the solution be centrifugation and discarded. The glucose
dehydrogenase remained in solution. The enzyme was precipi-
tated by increasing the concentration of (NH4)ZSO4 to 100%
saturation in order to effect a concentration of the enzyme.
This precipitate was resuspended in a 1/15 volume of
0.05 M imidazole chloride buffer pH 6.5 containing 4 M NaCl.
Because the enzyme was heat resistant in high concentrations
of NaCl, the preparation was heated to 60—65 C for 30 minutes
to effect a purification by differential thermal denaturation.
The denatured protein was removed by centrifugation and the
enzyme solution was dialyzed overnight against distilled
water to reduce the concentration of NaCl. The glucose dehydro-

genase was concentrated by "dialysis" against dry Sephadex,

17
G-25, medium grade (Pharmacia)..
Sedimentation Studies

Linear Gradients.—-The sedimentation properties of
glucose dehydrogenase were determined by the methods of
Hogeboom and Kuff (1954) and Martin and Ames (1961).

These procedures can be used with relatively impure prepar—
ations and utilize the enzymatic activities to locate the
position of those proteins under investigation. In the
method described by Hogeboom and Kuff (1954) a 0-3% sucrose
gradient was used to stabilize the protein boundary. The
runs were initiated with the protein distributed throughout
the centrifuge tube. Martin and Ames (1961) utilized a
5—20% sucrose gradient with the enzyme layered on the top
of the liquid column at the beginning of the sedimentation
experiment. Both procedures required the establishment of
linear sucrose gradients in the centrifuge tubes.

The preparation of linear gradients was accomplished
by using the device shown in Fig. 1. This apparatus,
described by HogeboOm and Kuff (1954), was constructed from
a lucite block in which were drilled two cylindrical chambers,
A and B, connected by a channel. A drain tube was mounted
in the bottom of chamber B. The two chambers contained
identical solutions except for the concentration of
sucrose, with chamber B always containing the solution of

higher sucrose concentration. The contents of chamber B

18

were stirred and progressively diluted as the sucrose
solution was conducted to the centrifuge tube. The gradients

were formed in 1/2 in. x 2 in. LusterOid centrifuge tubes.

Sedimentation Conditions.——The centrifuge tubes were
placed in a precooled SW—39 swinging bucket rotor which was
run at 38,000 RPM (158,000 x.g) at 20 C in a model L
preparative ultracentrifuge (Beckman Instruments, Inc.,
Spinco Division). The duration of the individual sedimentation
runs was 4-20 hours. At the completion of the experiment,
the rotor was allowed to "coast" to a stop. The time taken
to reach full speed and the time necessary for the centri-
fuge to stop was equivalent to 7 minutes at top speed
(Hogeboom and Kuff, 1954). The centrifuge tubes were removed
and sampled immediately by piercing the bottom of the tube
with a hypodermic needle using the device shown in Fig. 2
(Martin and Ames, 1961). Two drops were collected per sample
and these were assayed for enzymatic activity. The geometry
of the rotor and the volume in the centrifuge tube were
utilized in calculating the distance the protein moved

under the influence of the centrifugal field.

Theory and Calculation.--The sedimentation coefficient

 

is the velocity of a sedimenting molecule in a unit field

and is defined by

1

s =-——-
€02x

(l)

(Ila
rt ><

 

Figure 1.

Figure 2.

19

PM” .- \.— WVVVN w VVV ‘N ~~~~~
P VVVVV WV" .— v ‘— \—a'- P v ~—— V
r «\V V‘MVN r- ——\r\v VVVVV
v'vy'y ‘v'MW L— _.'\.—‘\a‘~ vvvvv
e. ‘V V‘VVWV‘N t, V-VV- iv v v \— 1
xv'v'ywwywy“. p v'v'w N r v v v j
. V‘V'V x..\/\_/'V y“, P v V» .a V V\—j
PVVVVVVVVH Bye,» pwvvq
y. VVVVV ‘w'- v e. V v\— \— ~ ‘1‘»
\ _ \4 x.— a v v .— \d F' V'V V vvvvv

~—-\—~—~\———\-—~—v--

- —’I \— —— -
v.’ - __ .. —\ . _ ._. _

 

 

 

 

 

 

((
l((
"(
(({
(.
{((

ll
(.(
(_LJ
{Tr—r
(((
l((

Lt "
((((

l((

Ill I

 

Apparatus for the preparation of linear sucrose
gradients. A modification of the device described
by Hogeboom and Kuff (1954).

11.
J’-

(l-‘ ..
ll .
_----

l

 

 

Centrifuge tube sampling device, similar to that
described by Martin and Ames (1961).

 

20

where d>is the angular velocity in radians/sec
t is the time in seconds, and
x is the distance of the protein boundary from the
center of rotation in centimeters.
The unit of sedimentation which is commonly employed is the
l

Svedberg which is equivalent to 10— 3 seconds. The integrated

form of equation (1) is:

s =-—l— ln-E- (2)

w
><

where x0 and x are the respective positions of the boundary
at time t0 and t. The angular velocity was easily calculated
from the rotational velocity of the centrifuge. The move—
ment of the protein boundary down the centrifuge tube was
detected by the sampling procedure described above.
Utilizing equation (2), it was possible to calculate the
sedimentation coefficient of the glucose dehydrogenase.

In order that data derived under various conditions
may be compared, it was necessary to correct the sedimentation

20,

water at 20 C. The following relationship was used.

coefficients to s w’ the equiValent sedimentation rate in

sn (1 —'pr)

s = __ (3)
20, w nw (1 _ VP)

 

where n and n are the viscosity of the medium and water at

W

20 C respectively

<|

is the partial specific volume of the protein
pw the density of water at 20 C

p the density of the medium.

