EXPERIMENTALLY INDUCED Vibrio fetus VAR. venerealis
INFECTION IN THE GUINEA PIG (Cavia porcellus L.)
ANTHONY F. WALSH
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OP FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
A. M. D. G.
To the memory of my parents
Sincere appreciation is expressed to the Chairman of the author's
Supervisory Committee, Dr. Franklin H. White, for his encouragement,
guidance and constructive criticism during the course of this study.
The author acknowledges the interest and valuable contributions
of the other members of his Supervisory Committee, Drs. George T. Edds,
George E. Gifford, James A.Himes, Charles F.Simpson and Alvin C. Warnick
The interest and assistance of Dr. Edward C. Schroeder in the
surgery, care and welfare of the experimental animals is greatly ap-
The author further acknowledges the interest and encouragement
of Dr. K.P.C. Nair, Visiting Professor in the Department of Veterinary
Science, as well as that of all the faculty members and technical staff
of the Department.
Sincere appreciation is expressed to Dr. Tirath S. Sarndu for
his helpful suggestions and valued technical assistance, to Miss Jean
Barry for valued technical assistance with the histology and to Mr. Ton.
Carlisle for the photomicrographs.
Deep love and gratitude are expressed to my wife Mamie whose love
loyalty, encouragement and tireless efforts in the support of our family
contributed so much to make this work possible.
TABLE OF CONTENTS
ACKNOWLEDGMENTS .................... ......... .......... iii
LIST OF TABLES ...... .. .......... ..... .... .......... v
ABSTRACT .......... ... ...... ........... ................... vii
INTRODUCTION ............ ............................ ... .. 1
REVIEW OF LITERATURE ..................................... 3
MATERIALS AND METHODS .................................... 22
RESULTS .. ............ .. .. .......... ............. ...... ......... 39
DISCUSSION .......... ................................... 72
SUMMARY .................... .................... .. .. 90
APPENDIX A .. -....- .....- ........... .... ........ .......... 92
APPENDIX B ..-... ......... ...... ..... ....... .. ........... 102
APPENDIX C ............... ......... ....................... 104
BIBLIOGRAPHY .................................. ........... 110
BIOGRAPHICAL DATA ........................................ 119
LIST OF TABLES
1 Optical density readings of V. fetus cultures with
erythritol added ....................................... 40
2 Optical density readings of V. fetus cultures wi th
estrone added ................... ..................... 41
3 Optical density readings of V. fetus cultures with
progesterone added .............. .... ................ 43
4 Optical density readings of V. fetus cultures with
estrone and progesterone added ....................... 44
5 Optical density readings of V. fetus cultures with
FSH added ................... ......................... 46
6 Vaginal cultures of guinea pigs during and after
intravaginal inoculation with V. fetus ............... 48
7 Reproduction efficiency of guinea pigs after intra-
vaginal inoculation with V. fetus followed by
natural mating ......................................... 50
8 Vaginal cultures of estrus induced guinea pigs
during and after intravaginal inoculation with
V. fetus ....................................... ...... 53
9 Vaginal cultures of progesterone treated guinea pigs
during and after intravaginal inoculation with
V. fetus ............................................. 55
10 Cultures taken at necropsy following the intra-
uterine inoculation with V. fetus ovariectomized
estrogen treated guinea pigs ......................... 57
11 Cell counts of the right uterine horns and secon-
dary organisms from the tissues of estrogen treated
ovariectomized guinea pigs following intrauterine
inoculation with V. fetus ............................ 59
12 Cultures taken at necropsy following the intrauterine
inoculation with V. fetus of ovariectomized progesterone
treated guinea pigs ........... .......... ............ G60
13 Cell counts of tne righ-. aterin:e hcn.- i. d secon-
dary orani-.ns frr. the tissues of progaserrone
treated ovarieczc.izad guinea pigs follow1ing intra-
uterine inoculation with V. fetus.................... 63
14 Cultites taken at necropsy fro- gravid guinea pigs
euthan-zed following intrauterine inoculation .,ith
V. fetus............................................. 66
15 Bacterial flora of aborted fetuses and fetuses in
uterc following intrau-erine inoculation of the
dan with V. fetus.................................... 69
Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of Doctor of Philosophy
EXPERIMENTALLY INDUCED VIBRIO FETUS VAR. VENEREALIS INFECTION
IN THE GUINEA PIG (CAVIA PORCELLUS L.)
Anthony F. Walsh
Chairman: Dr. F. H. White
Major Department: Animal Science
A study of Vibrio fetus var. venerealis infection of the female
guinea pig was undertaken to determine the value of this animal as an
experimental model and to assess the influence of estrogen and pro-
gesterone on the initial establishment and course of the infection.
Estrone in combination with progesterone, when incorporated into
the culture medium, had a growth stimulating effect on V. fetus var.
venerealis in vitro.
Guinea pigs with normal reproductive cycles failed to develop
a prolonged vaginal or systemic infection when inoculated by the intra-
vaginal route with viable V. fetus var. venerealis. The use of exogenous
estrogen or progesterone did not influence the course of vaginal infec-
tion. The organism was not isolated from the uterus and was rapidly
cleared from the vagina. A group of estrous guinea pigs inoculated
intravaginally with V. fetus var. venerealis and then mated showed a
decrease in litter size and an increase in neonatal deaths when compared
to a control group.
When inoculated into the uterus of ovariectomized guinea pigs
the organism spread throughout the reproductive tract and was isolated
from the blood, liver, spleen, gallbladder and peritoneal cavity. The
use of exogenous estrogen or progesterone did not influence the course
of the infection, which was transitory. The organism did not reproduce
in the female genital tract.
Intrauterine inoculation of the gravid female guinea pig with
V. fetus var. venerealis was followed by abortion within 30 hours. A
generalized infection of the aborted fetuses was noted together with
involvement of the reproductive tract, fetal-maternal membranes, blood,
liver, spleen and peritoneal cavity of the dam. Pregnant animals which
did not abort revealed transitory vaginal infection and fetal resorption.
No lesions were found on histopathological examination of the
uterus of normal intact guinea pigs and ovariectomized guinea pigs.
Histopathological examination of the uterus and placenta from the ani-
mals which aborted showed necrosis, hemorrhage, edema and inflammatory
The value of the guinea pig as a model for the study of V. fetus
var. venerealis infection was not established under the experimental
conditions used. The non-gravid guinea pig was refractory to the in-
fection when inoculated by the vaginal route. The use of intrauterine
inoculation in the non-gravid ovariectomized guinea pig resulted in an
infection which was well tolerated by the animal and failed to produce
pathogenisis. Exogenous hormones produced little effect on the course
of the infection. A marked predilection for the gravid uterus was
clearly demonstrated and the possible value of the pregnant female guinea
pig as a diagnostic aid was discussed.
The host specificity of V. fetus var. venerealis for the cow was
considered to be the factor largely responsible for the failure to es-
tablish a lasting and progressive infection in the non-gravid female
Since the first isolation and description of Vibrio fetus by
MacFadyean and Stockman (1913), genital vibricsis has become recognized
as one of the largest single causes of embryonal loss in cattle (Faulkner,
1968). It is a widespread venereal disease which results in considerable
economic loss to the cattle industry both in the United States and abroad.
Both the carrier bull and the infected cow are asymptomatic and re-
main apparently healthy. Vibriosis may be recognized in a herd by careful
appraisal of the breeding records. Indications of the disease are a re-
duction in calf crop from the previous year and the repeated return of
cows to service. The suspected diagnosis is confirmed by means of bac-
teriological procedures or serological methods (Lovell, 1964).
The specificity of present serological methods for the diagnosis
of vibriosis, in either the bull or the cow, is not wholly reliable
(Plastridge and Easterbrooks, 1953; Lawson and MacKinnon, 1953). The use
of vaginal mucus antibody titers to establish a diagnosis of genital
vibriosis in the cow, although more dependable than currently available
serological techniques (MacKinnon, 1954), is complicated by the occurrence
of false positive results (Boyd, 1955).
In spite of the improvements made in methods of isolation, the di-
agnosis of V. fetus in the bull remains a key problem and often requires
test-mating to virgin heifers with subsequent isolation of the organism
from the female reproductive tract (Adler, 1953). In vivo research of
bovine vibriosis is still mainly depending upon the cow. This factor
alone often presents problems of availability, cost and handling which
could be avoided if a reliable small laboratory animal were available.
