STUD-45$ ON THE WWI:

' RELATDNSHIPS FERMENF‘AWQN
REACTQONS AND DiSSOClATiON PATT’RmS
OF PASTEURELLA HEMOL‘YTtCA ESOLAFED A
FROM CHiCKENS

Thesis for #9109 Dogma 'af 4% S
MfCHQGAN SFA‘E‘E CG £335

Mary Louise Meyer
1933

15:38

This is to certify that the

thesis entitled

Studies on the Serological Relationships, Fermen-
tation Reactions, and Dissociation Patterns of

Strains of Pasteureila Hemolytica isolated from
chickens.

presented bl]

Mary Louise Mayer

has been accepted towards fulfillment

of the requirements for

Master of Science degfifi n: Bacteriology

./’ 2 (I .4
/7/l.{,/j;233é2;c22{

é Major prdfessor

 

Date firfl all: 1953

STUDIES ON THE SEROLOGICAL RELATIONSHIPS, FERMENTATION
REACTIONS, AND DISSOCIATION PATTERNS OF PASTEURELLA

HEMOLYTICA ISOLATED FROM CHICKENS

BY

MARY LOUISE MAYER

A THESIS
Submitted to the School of Graduate Studies of Michigan
State College of Agriculture and Applied Science

in partial fulfillment of the requirements
for the degree of

M ASTER OF SCIENCE

Department of Bacteriology and Public Healtlrt ” l‘x,‘

1953 r

TH E$1§

ACKNOWLEDGMENTS

The author wishes to express her gratitude to Dr. H. J.
Stafseth for his unfailing interest and encouragement in this inves—
tigation.

She also extends thanks to Dr. I. F. Huddleson for his sug-
gestions on the dissociation studies, Dr. M. Lois Calhoun for the
use of her camera, Philip Coleman, College Photographer, for taking
the photomicrographs of bacterial cells, and Dr. R. C. Belding for

many conside rations .

33432:

\ I u I l l [Ill Ill l. I ll‘. I I. II [III I ll [I I 'Illllllulll Ill! ll

TABLE OF CONTENTS .

- Page
LIST OF TABLES . . . . ......................... iv
INTRODUCTION AND HISTORICAL REVIEW ........... ' 1
EXPERIMENTAL STUDIES ....................... 12
Preliminary-«Selection of Media ' .............. . . . 12
Serological Relationships . ...................... 19
Fermentation Reactions ................... . . . . . 23
Dissociation Patterns ....... . ................. 27
Colonial Changes Occurring During Dissociation . . . . 3O
Sensitivity to Sodium Chloride . ................ 45
Growth characteristics in broth .............. 46
Stability'of organisms in saline suspensions ..... 46
Sensitivity in the presence of acid ............ . 47
Agglutination by acriflavine ................ ‘. 49 '
Presence of Capsules . . . . ................... 49
Variation in Biochemical ’Activity ............... 50
Variation in Morphology of Cells ............... 52
DISCUSSION AND SUMMARY ...................... 64
BIBLIOGRAPHY ............................... 68

iii

LIST OF TABLES

TABLE Page

I. Comparison of Fermentation Reactions Obtained
by Various Workers ........... . ........... 8

II. Studies of Growth on Various Media ............ 16
III. Agglutination Reactions of Several Strains

of Pasteurella hemolytica Against Antisera
Prepared from Strains 309 and 772M . . . ........ 22

 

IV. Fermentation Reactions of Twenty-Two
Strains of Pasteurella hemolytica Isolated

 

from Chickens , ............ . ...... . ....... 25

V. Comparison of Phases: Culture 309 ........... 36
VI. Comparison of Phases: Culture '684 ........... 39
VII. Comparison of Phases: Cultures 782 and 868 . . . . . 42
VIII. Comparison of Phases: Cultures 710 and 328 . . . . . 44

IX. Agglutination of Saline Suspensions Prepared
from Stable Phases Under Varying Conditions ..... 48

X. Fermentation Reactions of Stable Phases Studied
in Those Carbohydrates Showing Variability . ..... 51

XI. Opacity of Growth on Agar Slants and Description
of Cellular Morphology ..................... 53

iv

INTRODUCTION AND HISTORICAL REVIEW

While examining routine blood plates inoculated from diseased
birds submitted to the poultry diagnostic laboratory at Michigan State
College, conspicuous, hemolytic colonies were noted with significant
frequency. These colonies were 1.5 to 3 mm in diameter, circular
with an entire margin, convex, translucent, smooth and glistening,
butyrous in consistency and cream colored. I The average colony
appeared much like a hemolytic streptococcus colony although many
colonies were larger than a typical streptococcus colony. The cells
were gram negative, short cocco-bacilli with marked tendency toward
pleomorphism.

Attempts to identify the organism biochemically led to the

conclusion that it was Pasteurella hemolytica, although no reference

 

could be found describing its isolation from poultry or its signifi—
cance in poultry.

The earliest reference in the literature to a hemolytic pas-
teurella was found in the work of Jones (1921). At that time it was
customary to designate members of the hemorrhagic septicemia group
of pasteurellae by host affinity. It was becoming evident, however,

that host relationship was not the criterion for differentiation of

2
species; an organism isolated from one species was often pathogenic
for another. Jones separated the organisms within the species then
designated ”bovisepticus" into three groups on the basis of biochem-
ical and immunological differences. Prior to this time, fermentation
tests to characterize the pasteurellae, or any organism, were not
commonly employed. Immunological studies had been confined to
attempts to protect laboratory animals with immune sera against
small doses of virulent culture. For his studies Jones used strains
of pasteurellae which he had isolated from (1) cows during an out-
break of pneumonia in a dairy herd in 1920, (2) sporadic cases in
calves of the same herd following the main outbreak, and (3) strains,
isolated by Theobald Smith in 1917—1920, associated with a previous
outbreak of pneumonia in calves of the same herd. All the strains
isolated by Jones during the enzootic of 1920, and from the sporadic
cases following, were hemolytic and indol negative. They were iso-
lated in pure culture. Primary infection in the animals examined by

Smith was attributed to Bacillus actinoides. None of Smith's pas-

 

teurella cultures’was hemolytic and all were indol positive.

Group I was composed of the eight strains isolated by Jones.
These strains produced acid from dextrose, maltose, mannite, and
sucrose giving a pH of 6.1 to! 6.5 after incubation at 38 C for seven

days, and acid from lactose giving a pH of 6.4 to 6.8 on primary

I I l l I A itjll l l

isolation. After several transfers on artificial media, lactose was
fermented to the same extent as the other sugars. Indol was not
produced. The organisms were insoluble in ox or guinea pig bile.
Surface colonies were round, flattened, translucent, 3 to 5 mm in
diameter in 48 hours, and hemolytic. All were encapsulated. Viru—
lence was low for rabbits, mice, and cows; on subcutaneous injection
only a slight local reaction resulted. When introduced into the lungs
of cows, however, a severe diffuse pneumonia developed.

Group II was composed of the six strains isolated by Smith.
They produced acid from dextrose and sucrose, but not from lactose,
maltose, and mannite. The organisms produced indol. Broth cul-
tures were soluble in ox or guinea pig bile within ten minutes to
one hour. They were nonhemolyti’c, encapsulated, and relatively
nonvirulent.

Group III comprised two strains: one strain was isolated
by Smith at the beginning of the 1917 outbreak, and the other strain
was isolated in another state in 1913. These were like Group II
organisms except that they fermented mannite, were insoluble in
bile, and were virulent for rabbits. Injection of 0.25 ml of a 24-
hour broth culture killed rabbits in three to four days. Organisms
were present in the blood 24 hours before death. The presence of

capsules was variable in this group.

Immunological studies revealed groupings identical with the
cultural groupings. Agglutinins for Groups I and II were produced
with great difficulty to a titer of 1-200. Agglutinins for Group III
were easily developed to a titer of 1-500. No cross-agglutination
occurred between groups. Antigens were prepared by suspending
growth from 24 or 48 hour slants in 0.85 per cent sodium chloride
containing 0.25 per cent phenol. Tests were incubated at 38 C for
18 hours.

Jones concluded that Group I organisms differed enough to be
considered a separate species, and proposed the name PaSteurella

bovisepticus be used to identify them since infection occurred chiefly

 

in adult cows. Group II and III organisms he considered varieties

of the same species, and proposed_they be called Pasteurella vituli-

 

septicus. Although other serological types have since been identified
within these groups, this classification of Jones still serves to sep-
arate the members of the hemorrhagic septicemia pasteurellae into
their principal types. His work has been verified and extended to
include organisms isolated from other species of animals.

In 1923 Spray isolated three groups of pasteurellae with a
number of variants in each from an edematous type of pneumonia
in spring lambs as found under slaughterhouse conditions. His

Easteurella ovisepticum, S-8 type, belonged among the classical

 

pasteurellae and would be included in Jones' Group III based on bio-
chemical reactions. His second and third groups, isolated in pure

culture, were separated from the true Pasteurella ovisepticum by

 

their ability to hemolyze blood .and ferment maltose and glycerol.

The two types, S-3 and 5-14, produced a distinct ring of partial
hemolysis around the periphery of a rather flat, thin, translucent
colony. Nitrates were reduced to nitrites, and indol was not pro-
duced. Action in litmus milk was neutral with reduction after pro-
longed incubation. Spray found both types highly pathogenic for mice
and guinea pigs, but less so for rabbits. He considered them capable
of causing a septicemia.