 

21

The partial specific volume of the glucose dehydro-
genase was taken as 0.72 cm3/g based on the data of Fox and
Foster (1957).

The sedimentation velocity of a molecule can also
be described in terms of the properties of the molecule.

M (l — Vp)
Nf

 

(4)

S =

where M is the anhydrous molecular weight
N is Avogadro's number and
f is the frictional coefficient of the molecule in
solution.

The diffusion coefficient D is given by

DJ?"- ‘ '(5)
where k is Planck's constant, the gas constant per molecule
and T is the absolute temperature.

On solving equation (4) for molecular weight and
substituting for f in terms of the diffusion constant, the

following familiar equation for molecular weight is obtained

M 2 RTS (6)

D (l — Vp).

 

In order to confirm the sedimentation data obtained
by the procedures of Hogeboom and Kuff (1954), sedimentation
runs were also performed in 5-20% sucrose gradients (Martin
and Ames, 1961). In these runs, point to point variation of
the viscosity and density down the length of the centrifuge

tube had to be taken into account when the sedimentation

22

values Were based on the following considerations.
By combining equations (1) and (3)
n (1 — pr) dx

8 = 2 ._ (7)
a) x nw (l — Vp) dt

 

 

 

 

2 _
320’ w (D dt — A F(x) dx (8)
(l - pr)
where the constant A = n and (9)
w
F(x) = TL (10)
(1 - Vp)x
On integration
2 Xi+1
520’ w w t = A F(x)dx (11)
xi
Xi+1
A F(x)dx
Xi
s = (12)
20, w (”2t
Xi+1
The term F(x)dx can be evaluated by a trapezoidal
X.
i

approximation (Martin and Ames, 1961).

X.
1+1 i+1 1 1+1
F(Xi) (Xi+1) + E

OMx
OMX

F(X)dX = [F(Xi+l) " F(Xi)]

[X '- X-] (13)

i+1 1

An example of this calculation is presented in the Appendix.

23

Density and Viscosity Determinations.--A 25 ml

 

pycnometer was used for the density determinations and an
Ostwald viscosimeter was used for the viscosity determina-
tions. When using a pycnometer and a liquid standard

(water), the equation for the determination of density is:

 

p=w 'pHo (14)

where p and pH 0 are the densities of the solution measured

and water respectively and W and W are the weights of the

H20

solution and water in the pycnometer. Similarly, when using

an Ostwald viscosimeter the equation for viscosity is:

 

Tl=t . .nHO (15)

where n and nH O are respectively the viscosities of the
2

solution and water, and t and tH O are the times observed
2

for the solution or water to fall from one point to another

in the instrument. Density and viscosity determinations were
made on the solutions that were used in forming the 0-3%
sucrose gradients because over this short range the variation
in viscosity was essentially linear. These data are found

in the Appendix, The density and viscosity at any point in
the centrifuge tube could then be calculated because the
gradient was linear. In the 5—20% sucrose gradient, the
variation in density was linear but the variation in viscosity

was not. The viscosities and densities of buffered 3 M NaCl

24

solutions containing 0, 3, 5, 12.5 and 20% sucrose were

determined and are shown in the Appendix.

RESULTS

The yield of spores from 100 liter cultures of
_B. cereus was routinely 400 g wet weight. The activity of
glucose dehydrogenase in crude preparations extracted from
these dormant spores was 10,000 units per gram wet weight.
On the other hand, the amount of enzyme extracted from
germinated spores was 33,000 units per gram of original weight
of spores. This difference in yield can be ascribed, in
part, to the greater ease in breaking germinated spores.

The glucose dehydrogenase used in the sedimentation
studies was not extensively purified. Its specific activity
was 11,500 units per mg of protein, which was a fifty-fold
purification over the crude extract. The concentration of
enzyme used in the sedimentation runs was 192,000 units per
ml. This high concentration was necessary so that reasonable
levels of enzyme activity would occur in each fraction col—
lected from the centrifuge tubes.

Since Bach (1963) had found an apparent disaggre-
gation of the enzyme in solutions of NaCl, an attempt was
made to separate the aggregated from the disaggregated
molecules by gel filtration. Nine hundred milliliters of
G—lOO Sephadex gel, contained in a 5 x 60 cm column, were
used in this study. The buffer system was 0.05 M imidazole

25

 

26

chloride pH 6.5 which was supplemented with 4 M NaCl for some
experiments. The columns were run at room temperature be-
cause the enzyme was stable. Four milliliters of partially
purified enzyme were placed on the column and eluted with
either of the above buffers. The results of the experiments
with either solvent system were almost identical. Most of
the protein contained in the enzyme preparation was eluted
prior to the elution of the enzyme. A typical run using
0.05 M imidazole chloride buffer pH 6.5 is shown in Fig. 3.
The method was not sensitive enough to resolve the
various states of aggregation of the enzyme. The main
protein peak, as measured by 280 mu absorption, was, however,
completely separated from the activity peak. Since it was
recognized that future studies of glucose dehydrogenase would
necessitate the use of pure enzyme preparations, a more
extensive purification than that sufficient for the sedimen-
tation studies was attempted. A column of G-lOO Sephadex
and a buffer system of 0.05 M imidazole chloride pH 6.5
containing 0.5 M NaCl gave results similar to that seen
in Fig. 3. This NaCl concentration was used to insure
that the enzyme was heat resistant and that the column could
be run at room temperature. Preliminary purification studies
showed that a combination of ammonium sulfate fractionation,
heat denaturation of contaminating proteins, and gel
filtration resulted in an enzyme with a specific activity

of 500,000 units/mg or of 90—100% purity. The results are

 