Although Smith (1918) was the first to study the effects of
V. fetus infection in small laboratory animals, especially guinea pigs,
very little data have been reported since that time to indicate con-
clusively the value of these animals either as diagnostic aids or in
the area of research. At present there is apparently no suitable lab-
oratory animal available for these purposes. The need to evaluate
species of laboratory animals for use in vibriosis research was recog-
nized by McEntee (1958).
This study was undertaken to attempt to gain a greater understand-
ing of V. fetus var. veneralis infection in both gravid and non-gravid
guinea pigs. Emphasis was also placed upon the influence of the hormonal
condition of the female reproductive tract and the route of inoculation
on attempts to establish the infection in the reproductive tract.
REVIEW OF LITERATURE
Koch (1891) recognized the causal relationship between a specific
microorganism and a given disease and established three general criteria
to prove such a relationship. These criteria became popularly known as
"Koch's postulates." Certain limitations to the application of the
"postulates" were recognized by Wilson and Miles (1964) but in essence
they remain as Koch originally proposed them.
The dynamics of a host-parasite relationship were reported by
Garber (1960). This author noted that the host must provide all the sub-
strates necessary to satisfy the metabolic and reproductive processes of
the microorganism. Lichstein (1959) described cell adaptability, a suit-
able physical environment and, most importantly, the chemical composition
of the medium as important factors in the in vitro initiation of cell
Microbial pathogenicity is the ability of an organism to produce
disease in a susceptible host. The importance of in vivo studies of
microbial pathogenicity was emphasized by Smith (1960) who noted that
the peculiar nutritional and other growth requirements of various pathogens,
which led to their tissue specificity in disease processes, were not always
reproducible in vitro. He cited Brucella abortus infection of the bovine
cotyledon and Corynebacterium diphtheriae infection of the throat and
tonsils of man as typical examples.
The role of microorganisms as a cause of reproductive diseases of
cattle was first discovered by Bang (1897), with the isolation and
description of Brucella abortus. V. fetus was established as a cause
of infectious abortion in cattle by Smith (1919). The organism was
isolated from the fetal membranes of infected cows. The possibility of
V. fetus as a cause for low conception rates in cattle was first sug-
gested by Plastridge and Williams (1943). Other infectious diseases
which influence reproduction include those of viral, fungal and protozoal
origin. These are reviewed by Osebold (1969).
The bacterial flora of the reproductive tract may be classified
either as "normal" or potentially pathogenic. Bacterial populations of
the human vagina were found to be influenced by the hormonal condition of
the reproductive system (Huffman, 1959; Jones, Carter, Thomas, Peete and
Cherney, 1959). A constant bacteriological finding in the normal human
vagina is Doderlein's bacillus (Lactobacillus acidophilus) which, in
part, controls the micro-environment of other microorganisms present in
the vagina (Hunter, Long and Schuma:her, 1959). The presence of po-
tentially pathogenic organisms in the cervical flora of the human preg-
nant female was noted by White and Koontz (1968). These organisms were
considered as a hazard to pregnancy only if certain predisposing factors
were present in the host. These factors were an inadequate nutritional
status, anemia and difficult parturition. Complications in human pregnancy
due to V. fetus infection were described by Vinzent, Delarue and Hebert (1950)
The possibility that infertility in cattle was the result of bac-
terial flora other than V. fetus was suggested by Hatch, Feenstra and
Jennings (1949). Lindley and Iatfield (1952) isolated 26 separate species
of bacteria from the bovine uterus in a study of infertile cows. The
majority of these organisms were of the genus Niisseria. Such infections
were considered non-specific. The concept of a non-specific infection with
the production of endometritis as the major cause for infertility was
found untenable by Gibbons, Attleberger, Kiesel and Dacres (1959).
These authors described some 90 isolants of bacteria from bovine cervical
mucus which were classified into 15 genera. In contrast with Hatch et al.
(1949) and Lindley et al. (1952) a higher conception rate was observed
among those animals whichharbored potential pathogens than those found to
be bacteriologically negative or to carry non-pathogens.
The non-pregnant mammalian reproductive system undergoes cyclic
change in which the physical and chemical nature of the tissues and
lumen fluids is altered. These changes are mediated by steroid hormones.
Estrogens were found to give rise to increased vasodilation and increased
vascularity of the female genitalia (Hansel and Asdell, 1951). Histolog-
ical changes in the vaginal epithelium of laboratory animals were de-
scribed by Meyer and Allen (1933). The production of mucus by the vaginal
epithelium was due to the simultaneous action of estrogen and progesterone.
The production of cornified epithelial cells in the vagina was observed
by Meyer et al. (1933) as the result of an increase in the estrogen level
of the reproductive system. The use of vaginal cytology to determine the
hormonal condition of the female reproductive system of mammals with
short reproductive cycles was recommended by Nalbandov (1964). Asdell
(1946) reported the occurrence of vaginal epithelial cornification in
the guinea pig at estrus. An increase in leucocytic infiltration into
the uterine cavity following ovulation was also noted.
Asdell (1955) described an increase in the growth of the capil-
lary bed with subsequent thickening of the endometrium as a result of
estrogen stimulation of the uterus. Deposition of glycogen in the uter-
ine epithelium and musculature was another change observed by Boettiger
(1946). Progesterone was also found to influence the production of a
characteristic mucus-saturated stratified epithelium (Emmens, 1969). The
effect of progesterone on the non-pregnant uterus was to further increase
the endometrial thickening, already started by estrogen, and to cause
enlargement and branching of the uterine glands (Kraus, 1926).
The physiological effects of estrogen and progesterone cannot be
considered separately in the intact female due to a considerable over-
lapping in the secretion and action of the two hormones. Thus, the uterus
is never acted upon by either hormone alone (Nalbandov, 1964). Neverthe-
less, an increase in steroid hormone levels brought about by the use of
exogenous sources of these hormones has allowed the alteration of re-
productive cycle lengths and physical and chemical parameters. Smallwood
and Sorensen (1969) administered progestogens to feedlot heifers. At
slaughter, animals which were slaughtered 48 hours after the final feed-
ing showed an increase in the vascularity of the uterus. Uterine gland
epithelial height increased for a 96-hour period after the final feeding
but decreased rapidly thereafter. The presence of a very thick, clotted
cervical mucus in some of the heifers was considered abnormal. This con-
dition was not found in the untreated control animals. The continual
administration of progestins to heifers, when fed at a level of 0.6 mgm per
day, was shown effective in the prevention of estrus (Young, Cundiff and
Carroll(1942) noted an increase in both anaerobic and aerobic
glycolysis in uterine tissue as a result of exogenous estrogen stimulation.
Studies by Leonard and Knobil (1950) described an increase in the beta-
glucuronidase activity of the uterus and vagina of the rat when treated with
exogenous estrogen. Tissues of rats treated with estradiol were found to
incorporate labeled glycine into protein more readily in vivo (Mueller,
1953). A similar stimulation of protein synthesis in vitro was re-
ported by Brooks, Leitheiser, De Loecker and De Wever (1969). Signifi-
cant increases in the incorporation of labeled glycine and leucine into
the proteins of uterine microsomal supernatants were attributed to
estrone sulfate. This was the only estrogenic compound found to elicit
this particular response.
Histochemical changes within the cells of the mammalian repro-
ductive tract are also influenced by hormonally mediated cyclic activity.
Akins, Morrisotte and Cardeilhac (1969) reported a progressive rise in
the luteal tissue acid phosphatase activity in the pig as the corpus
luteum matured. A rapid rise in acid phosphatase activity occurred on
day 14 of the cycle and was indicative of impending corpus luteum re-
gression. Endometrial acid phosphatase activity was low throughout the
cycle. Alkaline phosphatase activity was decreased during the follicular
phase and elevated during the luteal phase of the cycle. Similar findings
were reported earlier by Goode, Warnick and Wallace (1965).
The lumen fluids of the mammalian reproductive tract comprise those
of the fallopian tubes, uterus, cervix and vagina. These fluids are sub-
ject to chemical and physical change during the reproductive cycle. Olds
and VanDemark (1957) conducted a series of slaughter experiments on cows
in which they determined locational differences in the percentages of ash,
total nitrogen, reducing sugars and ether extract. Concentrations of re-
ducing sugars were highest in the oviduct and uterus. Cyclic variations
in the sodium, potassium and calcium concentrations of the lumen fluids
were also noted. Uterine fluid sodium levels of 277 mgm per 100 ml were
present in the luteal phase of the cycle compared to 192 mgm per 100 ml in
the follicular and proestrus phases. Potassium levels remained virtually
unchanged in the uterus but were higher in the vagina during the luteal
phase. Uterine fluid was found to contain large numbers of extruded
epithelial cell nuclei in the luteal phase.