He separated these two hemolytic types from each other by
three tests: (1) S-3 fermented lactose after seven or eight days.
8-14 showed no action on lactose. (2) In 5 per cent glycerol-serum
broth, S-3 fermented glycerol readily and produced a heavy, flocculent
sediment which was not diffused by vigorous shaking. S-l4 fermented
glycerol after fourteen days and produced a moderate sediment which
was not flocculent and diffused readily. The serum broth was made
by adding an equal part of serum water (one part of serum to three
parts of distilled water) to standard veal infusion broth, adjusted to
pH 7.4, and containing 5 per cent glycerol and 1 per cent Andrade

indicator. (3) Direct agglutination and absorption tests identified

two distinct groups coinciding with those indicated by the lactose,
glycerol tests. No attempt was made to correlate these results with
the findings of Jones

In England, Edington (1930) described an outbreak of pneumonia
in seven cows and two calves caused by a hemolytic pasteurella.
He found by agglutination and absorption tests that these organisms
belong to JoneshGroup I. He described the organism in more de-
tail than did Jones and Spray. He observed that the organism showed
a decided preference for moisture. When surface growth from a
plate was collected in a heap, it had a yellowish tint with a slightly
waxy consistency. In broth it grew readily producing first a turbidity
and later a sediment. There was a tendency to pellicle formation.
It grew on MacConkey's bile salt agar. He reported that in severe,
rapidly fatal cases, as seen in the calves, the bacillus was cultured
from the blood stream. One of the strains failed to hemolyze blood
although all the other characteristics were the same.

In 1932, Newsom and Cross, working with strains of pasteurella,
both hemolytic and nonhemolytic, from sheep and cattle affected with
pneumonia, identified four serological groups. These authors pro-

posed the name Pasteurella hemcgytica for those strains showing

 

hemolysis on rabbit blood agar, negative indol production, avirulence

for rabbits, and acid from unheated dextrin, inositol, and maltose.

Bergey later adopted this name. Serologically, most of the hemolytic
strains fell into one group which was identical with Jones' Group I.
Two strains comprised the second group, but these were easily sep—
arated from the predominant group biochemically by their fermenta-
tion of mannose and inactivity on raffinose. Newsom and Cross
tried to duplicate Spray's separation of the hemolytic strains by
means of glycerol fermentation, but were unsuccessful since they
found fermentation slow and uncertain.

Six strains of E. hemolLtica were included by Rosenbusch

 

and Merchant (1939) in a study of the hemorrhagic septicemia group.
The biochemical results on the hemolytic strains, designated by them

as Group IV, were identical with those of Newsom and Cross. They

did not conduct serological tests on this group.

Routine diagnostic tests conducted on _13. hemolytica cultures
at Michigan State College revealed a disagreement with previous
1
workers concerning the sugars fermented. These differences are

shown in Table I. Jones reported dextrose, lacto'se, maltose, mannite,

l
The sugars used routinely for identification of organisms

are dextrose, lactose, maltose, mannite, and sucrose. These in con-
junction with colony morphology, Gram stain, motility, indol produc-
tion, and hydrogen sulfide production (Kligler Iron Agar) are usually
considered sufficient to identify the typical pathogens encountered in
P0111try diseases.

COMPARISON OF FERMENTATION REACTIONS OBTAINED

TABLE I

BY VARIOUS WORKERS

 

 

Jones

Spray

Eding- . News om Mer -

ton

Cross

chant:1

Be rgey ' 5
Manual

M.S.C.

 

Dextro s e
Lacto se
Malto se
Mannitol
Sucrose
Dextrin
Fructose
Galacto s e
Glyce rol
Ino sitol
Raff ino se
So rbitol
Xylos e
Arab ino se
Dulc itol
Inulin
Manno se
Rhamno s e
Salic in
Adonitol
Starch
Indol
I—Ie molys i s

>>>g>

+I

A
V
AI
A
A

A
A
AI

A
A

+1

<1<.'I>i>' >i>1>l>t>

I<>>I

<

+2

.F’

u<stn>><g<<>g>>g<

-I
+I

nn-<s<<>>>>>>>>>>>g>

>>1>i><1>i>i>i>>i><i>

+1

<<><<><><>><>>

I

<-

t<l

+

 

 

A x‘acid produced; V a variable acid production; + a positive;

negative, or no acid; I a considered an identifying characteristic.

Taken from Veterinary Bacteriology and Virology.

Some strains negative .

and sucrose consistently fermented and considered the slight and
sometimes slow fermentation of lactose characteristic. Newsom and
Cross reported lactose fermentation variable and considered maltose
fermentation an identifying characteristic. In the poultry diagnostic
laboratory, lactose fermentation, although slight, was consistently
positive, while maltose fermentation was variable.

Because each worker used a different method for demonstrating
acid production, a comparison of results was difficult. Jones recorded
all reactions in terms of final pH, the lowest of which was 6.1. The
pH produced in lactose broth by newly isolated strains was between
6.4 and 6.8. The method Jones used to determine pH was not given.
By adding brom phenol blue, pH range 6.3 to 7.6, to fermented sugar
broths at Michigan State College, the pH level given by Jones for
lactose broth was duplicated. The pH in the other common sugar
broths was below 6.3. The visible range of color transformation for
brom cresol purple, which was used by Newsom and Cross, is 5.0
to 6.4. The yellow color, which is characteristic in the acid range,
is barely visible through the purple at pH 5.8 and begins to predom-
inate at pH 5.4. Considering these facts, the variability of lactose
fermentation as recorded by Newsom and Cross was understandable.
Andrade indicator, used routinely in the diagnostic laboratory, turns

a definite pink at pH 6.8, and was used throughout these studies.

10
Variability of maltose fermentation, however, was not as
easily explained since it was recorded while using a more sensitive
indicator, and noted only in the poultry diagnostic laboratory. The

possible existence of two serological strains of 1:. hemgytica sep-

 

arable by the fermentation of maltose was considered since Spray,
and Newsom and Cross had apparently correlated serological proper-
ties with biochemical activity on carbohydrates. The thesis problem
as originally approved, consisted of an attempt to demonstrate the

presence of more than one serological type of 1:. hemolytica isolated

 

from chickens, and to correlate such results with the fermentation
of maltose and any other fermentable substance if such correlation
proved possible.

Soon after the serological phase of, the problem had been well
established, a laboratory mishap resulted in the dissociation of the
organisms under study. The distilled water supply became contam-
inated with tap water which was introduced through a leak in the
still. Before the contamination was discovered, this water was used
to prepare media on which the cultures for this project were trans-
ferred. Trace elements thus accidentally introduced into the media
caused dissociation and rendered many strains inagglutinable, includ-
ing those used in antisera production. By the time the source of the

toxicity had been identified, most of the cultures transferred to

11
uncontaminated media prior to the mishap were no longer viable
since cultures must be transferred once a week to insure survival.
A continuation of the problem was not possible until a new collec-
tion of cultures could be made. It was decided at this time, in
conference with the major professor, that a study of the dissociation

pattern of E. hemolytica was essential to an understanding of the

 

organism in view of its extreme sensitivity. One of the dissociated
cultures and seventeen strains isolated following the mishap were

examined for this purpose.

EXPERIMEN TAL S TUDIES

Preliminary- —Selection of Media

 

From the first it was apparent that the organism under study
was a fastidious one and that a medium for propagation must be se-
lected with care. Berkman (1943) included one strain of E. hemolytica
in a study of the accessory growth factor requirements of the genus

Pasteurella and reported good growth of this species in his basal

 

medium composed of hydrolyzed gelatin, inorganic salts, dextrose,

and amino acids without the addition of accessory growth factors.
Procedure

Nine agar media without the addition of any growth factor
were chosen for comparison. (1) Tryptose agar (Difco)1 containing
2.0 per cent tryptose, 0.1 per cent dextrose, and 0.5 per cent sodium
chloride, was chosen as a standard because it was known to be satis-
factory for growing many fastidious organisms. Prepared according
to directions, the resulting pH of the medium is 6.9. (2) Tryptose

agar adjusted with sodium hydroxide to pH 7.2 was used to see what

1
Difco Laboratories Incorporated, Detroit 1, Michigan.

13

effect raising the pH might have. (3) Nutrient agar (Difco) contain-
ing 0.3 per cent beef extract and 0.5 per cent peptone, was chosen
to represent a less nutritious medium. Prepared according to direc-
tions, the pH is 6.8. (4) Nutrient agar was also adjusted to pH 7.2.
Darby's lactose motility medium, a modification of the Difco product,
containing 1.8 per cent motility medium (Difco), 0.2 per cent beef
extract, 1.0 per cent lactose and 1.0 per cent Andrade indicator, had
been used routinely in the poultry diagnostic laboratory as a screen-
ing test to eliminate slow lactose fermenters and to classify lactose
negative bacteria by motility. Since it had been observed that 1:.

hemo_lytica grew well in this medium, two modifications of it were

 

included in the survey: (5) motility medium prepared according to
directions, containing 1.0 per cent tryptose and 0.5 per cent sodium
chloride, plus enough agar to make 1.5 per cent; and (6) motility
medium plus 0.2 per cent beef extract as in Darby's modification and
agar to 1.5 per cent. Lactose was omitted since a production of

acid in a medium for propagation was considered undesirable. (7)
Brain heart infusion agar (Difco) containing infusion from 200 g of
calf brains and 250 g of beef heart per liter, 1.0 per cent proteose
peptone, 0.2 per cent dextrose, 0.5 per cent sodium chloride, and 0.25
per cent disodium hydrogen phosphate plus 1.5 per cent agar, was

chosen because it is a much richer medium than tryptose. (8) Scott's

l4
salts agar was included because Eveleth had used it with success

for the maintenance of viability and pathogenicity of Pasteurella

 

multocida. It was hoped that this medium would prove equally satis-

factory for 1:. hemolytica. It contained 1.0 per cent peptone, 2.0

 

per cent agar, 0.2 per cent potassium citrate, 0.05 per cent potassium
bicarbonate, 0.06 per cent ammonium dihydrogen phosphate, 0.25 per
cent ferric ammonium citrate, and 0.05 per cent dextrose. (9)
Chicken infusion agar containing the infusion from 500 g of ground
chicken meat and bones per liter, plus 0.5 per cent sodium chloride,
1.0 per cent peptone, and 2.0 per cent agar, was tried to see if
chicken proteins would prove satisfactory.