27

 

 

3801 83:! ALIAILOV

 

 

O 280 MJIABSORBANCE
O ENZYME ACTIVITY
.20" 1.5000
.‘I
§ \. --4000
E .I5‘r . \
C)
C)
8 o
m "3000
(J
2!
g .lO-- \0 ’
a:
o / -
a) "2000
GD ..
< /0 O /
0515/0
‘0—0’0 «Iooo
‘0
’,/’.' \\\
*0
O .’.l g 1 l 0
I00 |20 I40 l60 l80
SAMPLE NUMBER
Figure 3. The purification of glucose dehydrogenase by gel

filtration. A 900 m1 column, which was run at
room temperature, contained G-100 Sephadex
suspended in and eluted with 0.05 M imidazole
chloride buffer pH 6.5. Four milliliters of
solution containing 29,000 units/ml were placed
on the column and 2 ml fractions were collected.

 

 

28

found in Table 1.

In the course of studying the heat resistance of
glucose dehydrogenase, Bach (1963) noted that pH 6.5 was
optimal for its thermal stability and that a three hundred—
fold difference in the stability of the enzyme occurred
between pH 6.5 and 7.5. Furthermore, in 0.05 M imidazole
chloride buffer, pH 6.5, increases in the sodium ion concentra—
tion produced dramatic increases in the thermal stability of
glucose dehydrogenase. Fig. 4 is a plot of the relationship
between the half-life of the enzyme at 55 C versus the pH
of the medium over the range 6.5—7.5. In Fig. 5 is shown
the relationship between the half—life of the enzyme at 85 C
and the concentration of the sodium chloride contained in
0.05 M imidazole chloride buffer at pH 6.5. The possibility
existed that the increase in thermal stability was related
to the state of aggregation of glucose dehydrogenase mole—
cules. Bach (1963) had shown that an apparent disaggregation
of the enzyme occurred when the sodium chloride concentration
of the suspending medium was increased from 0 to 5 M.

Sedimentation studies were made in an attempt to
determine the state of aggregation of glucose dehydrogenase
in solutions of various sodium chloride and hydrogen ion
concentrations. The results of a typical sedimentation
experiment are seen in Fig. 6 where the activity per 0.1 ml
of individual samples is plotted against the sample number.

The sedimentation run was performed at 20 C in 3 M NaCl

 

29

Table 1. Summary of the purification results for glucose
dehydrogenase from germinated spores of B. cereus.

 

 

 

 

 

 

Procedure Volume Total Purifi- Yield
(m1) Activity Protein Activity cation (%)
(units) (mg) (units/mg)
Initial
extract 1845 13,500,000 57,000 235 l 100
(NH4)2804
fraction—
ation
(60—100%
sat.) 374 13,450,000 2,240 6,000 26 100
Heat 302 14,000,000 757 18,500 79 104
Sephadex
treatment
on 4 ml 90 180,000 0.36 500,000 2,340 103
portion (effluent)

 

dissolved in 0.05 M imidazole chloride buffer pH 6.5. The SW—39
rotor was run at 38,000 RPM for 12 hours. A 0—3% sucrose
gradient was used to stabilize the protein boundary. Each
sample number represents a given position in the centrifuge

tube where 0 and 40 represent the top and bottom of the tube
respectively. In the figure there can be seen a zone

devoid of enzyme between samples 4 to 8, a zone of increasing
enzyme concentration between samples 8 to 20, a concentration
plateau region between samples 20 to 32 and finally a zone

where the enzyme had been concentrated due to sedimentation

 

3O

 

 

 

 

 

 

900-, C‘>
800"
700"
U)
m 600--
g.
:3
5 soc--
2
C)
400"
\N
¢a IBOTM-
200--
l00"
CL---__‘
0 I 1 1 I An?!
6.5 7.0 7.5
pH
Figure 4. The effect of pH on the half—life of glucose
dehydrogenase at 55 C. The enzyme was suspended
0.05 M imidazole chloride buffer at various
hydrogen ion concentrations. The values for t

were calculated from the original data 1/2

presented by Bach (1963).

 

 

 

31

 

 

200
I80"
I607
I40"

8
.- '20"
2)
_2_ I00«-
2
X“ 80"P
"' so--
401-
20w-
0
Figure 5.

 

 

l I

I I u

I 2 3 4 5

[Na cu] MOLES PER ,LITER

The effect of the NaCl concentration on the half—
life of glucose dehydrogenase at 85 C. The enzyme
was suspended in 0.05 M imidazole chloride buffer
pH 6.5 which contained various concentrations of
NaCl. The tl/2 values were calculated from the

original data presented by Bach (1963).

 

 

32

UwDDO©V uzflom coauooamsfl one
mmB mumpzson samuonm one

Eoonmmom mo nonumE osu >3 uzoEHHmmxm zoflumuzmfiflpom HMUHQSH m we muasmmu one

mmmZDZ u4i2<m
o.~ o..

d

A ACTIVITY

.Honfisc mamfimm may .m> onEmm mom mufl>flpum CH omsmso ecu mzHDDOHQ
an posaauouwp wm3 can humpsson saououm one mo Housoo onu muzomonmmn “mafia
.uzmflpmum omOHUSm Xmlo Hmozfla m an pwuflaanmpm
.Homz E m mzHchucoo ummmsn oUHHoHno maoumpflefl
S mo.o Ga 0 ON um meuomnom mm3 SSH coaumuzmefipmm one

o.»