Guay (1966) described a decrease in the calcium level of the cer-
vical secretion of the cow on the day of estrus followed by a further
steady decline during the post estrus period. The possible effect of
even slight changes in the composition of the lumen fluids on the fer-
tility of cows was reported by Guay and Lamothe (1969). A significant
decrease in the cervical fluid sodium concentration of infertile cows was
observed when compared with the sodium concentration of the cervical fluids
of fertile cows. The reproductive cycles of both test and control cows
were of normal duration.
The physical and chemical properties of the lumen fluids of the
female reproductive system affect the metabolism of the spermatozoon and
the ovum both before and after fertilization, and provide for the survival
of the blastocyst before nidation (Nalbandov, 1964). Uterine fluid was
found to stimulate spermatozoal respiration and to maintain sperm motility
(Murdoch and White, 1968). Olds and VanDemark (1957a) recorded differ-
ences in the survival times of motile spermatozoa in lumen fluids collect-
ed from different parts of the female bovine reproductive system. Motil-
ity persisted for 19 hours in follicular fluids but lasted only seven
hours in uterine fluids. Blastocyst survival is progesterone dependent.
However, Holland, Calhoun, Harris and Walton (1968) showed indirectly that
hyperthyroidism enabled blastocyst survival in the rat. This condition
stabilized the alkaline phosphatase levels in the uterus of the progesterone
In a study of the effects of increased body temperature on sheep
Ulberg and Burfening (1967) demonstrated damage to both ovum and sperma-
tozoon due to the physical stress of increased air temperatures. If
the physical stress was placed on the animals after fertilization the
embryo died in utero some time later in its development.
The preimplantation blastocyst depends for survival upon the
maintenance of a delicate balance of uterine fluid components. Kar,
Engineer, Goel, Kamboj, Dasgupta and Chowdhury (1968) determined the
principal effect of the intrauterine contraceptive device (IUD) to be
a change in the biochemical composition of the uterine fluid. In studies
of IUD fitted parous women a significant increase in the total protein and
non-protein nitrogen levels of the uterine fluid was described. Such
changes were brought about by the lysis of cast off epithelial cells and
infiltrated polymorphonuclear leukocytes. These changes produced an un-
favorable environment for the development and survival of the blastocyst.
A similar finding was reported by Kar, Goswami, Kunmboj and Chowdhury
(1964) in ovariectomized IUD fitted rats treaLed with estrogen. Inter-
ference with the response of target organs to estrogen was also noted.
IUD mediated interference with gonadotropin secretion in heifers was stud-
ied by Bhalla, Menon, Woody and Casida (1969). The IUD fitted animals
were found to have significantly heavier pituitary glands and lower lute-
inizing hormone levels than the control animals.
The presence of potential substrates for bacteria in the lumen
fluids of the reproductive tract has already been noted (Olds and VanDemark,
1957; Kar et al., 1968). A basis for the tissue specificity of some bac-
terial species was studied by Keppic, Williams, Witt and Smith (1965).
A growth factor, erythritol, which was found to occur in the placental
tissues and fetal fluids of the cow, enhanced the virulence of B. abortus
for the guinea pig in vivo. The authors fullher described growth stimu-
lation of B. abortus when erythritol was used as a substrate in vitro.
An earlier study by Pearce, Williams, Harris-Smith, Fitzgeorge and Smith
(1962) clearly demonstrated that erythritol enabled B. abortus to sur-
vive and multiply in bovine phagocytes.
In vitro studies of V. fetus by Osborne and Bourdcau (1955) indi-
cated an increase in the growth rate of the organism when it was grown in
a basal thioglycollate broth with separate additions of progesterone,
corpus luteum extracts and testosterone. It was suggested that a possible
relationship might exist between steroid honaone levels and V. fetus
virulence. Growth stimulation of V. fetus in the presence of 0.005% 17,
beta-estradiol was observed by Zemjanis and Hoyt (1960). From manometric
studies of possible energy sources used by V. fetus, Alexander (1957) de-
termined that lactate effectively stimulated the growth of this organism.
The presence of lactate in bovine follicular fluid was shown by Lutwak-Mann
(1954). The presence of lactate in the uterine fluid of the human female
was demonstrated by Kar et al. (1968).
Nalbandov (1964) noted distinct differences in the response of
uterine tissues to bacterial infection. The uterus was found to be more
resistant to infection during the follicular phase of the cycle than dur-
ing the luteal phase. The presence of antimicrobial substances in ex-
tracts of the human endometrium was discussed by Kozinn, Pomerance, Caro-
line and Taschdjian (1968). Extracts from females were pooled without
regard for cyclic variation in chemical composition. When tested in vitro
they inhibited the growth of Neisseria gonorrhoeae, N. meningitidis and
N. haemolysans. The antimicrobial activity wa- associated with a fraction
of the protein content of the extract. The inhibitory action was not
effective against Salmonella sp., Escherichia sp., Proteus sp., Pseu-
domonas sp. or Staphylococcus sp.
The apparent hormonal regulation of the uterine defense mech-
anism was observed to be organ specific by Hawk, Simon, Cohen, McNutt
and Casida (1955). It did not extend to other organ defense mechanisms
in the body. No significant differences were observed in the numbers
of E. coli recovered from the peritoneal cavities of either rabbits in
estrus or pseudopregnant rabbits 28 hours after intraperitoneal inocu-
lation. However, more organisms were recovered from the pseudopregnant
uterus (progesterone influence) than from the estrous uterus (estrogen
influence). The organisms gained access to the uterus of the animal
through the fallopian tubes. Inoculations made into the pleural cav-
ities of rabbits estrouss and pseudopregnant) revealed a similar clear-
ance rate existed between the two groups and that the clearance was in
no way related to the hormonal condition of the reproductive system.
McDonald, Black, McNutt and Casida (1952) studied the effects
of bacteria-free semen and bacteriologically contaminated semen upon
the rabbit uterus. Neither the estrous nor the pseudopregnant uterus
reacted to the bacteria-free semen. When bacteriologically contaminated
semen was used as the inoculum the uterus of the pseudopregnant rabbit
developed marked pyometra. In contrast, the uterus of the rabbit at
estrus showed only a mild inflammatory reaction.
The development of pyometra in heifers and its absence in "re-
peat breeder" cows as a response to uterine bacterial infection was de-
scribed by Black, Ulberg, Kidder, Simon, McNutt and Casida (1953). Heifers
inseminated during the postestrual period with an E. coli contaminated
semen developed pyometra. The "repeat breeder" cows failed to develop
pyometra after similar insemination. The authors suggested that the
"repeat breeder" cows lacked the progesterone levels necessary to in-
hibit the natural uterine defense mechanism and that this inadequate
progesterone level was also responsible for the "repeat breeder" con-
The effect of direct inoculation ofa potentially pathogenic
organism into the uterine horn was described by Black, Simon, Kidder
and Wiltbank (1954). Both the ligated and unligated uterine horns of
estrous rabbits inhibited the growth of E. coli whereas the uterine
horns of pseudopregnant rabbits not only supported the growth of E.
coli but allowed a significant increase in cell numbers.
Rowson, Lamming and Fry (1953) reported the control of bovine
uterine infections by means of exogenous estrogen therapy. The ad-
ministration of exogenous progestrone was found to promote conditions
within the uterus to favor the growth of potential pathogens.
A spermicidal activity due to certain bacterial enzyme systems
was observed by Marinov (1967). Part of this spermicidal activity was
attributed to the elevated glucose dehydrase activity of E. coli, Staph.
aureusor Bacillus prodigeosum. When any one of these organisms was
added to semen 18-24 hours prior to insemination, conception rates de-
Walsh, Hildebrandt and Prystowsky (1965) incubated species of
vaginal bacteria in antibiotic-free human amniotic fluid. A signifi-
cant increase in cell numbers was observed. In contrast, Galask and
Snyder (1968) reported an increase in the lag phase together with growth
inhibition in similar work with antibiotic-free human amniotic fluids.
The bacterial genera used in this study were Proteus sp., Pseudomonas
sp., coliformss," Staphylococcus sp.and Streptococcus sp.
The presence of steroid hormones in amniotic fluid was reported
by Schindler and Suteri (1968) and Schindler and Ratanasapta (1969).
A decrease in the amniotic fluid steroid hormone level was observed in
human pregnancies following fetal distress or death. Osburn, Staben-
feldt and Ewing (1969) reported a sharp decrease in the plasma pro-
gesterone levels of cows prior to abortion.
