EXPRESSQR AND PARTéAL PURiFECRi R33

8? THE VERU LENCE ANVGE RS
5 "7 tRSiRiA PES'E'ES

Efiesis fer the Re erase of R. S.
W J-R SERTE "R"’ERS&E‘:’
R’JRERT éeRR 1 R038? RR
1976

 

9

M IIIIIIIIIIIIIIIIIIIIIIIIIII

lllllHHlllllIIIIUIWHIIUIIINlllllfllIIWHIIIHIHIHHI ‘I

3 1293 10575 3663 4'

 

 

 

 

ABSTRACT

EXPRESSION AND PARTIAL PURIFICATION OF THE
VIRULENCE ANTIGENS OF YERSINIA PESTIS

By
Robert John Zahorchak

Purification of the V and w antigens from cell extracts
of 1. pestis was attempted. It was determined that more than
95% of the V antigen was associated with the cell. By employing
conventional techniques of protein purification (ammonium sulfate
fractionation, DEAE cellulose column chromatography, Sephadex
gel filtration, calcium hydroxylapatite column chromatography,
and preparative polyacrylamide gel electrophoresis), a prepar-
ation was obtained that contained two major proteins. One of
these proteins was identified as the V antigen. w antigen was
not purified from cell extracts by the use of similar methods
due to the inability to recover the antigen after adsorption

to calcium hydroxylapatite.

EXPRESSION AND PARTIAL PURIFICATION OF THE

VIRULENCE ANTIGENS OF YERSINIA PESTIS

 

By

Robert John Zahorchak

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of
MASTER OF SCIENCE
Department of Microbiology and Public Health

1976

DEDICATION

To Kaye, the mountains, and the songs they brought.

ii

ACKNOWLEDGMENTS

I wish to thank Dr. T. W. Burrows for the antiserum used
to initiate this work. Dr. R. N. Costilow, Dr. H. L. Sadoff,
and Dr. L. F. Velicar were all very generous with their labor-
atory eqiupment. I am grateful for the discussions and suggestions
that were openly offered by Dr. R. J. Patterson. A very special
thanks goes to Dr. R. R. Brubaker for his guidance and encour-
agment during these past three years, his help in the preparation
of this thesis, and for allowing me to learn by doing.

iii

TABLE OF CONTENTS

page
DEDICATION............................................... ii
ACKNOWLEDGMENTS.... ...................................... iii
LIST OF TABLES ........................................... vi
LIST OF FIGURES ......................... . ................ vii
INTRODUCTION ............................................. 1
LITERATURE REVIEW ....................................... . 3
MATERIALS AND METHODS .................................... 10
Bacteria .............................................. ' lO
Buffers .................. . ........................... . 10
Media.... ............................................. 10
Methods of Bacterial Cultivation ...................... 11
Preparation of the Spent Medium Concentrates
and Crude Cell Extracts ............................. ll
Immunological Methods ................................. 12
Column Chromatography ...................... ‘ ........... 13
Analytical Methods .................................... 13
Preparative Disc Gel Electrophoresis .................. 13
Chemicals ............................................. 14
RESULTS ................. .... .............. . .............. 15
Expression of the Virulence Antigens in
Oxalated N-Z Amine ....... . ..... . .................... 15
Separation of the V and W Antigens by DEAE
Cellulose Column Chromatography ..................... 15
Preparation of Crude V Antigen ........................ 15
Production of Anti-V Serum ............................ 20
Determination of the Relative Amounts of V Present
Extracellularly and Associated with the Cell ........ 26
Determination of the Optimum Inoculation Density
for Batch Culture Production of the V Antigen ....... 30

iv

Quantitative Studies of the Purification of

the V Antigen ................................... ..
DISCUSSION ....... O C 0000000000 O OOOOOOOOOOOOOOOOOOOOOO O O l O 0
LIST OF REFERENCES ............. . .........................

30
41
45

Table

LIST OF TABLES

The effect of the loss of various determinants

on the virulence of Yersinia pestis ...............
The effect of the cultural conditions on the

growth and virulence antigen expression of

Vwa+ and Vwa-‘X. pestis .............. ... ..........
Inoculation schedule for the immunization

against the crude V preparation ....... ... .........
Demonstration that absorbed anti-V reacts with

only Vwa+ strains of X. peggis and

1. pseudotuberculosis .............................
The effect of inoculation density on the
production of V antigen by M23 in oxalated

N-Z amine .........................................
Ammonium sulfate fractionation of M23 cell extract.

Purification data for the V antigen ............... .

vi

Page

25

29

31
33
37

Figure

1.

L‘

10.

ll.

12.

LIST OF FIGURES

Agar gel immunodiffusion analysis of the spent

media concentrates after cultivation of (a)

Vwa+ M23 or (b) Vwa- M23 in oxalated N-Z amine ..........

Agar gel immunodiffusion analysis of (NH4)ZSO4

fractions of (a) the spent medium concentrate

+
and (b) the cell free extract of Vwa M23 ...............

DEAE cellulose column chromatography of the

V and W antigens ........................................

Sephadex G-200 gel filtration of V antigen ...............

Calcium hydroxylapatite column chromatography of the

V antigen ...............................................

Analytical polyacrylamide gel electrophoresis of the

crude V antigen .........................................

Agar gel two dimensional immunodiffusion of antiserum

prepared against the crude V preparation ................

Immunodiffusion of an analytical polyacrylamide gel

of the crude V antigen against anti-V ...................

Outline of the methods used to purify the V antigen ......

Elution profiles of the V antigen from (a) DEAE

cellulose, (b) Sephadex 0-100, and (c) hydroxylapatite..

Analytical polyacrylamide gel electrophoresis

of the V antigen ........................................

Identification of a band of protein containing V

activity by polyacrylanide gel electrophoresis ..........

vii

Page

16

l7

18
21

22

24

27

38

39

INTRODUCTION

Expression of virulence by Yersinia pcstis, the causative
agent of bubonic plague, is dependent upon the genetic potential
to produce the virulence or V and W antigens described by Burrows
and Bacon (8, 12). These antigens are selectively expressed in an
environment that restricts the growth of virulent (Vwa+) cells 13
‘giggg (21). Although the growth of Vwa+ organisms that are cultured
at 37 C in a defined medium is normal only if Ca2+ (replaceable by
Sr2+ or Zn2+) is supplied at a concentration of 2.5 mM, the expression
of the virulence antigens is inhibited by the presence of this cation
(6, 24, 32). Vwa- mutants universally lack a nutritional require-
ment for CaZ+. When shifted from permissive to restrictive condi-
tions, Vwa+ cells cease to divide and their synthesis of DNA is
arrested after the current round of replication is completed (34, 41).
The Vwa+ determinant may promote resistance to phagocytosis by mouse
neutrophiles and free macrophages; it may also enhance survival and
multiplication of the bacteria during intracellular residence (8, 15,
29).

Lawton, Erdman, and Surgalla (33) have purified V and W from
culture fluids (100 and 1000 fold, respectively). These authors
reported that both antigens precipitated primarily between 1.3 and
2.2 M (NH4)2804 (24 to 41% saturation at 4 C) and that V elutes from
DEAE cellulose with 0.1 M NaCl and W elutes between 0.3 and 0.5 M
NaCl. They also determined that Vwasa protein (90,000 MW) and W
was a lipoprotein (145,000 MW). Although their preparations were
not homogeneous, they were able to demonstrate that anti-V serum
could passively protect mice against challenge with virulent organ-
isms. Anti-W offered no protection.

Virtually nothing is known about the biochemical role that the
virulence antigens play in restriction or virulence. The isolation
and purification of V and W must precede an attempt to specifically

identify their individual functions. Since the only assay system

that is presently available for the detection of the antigens is

an immunological one, monospecific antiserum to both V and W would
be invaluable to facilitate their purification. The studies
presented in this thesis were initiated in order to prepare
homologous antiserum to each antigen. Methods for the purification

of V from cell extracts are described.

LITERATURE REVIEW

.nggigig pgsgis is a facultative intracellular parasite that
is pathogenic for many mammalian species including man, in whom the
disease is infamously known as the ”black death" or bubonic plague
(17). Although formerly classified as a species of gaggegggllg,

.X; pestis now shares a genus with X; pseudotuberculosis and

 

14_enterocolitica due to the close relationship reported among these

 

organisms with respect to phage susceptibility, DNA homology, and

the common antigens (3). The members of the genus Yersinia appear

to be closely related to the other enteric bacteria and have therefore
Morphologically, the organism is a short blunt rod (0.5 to 0.8

X 1.5 to 2.0 um) (37). The cells are gram negative and are bipolarly

stained when treated with Wayson's stain or other analine dyes (30).