.Aemmav muss mam

 

0:
ON.
0m.
0?.
on.
00.
on.
00.

cm.

3.. om.

I

 

~.n ow Nm

0 0

di

 

 

A

A

A

.00.

.00N

.00m

.00?

room

.000

.005

.000

 

 

‘IW l'0 83:! SUN“

.0 ousmflm

 

33

against the bottom of the tube. It was the zone of increasing
enzyme concentration which was of interest in this study.

The inflection point which represents the center of the
boundary of the sedimenting protein was determined by

plotting the change in activity per sample vs. the sample
number. The sedimentation constants for glucose dehydrogenase
at the various conditions employed were calculated by equation
(2). They were then corrected to conditions in water at

20 C by equation (3).

The sedimentation constants for glucose dehydro—
genase in 0.05 M imidazole chloride buffer with pH values
6.5, 7.0, 7.5, and 8.0 are shown in Figure 7 and average 8.4 S.
The Appendix contains the data from which the points in this
curve were calculated. Although there was a slight upward
trend in the sedimentation rates with increasing pH values,
there was no indication that either aggregation or dis-
aggregation of the enzyme occurred.

The effect of various concentrations of sodium
chloride on the sedimentation properties of glucose dehydro—
genase is shown in Fig. 8. The data from which these points
were calculated are presented in the Appendix. In addition
to the NaCl, the medium contained 0.05 M imidazole chloride
buffer at pH 6.5. The sedimentation constant of the enzyme,
in the absence of NaCl, was 8.1 S and this increased to 8.7 S
at 1 M NaCl. An over-all decrease in the sedimentation rate

to 5.9 S occurred in the concentration range 1—5 M NaCl. Of

 

34

 

 

 

 

9 ..
o
o
o o
Szo,w a .. o
7 : : : :
6.5 7.0 7.5 8.0

Figure 7.

The effect of pH on the sedimentation properties
of glucose dehydrogenase. The enzyme was
suspended in 0.05 M imidazole chloride buffer at
various pH values. The protein boundary was
stabilized by a linear 0-3% sucrose gradient.

 

 

35

 

$20...

 

 

 

5 1 1 1 1

 

0 I 2 3 4 5
[No on] MOLES PER LITER

Figure 8. The effect of the NaCl concentration on the
sedimentation properties of glucose
dehydrogenase. The enzyme was suspended in

0.05 M imidazole chloride buffer pH 6.5
containing various concentrations of NaCl.
The protein boundary was stabilized by a
linear 0-3% sucrose gradient.

 

36

particular interest is the plateau region of 7.4 S which
occurs at about 3 M NaCl.

In order to confirm the above findings and to better
understand the results obtained with the intermediate NaCl
concentrations, sedimentation runs were made using sucrose
gradients of 5-20%. The medium contained 3 M NaCl in 0.05 M
imidazole chloride buffer and was saturated with respect

to sucrose and salt at the most dense portion of the gradient.

The results of the experiment are shown in Fig. 9. It should
be noticed that two peaks of enzyme activity occur. The
520 w values for these proteins were determined as shown in

the Appendix. The larger of the two peaks has a sedimentation
rate of 8.2 S, approximately the same as that found for the
enzyme in imidazole chloride buffer pH 6.5. The smaller

peak moved in the centrifugal field with a sedimentation

rate of 7.4 S which was approximately the same as that found
for the glucose dehydrogenase in 3 M NaCl solutions.

The molecular weights of the glucose dehydrogenase
were determined at 0, 3, and 5 M NaCl concentrations by
equation (6), using the diffusion constants obtained at 0
and 5 M NaCl by Bach (1963) and at 3 M NaCl by Sadoff

(unpublished). These values, corrected to water at 20 C,

6 6

were 1.9 x 10_ cmZ/sec, 2.7 x 10_6 cmZ/sec, and 2.8 x 10-
cmz/sec respectively. The molecular weights at.0, 3, and 5

M NaCl were respectively 37,000, 22,000, and 19,000. It is

 

37

 

2800

2600‘

\o

2400‘

2200‘

 

O“">

2000'

1

I800'

I600-

O

 

U

I400'

I2001

I000-

U

UNITS PER 0.l ML

800*

O>

U

600'

U

400q

200"t

 

 

l I l l l l l l I

I0 I2 I4 I6 I8 20 22 24 26 28 30
SAMPLE NUMBER

Figure 9. The effect of 3 M NaCl on the sedimentation
properties of glucose dehydrogenase. The NaCl
was suspendedin 0.05 M imidazole chloride
buffer pH 6.5. A 5-20% linear sucrose gradient
was employed and the enzyme was layered on the
top of the fluid column at the beginning of the
sedimentation run.

 

38

noteworthy that the sedimentation rate decreased from 7.4 S

to 5.9 S as the NaCl concentration increases from 3 to 5 M

without a change in the molecular weight.

 

 

DISCUSSION

The purpose of the research reported in this
thesis was to study the relationship between the heat re—
sistance and the state of aggregation of glucose dehydrogenase
in solutions of various salt and hydrogen ion concentrations.
Gel filtration was used in an attempt to separate the
dissociated from the undissociated enzyme. Sephadex G—100
was selected because the enzyme was in the size range that
would be “retained" by particles of this gel. Filtration,
under the conditions described, was not an adequate means
by which to differentiate between the states of aggregation
of the enzyme. However, a separation of enzyme and extraneous
proteins did occur during the filtration process. This
suggested that the purification of glucose dehydrogenase
could be readily accomplished by a combination of ammonium
sulfate fractionation, differential heat denaturation and
gel filtration.