V. fetus infection in the cow cannot be detected by clinical
examination due to the innocuous nature of the disease. Its presence
in a herd may be suspected when reductions of calf crops occur. Repeat-
ed returns of cows to service and mid term abortions are other signs of
possible herd infection (Laing, 1960). Accurate diagnosis requires
the isolation of the organism by cultural techniques (Lovell, 1964) or
the demonstration of significant antibody titers in serum or vaginal-
cervical mucus (Laing, 1960).
The true venereal vibriosis of cattle is caused by V. fetus var.
venerealis. Another variety of V. fetus called V. fetus intestinalis
causes some abortions in cattle but is primarily isolated from sheep.
Florent (1959) first described these two isolants thus;
V. fetus var. venerealis (Type I),
H2S negative, inhibited by 1% glycine or 1% sodium
selenite with failure to grow on a selective medium
containing brilliantgreen and bile salts.
V. fetus var. intestinalis (Type II)
H2S positive, able to grow on 1% glycine and in 1%
sodium selenite with subsequent reduction and with
growth occurring on a selective medium containing
brilliant green and bile salts.
Since certain isolants of V. fetus var. vencrealis were later found
to produce traces of H2S, a subtype 1 classification was suggested
by Florent (1963). The taxonomic position of V. fetus was challenged
by Sebald and Veron (1963). These authors recommended the introduc-
tion of a new genus, Campylobacter, to accommodate V. fetus and the
saprophytic V. bubulus.
Vibriosis is an asymptomatic infection in the bull. Samuel-
son and Winter (1966) studied the distribution of V. fetus var. vene-
realis in the preputial cavity and terminal urethra of the bull. The
organism was found throughout the preputial cavity with the greatest
number located in the preputial fornix and on the penis. The lumina
of the epithelial crypts were considered as the major sites of pro-
liferation. No lesions were observed. The authors were unable to de-
tect the presence of local antibody. V. fetus var. venerealis was not
considered an effective antigenic stimulus in the bull. Once estab-
lished, the infection in the bull may persist for many years (Laing,
The infection in the heifer, or cow, is apparently confined
solely to the reproductive tract and has been more widely studied than
the infection in the bull. The infection was noted to be acute at
first and later became chronic (Laing, 1960). A fresh isolant was
found to persist in the uterus of the virgin heifer for a period of
at least 13 weeks (Newsam and Peterson, 1964). The possibility of a
loss of virulence was suspected by these workers when a laboratory
adapted culture of V. fetus failed to presist when inoculated into the
uterus of the virgin heifer. Vanderplassche, Florent and Huysman (1957)
reported that V. fetus reached the uterus within five days after heifers
were bred to V. fetus infected bulls. The organisms were cleared from
the non-pregnant uterus within three weeks. Vaginal cultures of in-
fected cows were persistently positive for V. fetus for periods as long
as eight months.
The role of V. fetus as a disease of the fetal membranes of
cattle was early recognized by Smith (1918). From this work a series
of papers (Smith, 1919; Smith, Little and Taylor, 1920; Smith, 1923)
followed, in which the etiology of the infection and its role in bo-
vine abortion was described. V. fetus infection was not found to per-
sist with the tenacity shown by B. abortus (Smith et al., 1920). Cul-
tures of fetal stomach contents, spleen, liver and lungs were positive
for V. fetus. The experimental inoculation of pregnant cows resulted
in abortion with isolations of V. fetus from the stomach contents of
the aborted fetuses (Snith, 1919). The lesions described by Smith (1923)
as the result of V. fetus infection included mushy, yellowish cotyledons,
completely necrotic villi and edema of the chorion.
Smith (1918) stated that V. fetus showed a predilection for the
caruncles of the uterus and produced abortion by destruction of the
fetal-maternal vascular attachments. The evolution of a mild endome-
tritis, within three weeks of the establishment of the infection, in
the non-gravid uterus of the cow was described by Huysman (1957). It
progressed from the superficial layer and reached the spongiosa within
two months after the onset of the infection. The condition was found
to heal from the deeper to the superficial layers and healing was ap-
parently complete within three months from the onset of the infection.
Similar lesions and healing patterns were reported by Frank, Shalkop,
Bryner and O'Berry (1962). In addition to the endometrial changes
associated with a mild to moderate inflammatory reaction These authors
observed a definite neutrophilic and lymphocytic infiltration of the
endometrium. The relationship of the endometrial healing to the develop-
ment of an immune response was reported by Dozsa, Mitchell and Olson
(1962). Lymphocytosis was observed during the slow recovery of the
uterus. This was accompanied by an increased resistance to a rechal-
lenge inoculation with V. fetus. The new lesions produced by the
second infection were less severe than those initially observed when
the uterus first became infected. Peterson and Newsam (1964) described
lymphocytic and plasmacytic reactions in the infected endometrium of
the cow. No lesions were observed in the vagina. The presence of V.
fetus in the reproductive tract produced no lesions at all in some of
the experimental animals.
The occurrence of cervicitis and metritia as the result of
intravenous or intracutaneous injection of V. fetus growth products
into cows was demonstrated by Simon and McNutt (1957). Cervicitis
and metritis was observed in the non-gravid reporductive tract of
heifers in both naturally acquired and experimentally induced V. fetus
infections. Osborne (1965) studied the effects of whole cell and cell-
free supernatant fluids of V. fetus when injected intravenously into
calves, pigs, goats, sheep and rabbits. Both reversible and irreversible
bacterial shock resulted. Pregnancy was found to intensify the host
reaction to the toxic principle. Hypersensitivity was suggested as a
possible cause for abortion. A similar mechanism to explain V. fetus
associated abortion was proposed by Manclark and Pickett (1965). Dennis
(1959) isolated a toxic lipopolysacharide from V. feLus cells and dem-
onstrated its abortive action on rabbits, mice, guinea pigs and sheep.
Osburn et al. (1969) reported embryonal death within five to
20 days after the inoculation of V. fetus into the uterus of the
pregnant cow. Fetal death was followed by a decline in the plasma
progesterone level. This decline in the hormone level preceded abor-
tion and was attributed to placental dysfunction and luteolysis of
the corpus luteum.
The immune response of the cow to V. fetus infection was re-
ported by Manclark and Pickett (1965a). Cyclic variations of the
cervical-vaginal mucosal antibody titers were observed. The highest
antibody titers were determined at diestrus. Uterine infections with
V. fetus led to the appearance of earlier and higher genital mucus
antibody titers, which were greater than the antibody titers found in
Smith (1918) attempted to infect 40 guinea pigs with material
containing viable V. fetus cells. The animals were inoculated via the
subcutaneous and intraperitoneal routes. Cultures taken from the liver,
spleen and kidneys of each animal five days later were negative for
V. fetus. After this initial failure to recover the organisms, Smith
and Taylor (1919) and Smith (1923) inoculated guinea pigs intraperi-
toneally with tissues from V. fetus infected cows, and successfully
isolated V. fetus from the guinea pigs. Smith (1923) used three guinea
pigs, which were inoculated at the same time but euthanized on days
three, four and five post inoculation. Isolations of V. fetus were made
from the spleen, liver and kidney tissues of those animals euthanized
on days three and four post inoculation. Cultures taken from the animal
examined on day five post inoculation were negative for V. fetus. Smith
and co-workers (1918; 1919; 1920; 1923) made no reference to any gross
lesions in the guinea pigs studied. The use of the guinea piq in Lhis
early work was primarily to rule out the presence of B. abortus in the
cattle which were being examined for causes of unexplained abortion.
Lerche (1937) induced abortion in guinea pigs by means of intra-
peritoneal, conjunctival and subcutaneous inoculation of viable V. fetus
cultures. This author was unable to demonstrate V. fetus in the stomach
contents of an aborted fetus.
Ristic and Morse (1953) experimentally induced V. fetus infection
in the female guinea pig. A mixed inoculum of ovine and bovine isolants
was used. Seven pregnant and three non-pregnant guinea pigs were inocu-
lated intraperitoneally. V. fetus was either isolated by culture or
demonstrated microscopically from all the pregnant animals, four of which
aborted. The non-pregnant animals were negative on culture. Another
group of guinea pigs was inoculated with B. abortus, USDA strain 19 in
addition to V. fetus live cells. The pregnant animals were positive,
while all but one of the non-pregnant animals were culturally negative
In a second experiment (Ristic and Morse, 1953), pregnant and
non-pregnant guinea pigs were exposed to V. fetus by oral, subcutaneous
and intravaginal routes of inoculation. All the non-pregnant animals
were negative for V. fetus when cultured at necropsy 21-23 days post-
inoculation. Two of nine pregnant animals aborted following oral inoc-
ulation and two of seven pregnant animals aborted following subcutaneous
inoculation. Eleven pregnant animals delivered offspring after an un-
eventful gestation period. The intravaginal route of inoculation induced
a single abortion. V. fetus was isolated from the embryo. The re-
mainder of the animals inoculated intravaqinally were negative for
V. fetus when cultured 28 days after inoculation. It was proposed that
the guinea pig was a valuable animal in which to study the pathogenicity
of V. fetus if the intraperitoneal route of inoculation was used to
infect the animal.