The bacteria can grow at temperatures from -2 to 45 C, but the

Optimal temperature for growth lg yiggg ranges from 27 to 28 C (38).

The organism is highly virulent. For example, an L.D. f

0
less than ten is observed when nice or guinea pigs were chaIgenged
intraperitoneally with suspensions of wild-type X; pestis (16).

This high degree of virulence has been associated with the expression
of the following determinants: l) Pgm+; loss of the ability to
produce pigmented colonies on the hemin agar of Jackson and Burrows
(27) or on the congo red agar of Surgalla and Beesley (39) increases
the average lethal dose to greater than 107 for mice and greater than
108 for guinea pigs (16). 2) Fra+; cells that are unable to produce
the Fraction 1 or capsular antigen are decreased in virulence for

'3 I
guinea pigs (L.D. of 10 to 103) but are fully virulent for mice

50
(16). 3) Pur+; auxotrophy for purines resulting from a mutation
eliminating guanosine monophosphate synthetase activity increased the
average lethal dose to greater than 108 for both mice and guinea
pigs, whereas lesions prior to the synthesis of inosine monOphos-

2
phate lowered virulence only slightly (L.D. of approximately 10 )

50

(2, 9). 4) Pst+; non-pesticinogenic mutants are unable to prOduce
coagulase and fibrinolysin (1). These mutants are decreased in viru-
lence for both mice and guinea pigs if the animals are challenged
with subcutaneous injections of Pgm- cells (4, 7). 5) Vwa+; the
virulence antigens, V and W, are produced by virulent X; pgggig and
.X;.E§€!QQEQRQESL19£L§ (8, ll, 12). A considerable decrease in viru-
lence is observed if this detenninant is lost (16). These findings
are summarized in Table 1.

Avirulence for mice due to lesions in pigmentation and purine
independence was reversed by the injection of iron or purines,
respectively, prior to or simultaneously with the challenge dose (9,
28). As already mentioned, Fra- mutants retain their virulence for
mice. Non-pesticinogenic X;.2£§£i§ also exhibited wild-type virulence
if injected intravenously (4). However, avirulence due to the loss
of the Vwa+ determinant appears to be irreversible in the sense that
virulence has not been phenotypically restored by the injection of
any known nutritional supplement (3).

Burrows and Bacon (14, 15) first demonstrated a difference
between virulent and avirulent cells of 1; pgsgis with respect to
their ability to resist phagocytosis by polymorphonuclear leukocytes.
Avirulent cells cultivated i3,gigg or ig giggg at 37 C were shown to
be much more susceptible to phagocytic injestion both-ig_gigg and
under defined conditions ig_gi££g. Since phagocytosis-resistant cells
became sensitive when homologous antiserum was added to the 33 XEEEQ
system, the investigators embarked on a search for an antigenic
difference between sensitive and resistant isolates. Burrows (8)
detected an antigen present in cultures of virulent X; pgsgis that
had been incubated at 37 C for three hours with aeration. This
antigen, termed Vi, was not detectable in cultures of avirulent cells
incubated at 37 C or in cultures of virulent cells incubated at 26 C.
The antigen was redesignated V when further experimentation by Burrows
and Bacon (12) revealed the presence of a second antigen, designated
W, that was unique to cultures of virulent cells incubated at 37 C
after an increase in incubation time to six hours. The short incu-
bation time during which V and W were produced did not allow for the
expression of a visible capsule, the presence of which rendered both

+ -
Vwa and Vwa cells highly resistant to phagocytosis (12).

’4

Janssen et a1. (29) later showed that the high resistance to
phagocytosis by neutrophiles and free macrophages was indeed closely
related to the Fra+ determinant. Expression of V and W did render
Fra- cells somewhat more resistant to phagocytosis by mouse and
guinea pig neutrophiles and free macrophages procured from mice.
However, the susceptibility to phagocytosis by guinea pig free macro-
phages was not significantly different between Era- Vwa- and Fra- Vwa+
cells. These authors suggested that expression of the virulence anti-
gens may be more important for the survival and multiplication of
the bacteria during intracellular residence than for resistance
to phagocytosis.

Fukui et a1. (21) have shown that at the temperature at which
V and W are expressed, aerated cultures of virulent cells became
avirulent after serial transfers in a complex broth. The authors
demonstrated that the lowered virulence was due to a shift in pop-
ulation resulting from selection for avirulent mutants. Studies on
the growth of virulent and avirulent cells in a chemically defined
medium (23) resulted in the description of a nutritional requirement
for 2.5 mM Ca2+ (replaceable by Sr2+ or Zn2+) by virulent X4 pgggig
grown at 37 C with aeration (24). Avirulent cells, however, showed
no requirement for this cation under similar conditions.

The development of an agar medium consisting of blood agar base

with added MgCl and sodium oxalate (Mng agar) enabled Higuchi and

2
Smith (25) to assay for avirulent cells in virulent cultures. They
determined that mutation from virulence to avirulence occurred at
a spontaneous rate of 10-A. This high spontaneous mutation rate
plus the inhibition of growth of the virulent cells in a population
cultured at 37 C in a medium deficient in Ca2+ accounted for the rapid
attenuation of virulent cultures described by Fukui et al. (21).
The mutation appears to be irreversible, although one occurrence of the
reverse mutation has been reported (13). In this case, however,
the investigators cautioned interpretation of the results due to the
possibility of contamination by one of the many virulent strains of
1; pestis with which they were working at the time.

Due to the observations that avirulent cells lacked a nutri-

2+
tional requirement for Ca and also failed to express the virulence

, , 2+
antigens, there appeared to be a direct relationship between Ca -

5

Table l. The effect of the loss of various determinants on

the virulence of Xgrsinia.pgstis.

 

 

 

 

 

Determinant LDSOa

that is missing mice guinea pig

Pgm >107 >108
3 6
Fra ‘10 10 -10
7 /

Purb 2102 (>108) 2.10+ (>108)
Pstc «105 ~ 106
Vwa '107 ’108
none 410 ‘10

aIntraperitoneal challenge.

The values that are not in parentheses are the LD values
observed with organisms that have biosynthetic blocks prior to
inosine-S'-monophosphate. The values obtained when the challenge
organisms have no guanosine-S’-monophosphate synthase are enclosed
within the parentheses.

cIntravenously injected Pst-.X;_pe§gi§ exhibit wild-type virulence.

Table 2. The effect of the cultural conditions on

+ -
the growth and virulence antigen expression of Vwa and Vwa

X. Bestis.

 

 

Phenotype Virulence Cultural conditions
26 c 37 c + Ca2+ 37 c - Ca2+
VW growth VW growth VW Jrolth
Wra+ + - + - + + -
VWa- - - + - + - +

 

dependence, V and W production, and virulence (See Table 2).
Supportive evidence for this hypothesis came when Fukui et al. (20)
demonstrated that cells of X. pestis grown at 37 C were phenotypically
more virulent than those that were incubated at 5 C or 26 C. Further-
more, Lawton (32) and Brubaker and Surgalla (6) showed that the addition
of Ca2+ to cultures of virulent 1. pestis growing at 37 C lowered the
amounts of V and W produced.

The first evidence that Ca2+-dependence may be a separate deter-
minant of virulence was reported by Brubaker and Surgalla (5). By
selecting for mutants in a virulent culture that were resistant to
relatively high concentrations of streptomycin (5,000 units per ml),
Ca2+-independent mutants were isolated that retained the ability to
express V and W. These mutants were shown to be avirulent, suggesting
that Ca2+-dependence is more closely associated with virulence than
is the expression of V and W. On the other hand, growth at 37 C of one
Ca2+-independent Vwa+ isolate in a liquid medium devoid of free Ca2+
was intermediate between that of a Ca2+-dependent Vwa+ cells and
Ca2+-independent Vwa- cells. The authors later demonstrated that
cells of the Ca2+-independent Vwa+ phenotype expressed less V and W
then did CaZ+-dependent Vwa+ cells (6). The growth of Ca2+-inde-
pendent Vwa+ cells was similar to that of Ca2+-independent Vwa- cells
if Ca2+ was supplied in the medium, indicating that the former
phenotype was not completely devoid of a requirement for the cation.
The presence of Ca2+ also inhibited the expression of V and W by
the Ca2+-independent cells as it does when it is present in cultures
of wild-type cells.