The germination process in spores involves the
excretion of DPA, metal ions, and peptides into the
surrounding medium with the concomitant loss of the heat
resistance of the spores. Even though the germinated
spores are heat labile, they contain proteins which are
capable of "regaining" heat resistance at pH 6.5 and high

39

4O

NaCl concentrations. This conclusion is supported by the
results of the purification of glucose dehydrogenase. The
enzyme, along with other spore proteins, was precipitated by
ammonium sulfate in the concentration range 60-100 per cent
saturation. Of these proteins, 20—25 per cent remained
soluble after the heat treatment used in the enzyme purifi—
cation. Presumably, their solubility could be taken as a
criterion that the proteins remained in their native state.
Enzymes in this protein fraction may provide additional
models for the study of thermal resistance in spores. The
molecular volume of the glucose dehydrogenase must be
smaller than that of the other heat resistant proteins be-
cause it was retained on the G-100 gel and completely
separated from them. This purification technique will
permit future studies to be made of the chemical nature of
the protein.

The sedimentation studies reported in this thesis
were of an exploratory nature and were performed with
partially purified preparations in a preparative centrifuge.
The centrifuge tubes were uniformly cylindrical in cross
section, not sectors as in an analytical instrument, and
therefore, the sedimentation values must be considered very
close approximations. In the procedure of Hogeboom and Kuff
(1954), only the boundary of the most slowly sedimenting
glucose dehydrogenase molecules can be determined. If

two molecular sizes of the enzyme existed and their boundaries

 

41

overlapped, the sedimentation rate obtained was an average
value. The partial specific volume, V, was estimated and
errors in this estimate are multiplied approximately three—
fold in calculations involving this parameter.

Glucose dehydrogenase does not undergo a change in
its state of aggregation at 20 C over the pH range 6.5-8.0
as shown by the constancy of its sedimentation rate in a
centrifugal field. The enzyme was very unstable above pH 8
and this limited the pH range investigated. There is noIQ
priori reason to believe that aggregation or disaggregation
of the enzyme should occur over this pH range at 20 C even
though its heat resistance undergoes a 300—fold change
at 55 C.

Bach (1963) indicated, however, that the dissociation
of glucose hydrogenase did occur with increasing salt
concentrations at pH 6.5, conditions which also enhanced the
heat resistance of the enzyme. The extent of the dissociation
was determined ultimately by calculating the molecular weights
of the enzyme at various salt concentrations using sedi-
mentation-diffusion data. The sedimentation rate of the
protein decreased as the NaCl concentration in its environ—
ment was increased to 3 M concentration. Approximately
40% of the enzyme existed as a 7.4 S unit (wa. 22,000)
and 60% as an 8.2 S unit (M;w. 38,000) in 3 M NaCl at 20 C,
(see Fig. 9). As the sodium chloride concentration was

increased to 5 M, the sedimentation rate of glucose dehydrogenase

42

decreased to 5.9 S with no change in its molecular weight.
Undoubtedly, the shape and hydration of the enzyme subunit
underwent extensive changes as a result of the increased
salt concentration.

The sedimentation runs were performed for convenience
at 20 C. However, thermal agitation may play a considerable
role in determining the state of aggregation of glucose
dehydrogenase under the conditions of the above experiments.
Although the preparative ultracentrifuge has a limited
temperature range, sedimentation studies have been performed
at 5 and 27 C (Sadoff, unpublished). At 5 and 20 C, glucose
dehydrogenase has a sedimentation rate of 8.4 S in imidazole
chloride buffer pH 6.5. When the buffer system contains 3 M
NaCl, the enzyme sediments as a single peak with a rate of
8.6 S at 5 C and 7.2 S at 27 C. The enzyme disaggregates
completely in 3 M NaCl as the temperature is increased from
5—27 C.

These data suggest that the extent of dissociation
of glucose dehydrogenase is a function of the ionic con-
centration and temperature of its environment. The possi—
bility exists that at lower salt concentrations and tempera-
tures above 27 C, the enzyme could also exist in a completely
disaggregated state. If the subunits of the glucose
dehydrogenase are joined by ionic bonds, the function of
the salt in producing ion clouds and weakening those bonds is
readily appreciated. Thermal agitation could then more

readily bring about disaggregation of the protein.

43

The thermal resistance of glucose dehydrogenase
at 85 C increased approximately threefold when the NaCl
concentration was increased from 3 to 5 M. That is,
its half-life increased from 64 min to 187 min at 85 C.

Over this range there was no change in the state of
aggregation of the protein but some change in the shape of
the subunit must have occurred. Bach (1963) has pointed out
that the effect of NaCl on the heat resistance of the protein
is a continuous function. That is, the inactivation con-
stant of the denaturation process,k, varies as the reciprocal
of the square of the NaCl concentration over the
concentration range 0.5 to 5 M. If the effect of NaCl on

the protein in the concentration range 0.5—3 M is qualitatively
the same as that in the concentration range 3-5 M, the heat
resistant form of the enzyme in vitro must be the subunit.
Whether this is also true in the intact spore is a matter of
conjecture.

In order to achieve levels of heat resistance in vitro
equivalent to those in the spore, it is necessary to suspend
glucose dehydrogenase in 5 M NaCl solutions. It is unlikely
that salt solutions of this concentration ever occur in spores.
The bacterial endospore, like all cells, is an organized
structure. Its glucose dehydrogenase probably occupies some
relatively specific orientation as opposed to the positional
randomness that occurs in free solutions of the enzyme. It

is possible, that "point concentrations" of salts do occur

44

in spores and that these are adequate to satisfy the ionic

requirements for heat resistance.