Ristic, Wipf, Morse and McNutt (1954) described the lesions
observed in the preceding work (Ristic and Morse, 1953). Epithelial
sloughing, inflammatory cell reaction, hemorrhages, enlarged blood
vessels, myometrial and subendometrial edema, together with cystic
uterine glands were the major pathological changes listed. Histol-
ogical changes most frequently observed in the fetal tissues were
hemorrhages of the myocardium and liver. Black, Simon, McNutt and
Casida (1953) were unable to detect any appreciable pathologic change
in the estrous of pseudopregnant uterus or the rabbit after intra-
uterine inoculation with V. fetus.
Ristic, Morse, Wipf and McNutt (1954) were able to transmit V.
fetus from infected female guinea pigs to non-infected female guinea
pigs via coitus with a male as the carrier. In this study a sheep
isolant, strain K, was used as the inoculum. An attempt was made to
infect 10 females by this method and of these only two were found to
be positive when euthanized 43 days after mating. Four of the 10 ex-
perimentally infected females gave positive cultures for V. fetus. This
figure included two which aborted.
The use of the guinea pig as an animal for the diagnosis of V.
fetus infection was proposed by Adler (1953). This author attempted
to infect female guinea pigs by intravaginal inoculation during induced
estrus. A single intramuscular injection of estradiol benzoate f250 IU)
was given followed by another injection 72 hours later. Immediately
after the second injection of hormone the guinea pigs were inoculated
in:zavaginally either with a pure V. fetus culture cr eith V. fetus
contaminated seen. Six dais later the anirimls vir euthanized and
cultures ta!-en from the uterine 'lorns. when pure cultures of a fresh
isolant of V. fetus were inoculated, 10 of 11 animals were positive
for V. fetus. '. older, laboratory maintained stock culture of V.fetus
failed to infect guinea pigs. 'hnen infected bovine ser.un was used as
the inoculum, V. fetus was isolated from three of seven animals.
Further studies in the uso of the female guinea pig as a diag-
nostic animal were reported by Power (1954). A group of six females
was given diethylstilhestrol dipropionaee, intramuscularly, to induce
estrus. The animals were inoculated intravaginally with semen from a V.
fetus infected bull. At necropsy, seven days post inoculation, V. fetus
was isolated from three of the guinea pigs. Based on these results the
author recommended the use of at least six animals if any attempts at
diagnosis .were conserplated.
V. fetus was isolated from infected bull semen by intraperitoneal
inoculation into pregnant guinea pigs (Robinson, Van Rensburg, Van Heer-
den and Van Drirrmalen, 1956). The animals were euthanized five to seven
days after inoculaticn and the organism was recov'-ed ro:1 the sto-ach
contents of tne fetuses. Robinson et al. (1956) reported failure in
their attermts to infect two groups of stilbestrol treated female guinea
pigs with a viable culture of V. fetus.
Manclark and Pickett (1965a) stated that the changes which occurred
in the female guinea pig genital mucosa during the reproductive cycle com-
pared favorably :,.r.h those of the cow. From studies in which h the guinea
lc 'wa. used to observe the irmuse resornse to V. fetus infection, these
authors con ciud d that 7. fetus shc-:ed a predilecticn for the gravid
uterus, since pregnant or pseudopregnant guinea pigs were apparently
susceptible to infection. The non-pregnant guinea pig was not sus-
ceptible. The pattern of antibody formation was dependent upon the
route of inoculation. The inoculation of live cells by the intra-
peritoncal route led to the production of humoral antibody. The
organisms then localized in the uterus and stimulated the production
of mucosal antibody. Live antigen administered by the intravaginal
route stimulated the production of local, but not circulating, anti-
body. This same pattern of immune response was found to occur in the
The exact mechanism involved in the pathogenicity of V. fetus
and how it effects reproductive efficiency is not fully understood.
Bacterial species which produce infections of the female reproductive
tract and thus have the potential to adversely affect pregnancy may
be categorized into four major groups. 1) Those which produce endo-
toxins, i.e., Brucella sp., E. coli and V. fetus. 2) Organisms which
disrupt the pattern of endometrial metabolism by the production of
inflammatory reactions, i.e., Brucella sp. and V. fetus. 3) Organisms
which directly infect the fetus, i.e., Leptospira and 4) those organ-
isms which cause an imbalance of the uterine environment by the use
of available metabolites in the lumen fluids. Such competition could
affect the developing blastocyst before nidation.
V. fetus may apparently fall into all of the above listed cate-
gories. A study in which V. fetus var. venerealis is used to induce
an infection in the female guinea pig may help to clarify the mechanism
of pathogenesis of the organism on the reproductive tract and the embryo
and, at the same time, permit a reevaluation of the guinea pig as a
model in which to study the course of the infection.
MATERIALS Af:D lSTM'fDS
Principle characteristics of V. fetus var. ven realis used in Lnis
V. fetus var. venarealis is a Gram negative, motile, mono-
flagellat-d, slightly curved rod. It is catalase positive, H2S neg-
ative and does not grow in media which contain 3.5% sodium chloride
or 1.0% glycine. Growth occurs in 1.01 oxgall and in 0.1% sodium
selenite. A gas mixture of 85% N2' 101 CO2 and 5% 02 is required
in which the organism grows at 25 C and 37 C b-t not at 42 C.
Source and maintenance of the isolants
The isolants of V. fetus var. venarealis were recovered from
the reproductive tracts of cows slaughtered in Florida. These isolants
were characterized and kindly made available to the author by Dr. F. H.
White of the Department of Veterinary science, University of Florida.
After initial isolation and identification the isclants were preserved
in the lyophiliz-d state. To avoid repeated Iessaa of th- cultures
after recovery from lyophilization, batches of 24-48 hour thioglycollate
broth cultures were deep frozen at -80 C for subsequent use in the lab-
oratory and nr-iral studies.
Only smooth isolants were used throughout the study. Regular
checks were made on the cultures to ensure tile absence of dissociation.
These checks -3.r :v.e by pate r.-icrosco:i exsain.atl n cf bacterial
colonies isolaTad on Brucalla agar (\1--,.:. .
Basal media for growth. Thioglycollate broth (Difco) without
dextrose or indicator was distributed in 10 ml volumes in 16x150 mm
screw-cap culture tubes. Brucella broth with 0.16% added agar was
similarly prepared. Brucella broth (Albimi) without added agar was
distributed in 30 ml volumes to 50 ml capacity Ehrlenmeyer flasks and
in 45 ml volumes to Nepheloflasks.* The media were sterilized at 121 C
for 15 minutes
Solid media for the isolation of V. fetus and for viable cell
counts. Selective bovine blood agar plates containing brilliant green
and Novobiocin were prepared by the method of Lovell (1964). Brucella
agar plates were prepared by the addition of 2.0% agar to Brucella broth
(Albimi). The latter medium was sterilized at 121 C for 15 minutes.
Dilution blanks for viable cell counts
Deionized distilled water was added to screw-cap dilution bottles
in 99.0 ml volumes. The water was sterilized at 121 C for 15 minutes.
Production of the required gaseous environment for the growth of Vibrio
A large desiccator was used as a growth chamber. A desiccant of
one part calcium chloride and three parts activated silica gel** was added
to avoid excessive condensation of moisture. The growth chamber was used
for the isolation of v. fetus on solid media or to propagate the organism
in small flasks of Brucella broth without agar. The flasks or petri-
dishes were placed in the desiccator and a negative pressure of 590 rm Hg
*Corning Glass Company, New York, N. Y.
**Tel-Tale, a product of Davison Chemical, Inc., Baltimore, Md.
was produced with a vacuum pump. C02 was used to decrease the neg-
ative pressure to 490 inm Hg, The negative pressure was then further
reduced to 50 mm lig by the addition of N2. Each gas was delivered
from a separate cylinder. The final atmosphere in the desiccator
was approximately 10% CO2, 85% N2 and 5% 02, the latter derived from
the residual air. A partial negative pressure remained.