When typical Vwa+ cells of 1; pgggis were grown at 37 C in a
modification of the medium of Higuchi, Kupferberg, and Smith (24)
devoid of Ca2+, the cells underwent a dramatic physiological
change and entered a state called "restriction" (34). Restricted
organisms failed to synthesize significant amounts of DNA when
compared to growing cells (41). After Vwa+ cells were shifted
from permissive to restrictive conditions, the cells underwent
two doublings in mass during which time they divided once (34).
Morphologically, restricted cells appeared to be elongated when observed
microscopically (6) and lacked a dense nucleoid as elucidated by

electron microscopy (22). As mentioned earlier, it is this re-

7

strictive environment that stimulates V and W production. However,
if Ca2+ was included in the medium at a concentration of at least
2.5 mM, the cells divided normally and the virulence antigens were
not expressed (6). Therefore, a direct relationship seemed to
exist between cell division and the expression of the virulence anti-
gens. This hypothesis was strengthened when Brubaker and Surgalla
(6) showed a direct correlation between the stimulatory effect of
various energy sources on growth under permissive conditions and

on the expression of W under restrictive conditions. The more a
substrate stimulated growth when Ca2+ was present, the more it
stimulated W production in the absence of the cation.

In addition to the inhibitory effect of Ca2+ on the expression
of V and W, Brubaker and Surgalla (6) observed a stimulatory effect
of 20 mM Mg2+ on the production of these antigens. The W antigen
could not be detected in cultures that did not contain this high
concentration of Mg'+. It is interesting to note that the concentra-
tions of Ca2+ and Mgz+ that are necessary to promote the expression
of the antigens are remarkably similar to those reported for mammalian
intracellular fluids, whereas inhibitory concentrations of Ca2+
were present in the blood (31). These observations suggested that
the response to Ca2+ and Mg2+ may be related to the ability of the
bacteria to survive and multiply once it is phagocytized and that
the V and W antigens may play a role in this stage of infection.
However, the apparent paradox that cells cultured lg giggg must be
restricted in order for V and W to be fully expressed has not yet
been explained. Perhaps the restricted state is an abberant
manifestation of a physiological change which takes place 33
Iyiyg that renders the bacterial cells resistant to host cell
destructive mechanisms but still allows the bacteria to divide.

Attempts have been made to purify V and W in order to specify
their biological functions. Lawton, Erdman and Surgalla (33) have
purified V and W, 100 and 1,000 fold respectively, from the media
supernatant in which Vwa+lx._pg§gi§ were cultured. The chemical
composition of each antigen has been determined. V is a protein
of approximately 90,000 MW and W is a lipoprotein (38% lipid and
59% protein) of about 145,000 MW. These authors observed that

anti-V but not anti-W protected mice against challenge with

8

10

MATERIALS AND METHODS

Bacteria. Strain M23, 3 Vwa+ isolate of X; Egggis, and an
isogenic Vwa- mutant were employed throughout, unless stated other-
wise. This isolate was chosen because it is Pgm- and Fra- and thus
avirulent and unable to produce contaminating capsular antigen.
Stock cultures were maintained at -20 C in a phosphate buffer-
glycerol mixture as previously described (1), using a 0.033 M
buffer at pH 7.0.

lggffggg. In general the buffers employed in these studies were
either 0.05 M tris(hydroxymethyl)aminomethane buffer (Tris) or
phosphatgbuffered saline (PBS). The former was prepared by adding
concentrated HCl to Trizma Base (reagent grade, Sigma Chemical
Company, St. Louis, Mo.) until the pH was 7.85 at 4 C. PBS was
prepared by adding one ml of a solution containing 12.36 g of anhy-
drous Na HPO and 1.8 g of NaHZPO4.H20 per 100 m1 of distilled

2 4

water to 99 ml of 0.85% NaCl (final concentration of 6 mM P0 at pH 7.4).

'flgdia. The oxalated N-Z amine medium used to cultivate4
the organisms for V and W production was prepared as follows.
Stock solution A was prepared by bringing a 15% (wt/vol) solution
of N-Z amine in distilled water to 0.02 M with solid sodium oxalate.
The mixture was allowed to stand overnight at 4 C after which time
the precipitate formed was removed by centrifugation at 14,000 X g
for 20 min at 4 C in a refrigerated Sorvall centrifuge. The superna-
tant was carefully poured off and stored at 4 C with a small amount
of chloroform added to prevent microbial growth during storage.

The stock was routinely heated in a steamer and filtered through
Whatmann No. 4 filter paper prior to use. Stock solution 8
consisted of 0.25 M K HPO 0.1 M citric acid, 0.001 M Fec12, and

a
0.001 M MnCl2 in aqueius :olution. After addition of a few drops of
chloroform, the solution was stored at room temperature. For each
liter of medium, 200 ml of stock A were mixed with 100 ml of stock

B plus 600 ml of distilled water. This solution was neutralized

with NaOH and sterilized by autoclaving. When the medium had returned

to room temperature, 50 ml each of filtered sterilized 0.8 M
potassium gluconate and separately autoclaved 0.8 M MgCl2 were
added aseptically to yield a final concentration of 0.04 M apiece.

Methods of_bacte5ial cultivation. The starter cultures used to
inoculate fermenters consisted of 200 ml of oxalated N-Z amine medium
per 2 L flask into which was washed the growth from one slope of
blood agar base previously incubated with the appropriate organism
for 48 h at 26 C. The cultures were then cultivated overnight at 26
C during which time they were constantly shaken with the aid of a
model R-25 NBS Gyrotory Shaker (New Brunswick Scientific Co., New
Brunswick, New Jersey) at a setting of 84 cycles per min, which
promoted near optimal aeration. The resultant culture was aseptically
poured into fermenters (prewarmed to 37 C) containing 3 L of the
same oxalated N~Z amine medium until the cell concentration reached
a desired optical density (read On a Beckman DU spectrophotometer
at a wavelength of 620 nm). The concentration of cells obtained
in this manner will be referred to as the inoculation density and
is expressed in optical density units. The fermenters were then
incubated at 37 C with agitation and aeration for 20 h. Sterile
antifoam B emulsion (Dow Corning Corporarion, Midland, Michigan)
was added as needed.

Preparation_gf the spent medium conggn§§a£g§_and_g£udg_ggll
££££§€£§- The cells were harvested by centrifugation at 14,000
X g for 15 min at 4 C, washed in 0.033 M potassium phosphate buffer
(pH 7.0), and resuspended in Tris buffer at a ratio of 0.5 ml of
buffer per 1 g of wet packed cells. The Cells were then fractured
in a French pressure cell under a pressure of 7,000 psi and the
soluble fraction was separated from particulate debris by centri-
fugation at 27,000 X g for 20 min at 4 C. The reclaimed superna-
tant was designated the crude cell extract. The pellet was
resuspended in the original volume of buffer and the particulate
material was once again pelleted by centrifugation. The super-
natant from this step was designated the cellular debris wash.

The medium from which the cells had been harvested was brought
to 80% saturation with solid (NH/92804 and stirred at 4 C for 1 h.
The precipitated material was retrieved by centrifugation at 14,000

X g for 20 min and dissolved in a minimal amount of Tris buffer.
11

This solution was then dialyzed against Tris buffer at 4 C and
stored-at -20 C. The dialyzed preparation was termed the spent medium
concentrate.

Immunological methods. The antiserum used to detect the presence
of the V and W antigens (anti-VW) was a gift from Dr. T. W. Burrows.
Two dimensional agar gel immunodiffusion plates used in the qualita-
tive assay system were prepared as described previously (33). After
filling the wells with the appropriate antigen or antiserum, the
plates were incubated at room temperature in a sealed plastic bag
containing a wet paper towel to maintain a high humidity. Observa-
tions were made periodically with the final observations recorded
after 24 h of incubation. Relative concentrations of antigen obtained
during the course of purification were estimated by noting the posi-
tion of the precipitation band. Since the antiserum concentration
was constant, the closer the precipitation band was to the antiserum,
the higher the antigen concentration was in the sample. This method
of analyzing the plates aided in choosing fractions for further
purification.