Criteria for judging the suitability of the media were rate of
growth, amount and appearance of the growth present in 24 and 48
hours, sensitivity of the cells to physiological salt solution, morphology
of the cells as determined by staining, and the viability of cultures
after eight days, which was the longest period viability had been
maintained at that time.

Duplicate slants of each medium were heavily streaked with
24-hour cultures of 772M and 309 and incubated at 37 C. Slants
were examined at 4, 6, 8, and 24 hours for definite growth. At 24
hours, safranin stains were made from each medium showing a satis-

factory amount of growth. Physiological saline was added to the

15
remaining growth and the turbidity adjusted to tube 2 MacFarland
nephelometer. Ease of suspension and evidence of agglutination were
noted. Suspensions were held and observed for four days. The
duplicate culture in each case was reincubated another 24 hours and
appearance of growth noted. The culture tubes were then sealed
with parafilm and held at room temperature for six days. Trans-

fers were then made to the same medium in each case.
Results

The significant information obtained from this study is re-
corded in Table II. Nutrient agar at both pH 6.8 and 7.2 and Scott's
salts agar were discarded because of uniformly poor growth. Both
of the motility medium modifications were eliminated as neither
strain grew well on the basic motility medium, and strain 309 did
not grow well-on motility medium when beef extract was added.
Although growth of strain 772M was good when beef extract was
added, cells showed increased sensitivity to physiological saline.
Brain heart infusion agar was likewise eliminated because organisms
grown on it showed marked salt sensitivity. Chicken infusion agar
supported luxuriant growth as did tryp'tose agar at both pH 6.9 and
7.2. Stained‘organisms from chicken infusion agar were uniform

in size and shape at 24 hours. The size and shape of cells from

16

TABLE II

STUDIES OF GROWTH ON VARIOUS MEDIA

 

 

 

 

 

 

 

First Ap-
A h
pearance mount of Growt Growth on
A of Growth T ansfer
gal Within 24 Hours 48 Hours r
Media 8 Hours
7 2M 30
772M 309 772M 309 7 9
772M 309
1. Tryptose
(pH 6.9) ....... 6* 6 4+ 4+ 4+ 4+ + +
2. Tryptose
(pH 7.2) ...... 6 6 4+ 4+ 4+ 4+' + +
3. Nutrient
(pH 6.8) ...... 8 - + :I: + :i: ..
4. Nutrient '
(pH 7.2) ...... 8 - + i + :h +
5. Motility ....... 8 - 2+ + 2+ + +
6. Motility plus
beef extract . . . . 8 8 4+ + 4+ + ..
7. Brain-heart
infusion ....... 8 6 + 4+ 4+ 4+ - ..
8. Scotts ........ - - :1: — + _
9. Chicken infusion . - 6 4+ 4+ 4+ 4+ .- ..

 

 

* 3 hours; + a: growth visible; 2+
growth.

growth moderate; 4+ : maximum

17

TABLE II (Continued)

 

 

 

 

Saline Morphology of Cells in 24 Hours
Suspen-
sions 772M 309
M tl d' ' t t d—- -
HS OS ,Y 15111 egra e . re Coccobacilli, few bipolar
maining cells cocc01d.
HS Same as medium 1 Slender bac1111, few fila-
ments
Mostly coccobacilli——few
long filaments
Coccobacilli
Irregular in staining-- "Shadow” and involution
"shadow cells” forms
PA Slender bacilli pale Coccobacilli, few involution
staining forms
A Coccoid forms in clumps Size irregular
HS Slender bacilli Coccobacilli

HS a homogeneous, stable suspension; PA a partially agglutinated,
settled out in four days; A 8 strong agglutination.

strain 309 when grown on tryptose agar were uniform at 24 hours,
but those of strain 772M showed mostly disintegrated cells. Cells
from all media stained at 48 hours showed the same disintegration,
often with poorly stained "shadow cells" and involution forms ap-
pearing before complete disintegration. This process suggests the
presence of a very active autolytic enzyme. Viability was not main-
tained on chicken infusion agar, however. Subcultures made on the
eighth day did not grow. Tryptose agar was the only medium which
supported satisfactory growth of both strains. It also maintained
viability for the longest period of time and was selected for propa-

gation of E. hemolytica.

 

When the studies on dissociation were started, it was discov—
ered that although the newly isolated strains appeared to grow satis-
factorily on tryptose agar slants, cell suspensions plated for isolated
colony study on tryptose agar plates did not produce the expected
number of colonies as previously determined by plating dilutions of
strains of 309 and 772M. The same cell dilution plated on tryptose
blood agar base without blood yielded the expected number of colo-
nies. Tryptose blood agar base contains 1.0 per cent tryptose, 0.3
per cent beef extract, and 0.5 per cent sodium chloride, and is sim-
ilar to trial medium number.6. All the strains used for dissociation

Studies produced a larger number of colonies on tryptose blood agar

19
base than on tryptose agar except strains 309 and 328, which grew
more satisfactorily on tryptose agar. The medium which best sup-

ported the growth of the strain being examined was used throughout

the dis sociation studie s .

Serological Relationshigs

 

An attempt was, made to separate the strains of 1:. hemolytica
serologically using a maltose fermenting strain and a maltose nega—
tive strain. Strain 309 was selected to represent the maltose nega-
tive strains. At the beginning of this phase of the work, strain 309
had been freshly isolated in almost pure culture from a bird show-
ing hyperemia and patchy consolidation of the left lung as the only
Pathological change. Ten per cent of the flock of 3,500 were visibly
ill. A 6 per cent loss was reported in a three-day period. Strain
772M was selected to represent the maltose positive strains. The
1elfter M after any culture number indicates that the culture fer-
mented maltose. 772M was isolated six months before strain 309
from a bird affected with the visceral form of the avian leucosis
C—'<3t1'nplex. In addition to the neoplastic changes in the spleen and
ki(iney, the liver of this bird was ochre in color with necrotic foci,
one kidney was cystic, the ova discolored, and an abscess was noted

in the cecum. A moderate growth of 1:. hemolytica was obtained.

20
No other bacterium of significance was isolated. This strain was
chosen chiefly because a maltose negative strain was also isolated
from the same bird.

Both 309 and 772M were stable in physiological saline. In
addition to the maltose variation, these two strains also differed in
opacity of growth. The growth of strain 309 was almost transparent
on a tryptose slant like that of a streptococcus, while that of 772M
was more translucent. The majority of strains isolated were trans-
lucent and like 772M in appearance. Strains 772, maltose negative,
1138 and 1304 were less opaque than 772M. Strains 767, 818M, and

955 were more opaque and mucoid.
Procedure

Each of two adult rabbits per strain was given an intraperi—
toneal injection of 0.5 m1 of a 24-hour live, saline suspension of
Organisms. .A turbidity corresponding to tube 1 MacFarland nephel-
0“letter was used. A second intraperitoneal injection of 1.0 ml was
given on the third day. All injections were made at three-day inter—
Vills. The third, fourth, and fifth inocula were 0.5 ml introduced
intravenously. The sixth, seventh, and eighth inocula were 1.0 m1

giVen intravenously. The ninth and all subsequent injections were

21
1.0 ml of a live, saline suspension with turbidity corresponding to
tube 2 MacFarland nephelometer introduced intravenously.

One month after the first inoculation, blood was drawn from
the rabbits and agglutination tests conducted. The antigen used was
a 0.5 per cent phenolized saline suspension with a turbidity correspond-
ing to tube 1 MacFarland nephelometer. Tests were incubated for
18 to 20 hours at 37 C. Serum from all four rabbits showed a weak
agglutination with homologous serum through a dilution of 1—40. In-
jections were continued every three days for two additional months.
The titers of sera pooled from the two rabbits receiving the same
inoculum were 20,000 and 10,000 respectively for 772M and 309.