To grow the organism in volumes of fluid media greater than
30 ml and to allow the production of a gaseous environment in the
Nepheloflasks, a commercially manufactured gas mixture* of 10% C02,
85% N2 and 5% 02 was used.
To sterilize the gas mixture a filter assembly was constructed
in the laboratory. A large glass combustion tube, 100 cm x 4 cm,was
packed with glass wool. Non-absorbent long fiber cotton was placed
in each end at the top and bottom of the glass wool column. The ends
were closed tightly with black rubber stoppers through which fire
polished glass tubing, 0.5 cm outside diameter, was passed. One end
of the filter carried a length of rubber tubing with the free end
closed with a spring clip. The other end of the filter was left open.
The entire assembly was wrapped in Kraft paper and sterilized at 121 C
for 15 minutes.
To gas a fluid culture the filter was connected by the open end
to the gas cylinder reducing valve. A sterile pipette was attached to
the clamped rubber tubing and the clamp was released. After inoculation
the flask was opened aseptically and the gas mixture passed into it with
the tip of the pipette held just above the surface of the medium. The
4Airco, Inc., Jacksonville, Fla.
flask was gassed for 30 seconds with the flow regulated at two pounds
per square inch. The pipette was removed from the flask which was
closed tightly with a sterile rubber stopper. A new sterile pipette
was used for each flask gassed.
Hormones and organic compounds used in the in vitro growth experiments
and for the regulation of the hormonal states of the guinea pigs
In vitro growth studies.
Follicule stimulating hormone (F.S.H.), porcine*
Estrone (1,3,5, (10)-estratrien-3-ol-17-one)*
Regulation of the hormonal states of the guinea pigs.
Progynon Benzoate*** (estradiol benzoate in oil 1 mgm per ml)
Progesterone solution (Repository type, 25 mgm per ml)****
Histological examination of tissues
All tissues were fixed in formalin, embedded in paraffin and
sectioned at six microns. Sections were stained with hematoxylin and
Mature, non-gravid female and mature male guinea pigs***** were
used in mating studies and experiments in which the intravaginal route
of inoculation was used. Ovariectomized and pregnant guinea pigs******
were used in studies which involved intrauterine inoculation. The animals
were of mixed breed and English short-hair varieties. They were identified
*Mann Research Laboratories, Inc., New York, N. Y.
**Matheson, Coleman and Bell, East Rutherford, N. J.
***Schering Corporation, Bloomfield, N. J.
****Norden Laboratories, Lincoln, Neb.
*****Kel Farm, Alachua, Fla.
******Camm Research Institute, Wayne, N. J.
by color and cage number. White animals were stained on the back to
The animals were housed in stainless steel cages with one-half
inch mesh floors, except for pregnant females in the late stages of
gestation. These animals were housed on solid floors.
A diet of Purina guinea pig chow was fed, supplemented with
fresh cabbage and hay twice weekly. Water was given ad libitum.
The animal quarters were maintained at a temperature of 68 F
+ 5 F. Normal daylight was provided in the animal room. Male animals
were housed in separate facilities but under the same conditions as
the females. Females were taken to the males for breeding experiments.
Mating was polygamous with two to three females to each male.
Determination of female estrus cycles. Estrus cycle lengths
were determined by examination of the vaginal cytology (Asdell, 1946).
Daily vaginal smears were taken with a small blunt-ended dropper fit-
ted with a rubber bulb. A small volume of 0.85% saline was washed
into the vagina and withdrawn. The material collected was transferred
to a clean glass slide and allowed to air dry. The smears were stain-
ed with a modified Wright Giemsa stain technique. Wright stain was
applied to the smear for one minute to effect fixation. The Wright
stain was diluted, 1:4, on the slide with Wright buffer and the smear
stained for six minutes. The smears were washed with distilled water
and the excess water drained off. The smears were next stained with
a dilution of 1:50 Giemsa stain and Wright buffer for 1 1/2 hours. Fi-
nally the smears were rinsed in distilled water and air dried. The
smears were examined microscopically at magnifications of X450 and X950.
Cytological data for each animal were collected over a period of 20
days and recorded. The animals were then placed into proestrus,
luteal and follicular phase groups. The mean cycle length was
16 + 2 days.
In vitro growth studies V. fetus var. venerealis
Experiment 1. Erythritol as a growth factor. Stock solu-
tions of erythritol were prepared. Solution 1 contained 2 gm of
erythritol in 4 ml of distilled water (W/V). Solution 2 contained
0.3 gm erythritol in 10 ml of distilled water (W/V). These solutions
were sterilized by passage through a 0.22 pore size membrane filter.*
Volumes of Brucella broth (Albimi) were prepared in duplicate Nephelo-
flasks so that the final volume of liquid medium after the addition
of the inoculum and erythritol solution was 50 ml. The broth was
added to the flasks before sterilization and the erythritol was added
aseptically after the medium was cooled. The flasks of medium were
prepared as follows.
ml of stock ml of Final cone
solution and Brucella ml of erythritol
Flask No. solution No. broth inoculum per ml
1 and 2 0.1, sol 2 44.9 5.0 0.06 mgm
2 and 3 0.1, sol 1 44.9 5.0 1.0 mgm
4 and 5 0.3, sol 1 44.7 5.0 3.0 mgm
5 and 6 0.5, sol 1 44.5 5.0 5.0 mgm
7 and 8 1.0, sol 1 44.0 5.0 10.0 mgm
9 and 10 control 45.0 5.0
Seed cultures of V. fetus were prepared in 30 ml of Brucella broth
without added agar. These cultures were placed in the gas jar and
incubated at 37 C on a New Brunswick Model G incubator shaker at 100
oscillations per minute. After an incubation period of 36 hours the
*Millipore Filter Corporation, Bedford, Mass.
seed culture ..as checked for r.ozility, .c:rphol.or- and -I rity. Syn-
chronization of the culture was effected by .-s2ge through five
layers of sterile -hat-nan r.nub_- -:. filter _.- -' under as-ptic con-
An inoculu of 5 ml synchr-nized cells suspended in -ru-
cella broth was added to the test and control "epheloflask,. These
were gassed and stopnered tightly with black rubber stolpers. The
optical density (OD) c_ e-ch flask was read at tire zero on a Bausch
and Lomb Spectrcnic 20 spectrophotom.eter at a wavelength of 525 mr .
Brucella broth was used as a cclorie.ter blank.. The !.-epheloflasks
were incubated on the shaker at 37 C and s-aken at 100 oscillations
per minute. The OD of each flask was read at intervals bet.wee eight
to 12 hours until the exponential growth phase of the cultures was
completed. A final check of cell morphology was made at the end of the
Graphs of OD against time were plotted for each test dilution
and control. The slopes of these graphs were determined. The data
were examined for significant differences between the test and con-
trol growth curves by an analysis of variance and Duncan's 'ultiple
Range Test (Steel and Torrie, 1960).
Experiment 2. Estrone as a qro:t"' factor. A stock solution
of estrone was prepared by the addition of 15 ml of analytical grade
chloroform (99.9% pure) to 0.06 gm of estrone. Nepheloflasks which
contained 35 -.1 of Brjcella broth, without add.- agar, were sterilized
at 121 C for 15 minutes and allowed to cool. Stock estrone solution
was alf e: a=-_ttic]-'- to the' flas, is follo- s.
Volume of estrone ml of final cone
Flask No. solution il inoculum of estrone 5
1 and 2 1.0 5.0 0.01
3 and 4 0.9 5.0 0.009
5 and 6 0.7 5.0 0.007
7 and 8 0.5 5.0 0.005
9 and 10 0.3 5.0 0.003
11 and 12 0.1 5.0 0.001
13 and 14 0.05 5.0 0.0005
Controls (2) 5.0
A series of reagent blanks were prepared by the addition of estrone in
the above volumes to Nepheloflasks which contained 40 ml of sterile
Brucella broth but were not inoculated with V. fetus. The chloroform
was removed from the test and reagent blank flasks by gentle steam
heat applied to the outside of the flasks. The hormone precipitated
in the broth as the chloroform was removed. The test and control flasks
were inoculated with 5 ml of a 36-hour synchronized seed culture of
V. fetus. Each was gassed and tightly stoppered.
An initial OD was read at time zero against a blank of Brucella
broth. The optical densities of the test hormone dilutions were cor-
rected by the subtraction of the reagent blank OD from its respective
test flask OD. The flasks were incubated on the gyrotary shaker at
37 C. The OD was determined at regular intervals throughout the ex-
ponential growth phase of the organism.