The semi-quantitative assay system employed in some of the
experiments was performed on glass slides (l by 3 in) overlayed
with agar. The sane agar medium was used as was for the plates
except that merthiolate was omitted. To eliminate microbial growth,
the slides were placed under an ultraviolet light for 10 min immedi-
ately after the agar had solidified and before filling the wells. The
wells were punched with the aid of a Gelman immunodiffusion punch
and the agar was removed with a Pasteur pipet attached to a vaccuum
line. The well pattern was octagonal with the wells being 3 mm in
diameter and 3-mn apart; the outer well were 6 mm from the center
well. The slides were incubated for 24 h at room temperature in
a plastic box with a small amount of water added to prevent dessica-
tion. .

Antigen concentrations were determined as described by the
method of Lawton, Erdman, and Surgalla (33) and consisted of placing
serial two-fold dilutions of antigen in PBS in the outer wells.

The center well was filled with anti-V serum. The reciprocal of
the highest dilution showing a line of precipitation was designated

the titer. Arbitrarily, a concentration of one unit per ml was

12

assigned to a sample that had a titer of one.

Column chromatography. Prior to any chromatographic step,
the sample to be fractionated was dialyzed against the appropriate
starting buffer. Samples were concentrated either by precipitation

with 80% (NH and resuspension in a minimal amount of buffer

4)2504
or by ultrafiltration through a Diaflo UM lO ultrafilter (Amicon
Corp., Lexington, Mass.) under a pressure of 38 psi. Gradient
elution of samples adsorbed to diethylaminoethyl (DEAE) cellulose
(Whatman Ltd., Springfield Mill, Maidstone, Kent.) and hydroxyl-
apatite (Bio-gel HTP, Bio-Rad Laboratories, Richmond, Ca.) was
accomplished with the aid of an Autograd gradient maker (Technicon
Instruments Corp., Chauncey, N.Y.). The method of using four of the
cylinders to achieve a non-linear gradient has been described (26).
Sephadex (Pharmacia Fine Chemicals, Piscastaway, N.J.) and DEAE
cellulose fractionations were performed at 4 C. Hydroxylapatite
column chromatography was performed at room temperature. The
absorbancy at 280 nm of the eluant from each fractionation was
routinely monitered with the aid of an 1800 Model UA-2 Ultraviolet
Analyzer (Instrument Specialties Company, Lincoln, Nebraska).

‘Agglxti-_LJ@gghgd§. Protein concentrations of samples were
determined either by the method of Lowry et a1. (35) using bovine
serum albumin as the standard or by measuring the absorbancy at
280 and 260 nm and reading the protein concentration from a nomo-
graph (40). Inorganic phosphate concentrations were determined
by the method of Fiske and SubbaRow (l9). 'NaCl concentrations were
determined by measuring the conductivity of samples with a Wheat-
stone bridge (Model 31 Conductivity bridge, Yellow Springs Instru-
ment Co., Yellow Springs, Ohio) and converting the value obtained
to salt concentration with the aid of a standard curve. Analytical
polyacrylamide gel electrophoresis was performed according to the
method of Davis (18).

Preparative disc_ggl_§lgggggph9rg§i§. Preparative polyacryl-
amide disc gel electrophoresis was performed with the aid of a
"Prep~Disc" electrophoresis apparatus, employing the PD-2/7O upper
column (Canal Industrial Corp., Rockville, Md.). The separating

system consisted of a 2.5% stacking gel (8 mm in length) and a 7%

separating gel (97 mm in length). The gels were prepared as described

13

by Davis (18) with ammonium persulfate as the catalyst for poly-
merization of both gels. Tris-glycine buffer (the 10X stock con-
sisted of 6 g tris(hydroymethyl)aminomethane and 28.8 g of glycine
diluted to 1000 ml with distilled water) was used as both the top
and bottom electrode buffer as well as the elution buffer. The
current was maintained at 10 ma and 5 ml fractions were collected
at a flow rate of one ml per min.

.thmigglg. The sources of the chemicals used in these studies
were: N,N'-methylenebisacrylamide and acrylamide, Eastman Kodak
Company, Rochester, N.Y.; N,N,N',N',-tetramethylethylenediamine, Canal
Industrial Corp., Rockville, Md.; potassium gluconate, Calbiochem,
San Diego, Ca.; Ionagar No. 2, Colab Laboratories, Inc., Glenwood,
Ill.; sodium oxalate, J. T. Baker Chemical Co., Phillipsburg, N.J.;
N-Z amine type A, Sheffield Chemical, Norwich, N,Y.; Freund's

compete adjuvant, Difco Laboratories, Detroit, Michigan.

14

15

RESULTS

 

+
At least one antigen was produced by cells of Vwa M23 during
cultivation in the oxalated N-Z amine medium which was not detected

in similar cultures of Vwa- M23 (Fig. 1). After (NH 304 frac-

4)2
tionation, two unique antigens were resolved in the immunodiffusion

system (Fig. 2). The material from the spent medium concentrate

which precipitated between 20 and 50% saturated (NH4)2804 (MC 20-50)

and that from the cell extract which precipitated between 20 and

60% saturated (NH4)2804 (CE 20-60) were used in subsequent frac-

tionations.

.SeeetetieuatklesLW antigens bx DEAE celleleeeeqlsme
_ghggm§£ggggphy. Elution of MC 20-50 from DEAE cellulose with a
linear gradient of NaCl in Tris buffer resulted in the profile
shown in Figure 3a. The V antigen eluted at a NaCl concentration in
the range of 0.1 to 0.15 M. W eluted in the range of 0.22 to 0.28 M
NaCl. Chromatography of CE 20-60 resulted in similar separation of
the antigens (Fig. 3b). It should be noted that the W antigen could
be detected in fractions from DEAE cellulose column chromatography
of CE 20-60 only if the anti-VW serum was diluted ten-fold. No
difficulty was encountered in detecting W in the eluant of the
MC 20-50 fractionation when undiluted anti-VW was employed. This
finding suggests that the W antigen was present in much lower concentrations
in the CE 20-60 fractionation even though almost four times as much
protein was initially applied to the column.

.ggeparation of the crude_y_aggiggg. The assay system used in
the previously described studies was required because of the limited
quantity of anti-VW serum available. I thought it desirable, if
not necessary, to prepare.antiserum to each antigen individually
in amounts large enough to be employed in a more quantitative assay
system. I, therefore, proceeded to purify each antigen using the
qualitative assay system so that antiserum to each antigen could

be obtained.

@G
(D

Fig. 1. Agar gel immunodiffusion analysis of the spent
. +
media concentratesobtained after the cultivation of (A) Vwa M23
or (B) Vwa — M23 in oxalated N-Z anine. Well S contained anti-VW

serum .

l6

0-20 <:::> (:::> 30-40

Fig. 2. Agar gel immunodiffusion analysis of (Nd/2230‘,+ fractions
of (a) the Spent medium concentrate and (b) the cell free extract of
Vwa+ M23. The center wells contained anti-VW serum. The numbers
assigned to each of the outer wells indicate the upper and lower
(Nd4)2S04 concentrations, in percent saturated (NH4)2804, between

which the material precipitated.

17

Fig. 3. DEAE cellulose column chromatography of the V and W
antigens. In (a) 564 mg of protein of MC 20-50 were adsorbed to the
resin and in (b) 2,200 mg of CE 20-60 were adsorbed. The sample was
eluted with a linear gradient from 0.0 to 0.5 M NaCl in 0.5 M Tris
buffer, pH 7.85, at a flow rate of 2 ml per min. The gradient was
applied as soon as the sample was adsorbed to the column. The packed
resin dimensions were 2.5 by 40 cm. In (a) the relative absorbancy
at 280 nm was continuously monitered. In (b) the absorbancy of each
fraction was measured at 280 nm with the aid of a Beckman DU Spectro-
photometer. The horizontal bars indicate the fractions (5 ml) which

exhibited antigenic activity when tested against anti-VW serum.

18

Relative absorbancy

Absorbancy
280 nm

Fig. 3.

280 nm

 

H

P---—-l

11411 lllllllllilll

 

18'

14’
12

I

10

I I

 

10 20 3o 40 50 605*70
Fraction number

 

10 is 5050 30 60 70 s

Fraction number

19

80 90 100

H

0 90 100 Tio T'zo

be obtained.

The V preparation obtained from DEAE cellulose chromatography
was first filtered through Sephadex G-200. Antigenic activity
eluted in a broad peak, indicating that optimal separation was not
achieved by this technique (Fig. 4). However, this step did provide
a means to eliminate contaminating molecules of both very high and
low molecular weights. The V positive fractions were pooled and con-
centrated. This concentrate was then applied to an hydroxylapatite
column and eluted with a linear sodium phosphate buffer gradient
(0.001 to 0.5 M, pH 6.8). V was detected in fractions containing
between 0.075 and 0.15 M phosphate (Fig. 5a). When a more shallow
non-linear gradient was used to elute the sample, the antigen was
detected in fractions that reflected eluting protein (Fig. 5b).