 

 

 

SUMMARY

The relationship of the heat resistance of the spore
enzyme, glucose dehydrogenase, to its state of aggregation
has been investigated in solutions of various salt and hydrogen
ion concentrations. Gel filtration was used in an initial
attempt to separate the aggregated from the disaggregated
enzyme. Although Sephadex G—100 was not suitable for this
purpose, it was found that upon filtration through this gel
there occurred an excellent separation of glucose dehydrogenase
from contaminating proteins. Ninety to one hundred per
cent purification of the enzyme was obtained when the
protein solution was subjected to (NH4)ZSO4 fractionation,
differential heat denaturation, and gel filtration. The
enzyme yield from these procedures approached 100%.

Glucose dehydrogenase has a constant sedimentation
rate of 8.4 S at 20 C over the pH range 6.5-8.0. However,

a dissociation.of the enzyme from molecular weight 37,000 to
subunits of 19,000 molecular weight was observed when the
NaCl concentration was raised to 3 M. By increasing the
salt concentration to 5 M NaCl, a decrease in the sedimen-
tation constant occurred with a corresponding decrease in
the diffusion constant. The data indicated that the shape

of the protein changed but that its molecular weight was

45

 

46

constant.

The hypothesis was advanced that the property of heat
resistance resides in the subunit of glucose dehydrogenase.
Furthermore, the disaggregation which produces these subunits
occurs under conditions of increased temperature and

increased NaCl concentrations.

 

 
 

REFERENCES

Adye, J., H. Koffler, and G. E. Mallett. 1957. The relative
thermostability of flagella from thermophilic bacteria.
Arch. Biochem. Biophys. 61:251-253.

Bach, J. A. 1963. Some properties of the glucose dehydrogenase
from spores of Bacillus cereus. Ph.D. Thesis.
Michigan State University, East Lansing.

 

Black, S. H. and P. Gerhardt. 1962. Permeability of
bacterial spores. IV. Water content, uptake, and
distribution. J. Bacteriol. ‘83:960-967.

Black, S. H., T. Hashimoto, and P. Gerhardt. 1960. Calcium
reversal of the heat susceptibility and dipicolinate
deficiency of spores formed "endotrophically" in
water. Can. J. Microbiol. 6:213-224.

' Brohult, S. 1937. Splitting of the haemocyanin molecule
by ultrasonic waves. Nature 140:805.

Campbell, L. L. 1954., Crystallization of alpha-amylase
from a thermophilic bacterium. J. Am. Chem. Soc.
16:5256.

- Church, B. D. and H. Halvorson. 1959. Dependence of the
heat resistance of bacterial endospores on their
dipicolinic acid content. Nature 183:124-125.

Curran, H. R., B. C. Brunstetter, and A. T. Myers. 1943.
Spectrochemical analysis of vegetative cells and
spores of bacteria. J. Bacteriol. 45:484-494.

Davenport, C. B., and W. E. Castle. 1895. Studies on
morphogenesis. III. On the acclimatization of
organisms to high temperatures. Arch. Entwicklungsmech.
Organ..g:227—249..

Deal, W. C., W. J. Rutter, V. Massey and K. E. Van Holde.
1963a. Reversible alteration of the structure of
enzymes in acidic solution. Biochem. Biophys. Res.
Comm.‘lg:49-54.

47

48

Deal, W. C., W. J. Rutter, and K. E. Van Holde. 1963b.
Reversible dissociation of aldolase into unfolded
subunits. Biochemistry .2:246-251.

Fischer, E. H., and E. A. Stein. 1960. Alpha-amylases,
p. 313-343. .In P. D. Boyer, H. Lardy and K. Myrback,
[ed.], The Enzymes. 2nd. ed. Academic Press, New
York. 7 v.

Fox, W. F and J. F. Foster. 1957. Introduction to
protein chemistry. John Wiley and Sons, Inc.,
New York. 459 p.

Gerhardt, P. and S. H. Black. 1961. Permeability of bacterial
spores. II. Molecular factors affecting solute
permeation. J. Bacteriol. .82:750—760.

Halvorson, H. and C. Howitt. 1961. The role of DPA in
bacterial spores. .In H. O. Halvorson [ed.],
Spores II. Burgess Publishing Co., Minneapolis.

Hashimoto, T., S. H. Black, and P. Gerhardt. 1960.
Development of fine structure, thermostability, and
dipicolinate during sporogenesis in a bacillus.
Can. J. Microbiol. 16:203-212.

Hogeboom, G. H. and E. L. Kuff. 1954. Sedimentation behavior
of proteins and other materials in a horizontal
.preparative rotor. J. Biol. Chem. 210:733-751.

Koffler, H., G. E. Mallett, and J. Adye. 1957. Molecular
basis of biological stability to high temperatures.
Proc. Natl. Acad. Sci. U.S. 43:464-477.

Kools, J. W. 1960. Studies of a heat resistant catalase
from cells of Bacillus cereus. M.S. Thesis, Michigan
State University, East Lansing.

 

Lewith, S. 1890. Uber die Ursache der widerstandsfahigkeit
der Sporengegen hohe temperaturen. Arch. Exptl.
Pathol. Pharmakol. .g§:341.

Lundgren, H. P. 1936. Association and dissociation reactions
of thyroglobulin. Nature 138:122.

Martin, R. G., and B. N. Ames. 1961. A method for deter—
mining the sedimentation behavior of enzymes:
Application to protein mixtures. J. Biol. Chem.
.236:1372—1379.