A growth control to determine the effects of the chloroform on
the growth of V. fetus was prepared by the addition of chloroform to
Brucella broth followed by its removal by gentle stean heat and
inoculation with 5 ml of the synchronized culture. This control was
incubated under the same conditions as the tests and non-chloroformed
controls. The results of both controls were subsequently compared
(see Appendix A).
Data collection, statistical analyses and .:-rphological
checks of the cultures w .re identical to those described in Experi-
Experir. nt 3. Procesterone as a growth factor. A stock
solution of progesterone was prepared by the addition of 15 nl of
analytical grade chloroform to 0.06 gn of progesterone. The test was
performed in accordance with the procedure described in Experinent 2.
Experirent 4. Estrone and progesterone as cr i'inrd ro:th
factors. 'epheloflasks of 35 ml of Brucella Lro th without added agar
were sterilized at 121 C for 15 minutes. Stock solutions of estrcr-o
and progesterone were prepared (see Experiments 2 and 3 above). These
were added to the flasks of broth in the following manner.
Vol. Estrone terone Vol. of Final estrone
stock sol stock sol inoculum progesterone
Flask No. ml ml n] cone %
1, 2 and 3 0.1 0.9 5.0 0.001/0.009
4, 5 and 6 0.3 0.7 5.0 0.003/0.007
7, 8 and 9 0.5 0.5 5.0 0.005/0.005
10, 11 and 12 0.7 0.3 5.0 0.007/0.003
13, 14 and 15 0.9 0.1 5.0 0.009/0.001
Control cultures, reagent blacks and colorimeter blanks were prepared
by the methods described in Experiment 2. Inoculation, incubation and
data collection and examination were similarly performed.
Experi-ent 5. Follicule Stimulating Horrone (FSH) as a growth
factor. Nepheloflasks of Brucella broth (w/o agar) were sterilized at
121 C for 15 vinuLes. A vial of FSH (pcrcine) which contained 50 ngn
of hormone was reconstituted with 10 ml of sterile distilled water.
A series of dilutions o' FSii Et: broth was prepared i.i o-,c cooled mediri
vol. of Final F3i:
Stock FS'. Vcl-e of inoculum cone %
Flask No. sol ml broth n! ml
1, 2 and 3 0.8 34.20 5.0 0.01
4, 5 and 6 0.4 34.60 5.0 0.005
7, 8 and 9 0.24 34.76 5.0 0.003
10, 11 and 12 0.08 34.92 5.0 0.001
13, 14 and 15 0.04 34.96 5.0 0.0005
Controls 35.00 5.0
Test and control flasks were inoculated with 5 ml of a 36-hour
synchronized culture, gassed and tightly stoppered. incubation,
data collection and culture checks were performed by the methods
described in Experiment 1. A statistical analysis of the data was
also made by the methods outlined in Experiment 1.
V. fetus var. venerealis infection of the non-gravid guinea pig
Excerirent 6. Intrav qinal inoculation d'rinr normal follic-
ular and luteal reproductive cycle phases. Two groups of three vir-
gin female guinea pigs each were selected on the basis of vaginal
cytclccy into follicular ard luteal phase groups. All the animals
were mature and weighed between 450-500 gm. A control group comprised
two animals, one in the follicular chasee of the reproductive cycle and
the other in the luteal phase. Preliminary vaginal cultures of all
the animals were plated to selective blood agar plates which were
incubated in the gas mixture a- 37 C.
A culture of V. fetus was prepared in Brucella broth for use
as the inoculum. A check of motility, norpholog/ and cultural char-
acteristics was made before the inoculation procedure was undertaken.
A viable cell count 'as vrfo-n-ed on the broch culture.
One ml of the broth culture for each inoculuj required was
was centrifuged at 3000 G to concentrate the cells. The super-
natant fluid was decanted and the cell pellet resuspended in 0.3 ml
of Brucella broth. The animals in each test group were inoculated
intravaginally on each of three days with 0.3 ml of concentrated cell
suspension (approximately 1 x 107) cells. The inoculum was placed in
the vagina with a sterile glass serological pipette fitted with a
rubber bulb. The control animals were inoculated with 0.3 ml of cell
free Brucella broth over the same time period.
Before the administration of the inoculum on the second and third
days of the inoculation period, a vaginal culture and smear for Gram
stain was taken from each test and control animal. Vaginal cultures
and smears were also taken from the test and control animals for three
consecutive days following the final inoculation.
Five days after the final inoculation each test and control ani-
mal was euthanized. The entire reproductive tract was aseptically
removed and placed in a sterile petri dish. Cultures were taken from
the vagina, proximal and distal cervical area, both uterine horns, both
ovaries, liver, spleen and blood from the heart. These were plated on
selective blood agar plates and into thioglycollate broth. Portions
of the uterine horns were opened with sterile scissors and placed into
thioglycollate broth without dextrose or indicator. Vaginal and cer-
vical cultures were plated to selective blood agar plates and to thi-
oglycollate medium. The ovaries were cultured in thioglycollate broth.
Any organisms which grew in the thioglycollate broth cultures wore sub-
cultured to selective blood agar. All solid mediums were incubated
at 37 C in the gas mixture. Uterine and cervical tissues were pre-
served in 10% formalin for histological examination.
Experiment 7. Intravaginal inoculation of the normal estrus
guinea pig. Twenty-four mature virgin female guinea pigs between
500-600 gm in weight were divided into a test group and a control
group with 12 animals in each group. A preliminary vaginal culture
was made on each animal to determine a vibrio free status (see Ex-
periment 6). Vaginal cytological examinations were performed and
each test animal in the latter stages of the follicular phase of
the cycle was inoculated with approximately 3 x 107 viable cells
daily until estrus was observed. Control animals at the same phase
of the reproductive cycle were inoculated with sterile Brucella broth.
The inoculum was prepared and given by the methods described in Ex-
Both test and control animals were mated to fertile males in
the ratio of two or three females to a single male. The females were
examined weekly for pregnancy by gentle abdominal palpation and re-
moved from the male as soon as signs of pregnancy were observed. All
pregnancies which resulted were allowed to continue to term. Maternal
or fetal deaths, stillbirths or other complications of pregnancy were
investigated by necropsy and extensive cultural examination. In the
case of maternal death in the test group, the stomach contents of all
the embryos in utero were cultured in thioglycollate broth. The ab-
dominal and thoracic viscera of the embryos were similarly cultured.
Selective bacteriological examination of the fetal-maternal membranes
and reproductive tract of the dam was made for V. fetus. In addition,
routine bacteriological culture techniques were used to enable the
isolation of other potentially pathogenic organisms. Stillborn off-
spring and offspring which died within 48 hours of birth were necropsied
and the stomach contents cultured for V. fetus. Routine bacteri-
ological examination was also made of the liver and spleen. In
cases of maternal death, histological examinations were made of
the fetal-maternal membranes and the uterus of the dam. Any com-
plications of pregnancy which occurred in the control animals
were similarly investigated.
The reproductive capability of the test and control ani-
mals was examined for significant difference by the Student's "t"
test (Steel and Torrie, 1960).
Experiment 8. Intravaginal inoculation during induced
estrus. Six virgin female guinea pigs weighing between 500-600 gm
were used as test animals. Two similar animals were exogenous
hormone-free controls. The vibrio-free status of all the animals
was confirmed by vaginal cultures which were plated on selective
blood agar and inoculated into thioglycollate broth.
All the test animals received two intramuscular injections
of 300 TU estradiol benzoate, the second injection 72 hours after
the first. Vaginal smears for cytological study were taken at the
time of the second injection to determine the presence of induced
estrus. Cytological studies were also made on vaginal smears of
the control animals to detect proestrus conditions in their repro-
Immediately after the second injection of hormone an inoc-
ulum of approximately 3 x 105 viable cells was given intravaginally
to both test and control animals. The inoculation was repeated after
24 and 48 hours. Vaginal smears for Gram stain and vaginal cultures
were taken at the same time intervals described in Experiment 6.
Five days after the final inoculation the animals were euthanized.
Necropsy, bacteriological and histological examinations of test and
control animals were made by previously described methods (see Ex-
Experiment 9. Intravaginal inoculation of progesterone treated
females. Six virgin female guinea pigs weighing between 500-600 gm
were used as test animals. The control group comprised two similar
animals. Each test animal was given 2.5 mgm of progesterone intra-
muscularly daily for three days. Preliminary vaginal cultures were
taken on all the animals to confirm a vibrio-free status. A check
of vaginal cytology was made to ensure the effectiveness of the
exogenous hormone treatment.