These fractions were pooled, concentrated, and dialyzed against
PBS. The resultant preparation, containing 2.5 mg of protein per
ml, was analyzed by polyacrylamide gel electrophoresis. Stained
gels revealed the presence of one very prominant band and at least
13 minor bands of protein (Fig. 6). This preparation was termed
the crude V antigen.

Similar manipulations of the W antigen did not yield positive
results. After Sephadex G-200 fractionation, the antigen was adsorbed
to hydroxylapatite. No W activity was detected in the eluant when a
linear sodium phosphate gradient was applied (0.001 to 0.5 M, pH 6.8).
Various other methods of elution, including 2.0 M sodium phosphate buffer
and NaCl solutions at concentrations up to 3.0 M,failed to yield detect-
able W activity, although these processes did release some UV absorbing
material.

Production of anti-V serum. Anti-V serum was obtained by inducing
antibody formation to the crude V antigen in a New Zealand white rabbit.
Prior to immunization, the preparation was mixed with an equal volume
of rabbit anti-Vwa- M23 cell extract and incubated at 37 C for l h.

The immunoprecipitate formed was pelleted by ultracentrifugation at

149,000 X g for 30 min at 4 C. An emulsion of the supernatant and Freund's
complete adjuvant (mixed in equal proportions) was injected according to
the schedule presented in Table 3. The rabbit was bled through the
marginal ear vein on the sixth, eighth, tenth, and twelfth days after

the final injection. Anti-V activity was discerned after absorbing out

contaminating antibodies with MK)mg of lypholized Vwa- cell extract per 1.0

20

.umn ~mucoufluos ecu he voowowwcw mum moa>woom

owcmwuocm megawmucOo mcowuumpm one .wmucmmoua mm oflwwoua comesfio use no mcfiomuo a .oououwcofi xfimsoocHonoo

mm: Ec owm um xoconpomnm one pro wouoofifioo mum3 HE m we mcoHuumum .Eo 0H «0 wuommmum owumumouvm: m pope:

wwoofim paw AEo 00 we m.m we mcoflmcmeww omev cEsfioo ecu 0u powfiamm mmB Amw.n any acumen mane 2 mo.o cw
samuoum mo we 00 wowcwmoooo HE m we oHQEmm < .cmwwucm > we cowumuufiwm How oomuw xopmsaom .e .Mwm

umcfiac COMuomuh

 

 

 

 

e e s a s e. e a om a. s-
< (I! L

mu 093

Aoueqaosqe snideIeg

21

Fig. 5. Calcium hydroxylapatite column chromatography of the
V antigen. After (NH4)280A fractionation, DEAE cellulose column
chromatography, and Sephadex gel filtration, the V antigen prepara-
tion was adsorbed to a column containing packed calcium hydroxylapatite
(column dimensions of 1.5 by 26 cm). The sample was then eluted with
either (a) a linear gradient (0.001 to 0.5 M sodium phosphate buffer
at pH 6.8) or (b) a non-linear gradient formed by using four cylinders
of the gradient maker (cylinders one through four contained 0.001,
0.1, 0.1, and 0.25 M sodium phOSphate buffer at pH 6.8, respectively).
In (a) 55.5 mg of protein were adsorbed to the resin and in (b) 15 mg
of protein were adsorbed. The flow rate was maintained at 1 ml per
min, 5 ml fractions were collected, and the absorbancy at 280 nm was
continuously monitered. Tracings of the elution profiles are presented.
Fractions containing V activity are indicated by the horizontal

bars.

22

access coHuoouh
mm on me a mm om mm ON ma OH m
d 4 4

d d u a d

--
ul—

umnEnc newuomum
om mm On no 00 mm cm mm on mm 0m mm om ma OH m
u q q q 4 - .- q q u q q «

 

um oez

Aoueqaosqe 3A138133

 

 

 

“ID 082

Aoteqlosqe aAiJBIeu

 

Fig. 6. Analytical polyacrylamide gel electrophoresis of the
crude V antigen. The sample analyzed contained 496 pg of protein.
The positive electrode was at the bottom. The gel was stained with
1% Amido black in 7% acetic acid after electrophoresis was completed,
as indicated by the fact that the tracking dye had migrated to the

bottom of the gel.

24

Table 3. Inoculation schedule for the

hmnunizatiqq_against.5}c_gru1g_!_prgpara£ign.

--- -- --

 

Day Volume ' Routea Location
1 0.1 ml S.Q. each foot padb
8 0.1 ml S.Q. each foot pad
15 0.3 ml J.M. each thighC

 

aS.Q., subcutaneously; I.M., intramuscularly.
btotal of 0.4 m1 injected.

c . . . ‘
preparation not mixed With adJuvant.

25

ml of serum and incubating the mixture at 37 C for 1 h. The absorbed
antiserum showed a line of identity with anti-VW serum in the Ouchter-
luny test (Fig. 7). A single arc of precipitation was observed

when absorbed anti-V was diffused against 3 polyacrylamide gel into
which was electrophoresed the crude V antigen (Fig. 8). It was

not possible to identify which band of protein contained the anti-
genic activity, although the major protein was ruled out as the source
of activity.

To show that the antigen which was being detected was not unique
to strain M23, cell extracts of various strains of X; pgggig and
described in Materials and Methods and tested against the anti-V
serum. The results of those tests are tabulated in Table 4. Only
the extracts of Vwa+ cells showed a reaction with the absorbed anti-V.
This antiserum made it possible to employ a semi-quantative assay
system in the following studies.

Determination of the relativg_amounts of V_present extra-
cellularly_and assogiaggd;with_thg_ggll. I performed one experiment
to determine the relative amounts of V that were present either extra-
cellularly or associated with the cell. Fermenters containing 3 L
of oxalated N-Z amine were inoculated at an optical density of 0.35
and incubated at 37 C for 20 h. After harvesting the cells, the
medium supernatant volume was measured. A 100 ml aliquot of the
supernatant was filtered through a Nalgene filter unit (pore
diameter of 0.20 um) to remove any residual cells. A crude medium
concentrate was prepared from the filtrate. The cells were ruptured
in a French pressure cell in order to obtain the extract as described
in Materials and Methods. The concentration of V antigen in both
preparations was determined and the total units per culture was cal-
culated. A total of 98 units were present extracellularly, whereas
1700 units were detected in the cell extract.

To eliminate the possibility that these results were due to
adsorbtion of V to the filter, a reconstruction experiment was
performed. After diluting 1.0 ml of a protein preparation contain-
ing V to 100 ml with distilled water, the solution was filtered in
the same manner as described above. The filtrate was concentrated

by adding (NH4)2804 to 80% saturation. The precipitate was pelleted

26

Fig. 7. Agar gel two dimensional immunodiffusion of antiserum
prepared against the crude V preparation. A volume of 1 ml of
antiserum was absorbed with 100 mg of lyophilized Vwa- M23 whole cell
extract as described in Materials and Methods. Symbols: i, absorbed
anti-VW; ii, absorbed anti-V; iii, absorbed anti-V diluted 1:2 in

PBS. The center well contained crude V antigen.

27

45.55 mm

23 .35 min-g!

 

[~ gel 34]

V ( -- ---- - f-precipitation arc

[— Anti-V

Fig. 8. Immunodiffusion of an analytical polyacrylamide gel
of the crude V antigen against anti-V. A sample of the crude V
antigen was electrophoresed into a polyacrylamide gel until the
tracking dye migrated to the bottom of the gel (dotted line at the
right of the gel). The gel was then placed in a glass
petri dish and molten agar was poured to cover the bottom of the
dish. After the agar had solidified, a rectangular well was cut
out of the agar parallel to the gel. The well was filled with
anti-V diluted 1:2 in molten agar. The plate was then incubated
for 24 h at room temperature. The calculated mobility of the antigen
antigenic activity was approxiamately 0.5. This mobility is not
the same as that observed for the major protein shown in Fig. 6

(mobility = 0.25).

28

Table 4. Demonstration that absorbed anti-V reacts with only

Vwa+ strains of X. pestis and X. pseudotuberculosis.