Agglutination tests were conducted using five strains of B.
hemolytica against each antiserum. Before adsorptions and further
agglutinations were completed, the twenty-two cultures being studied

We re accidentally dissociated as previously described.
Results

Table III shows the results of the agglutinations that were
Completed. Although some of the tubes showed clearing with com~
plete precipitation of antigen, the clumps were small when the reac-
ti011$ were shaken. When 772M was used as antigen against antiserum

309, agglutination occurred through a serum dilution of 1-50 in contrast

22
TABLE III
AGGLUTINATION REACTIONS OF SEVERAL STRAINS OF

PASTEURELLA HEMOLYTICA AGAINST ANTISERA
PREPARED FROM STRAINS 309 AND 772M

 

 

5...... A3532?"
Antigen Se rum Ab- Titer Complete Con-
sorbed trols

With Through

Dilution
772M 772M 1,600 1-200 -
309 772M 0 - -
309 309 3,200 1-800 -
772M 309 50 Inc. -
772 772M 0 - -
772. 309 6,400 1-400 —
8 18M 772M 0 - —
818M 309 400 1-100 —
818 772M 100 Inc. +
818 309 1,600 Inc. +
784 772M 0 - —
784 309 0 - -
767 772M 3,200 Inc. -
767 309 1,600 1-800 -
309 309 772M 3,200 1-800 —

 

 

 

 

he. a incomplete clearing of supernatant fluid with definite but in-
complete agglutination; - 8 negative agglutination; + a positive ag-
glutination.

Complete clearing of supernatant fluid with complete ag-
glutination.

Both antigen and negative serum controls used.

23

to the l-3,200 recorded when homologous antigen was used. No ag—
glutination occurred when 309 was used against 772M antiserum,
however. It is interesting that antigen 772 was not agglutinated by
antiserum 772M although both cultures were present in the same
bird, but was agglutinated by antiserum 309. Neither 772 nor 309
fermented maltose. Culture 818 apparently was rough since the
antigen control showed slight agglutination following incubation. Cul-
ture 767 was agglutinated by both antisera although the reactions
we re more complete with antiserum 309. Culture 818M agglutinated

with antiserum 309 only.

Fe rmentation Re ac tions

 

Procedure

The base medium used in studying the fermentative ability of

E- I'Lemolytica consisted of 1.0 per cent tryptose, 0.5 per cent sodium

 

Chloride, 1.0 per cent Andrade indicator, and 0.2 per cent agar.
Growth was more uniform in this semisolid medium than in the
u$41211 broth medium. This base was autoclaved at 15 pounds steam
preSsure for 20 minutes. One per cent of the desired sugar was
added to the base, and the medium was resterilized at 12 pounds

Pressure for 20 minutes.

24
The twenty-two strains used in this 'study were not inoculated
into all the sugars immediately upon isolation, but were carried on
tryptose agar until a number of strains were available. Inoculations
were then made at one time. Each sugar was heavily seeded with
a loopful of 24-hour tryptose broth culture and observed daily for
two weeks. The cultures were transferred once a week for several
months and reinoculated into sugars for comparison with the first

set of re sults.
Results

Acid was produced from dextrose, maltose, mannitol, sucrose,
galactose, xylose, mannose, and trehalose in 24 hours. Lactose was
positive in 24 or 48 hours, but the acidity produced was slight. A
small amountof acid was produced from sorbitol in 48 hours; this
reaction faded in four days and appeared negative thereafter. Glyc-
erol and raffinose were fermented in three to six days. Production
of acid from dextrin and starch took five'ito seven days in the few
instances that these substances were fermented. Table IV shows the
substances fermented.

It was observed that two of the maltose—fermenting strains
lost the ability to ferment several of the sugars after several months.

Strain 772M failed to ferment sorbitol on the second study. This was

25
TABLE IV

FERMENTATION REACTIONS OF TWENTY-TWO STRAINS OF
PASTEURELLA HEMOLYTICA ISOLATED FROM CHICKENS

 

 

Dex- Lac— Malt— Man— Su— Glyc- Raf— Man— Galac—
trose tose ose nitol crose erol finose nose tose

 

309
767
772
784
818
955
1138
1304
75
698
540
601
579
556
541
94

76
699
772M
818M
540M .
541M
3377M* A
3382M* A

>>>

>>>>>>>>>>>>>>>>>>>>>>

>>>>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>>>>>>
>>wmm>mmmm>mmmmmmm>>mm>>
mmmmmmmmmmmmmmmmmmmmmmmm
>>>>>>>>>>>>>>>>>>>>>>>>

>>>>>>>>>>>>>>>>>>>>'

'>>>>>>>>>>>>>>>>>>>>>>>>

 

 

 

A " acid produced within two days; 5 3 acid produced in three to

seven days; - 2 no acid; ' = lost ability to ferment after several

transfers; * :- porcine strains.

TAB LE IV (Continue d)

26

 

 

 

Sor- Xy- Tre— Dex- Arab- Dul- In— Rham- Sal- Adon-
. . Starch . . . . . .
bItol 10 s e halo S e trIn Ino s e c Itol 1.11111 nos e 1c In Itol

A .. _ - - - .. .. _. - -
A A A - - - - - - - -
A A S - - - - - - - -
A A A - - — - - - - -
A A A - - - - - - — -
A A S - - - - - - - -
A A A - - - - - - - -
A A A ~ - - - - - - -
A A A - — - - - - - —
A A A - - - - - - - —
A A A - - - - - - - -
A A A - - - - - - — -
A A A - - - - - — - -
A — S - - - - - - - -

" A A A - - - - - - - -
A A A - - - - - - - -
A A A - - — - - - - -
A A. A - - - - - - — -
A' A S S S - - - - - -
- A' A .. _. ._ .. _ .. _ ..
- A A S S - - — - - -
- A A S S S «- - - - -
A A A S S S - - — - -
A A A S S S - - - - —

 

 

27
of interest since all the other maltose positive strains failed to fer-
ment sorbitol at any time. It was noted during the subsequent work
on dissociation that all newly isolated strains fermented sorbitol
slightly. Strain 818M failed to ferment raffinose, galactose, xylose,
and mannose after a period of artificial cultivation. Occasionally a
strain failed to attack a substance utilized by the majority of strains,
but no pattern was observed. Strain 309 was of interest because it
fermented fewer sugars than any other strain.

Cultures. 3377M and 3382M were obtained from Dr. John
Collier at Iowa State College. They were isolated from lung lesions
in pigs and are included for comparison with the strains isolated
from chickens. These two strains were not used in any other studies
for this thesis because they agglutinated rapidly in 1—500 acriflavin.

The strains used by Jones and others were not available.

Dis sociation Patterns

 

The term ”dissociation” is used to denote one of several
types of variation which occur in bacteria. It refers specifically to
changes in colonial morphology from smooth (S) or mucoid (M) to
intermediate (1) or rough (R) and other altered characteristics which
may be associated or appear independently. The typical smooth (S)

colony is round with an entire margin, soft in texture and glistening

28

in appearance. It is composed of cells which Show little variation
in size or shape, make homogeneous suspensions in saline and agglu—
tinate readily in specific serum. The rough (R) colony is larger
and flatter with a rough or corrugated surface, not soft and often
dull in appearance. It is composed of cells which are longer, pleo-
morphic and agglutinate spontaneously in saline. Mucoid (M) types
take their name from their mucoid texture which is dependent on a
well-developed capsule. It may represent the virulent phase of the
organism. Intermediate (1) types are usually less stable than S, R,
or M, and represent stages between these phases.

Braun (1947) states:

. the phenomenon of dissociation appears to resolve
itself into merely another manifestation of causally related
mechanisms of discontinuous bacterial variation which involve
the spontaneous occurrence of undirected variants (mutants) and
their subsequent establishment within a population under the con-—
trol of inherent and environmental factors governing population
dynamics.

Braun further states that the bulk of the evidence available supports
the mutation concept of dissociation. The occurrence of mutants in
bacteria has been established by Zelle and others as comparable in
rate to those of higher organisms. The inherent factors involved in
the establishment of a mutant are the genetic ratio, the rate with

which stable cells split off from unstable cells, and the relative

growth rate of the two types in the environment being used. When

29
the environment is less than optimum, a high differential between the
total number of cells present in the medium and the number cf viable
cells leads to population pressure which permits the selection and
establishment of any mutant with a higher growth rate or viability.

It has been proved with Brucella abortus that an ideal medium pro-

 

duces a low differential with little or no population pressure result-
ing in the reduction or absence of dissociation. In some cases the
addition of 2.0 per cent normal serum to a medium may completely
inhibit dissociation.

Since mutation may take place at different levels, variations
may occur independently or in combination. Changes in biochemical
activity and cell morphology often take place independently, but the
loss of polysaccharide surface antigen is usually responsible for loss
of type specific antigenicity, increased agglutinability by salt, a1—
tered sensitivity to bacteriophage and loss of virulence. Many im-
portant exceptions to these general rules are known.

The following characteristics were chosen for this study on

the dissociation of 1:. hemolLtica: (1) colonial changes, (2) sensitivity

 

to sodium chloride as evidenced by growth characteristics in broth,
stability of organisms in saline suspensions at room temperature and
80 C, sensitivity in the presence of acid and agglutination by acri—

flavin, (3) presence of capsules, (4) variation in morphology of cells,

30
and (5) variation in biochemical activity. Virulence studies were not
included because the organism is normally a weak pathogen, and
housing facilities for the number of chickens required for statistical

data were not available at the time.
Colonial Changes Occurring During Dissociation

The phases present in culture 309 following accidental disso-
ciation and those arising during the study were described first. Then
as new cultures were isolated, the phases present at the time of iso-
lation were described. It was hoped that S and R types would be

identified.