49

McFarlane, A. S. 1935. An ultracentrifugal investigation
of the serum proteins. Biochem. J. 29:407-29.

Murrell, W. G. 1961. Discussion. 13 H. O. Halvorson
[ed.], Spores II. Burgess Publishing Co.,
Minneapolis.

. O'Connor, R. J., R. H. Doi, and H. Halvorson. 1960.

Preparation of bacterial spore extracts by use of
the Eppenbach colloid mill. Can. J. Microbiol.
6:233—235.

Pedersen, K. O. 1940. The protein molecule, p. 406-415.
.In T. Svedberg and K. O. Pedersen [ed.], The
Ultracentrifuge. Oxford University Press, New York.

Pedersen, K. O. 1950. Size relationship among similar
proteins; Association and dissociation reactions of
protein units. Cold Spring Harbor Symp. 14:140—152.

Powell, J. F. 1953. Isolation of dipicolinic acid
(pyridine 2:6-dicarboxylic acid) from spores of B.
megaterium. Biochem. J. 54:210-211.

Powell, J. F. and J. R. Hunter. 1956. Adenosine deaminase
and ribosidase in spores of Bacillus cereus.
Biochem. J. (London) 62:381—387.

 

Samejima, T., and K. Shibata. 1961. Denaturation of catalase
by formamide and urea related to the subunit make-
.up of the molecule. .Arch. Biochem. Biophys. 93:407—412.

Stanier, R. Y., M. Doudoroff, and E. A. Adelberg. 1963.
The Microbial World. 2nd ed. Prentice—Hall,
.Englewood Cliffs, N. J., 753 p.

Stewart, B. T. and H. O. Halvorson. 1953. Studies on the
spores of aerobic bacteria. I. The occurrence of
alanine racemase. J. Bacteriol. 65:160-166.

Stewart, B. T. and H. O. Halvorson. 1954. Studies on the
spores of aerobic bacteria. II. The properties of
an extracted heat stable enzyme. Arch. Biochem.
Biophys. .4g:168—178.

Svedberg, T. and S. Brohult. 1938. Splitting of the haemo-
cyanin molecule by ultraviolet light. Nature 142:
830—831.

 

 

50

. Vinter, V. 1961. The formation of cystine—rich structure
in sporulating cells and its posSible role in the
resistance of spores. In H. O. Halvorson [ed.],
Spores II. Burgess Publishing Co., Minneapolis.

Young, I. E. 1959. A relationship between the free amino

acid pool, dipicolinic acid, and calcium from resting

spores of Bacillus megaterium. Can. J. Microbiol.
.§:197—202.

M

 

APPENDIX

 

 

52

 

 

 

 

I.2-- >
o
o
.l
S
\ o
3 3 Va SUCROSE—> o
f; I.I--
6 <—— 0 °/o SUCROSE
z
m
o
o
|.O£ l l 1 1
O l 2 3 4 8

[NaCl] MOLES PER LITER

Figure 1. A plot of the minimum and maximum densities as
found in the centrifuge tubes vs. the NaCl
concentrations. These data were used to
calculate the density of the suspending medium,
at any point, in centrifugation studies utilizing
0—3% linear sucrose gradients.

53

 

 

 

 

2M
20:»
I9"
)
In J.
0) I8
0
E I7» 0
.J
:! 16..
2
y d-
b 15 o
{,5 3°/o SUCROSE—9
0 l4“
3 o
S; '3." <3
<—- 0 °/o SUC ROSE
Ian/0 0
II
0'
I0, . 1 1 1
I 2 3 4 5
[NaCl] MOLES PER LITER
Figure 2. A plot of the minimum and maximum viscosities as
found in the centrifuge tubes vs. the NaCl
concentrations. These data were used to calculate

the viscosity of the suspending medium, at any
point, in centrifugation studies utilizing 0—3%
linear sucrose gradients.

 

54

Appendix Table l. A summary of the data used for calculating
‘ the sedimentation constants of glucose
déhYdrogenase in‘U‘TUS'Mvi'midazole'buff'er~
and various hydrogen ion concentrations.*

 

 

 

 

 

 

 

 

 

 

 

Time at top speed (t) in a 2.2 1.5 1.5 1.1'
sec x 10 b 1.5 1.5

Angular velocity as) in a 4.03 4 07 4.07 4.03
radians/sec x 103 b 4.03 4.05

Number of samples a 30 39 37 37

b 37 37

Sample where boundary a 11 5 10.5 10.5 7.5
occurs b 9 5 9.5

Sedimentation constant in a 7.6 8.1 8.3 8.4
given solution at 20 C b 7.9 7.9
in Svedbergs

Solvent density (p) at a l 1 l l
boundary in g/cm3 at b 1 1
20 C

Solvent viscosity (n) at a 1.05 1.04 1.05 1.04
boundary in centipoise b 1.05 1.05
at 20 C

320 w in Svedbergs a 7.9 8.4 8.6 8.7
’ b 8.2 8.2

 

*A 0~3% sucrose gradient was used to stabilize the
enzyme boundary in the centrifuge tube.