 

 

m.—

 

b
Speciesa Strain Virulence Determinant Reactionc

Vwa Pgm Pst Fra

 

p M23 - +
’ D1

032
MD31
A12
A4 - -
KIM
EV76
K [M - -
KUMA -

"O
+ + + +
I

+-r++

"U
0.1

+ +

I
+ + + +
+ + + + + + + :3 + +
+ +

l
T

+
Salazar - +
M23 - +
A1122 - - + + -
JAVA - - - - -
G35 - - -
PBl/O - O O
KUMA - -
632 -

TS -

'O'U'U'U'U'U’U'U'U'U'U

'U
U)
C)

I

T
+ + + + +
I

T
+
Salazar - +
Yokohama - +

+

+ + +

Kimberly -
632. - - -
M23 - - + - -

+
I

Dodson -
A12 -
PBl/+ +

'O'U'U'U'U'U'U'U'U'U

<3 + +
<3 + +
I

0

“U
m

 

ap,'X.ng§£i§; ps, 1. pseudotuberculosis.
+, expressed; -, not expressed; 0, not expressed by the species.

c . .

+, cell extract shows a line of precipitation in the Ouchter-
lony test when diffused against anti-V; -, no reaction observed under
identical conditions.

29

by centrifugation, redissolved, and dialyzed against PBS. The volume
of the dialysate was measured and the concentration of V was determined.
After diluting 1.0 ml of the original protein preparation to a volume
equal to that of the filtered sample, the V concentration was deter-
mined. There was no decrease in the antigen concentration due to
filtration and subsequent (NHQZSO!+ precipitation as indicated by

the fact that the V titer in the two preparations was identical.
Igglggggvpqugqgiqn_gf_£hg_!_aq£iggn. I thought it desireable to ino-
culate the cultures at as high a concentration of cells as possible
in order to obtain the maximal amount of V per batch of cells culti-
vated. Therefore, fermenters were inoculated at different densities
and then incubated at 37 C for 20 h to see if the inoculation
density had an effect on the expression of the V antigen. Deter-
mination of the concentration and specific activity of the antigen
revealed that this paramenter did affect the relative amounts of

V produced in the culture. Although the amount of V produced per
culture appeared to be proportional to the inoculation density
except at the highest density, the specific activity of the antigen
was higher at an inoculation density of 1.4 than at the other
densities tested (Table 5).

 

~—---—-----‘ 00-.n— ..-—'—

The purification scheme that resulted in the least contaminated
preparation of V is outlined in Fig. 9. With the use of the semi-
quantitative assay system it was observed that approximately 90%
of the antigen precipitated between 30 and 50% saturated (NH4)ZSO4
(Table 6). Therefore, a 30 to 50% saturated (NH4)2804 fraction of
Vwa M23 cell extract (3.7 g of protein) was applied to a DEAE
cellulose column (2.5 by 43 cm). The column was then washed with
Tris buffer at a flow rate of 2 ml per min until no protein was
detected in the eluant. At this time, a linear gradient was applied
(0.0 to 0.3 M NaCl in Tris buffer) maintaining the flow rate.
Fractions of 10 ml were collected and analyzed for V (Fig. 10a).

V antigen eluted along with the first protein peak. Although the
elution profile from DEAE suggested that this technique was desire-
able to include in the purification of the antigen, a more stringent

analysis revealed that no net purification of V was acheived and more

30

Table 5. The effect of inoculation density on the

production of V antigen by M23 in oxalated N-Z amine.

 

—- - .- v *-

 

Inoculatign Total Vb Specific activityC
densu” CE CD CE+CD
0.385 1,696 1.8 1.9 ND
0.389 3,008 1.8 1.8 ND
0.743 ND ND ND 3.1
1.4 ND ND ND 4.5
1.4 12,096 3.3 7.1 ND
2.65 9,660 2.0 2.5 ND

 

aOptical density at 620 nm.
bTotal units of V calculated for both CE and CD.

CUnits of V per ml / mg of protein per m1; CE,.
crude cell extract; CD, cellular debris wash'
(see Materials and Methods).

31

(NH4)ZSO4 fractionation

(30-50%)

1

DEAE cellulose

(linear gradient, 0.0 to 0.5 M NaCl
in 0.05 M Tris buffer, pH 7.85)

Sephadex G-100 gel filtration

Hydroxylapatite column chromatography

(non-linear gradient,
0.001 to 0.25 M sodium phosphate buffer, pH 6.8)

Preparative polyacrylamide gel electrophoresis

Fig. 9. Outline of the methods used to purify the V antigen.

Table 6. Ammonium sulfate fractionation of

M23 cell extract.

 

Fractiona Resuspended V Total units
volume titer of V
(m1) (units/ m1)

20-25 5.7 16 91.2
25-30 5.6 32 179.2
30-35 8.0 128 1,024.0
35-40 8.1 128 1,126.4
40-45 6.6 128 844.8
45-50 5.3 64 339.2
50-55 5.1 16 81.6
55-60 5.2 2 10.4

 

8A 20-60% (NH ) so, fractionation of M23 cell
extract was brought 0 25%.saturation with solid
(NI-1402804 and the solution was stirred on ice for
l h. The precipitate was pelleted by centrifuga-
tion at 39,000 X g for 15 min and resuspended in
5 ml of 0.05 M Tris buffer at pH 7.85. This was
designated the 20-25 fraction. The supernatant
was then brought to 30% (NH,) 80, saturation and
treated similarly (fraction425-30). The other
fractions were prepared and designated in a
similar fashion. All fractions were dialyzed
against 0.05 M Tris buffer at pH 7.85 before
being analyzed for V antigen.

33

Fig. 10. Elution profiles of the V antigen from (a) DEAE
cellulose, (b) Sephadex G-100, and (c) hydroxylapatite. The
methods used in these fractionations are detailed in the Results
section of this thesis. The relative absorbancy at 280 nm was
continuously monitered in (a) and (b). The protein concentration
in each fraction from hydroxylapatite (c) was determined by measuring
the absorbancy at 280 and 260 nm and reading the protein concentration
from a nomograph. Tracings of the elution profiles are presented.

The vertical bars indicate the titer of V in each fraction.

 

 

 

 

 

 

 

 

 

A~E\muwcsv
umuwu >
4. a.
6 H m 42
q \ — — - w
.\.
A
II. you
5
lo
I” I4
L
AMIIMHHMHHH
unwmqmau
I3
uu
.2
no

 

 

Ea owm
mucmnuomnm m>wumfimm

 

Fraction number

Aae\muacsv
amuse >

6 8’42

2
3 1 .
.1 .1 ~ ~

5
3

 

71

 

 

Ill,

E: owu
mucmnuOmnm m>fiumfimm

Fraction number

Ase\muacsv

 

  

 
 

 

 

umuau >
2 6 84%
)J 1
- _ . 44¢ u
5b
‘5 m
t u
I n
5
I'll
'5‘qu n
HHHUr§.t
Ill]. 3 .c
c IV a
r
.3 fl. .3 nu .3
2 2 .l. 1 0
O O 0 0 0
hae\wev

sewumMucmuaou ewmuoum

10.

Fig.

35

than 70% of the antigen adsorbed to the column was not recovered
(Table 6).

The fractions from DEAE that contained the highest concentra-
tions of antigen were pooled, concentrated and dialyzed. This
concentrate was filtered through Sephadex G-100 (2.5 by 90 cm) in
two separate hatches (2.0 ml each, containing 145 mg of protein) under
an operating pressure of 30 cm. The fractions containing the bulk of
the antigen were pooled and, after concentrating the solution, the
preparation (69 mg of protein) was adsorbed to an hydroxylapatite
column (1.5 by 28 cm). A gradient (non-linear with cylinders one
through four containing 0.001, 0.1, 0.1 and 0.25 M sodium phosphate
buffer at a pH of 6.8, respectively) was applied at a flow rate of
one ml per min. Each cylinder contained 100 m1 of buffer resulting
in a total gradient volume of 400 ml. The 5 m1 fractions from this
separation that contained V were concentrated and dialyzed against
PBS. The elution profiles from Sephadex G-100 and hydroxylapatite
are presented in Fig. 10b and 10c. The antigen eluted as a single
symmetrical peak from each of the columns. Both techniques resulted
in a net purification of V (Table 7). Preparative disc gel electro-
phoresis of the V antigen obtained by the methods described above
resulted in an alteration of the relative amounts of the proteins-
in the preparation (Fig. 11).