Procedure. Both tryptose agar and tryptose agar base were

 

used. The medium which produced the larger number of colonies
from a given dilution of that culture was selected. The agar content
was adjusted to 2.0 per cent so that a firm surface was obtained
for streaking cultures with a wire loop. Agar plates were poured
to a thickness of one-quarter inch and incubated 24 hours to dry
the surface before use.

No single procedure of seeding cultures on the plates to Ob-

tain adjacent and contiguous colonies proved superior. Two methods

were used. The least time-consuming was direct streaking of a

31
small inoculum with a wire loop over the entire surface of a plate.
This produced confluent growth, contiguous, adjacent, and well—isolated
colonies provided the amount of inoculum was carefully controlled. A
colony selected for study was picked by use of the stereoscopic
microscope using a fine, flexible, one-inch wire which had been re—
moved from a light bulb, flattened on one end; and inserted into an
inoculating—needle holder. The amount of inoculum transferred to
the plate was just barely visible to the unaided eye. The inoculum
was then streaked over the entire surface of the plate using a stand-
ard-type loop. This was Satisfactory when single phases were streaked
for descriptive purposes. Since coloring of colonies is dependent on
size and proximity of the colonies, this procedure allowed one to be-
come familiar with the variations of each phase. The method was
also satisfactory for streaking two phases together for comparison
or verifying that they were in the same phase. Contiguous colonies
are best suited for this purpose. The other method consisted of
diluting the culture in saline to such a concentration that 1.0 ml of
suspension distributed over the plate would produce sufficiently iso-
lated colonies. The advantage here lay in greater uniformity of
coloring which resulted in greater ease of examination. On the other

hand, closely related phases are often indistinguishable unless the

colonies are contiguous, and the uniform distribution expected by this

32
method usually produced few such colonies. Dilution of the broth
culture for optimum concentration of colonies is dependent on colony
size, strain, and phase. Culture 309 was diluted l—500,000 while
culture 684, phase B, required dilution to 1-4,000,000 for satisfactory
distribution. Forty—eight-hour broth cultures were used Since 24-
hour cultures gave variable results. This method was very time-
consuming. A colony picked from a plate was placed on a tryptose
agar slant, incubated for 24 hours, then transferred to tryptose broth,
incubated for 48 hours, and diluted in saline for inoculation onto
plates. All plates were incubated 48 hours before examination. Thus,
five days were required to obtainpthe same information obtainable in
two days by direct streaking. This method had the further disad-
vantage that cultures dissociate more rapidly in liquid media than
in solid.

Cultures were examined using a stereoscopic microscope at
magnifications of 9X and 18X. The source Of illumination was an
American Optical Spencer microscope illuminator No. 735B set on a
block 4-1/2 inches high and placed at a distance of 10-3/4 inches
from the microscope. The adjustable condensing lens of the lamp
was so positioned as to produce maximum color differential in the
bacterial colonies being examined. The lamp was tilted until the

outside rim holding the filter in place rested on the front edge of

33
the supporting block. The diaphragm was opened as far as possible
and the blue glass filter removed. A thin photographic plate with
the emulsion removed was substituted for the glass stage of the
microscope So that no light rays would be absorbed. A five-centi-
meter concave mirror was placed about 8-3/4 inches from the sup-
porting block for maximum brightness. The angle of reflection was

approximately 4 5 °.

Results. All colonies from cultures 309, 710, 684, 782, 868,
and 328 were circular, glistening, and ranged from 1.0 mm to 2.5
mm in diameter. Most colonies were convex and butyrous. The
opacity of all 309 phases was grossly much less than that of the
other cultures except 684 A, Avl, AvZ, and 328, which were more
opaque than the 309 phases when viewed under the microscope. Dife
ferences in opaqueness of other phases were not visible to the un-
aided eye.

Culture 309 was studied over a period of four months. There
were three phases present when first examined: Y, YO, and some
striated colonies, St. The striated colonies, when restreaked, yielded
St colonies plus a third stable phase, YT. Phase YT occasionally
Pro'duced St colonies which were at first thought to be a transition

stage between YT and another phase. NO phase arose from either

34
St or YT during the period of study, however. Other stable and un-
stable phases arose from Y and Y0 in sequence as shown in Plate
1. Descriptions of the stable phases appear in Table V.

Other rapidly changing phases were also found. A phase
designated YO'vb was occasionally isolated from YO'. It was 1 mm
in size, pulvinate, mucoid, and cottony in appearance under the micro-
scope. The color w'as yellow buff, and the colony was more opaque
than YO' and more granular than any of the stable phases. When
restreaked, it gave rise to YO'vb and YO' colonies plus some light
unorganized, mush-appearing growth in the background. If held on
slants and transferred several times, it reverted to YO'. Morphology
of cells did not differ from YO'.

Another group of transitory colony types, designated the V
group, originated from phase Y. All these types had an irregular
margin. The first to appear was a flat, translucent colony with a
bright refractive margin. The cells from this colony when stained
Were found to consist of long curved filaments which may be seen
in Plate VI, Figure 4. This colony, when transferred to another
Plate, broke down into three related types: V-l was a clear, gray,
evenly granular colony with a very irregular margin. This colony
When streaked produced like colonies which did not grow on transfer.

V-2 was similar to V-l except for a heavy, prismatic strip appearing

(

35

PLATE 1

DISSO CIA TIVE PA TTERNS

(ST) YO -——> YO' H (YO'VB)

.1“ \

YT YOMO ——-> (SP)

Culture 309

 

4. (PC) ————+ B ———-> BVI

/\\1 1\ \/

(A0) (AS) (FCO) ——-)(F)C (BIR)
AVI AVZ (FCO)
Culture 684

) = unstable phases.

COMPARISON OF PHASES:

36

TABLE V

CULTURE 309

 

 

 

Phase Ap:::::ce Opacity Coloring

Y 1.5 mm, convex, L YO’F Centerw-more opaque, yellow
entire, butyrous, G YO'* buff; margin-—thinner with blue
granules fine. to red rainbow effect; confluent

colonies appear pink; red pre-
dominates more than in Y0 or
YO'.

YO 1.5 mm convex, G Y, Center--warm cream; margin
entire, butyrous, YO' --clear, narrow, and bluish;
granules finer little red in coloring; yellows
than Y. predominate.

YO' 1.5 mm, convex, least Rainbow effect throughout col-
entire, butyrous, opaque ony from blue on bottom to
granules finest. yellow to red on top; blue—

greens predominate.

YT 1.5 mm, convex, G YO' Center--—orange yellow for 1/2
entire, butyrous, L YO area; margin--green-blue; some
granules as YO. red on top; closer in hue to

YO' than YO.

YOMO 1.0 mm, entire most Cream white.
colony lifts from opaque
agar when 309
touched, gran-
ules as YO.

YG 1.5 mm, convex, same Appears like Y in texture, but
entire, butyrous, as Y green-yellows predominate; hue

granule s as Y.

*L:

closest to YO'.

less than; G = greater than.

37
along one side. When transferred to another plate, this colony pro-
duced V-2 and V-3 type colonies, but when the resultant V-Z was
streaked, the colonies were small, gray, flat, and did not grow when
subcultured. The V-3 colony was entirely prismatic and more opaque
than V-l. When streaked on another plate two colony types appeared
which did not grow on transfer. One was V-3 type; the other was
more opalescent with a blue edge. The surface was smooth in re-
flected light but appeared to have eccentric lines when viewed with
transmitted light.

A colony similar in appearance to the original V-type colony
developed from phase YOMO. It was streaked on a fresh plate to
see if V—l, 2, or 3 would be produced. After two transfers the
crystalline edge disappeared. On the following transfer three phases
were present. One seemed to be the same as the previous type.
Both other types were larger than the stable 309 phases and spread-
ing in nature. Sp-l was a flat, orange colony with an irregular edge.
Sp-2 was raised, more buff colored, and smooth with a thin edge.
Transfers from these colonies did not grow. The six stable phases
were lyophilized for reference using sterile milk as the vehicle.

Culture 684 was isolated on blood agar from the liver of a
chicken affected with Newcastle disease. Two colonies were trans-

ferred from the blood plate to lactose motility tubes and then streaked

38
on tryptose agar plates. Growth was not satisfactory as previously
indicated, and tryptose blood agar base was substituted. Four colony
types were present: A, B, FC, and FCO. Stable colony types pres-
ent at the time of isolation were designated by letter (A, B, etc.),
and stable phases arising from them later were designated by the
phase letter from which the new phase arose plus v (variant) and
numbers to differentiate these new variants. (When two phases were
found to develop from 684 A, they were labeled Avl and Av2.) All
new colony types were streaked with all previously described stable
phases, and if they were in the same phase, were given the Same
identification. All results were checked many times. Descriptions
of phases A and B together with the other stable phases arising from
them appear in Table VI. Sequence of appearance of variants ap-
pears in Plate 1. FC and FCO were transitory phases originating
from A. FC was 1.0 to 1.5 mm in diameter, flat, gray, granular,
and nearly transparent. When transferred to another plate, phases
FC, FCC, and B resulted. FCO appeared the same as FC except
for a raised, opaque area varying in size and yellow cream in
color. It gave rise to FC, FCO, and A colonies.