 

55

. .mnsp mmsmwuucoo
esp Gd mumccnon mfimncm map muflaflnmum on poms mmB ucoflpmum omouosm XMIO d4

 

 

 

 

 

 

 

 

 

H.o m.e 6.5 m.m ~.m n 3 .ON
s.m m.n ~.R 4.s w.m m.s m mmumnow>m as m
o om um
Hm.a mm.a o4.a NH.H mo.H n mmnoaspamo an sumecson
Hm.H mm.H o4.a 4N.H NH.H mo.a m um Ace suamooma> pam>aom
0 ON
o~.H 4H.H NH.H 4o.a a a pm mao\m as sumecson
o~.a GH.H NH.H mo.a 4o.H H m um lav suamcme ucm>aom
wmnmnpm>m CA
e.H e.m s.m m.e m.n n 0 ON pm coausHOm cm>am
m.a m.m m.m m.4 o.n e.s m an ucmumaoo ceaumucmsaemm
m.m o.ma o.eH o.mH m.m A Resume
o.oa , m.4H o.ma m.ma m.ma m.HH m ssmecson mums; maasmm
44 44 «4 mm am 9
m4 m4 N4 84 mm cm 4 mmamsmm no impasz
40.4 40.4 mo.4 40.4 mo.4 9 mos x omm\mcmaemu
40.4 40.4 mo.4 mo.4 40.4 mo.4 m an Ass suaoon> “magmas
«.4 H.m 4.4 m.m m.H n 40H x new
«.6 H.m 4.4 s.m m.m m.m m an luv emwmm no» um mane
m 4 m m a o

 

Hmuaa\mmaoz _Homzi

 

 

*.mcoflumnuzoucou HOmz msoflum> 0cm m.m mm
Hmwmnn oHONMUHEH 2 mo.o CH mmmzmmoupwsop wmoodam mo mucmpmcou
coaumuzmfiflpom may mcflpmHsoawo now new: sump ecu mo hhmfifism a .N manna xflpcmmmd

 

56

 

I.2-

 

l l l l L 1 l

 

 

.J
2
‘~
(9
~v C
{I l.l‘
27$
:5
u:
(3
'00
Figure 3.

0 2 4 6 8 I0 l2 I4 I6 I8 20

PERCENT SUCROSE

The densities at 20 C of 3 M NaCl solutions
dissolved in 0.05 M imidazole chloride buffer
pH 6.5 vs. various concentrations of sucrose.
These data were used to calculate the density
of the suspending medium, at any point, in
centrifugation studies utilizing 5—20% linear
sucrose gradients.

 

57

 

 

 

 

L
28 (D
26...
A 24..
“J
2
o 224-
E
.J
a 20...
2
>_ '8"
I:
(0 l6"
(3 C)
:3 o/
s; I4fy4’EII’
IZ‘P
l0...»
1 1 1 1 1 1 1 1 1
2 4 6 8 I0 l2 I4 l6 I8 20
PERCENT SUCROSE
Figure 4. The viscosities at 20 C of 3 M NaCl solutions

dissolved in 0.05 M imidazole chloride buffer
pH 6.5 vs. various concentrations of sucrose.
These data were used to calculate the viscosity
of the suspending medium, at any point, in
centrifugation studies utilizing 5—20% linear
sucrose gradients.

 

SAMPLE CALCULATION BY THE METHOD OF MARTIN AND AMES (1961)

The enzyme was layered on the top of a 5-20% linear
sucrose gradient containing 3 M NaCl dissolved in 0.05 M

imidazole chloride buffer pH 6.5. The enzyme peak was found

in the 22nd of 44 samples, representing a distance of 7.6 cm

from the axis of rotation. At this point the sucrose con-

centration was 12.5%. In order to determine the sedimentation

constant (8 ), equation (12) is used:

20,
Xi+1
F(x)dx

Xi+1 -
F(x)dx term,

In order to evaluate the

X .
l

0.2 cm segments (xi) were taken
column (5.6 cm) to the position
At each segment the density (p)
mined and from these data F(x)

F(x)~
(1 -

58

from the top of the liquid
of the enzyme boundary (7.6 cm).
and viscosity (n) were deter—
was found from equation (10).

0
Vp)xi

59

 

 

 

 

 

 

 

 

 

Table 3. Data used for the Martin and Ames calculation.
xi Sucrose p 3 n F(x) F(Xi+1) _ in
(cm) (%) (g/cm ) (poise)
5.6 5.00 1.134 .0156 .0152
5.8 5.75 1.137 .0161 .0153 .0001
6.0 6.50 1.140 .0165 .0153
6.2 7.25 1.143 .0169 .0154 .0001
6.4 8.00 1.145 .0174 .0154
6.6 8.75 1.148 .0178 .0155 .0001
6.8 9.50 1.151 .0183 .0157 .0002
7.0 10.25 1.154 .0188 .0159 .0002
7.2 11.00 1.157 .0192 .0160 .0001
7.4 11.75 1.159 .0197 .0161 .0001
7.6 12.50 1.162 .0203 .0164 .0003
.1722 .0012
Xi+l
The term F(x)dx can then be evaluated by
x.
1
equation (13).
Xi+l Xi+l 1 Xi+1

F(x)dx = : [F(xi)(xi+1 — Xi)] +-§ : [F(xi+l)

x. i i

F(Xi)][xi+l ‘ Xi]

.1722 x .2 +-% (.0012)(.2)
= .03456.
The constant, A, in equation (12) is evaluated by the

use of equation (9).

A = (1 -'ng) = 11 - L 72)(.99811 = 28
nw .01005

 

60

 

 

Thus, 820 w of equation (12) can be found:
’ x.
1+1
A F(x)dx
x.
s = l
20, w 2t
2 (28)(.03456)
(1.62 x 107)(7.27 x 104)

8.2 x 10‘13 sec“1 2 8.2 s

 

 

ROOM USE ONLY

ROOM USE 0N1?-

 

 

   

E

M'lI'IIfiIIIIINIIIIIflIgIIInfillijljnjillmflliflfliS

L