In order to obtain a better separation of the two major proteins
and identify the band associated with V activity, 80 ug of the
purified preparation were electrophoresed into a 10 cm polyacryl-
amide gel (10% acrylamide) for 5.5 h. The gel was scanned at 280
nm and then sliced into 1 mm slices. Each slice was placed in the
cold overnight in a test tube containing 0.1 m1 PBS to allow the
protein to diffuse out of the gel. The concentration of V in each
tube was determined. A peak of antigenic activity was detected in the
region of the gel that contained the major protein when the slices
were matched up with the scan (Fig. 12). Although the evidence is
not conclusive, it appears that the major protein in the preparation

exhibits V activity (see Fig. 11 and 12).

36

 

muwuwmm

 

 

H «.0 o.oH as a «ma so s.~ -Haxouea;
Edwufiwo
m.s m.o o.~H om aa eon mNH m.o ooH-o
xmwmsawm
e N.~ a.m now so mao.H omw ~.q macanaamu
mama
. . a N a
an ~.N w.m Nae m we «mm ma emm em cm A sz
OOH H N.H os©.oH mm omo.sfl so 0mm uumuuxm
same
Aeaouv Awev Aae\wav Aae\anasv

huosoomu xUfl>fluum cHououa mugs: doom

ucmopom cofiuoUHMMusm owmwooam HmuOH cwmuoum fimuoe >

.mewucm > ago now muwv COwUmUHMflusm .h mfinme

37

 

 

 

 

 

 

 

 

 

 

 

Fig. 11. Analytical polyacrylamide gel electrophoresis

of the V antigen. A sample containing 160 ug of protein obtained
either before (left gel) or after (right gel) preparative polyacryl-
amide gel electrophoresis was electrophoresed into the analytical
gel until the tracking dye reached the bottom of the gel. After
electrophoresis, the gels were stained with 1% Amido black in 7%
acetic acid. Presented above are full scale drawings of the stained
gels. The mobilities of the proteins that migrate shnilarly to

the V antigen (see Fig. 8) are indicated.

38

Fig. 12. Identification of a band of protein containing
V activity by polyacrylamide gel electrophoresis. The figure
is a tracing of a gel scan made with the aid of a Beckman
DU Spectrophotometer with an attached linear transport. The
inlay indicates the titer of the slices that contained V
activity (vertical bars). The diagram on the top depicts a
stained gel of the same preparation that was electrophoresed
for a shorter period of time (ca. 2 h). The arrows indicate
the staining band that correlates with the 280 nm absorbing

band. For a more complete description of the methods see the text.

3')

l

 

 

 

 

 

top

 

Q)
U
"-4
v—i
U)
\o (D ~:r N
....
(Iw/snlun)
13312 A
we 082

Anucqlosee aniielau

bottom

12.

Fig.

DISCUSSION

In my first attempts to purify the virulence antigens, I
cultivated the bacterial cells in the medium of Lawton, Erdman,
and Surgalla (33) and used the spent medium concentrate as the source
of the antigens. However, many medium constituents precipitated
upon (NH4)2304 fractionation. This resulted in difficulties in
resuspending the precipitate in a workable volume and in subsequent
separation of the antigens from the medium constituents. I, therefore,
employed the oxalated N-Z amine described in the Material and Methods
section of this thesis. The medium proved to be desirable in that;
(i) the virulence antigens are produced in the medium, (ii) it
2+ 4)2304 fraction-
ation, (iii) any Ca ,which would inhibit the expression of V and

contains few constituents that precipitate upon (NH

w, was removed by the addition of sodiun oxalate, and (iv) the

N-Z amine is less expensive than the Bacto-casitone previously
employed for such studies. Although no studies were done comparing
the efficiency at which V and W were produced in oxalated N-Z amine
verses their production in Bacto-casitone, it is obvious that the
former promotes the expression of sufficient V and W to warrent its
use for purposes of antigen production and purification.

The observation that the V antigen is cell associated to some
extent has been reported by other investigators. Burrows (10) states
that 50% of the antigen was deposited with the cells upon centri-
fugation. Pirt, Thackeray, and Harris-Smith (36) were unable to
detect V extracellularly in studies on antigen production in contin-
uous culture. They observed that the antigen was associated with
the cell. On the other hand, Brubaker and Surgalla (6) were unable
to detect V and w in lysates of virulent cells. These authors
were able to demonstrate the presence of both antigens in whole
cultures.

The discrepancies in the results reperted may be due to

differences in the methods used to prepare the culture fluids.

The presence of 0.2 mM Ca2+ in the medium of Pirt, Thackeray,

and Harris-Smith (36) should not allow significant growth of the
Vwa+ cells (24). However, this concentration of Ca2+ may inhibit

V antigen expression to some extent. It is also possible that these
authors were unable to detect extracellular V due to the continuous
dilution of the culture. The expected transition of the cell
population to the Vwa- phenotype may also have adversly affected
their results.

Under the conditions specified in this thesis, more than 95%
of the V antigen was associated with the cells. The lysates of all
the Vwa+ strains tested in this work exhibited V activity. The anti*
gen, therefore, is probably associated with the cell.

The assay system used in this research is useful in that it is
reliable,, specific, semi-quantitative, and economical. The
employment of this assay system allowed various methods of protein
separation to be evaluated as to their applicability to the purifi-
cation of V. It was observed that the DEAE cellulose step was re-
sponsible for the loss of a large amount of the antigen. In addition
to the poor recovery of V, the method resulted in no net purifi-
cation. Lawton, Erdman, and Surgalla (33) experienced similar
difficulties in their attempts to purify V by batch elution of
culture fluids from DEAE cellulose. They report acheiving an
average of a 3.4 fold purification with recoveries ranging from
5 to 63%. The recovery of the antigen in their studies appeared
to be a function of the amount of protein initially adsorbed to the
resin. More recent experiments done in this laboratory have indi-
cated that the phenomenon is also true for the recovery of V from
cell extracts; increasing the amount of protein adsorbed to DEAE
cellulose decreased the amount of antigen recovered.

I also observed that a large amount of the antigen is lost
upon hydroxylapatite column chromatography. ‘Two common character-
istics of DEAE and hydroxylapatite column chromatography are that
(i) localized high concentrations of protein occur during adsorbtion
and (ii) a certain amount of physical stress probably occurs during
elution. Possibly, antigenic activity is lost due to one or both of
these phenomena. The former may promote aggregation of the antigen
whereas the latter may result in the removal of subunits or in the

I

42

alteration of the conformation of V. In any case, these methods
should not be used in the purification of the antigen unless the
conditions are altered in such a way that recoveries and purifi-
cations are optimized.

It has been demonstrated that antisera prepared against a
partially purified preparation of V passively protects mice against
virulent X. pestis (33). It is not likely that anti-V protects against
plague in the manner that antibody to somatic antigens protects against
infection by many gram negative organisms. In these cases the
reaction between antibody and antigen promotes subsequent phagocytosis
or even direct killing of the invading organism via the action of
the complement system (17). in that the V antigen is a cytoplasmic
component, it is difficult to envision these types of mechanisms
occuring with the V-anti-V system. Also, anti-V probably does not
act by neutralyzing an exotoxic activity of V because Vwa+ organisms
that lack one of the other virulence determinants do not appear to
be toxic to the host (except when injected in very high concentrations
where the toxicity is attributed to endotoxin or murein toxin) (3).
Toxicity studies with purified V should reveal the relationship
of this antigen to other bacterial toxins.

Although the precise role that V plays in the
pathogenesis of plague remains obscure, the protective capacity
of anti-V strongly suggests that its function it necessary to the
disease process. It will now be possible to use the purified
preparation of V that was obtained to stimulate the production
of monospecific anti-V in rabbits. This antiserum will be useful
in studying the pretective role of anti-V as well as in studying
the role of V in the pathogenesis of plague. The development
of a quantitative assay system,such as quantitative immunoelectro-
phoresis,would make studies on the expression of the antigen
possible. Such an assay system requires monospecific antiserum.

It would also be possible to employ this assay system which,

along with optimized or deleted DEAE and hydroxylapatite steps and
the other techniques described herein, should result in the
purification of enough homogeneous V to extensively characterize
the antigen both physically and chemically. The information

obtained from these proposed investigations should aid not only

43

in the understanding of the relationship of the V antigen to
virulence but also in the understanding of the role of host

defense mechanisms, such as phagocytosis, in the prevention

of disease.

LIST OF REFERENCES

10.

11.

LIST OF REFERENCES

Beesley, E. D., R. R. Brubaker, W. A. Janssen, and M. J.
Surgalla. 1967. Pesticins. III. Expression of coagu-
lase and mechanism of fibrinolysin. J. Bacteriol.
23:19-26.