Two other transition colonies arose from A. A0 was the
same size as A but more opaque and buff in color. When streaked,

A and Avl colonies were present. Phase AS was half the size of

CO MPARISON OF PHASES:

39

TABLE VI

CULTURE 684

 

 

 

Phase c°l°ny Opacity Coloring
Appearance

A 1.0-1.5 mm, G any’i‘ Buff yellow with orange center,
convex, some- 309 some red on top edge, blue on
times umbo- bottom; when plated with B is
nate, entire, smaller than typical colony and
butyrous gran- loses central raised area.
ules fine.

Avl 1.0—1.5 mm, G A Color and density more homo-
convex, entire, geneous than in A; yellow-
butyrous, gran- orange throughout.
ules fine.

AvZ 1.5-2.0 mm, L Avl Color and density more homo-
convex, slightly geneous than in A: color paler
irregular, bu- than A, cream yellow.
tyrous, cottony
texture.

B 2.0-2.5 mm, G A Color much like A; B gives ap-
raised, entire, pearance of being more orange
butyrous, gran- when viewed alone, probably due
ules fine. to size; red fringe on top edge

pronounced.

Bvl 2.0—2.5 mm, same Homogeneous pink—cream with
raised, entire, as B faint yellow-green margin; when

butyrous, gran—
ules fine.

colonies well separated, looks
like B; when streaked with B,
looks pale, chalky, and more
opaque.

 

 

* L = less than; G

= greater than.

40
A, more opaque and buff in color. It produced A and Av2 colonies.
Avl and AvZ phases were also found without appearance of the tran-
sition phases. These stable phases may be seen in Plates II and
IV.

Phase BIR was first seen on a saline dilution plate that was
made from a brothiculture of phase B that had been exposed to
garlic vapors during incubation. It was identical to phase B except
for the irregularity of the margin. This is shown in Plate IV,
Figure 3.

Culture 782 was isolated from a bird submitted as a pullorum
reactor. The bird gave a positive whole blood agglutination in the

laboratory with polyvalent Redigen, but Salmonella Ellorum was not

 

isolated. Two phases were present which proved to be indistinguish-
able from 684 A and B, and were so designated 782 A and B. Stock
cultures of these phases were being incubated the two succeeding
week ends when garlic antibiotic cultures were placed in the incu-
bator without permission of the persons using the incubator. Garlic

appeared to have a marked antibiotic effect on 1:. hemolytica. Most

 

of the dilution plates incubated at this time were sterile; the others
had only a few, abnormally small colonies present. Agar slants of
phases 782 A and B grew very poorly. When these cultures were

plated following the first exposure to garlic vapors, the colonies

41
produced were atypical, and bacterial cells from them were extremely
pleomorphic. After several subcultures in a normal atmosphere, the
phases again appeared typical. Following the second exposure, how-
ever, two phases differing frOm A and B were found. The new
phases were called C and H. Descriptions of these colonies are
recorded in Table VII. Later another phase originated from C and
was designated Cvl. None of these phases resulting from exposure
to garlic vapors coincided with any of those formed without induced
dissociation from 684 A and B.

Culture 868 was isolated from rachitic five-day old chicks.
At first only one phase was thought to be present, but after several
platings two phases called E and F were separated. These were dif-
ferentiated chiefly by opacity. Phase E was found to vary in size
from pin point to 1.5 mm with regularity, while phase F colonies
ranged from 1.5 to 2.0 mm and had a slightly irregular margin.
Neither phase corresponded with any phase previously described. A
third phase arising from E was found to be identical in colonial ap-
pearance with 782 C and was designated 868 C. A description of
these phases appears in Table VII. Pictures of phases E and F
may be seen in Plates III and IV.

Culture 710 was isolated from a chicken affected with visceral

1Ymphomatosis. Two colony types were present and are described in

TABLE VII

COMPARISON OF PHASES:

42

CULTURES 782 AND 868

 

 

 

Phase Colony Appearance Opacity Coloring
782 indistinguishable from 68.4 A
A
782 indistinguishable from 684 B
B
782 2.0 - 2.5 mm, raised, slightly Buff-yellow orange; hue
C entire, butyrous, gran— L 684 B closest to 684 B, but
ules fine. slightly more blue-green
present.
782 2.0 - 2.5 mm, raised, same as Homogeneous yellow
Cvl entire, butyrous, gran— 684 B cream with some blue
ules fine. G 782 C present.
782 2.0 mm, convex, en- grossly Ivory yellow with blue-
H tire, butyrous, gran- L 782 C green margin which
ules fine. tapers out thin if colo—
nies are well separated.
868 size varies from pin G 782 C Ivory buff with some
E point to 1.5 mm, con- L 782 blue tinge; coloring
vex, entire, butyrous, Cvl closest to 782 H.
granules fine.
868 1.5 - 2.0 mm, convex, L 868 E Pale ivory-yellow with
F Slightly irregular, same as blue-green tinge; very
butyrous, cottony 684 Av2 close to 868 E in hue,
texture. color more definite than
in 684 Av'2.
868 indistinguishable from 782 C
C

 

 

L : less than; G a greater than.

43

Table VIII. Both of these phases agglutinated spontaneously in phys-
iological saline. They were considered rough, and were designated
R1 and R2. The morphology and growth in broth were described,
capsule stains made, and the phases inoculated into carbohydrates.
They were transferred to brain-heart infusion (Difco) with 0.2 per
cent agar and stored at -40 C. It was hoped that they would remain
inactive yet viable at this temperature until similar phases could
be isolated from other cultures for comparison. The frozen prepara-
tions did not grow when thawed one month later. Other cultures did
not change appreciably when stored in this manner up to three months.

None of the phases thus far described was found to be smooth
when studied further. Colonies were pickedfrom twelve other cul—
tures upon isolation. None of these were smooth either. Two of
the twelve cultures agglutinated spontaneOusly in 0.85 per cent saline.
The remaining ten cultures agglutinated in acriflavin; three were
positive in the rapid slide test, and the other seven were positive
in the six-hour tube agglutination.

Finally, one phase which was considered smooth was isolated.
The growth was quite transparent, the colonies were small, and re-
sembled most closely those of culture 309. Growth on plates was
very poor and isolated colonies rarely developed. The presence of

blood markedly enhanced growth. Tryptose agar proved to be the

TABLE VIII

COMPARISON OF PHASES:

44

CULTURES 710 AND 328

 

 

 

Phase Colony Appearance Opacity Coloring

710 2.0 - 2.5 mm, convex, L 684 B Pale blue-yellow; con-

Rl slightly irregular, centric rings appear in
butyrous, granules transmitted light; sur-
fine. face smooth.

710 1.0 - 1.5 mm, flat, irides- All colors reflected by
entire, brittle, pow- cent the particles; appears
dered glass texture. very bright.

328 1.0 mm, mucoid, G 309 Y Top half pink, bottom

5 slightly irregular, half yellow; coloring
granules fine, brighter than 309Y;
like 309 Y. margin - very thin,

narrow, clear.

328 1.0 mm, slightly slightly Light pink throughout;

I irregular, brittle, L 328 S yellow and red re-

powdered glass
texture, slightly
more granular

than 328 S.

flected from granules;
margin almost pris-
matic.

 

 

L = less than; G a greater than.

45
most satisfactory medium without blood, but dissociation was rapid.
Descriptions of the smooth phase, S, and another phase, I, appear
in Table VIII.

An attempt was made to utilize 2,3,5 triphenyl tetrazolium
chloride for greater differentiation of phases. It proved to be mark-
edly inhibitory even at a concentration of 0.00125 per cent. The
recommended concentration for distinctive colony coloring is 0.005

to 0.01 per cent.
Sensitivity to Sodium Chloride

Since the first studies on dissociation, auto-agglutination in
phy'rsiological saline has been accepted as a means of confirming the
roughness of a colony. At present,-the phenomenon is generally
regarded as the result of an increase in lipoidal substance on the
surface of the bacteria due to the loss of normal polysaccharide
antigen. The intermediate phases between S and R become increas-
ingly sensitive to sodium chloride as more of this lipoidal substance
is exposed. Intermediate phases which are stable in physiological
saline at room temperature may often be detected by auto-agglutina-
tion at 80 C or in the presence of acid. The granular appearance
of rough variants in broth cultures is attributed to the same phe—

nomenon. The acriflavin test, which was first used by Pampana, is

believed by Edwards and Bruner (1942) to follow more closely the
serological behavior of S and R antigens than any other indicator
of roughness. Smooth phases are not agglutinated in acriflavine

solution.

Growth characteristics in broth. All stable phases described

 

were grown at 37 C in tryptose broth and examined daily for six
days. Most phases produced a light uniform turbidity with some
sediment and' a ring on the surface of the broth in 48 hours. A
partial pellicle formed in three to six days. No sediment or pel-
licle was formed in the broth inoculated with 328 S, the one smooth
phase. Increasedgranularity was not noted grossly in the rough
phases 710 R1 and R2, but the tendency to pellicle formation was
more pronounced, and stains made from broth cultures revealed the

pre sence of clumps .

Stability of organisms in saline suspensions. The growth of

 

a 24-hour tryptose agar slant was suspended in sterile, 0.85 per cent
sodium chloride and the turbidity adjusted to tube 1 MacFarland
nephelometer. One ml of the suspension was pipetted into a 12 by
75 mm agglutination tube and incubated in a water bath at 80 C for
one hour. Five ml of suspension were placed in a 16 by 150 mm

test tube and examined daily for one week for evidence of agglutination.

47
None of the phases except 710 R1 and R2 showed evidence of
agglutination. The rate at which suspended cells settled to the bot-
tom of the test tube varied, however. Cells from phases 684 Av2
and Bvl were partially settled within 24 hours. Phase 782 H was
completely settled, and the 309 phases and 868 E were partially
settled in one week. Cells from phases 684 A, Avl, and B, 782 C
and Cvl, and 868 F were still in suspension after one week. None

of the phases stable in saline at room temperature agglutinated at

80 C.

Sensitivity in the presence of acid. All phases studied were

 

suspended at a turbidity corresponding to tube 1 MacFarland nephel-
ometer. in 5.0 m1 of sterile, 0.85 per cent sodium chloride buffered
at pH levels of 3, 4, 5, 6, 7, 8, 9, and 10 with "pHydrion Buffers'l
manufactured by the Micro Essential Laboratory, Brooklyn 10, New
York. Suspensions were held at room temperature and examined‘in
24 hours and again after one week.
Results are tabulated in Table IX. Phases 309 Y, YG, and

328 were stable in saline at all pH levels. Most phases agglutinated
at pH 3 after one week although phases 309 YO, YOMO, and 782 C

were stable in 24 hours. Fourteen of the eighteen phases showed

agglutination. at pH 4, but the Size of the agglutinated particles varied.

TABLE IX

48

AGGLUTINATION OF SALINE SUSPENSIONS PREPARED FROM

STABLE PHASES UNDER VARYING CONDITIONS

 

 

Buffered Saline *

 

 

 

 

 

 

Phase 035% Acnflam
Saline 24 Hrs. 1 Week Plate Tube
(pH) (RH)

684 A - 3 3, g - +
684 Avl - 3, 4, 5 3, 4, 5 + +
684 sz — 3, 4 3, 4 + +
684 B - 3, 4' 3, 4_, _5_ — +
684 Bvl — 3, 4 3, 4 - +
710 R1 +

710 R2 +

782 c - — 3, g + +
782 Cvl — 3, g 3, 4_ + +
782 H - 3, 4 3, 4, g + +
868 E - 3, 4, 5 3, 4, g + +
868 C - _3_' }_ + +
868 F - 3, 4 3, 4 + +
309 YT -. 3, g 3, 4_ .— +
309 YO - - _3_ + +
309 YO' - 3, 4 3, 4 + +
309 Y - - — + +
309 YG - - - + +
309 YOMO - - _33 i + +
328 S - - — - -
* a Phase showed agglutination with settling of particles at pH

level indicated.

Agglutinated particles remained in suspension.

Agglutination weak, particles small.
+ 8 positive agglutination
- a negative agglutination

49
Four of the eighteen phases agglutinated slightly at pH 5. No agglu-
tination was noted above pH 5. In most tubes above pH 5 the cells
had settled after one week. Maximum stability was observed at pH

10 although all phases stayed in Suspension at pH 8 for 24 hours.

Agglutination by acriflavine. Both slide and tube methods

 

were used. In the slide test, one loopful of heavy saline suspension
of organisms was mixed with one loopful of 1-500 acriflavine in 0.85
per cent saline and watched for two or three minutes for agglutina-
tion. For the tube test equal parts (0.5 ml) of l-l,000 acriflavine
in saline, and suspended organisms of turbidity corresponding to tube
1 MacFarland nephelometer were incubated at 37 C for six hours.
The tube method produced more positive agglutinations than
the slide method. Results are recorded in Table IX. No phases
were studied which did not agglutinate in acriflavine except 328 S.
Phases 684 A, B, Bvl, and 309 YT were positive in the tube test

only.
Presence of Capsule s

The method employed for demonstration of capsules is a
modification of one described by Huddleson (1940) as recorded in

Stafseth's laboratory manual (1953). Organisms were grown on

50
blood agar for at least two transfers and cells from a 24-hour cul-
ture suspended in a small drop of water using an inoculating loop.
The drop was then smeared on the Slide with a second slide as if
preparing a blood film. When dry, this film was covered with a
smear of India ink made in the same manner and then fixed by
flooding for five seconds with absolute methyl alcohol containing 0.25
per cent acetic acid. The preparation was then washed in water
and stained with a saturated solution of gentian violet in aniline di-
luted with an equal part of distilled water. The capsules appear as
a clear area around the purple bacteria.

Phases 684 A and B, 782 H and Cvl, and 868 F were thought
to have small capsules. Although this method is considered by many
to be the most satisfactory staining method for the demonstration
of capsules on small bacteria, the results obtained here were not

consistent. The absence of capsules was not considered reliable.
Variation in Biochemical Activity

All stable phases were inoculated into carbohydrate media
prepared as previously described and incubated at 37 C for two weeks.
Results are recorded in Table X. Acid was produced consistently
from dextrose, mannitol, sorbitol, sucrose, mannose, glycerol,

fructose, and galactose, No change was noted in rhamnose, inulin,

51

TABLE X

FERMENTATION REACTIONS OF STABLE PHASES STUDIED
IN THOSE CARBOHYDRATES SHOWING VARIABILITY

 

 

 

Phase Lac- Malt- Raf - Denx- Starch Arab- Inc-

tose ose finose trm Inose Sitol
684 A . . . . A S - - - S A.
684 Avl . . - S - - - S A
684 Av2 A “A S - - S A
684 B A A - - - S A
684 Bvl A A — - - S A
710 R1 - A S - - - —
710 R2 A A S - - - -
782 C A - S S S S A
782 Cvl A - S S S S A
782 H — S S S A - -
868 E . . . . - S S - - - -
868 C . . . . A S S S S S A
868 F . . . . 4- S S S - - -
309-all . . . A - S - - - A
328 S . . . . A A S S S - -

 

 

A an acid produced within two days. '
S 8 acid produced in three to seven days.
- 3 no acid.

52
salicin, dulcitol, or adonitol. Variable reactions were obtained in
lactose, maltose, raffinose, dextrin, starch, arabinose, and inositol.
Variability in fermentation did not appear correlated with colonial
morphology since identical colony types produced from different

cultures did not give identical fermentation reactions.
Variation in Morphology of Cells

Gram stains were made from 24 and 48 hour tryptose agar
slants. All phases were gram negative, but much variation in size
and Shape of cells was noted. Phases 684 B, 782 H, 868 E, and 309
YT were cocco-bacilli typical of pasteurellae. Variation in Size of
bacilli occurred in 684 Bvl, 782 Cvl, 309 YO', 309 Y, 309 YG, 868
C, and 710 R1. Filaments appeared in 684 Avl, 684 Av2, 782 C,
782 Cvl, 868 F, 868 C, 309 YO, 309 Y, and 309 YG. Involution
forms were noted in 684 A, 684 Avl, 684 Av2, 309 Y, 868 C, and
710 R2. 309 YOMO was composed of deep-staining, almost circular
forms. Some cells seem to have been lysed in 48 hours leaving a
formless residue which stained with safranin. Descriptions of mor-
phology of these phases together with a listing of the gross Opacity
of growth on an agar Slant appears on Table XI. Pictures illustrat—

ing the cell types noted will be found in Plates V and VI.

53

TABLE XI

OPACITY OF GROWTH ON AGAR SLANTS AND DESCRIPTION
OF CELLULAR MORPHOLOGY

 

 

 

Phase 0::th: 24-Hour Morphology

684 A T Small ovoid bodies, occasional involution
form.

684 Avl T Pleomorphic, mostly involution forms, few
filaments.

684 Av2 T Mostly bacillary forms, few involution and
filamentous forms.

684 B M Regular cocco-bacilli.

684 Bvl M Regular, cocco—bacilli larger than B.

710 R1 M Clumped, slender, bacilli of varying length.

710 R2 T Pleomorphic, mostly involution forms, marked
autolysis.

782 C M Long, slender bacilli, 50% filaments, few
bacilli in chains.

782 Cvl M Bacilli varying in length, short filaments

_ present.

782 H M Small cocco-bacilli, a few chains.

868 E M Short, regular bacilli.

868 C M Bacilli, irregular in size, some involution
and filamentous forms.

868 F M Short bacilli, some long filaments.

309 YT T Regular cocco-bacilli.

309 YO T Short bacilli, few slender filaments.

309 YO' T 'Fat bacilli, varying in length.

309 Y T Chiefly fat bacilli, some involution and fila-
mentous forms.

309 YG T Short bacilli, vary in length, some filaments.

309 YOMO T Almost circular forms, deeply stained.

328 S T Variation in size from frank coccus to def-

inite bacilli, few long filaments.

 

 

T 3 growth transparent.
M 1 growth more translucent, heavier than T.

54

PLATE 11

Figure l. Phases 684 A and 684 Avl.
Note the raised center in the three 684 A colonies and
the more opaque and homogeneous appearance of 684 Avl.