Brubaker, R. R. 1970. Interversion of purine mononucleotides
in Pasteurella.pg§§i§. 1nf. Immun. 12446-454.

Brubaker, R. R. 1972. The genus Yersigia: biochemistry and
genetics of virulence, pp. 111-158. In: W. Arber et al.
(ed.), Current topics in microbiology and immunology,
vol. 57. Springer-Verlag, New York.

Brubaker, R. R., E. D. Beesley, and M. J. Surgalla. 1965.
Pasgeurella pestis: role of pesticin I and iron in exper-
imental plague. Science. 149:422-424.

Brubaker, R. R., and M. J. Surgalla. 1962. Genotypic altera-
tions associated with avirulence in streptomycin-resistant
Pasteurella pestis. J. Bacteriol. .83:615-624.

. ++
Brubaker, §+ R., and M. J. Surgalla. 1963. The effect of Ca
and Mg on lysis, growth, and production of virulence
antigens by Pasteurella_pestis. J. Infect. Dis. 113:13-25.

Brubaker, R. R., M. J. Surgalla, and E. D. Beesley. 1965.
Pesticinogeny and bacterial virulence. Zentr. Bakteriol.
Parasitenk, Abt. Orig. 126:302-312.

Burrows, T. W. 1956. An antigen determining virulence in
Pasteurella pestis. Nature. 112:426-427.

Burrows, T. W. 1955. The basis of virulence for mice of Pasteur-
ella pcstis, pp. 151-175. In: J. W. Howie and A. J. O'Hea
(eds.), Mechanisms of microbial pathogenicity. Fifth
Symposium of the Society for General Microbiology. Cambridge,
England: Cambridge University Press.

Burrows, T. W. 1960. Biochemical properties of virulent and
avirulent strains of bacteria: Salmonella typhosa and Pastegg-
ella pestis. Ann..N. Y. Acad. Sci. ‘§§:1125-ll35.

Burrows, T. W. 1960. V and W antigens in strains of Pasteurella
pseudotuberculosis. Brit. J. Exptl. Path. 31:38-44.

 

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

Burrows, T. W., and G. A. Bacon. 1956. The basis of virulence
in Pasteurella pestis: an antigen determining virulence.
Brit. J. Exptl. Path. élc481-493.

Burrows, T. W., and G. A. Bacon. 1954. The basis of virulence
in Pasteurella_pestis: attempts to induce mutation from
avirulence to virulence. Brit. J. Exptl. Path. _3§:129-133.

Burrows, T. W., and G. A. Bacon. 1954. The basis of virulence
in Pasteurella pestis: comparative behavior of virulent and
avirulent strains 13 vivo. Brit. J. Exptl. Path. 35:134-143.

 

Burrows, T. W., and G. A. Bacon. 1956. The basis of virulence
in Pasteurella pestis: the development of resistance to
phagocytosis.ig yitro. Brit. J. Exptl. Path. _3Z:286-299.

 

Burrows, T. W., and G. A. Bacon. 1958. The effects of the loss
of different virulence determinants on the virulence and
immunogenicity of strains of Pasteurella pesgis. Brit.

J. Exptl. Path. 39:278-291.

 

Davis, B. D., R. Dulbecco, H. N. EiSen, H. S. Ginsberg, and W. B.
Wood. 1970. Microbiology. Harper and Row, New York,
Evanston and London. pp. 802-807.

Davis, B. J. 1964. Disc electrophoresis II: methods and appli-
cations to human proteins. Ann. N. Y. Acad. Sci.
121:404-427.

Fiske, C. H., and Y. SubbaRow. 1925. The colorimetric deter-
mination of phosphorous. J. Biol. Chem. ‘§§:375-400.

Fukui, G. M., W. D. Lawton, D. A. Ham, W. A. Janssen, and M. J.
Surgalla. 1960. The effect of temperature on the synthesis
of virulence factors by Pasteurella pgstis. Ann. N. Y.
Acad. Sci. _§§:ll46-1151.

 

Fukui, G. M., J. E. Ogg, G. E. Wessman, and M. J. Surgalla.
1957. Studies of the relation of cultural conditions and
virulence of Pasteurella_pgstis. J. Bacteriol. .14:7l4-717.

Hall, P. J., G. C. H. Y pg, R. V. Little, and R. R. Brubaker.
1974. Effect of Ca on morphology and division of Yersinia
.pgstis. Infect. Immunity. 2:1105-1113.

Higuchi, K., and C. E. Carlin. 1958. Studies on the nutrition
and physiology of Pasteurella pgstis: li. a defined media
for the growth of gaggeurellg pestis. J. Bacteriol. .15:
409-413.

Higuchi, K., L. L. Kupferberg, and J. L. Smith. 1959. Studies
on the nutrition and physiology of Pasteurella pesgis: III.
effects of calcium ions on the growth of virulent and avir-

ulent strains of Pasteurella pestis. J. Bacteriol. 11;
317-321.

 

 

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

Higuchi, K., and J. L. Smith. 1961. Studies on the nutrition
and physiology of Pasteurella pestis: VI. a differential
plating medium for the estimation of mutation rate to aviru-
lence. J. Bacteriol. .81:605-608.

 

Hu, P. C., G. C. H. Yang, and R. R. Brubaker. 1972. Speci-
ficity, induction, and absorption of pesticin. J. Bacter-
iol. .lLZi212‘219'

Jackson, S., and T. W. Burrows. 1956. The pigmentation of
gagggurella pgggis on a defined media containing hemin.
Brit. J. Exptl. Path. ‘37:570-576.

Jackson, S., and T. W. Burrows. 1956. The virulence-enhancing
effect of iron on non-pigmented mutants of virulent strains
of Pasteurella pgsgig. Brit. J. Exptl. Path. .QZ:577-583.

Janssen, W. A., W. D. Lawton, G. M. Fukui, and M. J. Surgalla.
1963. The pathogenesis of plague. I. a study of the corre-
lation between virulence and relative phagocytosis of some
strains of Pasteggella‘peggis. J. Infect. Dis. .llé‘139'143-

Jawetz, E., J. L. Melnick, and E. A. Adelberg. 1972. Review
of Medical Microbiology, pp. 209-212. Lange Medical Publi-
cations, Los Altos, California.

Kugelmass, N. I. 1959. Biochemistry of blood in health and
disease. Ch. C. Thomas, Inc., Springfield, Illinois.

Lawton, W. D. 1960. Antigens associated with virulence in
.Pgsggurella pestis. Doctoral Dissertation. George Wash-
ington University, Washington D.C.

Lawton, W. D., R. L. Erdman, and M. J. Surgalla. 1963. Bio-
synthesis and purification of V and W antigen in Pasteurella
ngggis. J. Immunol. 21:179-184.

Little, R. V. 1974. Temperature-dependent regulation by Ca2+
of macromolecular synthesis in.XSE£lfllé Egsgis. Doctoral
Dissertation. Michigan State University, East Lansing,
Michigan.

Lowry, O. H., N. N. Rosebrough, A. L. Farr, and R. J. Randall.
1951. Protein measurement-with the Folin phenol reagent.
J. Biol. Chem. .193z265-275.

Pirt, S. J., E. J. Thackeray, and R. Harris-Smith. 1961. The
influence of environment on antigen production by Pasteurellg
pesgis studied by means of the continuous flow culture
technique. J. Gen. Microbiol. .252119rl30.

 

Poland, J. D. 1972. Plague, pp. 1141-1148. In: D. D. Hoeprich
(ed.), Infectious Diseases. Harper and Row, Hagerstown, Md.

Sokhey, S. S., and M. K. Habhu. 1943. Optimum and limiting
47

temperatures for the growth of the plague bacillus in broth.
J. Bacteriol. 46:25-32.

39. Surgalla, M. J., and E. D. Beesley. 1969. Congo red agar
plating medium for detecting pigmentation in Bagggugella
Bestis. Appl. Microbiol. .18z834-837.

40. Warburg, 0., and Chrkstian. 1942. Isolierung und kristallisation
des garungsfermcnts unolase. Biochem. Ztschr. _319:384.

41. Yang, G. C. H., and R. R. Brubaker. 1971. Effect of Ca2+ on

the synthesis of deoxyribonucleic acid in virulent and
avirulent-Igggigia. Infect. Immunity. 3:59-65.

48

HICHIGQN STQTE UNIV. LIBRRRIES

"I“ fill ”Iii“ I) lllll IlHl NH

31293105753663