Figure 2. Phases 684 A and 684 B.
Note the larger size of 684 B and the homogeneous ap-
pearance and reduction in size of 684 A in the presence
of B.

Figure 3. Phases 684 B and 782 C.
Note similarity of opacity and contour. Phase C is the
colony farthest right.

Magnific ation: 16X

 

56

PLATE III

Figure 1. Phases 782 C and 782 H.
Note the thin, narrow margin of Phase 782 H in the lower
right corner.

Figure 2. Phases 782 C and 868 E.
Note the greater opacity and very narrow, clear margin

of Phase 868 E, the colony to the right and contiguous
with 782 C.

Figure 3. Phase 782 C.
Note the pronounced ring margin often seen in well-iso-
lated colonies of this phase and 684 B.

Magnification: 16X

 

58
PLATE IV

Figure 1. Phases 782 C and 868 F.
Note the softer, looser texture and very slight irregularity
of margin in Phase 868 F to the right.

Figure 2. Phases 684 Avl and 684 Av2.
Note the soft texture and irregular margin of the two
(larger) 684 Av2 colonies, and their similarity to 868 F.

Figure 3. Phase 684 BIR. and 684 B.
Note reduction in size and irregular margin typical of
Phase BIR.

Magnification: 16 X

 

60

PLATE V

Figure 1. Phase 868 E showing short, regular bacilli typical of
pasteurellae.

Figure 2. Phase 868 F showing short bacilli and some slender fila-
ments.

Figure 3. Phase 782 H (48 hours) showing autolysis with disintegra-
tion of cellular structure. ‘

Magnification: 1,000X

 

\I‘
.o.‘ .0.

I .. N..\‘.N..- ? . L

tau...” ....

I On

 

 

 

62
PLATE VI

Figure 1. Phase 782 H (24 hours) showing coccoid cells, a few short
chains of bacilli, and a clubbed form.

Figure 2. Phase 309 YOMO showing the almost circular and deeply
staining cells seen in this phase.

Figure 3. 684 Avl showing involution forms; note the pale-staining,
swollen cells as well as the more prominent irregular
filaments.

Figure 4. 309 V-Z showing curved filaments.

Magnification: 1,000X

 

  

  

324

«so; {‘37}? .-
' -. .‘. .

  
   

DISCUSSION AND SUMMARY

Studies were made on the serological relationships, fermenta-

tion reactions and dissociation patterns of Pasteurella hemolxgca

 

strains isolated from chickens. The cultures studied produced acid
readily from dextrose, lactose, mannitol, sucrose, galactose, xylose,
and mannose. On primary isolation, only a small amount of acid
was produced from lactose. A small amount of acid was also pro-
duced from sorbitol in 48 hours; this reaction faded and appeared
negative after four days. Acid was produced slowly from glycerol,
raffinose, and trehalose. A few strains fermented dextrin, starch,
and arabinose. Maltose fermentation was variable.

A study of the dissociation phases present in newly isolated
cultures revealed a high proportion of intermediate (1) phases. Out
of eighteen cultures, one was smooth (S) as evidenced by stability
in acriflavin, three were rough (R) as evidenced by spontaneous ag-
glutination in physiological saline, and the remaining fourteen were
intermediate. The smooth phase produced a uniformly turbid growth
in broth with no pellicle or sediment, remained stable in physiologi-
cal saline and acriflavin, and was not agglutinated in acid solution.

Cellular morphology varied from coccoidforms to definite bacilli.

65

This variation may have been environmentally induced since the
medium used in these studies was not optimum for the growth of this
phase. The intermediate phases were very unstable and dissociated
into other I phases readily on agar media. They also showed a
greater tendency toward sediment and pellicle formation than did the
smooth phase. The I phases did not agglutinate in physiological
saline, and most of them. remained in suspension for 24 hours. Ag—
glutination in acidified saline took place at pH 3 and 4 for most of
the phases. Five of the eighteen phases also agglutinated at pH 5.
Five other I phases did not agglutinate in acidified saline at any pH.
None of the phases stable in saline. agglutinated at 80 C. Four of
the intermediate phases were stable in l-500 acriflavine on the slide
test but agglutinated in l-l,000 acriflavine when incubated in tubes
at 37. C for six hours.

Variation, independent of colonial changes, occurred in the
fermentation of carbohydrates and in cellular morphology. The
changes in cellular morphology were marked. Both coccoid and
bacillary forms were seen. Filaments were often present, and invo-
lution forms, which are typical in old cultures of most pasteurellae,
were sometimes present in 24 hours.

Although studies were incomplete due to a laboratory mishap,

the presence of more than one serological type of E. hemolytica was

 

66

demonstrated by agglutination tests. These types could not be dif-

ferentiated by correlation with biochemical activity on carbohydrates
as were the types demonstrated by Spray and Newsom and Cross in
their work with sheep and bovine strains. Inasmuch as the dissocia-

tion studies indicated that _13. hemolytica strains were seldom smooth

 

when isolated, these serological data are of limited value. The use
of I phases for such study may partially account for poor antigenicity
noted by most investigators.

Two of the cultures studied exhibited nutritive requirements
different from those of the other strains. When 1.0 ml of a suitable
saline dilution of broth culture was inoculated onto the two media
being used, most of the cultures produced a larger number of colonies
on tryptose blood agar base .(Difco), containing 1.0 per cent tryptose,
0.3 per cent beef extract, and 0.5 per cent sodium chloride. Strains
309 and 328 evidenced a larger number of viable cells on tryptose
agar (Difco) which contains 2.0 per cent tryptose, 0.1 per cent dex-
trose, and 0.5 per cent sodium chloride. It is not known whether

this difference in nutritive requirements is correlated with distinct

strains of E. hemolytica or is a characteristic of certain dissociative

 

phases. Strains 309 and 328 were also decidedly more transparent
than the other cultures both grossly and microscopically. Whether

or not this difference in opacity would always coincide with different

67
nutritive requirements cannot be stated on the evidence of two cul-
tures. These transparent cultures are isolated from chickens far
less frequently than are the more opaque strains.

A truly adequate method for propagating _P_. hemolytica was

 

not found. The use of' artificial media for holding cultures was pre-
carious and time-consuming. Subcultures had to be made every five
days to insure survival. The addition of 5.0 per cent blood markedly
enhanced growth and facilitated recovery from cultures with few
viable cells but did not lengthen the interval between subcultures.
The high rate of dissociation as well as difficulty in obtaining suit-
able growth, difficulty in maintaining viable cultures, and cellular
pleomorphism all indicate that the media used were not optimum.
Attempts were made to preserve cultures by holding semisolid cul-
tures at —40 C and by lyophilizing in milk, but these methods were
not entirely satisfactory either. Recovery on subculture was un-
certain, and the presence of whole blood was necessary. A better
understanding of the nutritive requirements and more information on
enzymes and toxin production should be sought in further studies of

B. hemo_1ytica.

 

BIB LIOG RAPHY

Breed, R. S., Murray, E. G. D., Hitchens, A. P. 1948. Bergey's
Manual of Determinative Bacteriology. Sixth Ed. The Wil-
liams and Wilkins Company, Baltimore, Md. 549.

Berkman, S. 1942. Accessory growth factor requirements of the
members of the genus Pasteurella. J. Infectious Diseases,
71, 209-211.

Braun, W. 1947. Bacterial Dissociation, A Critical Review of a
Phenomenon of Bacterial Variation. Bact. Rev., 11, 75—104.

Edington, J. W. 1930. Pneumonia of Bovines due to Pasteurella
boviseptig, Pathological Report. J. Comp. Path. Therap.,
63, 239-252.

 

 

Edwards, P. R., Bruner, D. W. 1942. Serological Identification of
Salmonella Cultures. Ky. Agr. Exp. Sta. Circ. 54, 4-6.

Eveleth, D. F., Goldsby, A., Nelson, C. I. 1949. Fowl Cholera
(Pasteurella multocida). Vet. Med. 44, 73—78.

 

Huddleson, I. F. 1940. The Presence of a Capsule on Brucella
Cells. Mich. Agr. Exp. Sta. Bull. 177, 11-14.

Jones, F. S. 1921‘. A Study of Bacillus bovisepticus. J. Exptl.
Med., 34, 561-577.

 

Merchant, I. A. 1950. Veterinary Bacteriology and Virology. Fourth
Ed. Iowa State College Press, Ames, Iowa. 436-438.

Newsom, I. 13., Cross, F. 1932. Some Bipolar Organisms Found
in Pneumonia in Sheep. J. A. V. M. A., 'N. s. 33, No. 5,
711-719.

Packer, R. A., Merchant, I. A. 1946. Bovine Mastitis Caused by
Pasteurellae. No. Am. Vet., 27, 496-498.

69

Rosenbusch, C. T., Merchant, I. A. 1939. A Study of the Hemor-
rhagic Septicemia Pasteurellae. J. Bact., 37, 69-89.

Spray, 'R. S. 1923. Bacteriologic Study of Pneumonia in Sheep. J.
Infectious Diseases, 33, 97-112.

Stafseth, H. J., Newman, J. P., Stockton, J. J. 1953. A Manual for
Pathogenic Bacteriology. Burgess Publishing Co., Minneapolis,
Minn., 8.