After the third hormone injection each test animal was inoc-
ulated intravaginally with an inoculum of approximately 3 x 106 viable
cells. The control animals were similarly inoculated. The inocula-
tion was repeated at 24 and 48 hours. The study was completed by
the methods employed in Experiment 6.
Intrauterine inoculation of ovariectomized guinea pigs treated with
Experiment 10. Inoculation during exogenous estrogen admin-
istration. Twelve ovariectomized guinea pigs of 500-550 gm weight
were given a single intramuscular injection of estradiol benzoate
(1000 IU). The animals were prepared for surgery 72 hours after the
hormone injection. The hair coat was clipped and shaved from the
abdomen and solid food was withheld for a 24-hour period. Anesthesia
was induced with methoxyfluorene* and the animal placed in a position
*Metofane, a product of Pitman-Moore, Dow Chemical Co., Indianapolis,
of dorsal recumbancy on a small laboratory animal board.* A 3.0 cm
incision was made in the abdominal wall and the uterus exposed through
the incision. An inoculum of 1.2 x 106 viable cells of V. fetus con-
tained in 0.2 ml of Brucella broth was introduced into the right horn
of the uterus midway between the point of bifurcation and the fallopian
tube. The inoculum was delivered from a tuberculin syringe fitted with
a 27-gauge needle. The left horn of the uterus was injected with 0.3 ml
of sterile Brucella broth for use as a control. The incision was closed
with 000 gut and surgical clips. The animals were observed until re-
covery from anesthesia was complete.
The guinea pigs were euthanized in groups of four at 24, 48 and
72 hour intervals post-surgery. A necropsy was carried out under
aseptic conditions. The viscera were examined for gross lesions and
any fluid in the peritoneal cavity was cultured in thioglycollate
broth. The entire reproductive tract was removed aseptically. A
ligature was tied around the right uterine horn at its junction with
the fallopian tube to facilitate orientation of the tract after its
removal. The two uterine horns were dissected free from the cervix and
the fallopian tubes. Both horns were flushed with 5 ml of sterile
phosphate buffered saline (pH 7.2) and the washings collected into
sterile centrifuge tubes. A viable cell count was performed on 1 ml
of the saline washings from the right horn. The 4 ml volume which re-
mained was centrifuged at 3000 g and the sediment cultured on selective
blood agar plates and in five tubes of thioglycollate broth. Smears
were also prepared for Gram stain and cytological study. Transverse
*Germfree Laboratories, Inc., Miami, Fla.
sections of both uterine horns were prepared for histopathological
The cervix was cultured on selective blood agar plates and in
thioglycollate broth. Transverse sections were taken for histopath-
ological study. The vagina was similarly cultured but was not his-
tologically examined. The liver, spleen and blood from the heart of
each animal were cultured for V. fetus. In addition, cultures were
taken from the liver, spleen, vagina, uterus, cervix and peritoneum
for organisms other than V. fetus. These were plated to bovine blood
agar plates and Eosinmethylene blue medium and incubated at 37 C.
Experiment 11. Inoculation during exogenous progesterone ad-
ministration. Twelve ovariectomized guinea pigs weighing 500-550
gm were primed with 50 IU of estradiol benzoate given intramuscularly.
Twenty-four hours later an intramuscular injection of 2.5 mgm pro-
gesterone was given. This was repeated 48 hours later. Twenty-four
hours after the second injection of progesterone the animals were taken
Preoperative preparation, operative technique, inoculation and
subsequent cultural and histological examinations were identical to
those employed in the exogenous estrogen study (see Experiment 10).
Exposure of the gravid guinea pig to V. fetus
Experiment 12. Intrauterine inoculation of guinea pigs. Thir-
teen pregnant guinea pigs in the 30th day of gestation were separately
housed and identified. An inoculum of 3 x 106 viable cells in 0.2 ml
of Brucella broth was prepared (see Experiment 6). Eleven of the ani-
mals were inoculated directly into the right uterine horn with 0.2 ml
of the whole broth culture. The inoculum was placed in the amnionic
cavity but directed away from the embryo. The control animals were
sham operated and inoculated in a similar manner with 0.2 ml of
sterile Brucella broth. The surgical technique employed was described
in Experiment 10.
Two animals which aborted within 24 hours after inoculation
were euthanized. A necropsy was performed under aseptic conditions
and tissues from the reproductive tract, placentae, liver and spleen
of the dams were cultured for V. fetus and other potential pathogens.
The thoracie and abdominal viscera of all fetuses, whether aborted
or in utero,were cultured in thioglycollate broth. If the stomach of
an embryo contained fluid this was cultured separately in thioglycollate
broth. Tissues from both the dams and the aborted fetuses were pre-
served for histological examination. A single animal which aborted
48 hours after inoculation was euthanized and cultured aseptically
at necropsy. Tissues were saved for histological examination.
Two animals were euthanized 72 hours after inoculation and
bacteriological and histological examinations wore made. Two animals
which did not abort after inoculation were allowed to go to term.
The animals which aborted after inoculation and were not euthanized
were cultured daily from the vagina until negative cultures for V.
fetus were obtained.
In vitro studies of V. fetus var. venerealis
Experiment 1. Erythritol as a growth factor. The recorded
optical densities and mean changes in OD per unit time of the V.
fetus cultures are given in Table 1. The slopes of the curves de-
rived during the unit time interval are given in Appendix A. The
analysis of variance of these slopes gave a calculated F value of
4.11. The tabulated F values at the 0.01 and 0.05 levels of sig-
nificance were given as 8.75 and 4.39 respectively. No signifi-
cant differences were found between the test and control slopes.
The organisms maintained their typical curved morphology through-
out the experiment and no apparent change in the size of the cells
Experiment 2. Estrone as a growth factor. The recorded
optical densities and mean changes in OD per unit time of the V.
fetus cultures are given in Table 2. The slopes of the curves
derived during the unit time interval are given in Appendix A.
The analysis of variance of these slopes gave a calculated F. value
of 3.32. The tabulated F values at the 0.01 and 0.05 levels of
significance were given as 6.18 and 3.50 respectively. No sig-
nificant differences were found between the test and control slopes
(see Appendix A). There was no apparent change in the morphology or
size of the cells during the experiment.
Table 1. Optical density readings of V. fetus cultures with erythritol added.
Erythritol concentrations (mgm per ml)
Time 0.06 1.00 3.00 5.00 10.00 Controls
hours A B A B A B A B A B A B
0 .032 .040 .040 .030 .028 .029 .040 .045 .030 .030 .030 .032
[8 .130 .130 .142 .140 .128 .135 .140 .140 .120 .120 .140 .1451
16 .260 .245 .315 .345 .300 .355 .330 .300 .300 .330 .330 .350
[24 .450 .405 .530 .470 .508 .470 .470 .560 .500 .470 .550 .508J
32 .560 .540 .530 .470 .520 .470 .470 .600 .510 .460 .600 .500
44 .600 .590 .590 .450 .500 .440 .430 .610 .460 .420 .700 .470
.297 .359 .357 .375 .360 .386
A and Bare replicates of each erythritol concentration.
*Mean change in OD in 16 hours determined between [8 and 24] hours.
OD readings taken at 525 mp .
Estrone concentrations (%)
0.01 0.009 0.007 0.005 0.003 0.001 0.0005 Controls
A B A B A B A B A B A B A B A B
.120 .120 .120 .120 .120 .120 .120 .120 .120 .120 .120 .120 .130 .120 .120 .120
.180 .170 .180 .180 .170 .170 .180 .180 .170 .170 .170 .170 .160 .180 .160 .1601
.240 .240 .240 .240 .210 .220 .220 .220 .220 .230 .220 .230 .230 .230 .220 .2201
.310 .310 .300 .300 .250 .290 .280 .280 .280 .280 .280 .280 .290 .290 .270 .280
.360 .340 .350 .350 .300 .330 .330 .340 .330 .330 .330 .320 .340 .330 .310 .300J
.380 .360 .370 .360 .320 .380 .370 .370 .360 .360 .360 .360 .370 .350 .350 .350
.460 .440 .450 .444 .380 .450 .450 .430 .430 .420 .430 .430 .440 .440 .440 .460
.510 .490 .480 .470 .430 .470 .490 .480 .510 .500 .500 .480 .510 .490 .510 .500
.490 .490 .480 .490 .420 .490 .490 .500 .510 .500 .500 .480 .510 .500 .520 .520
* .175 .170 .145 .155 .160 .155 .165 .145
Experiment 3. Progesterone as a growth factor. The recorded
optical densities and mean changes in OD per unit time of the V. fetus
cultures are given in Table 3. The slopes of the curves derived dur-
ing the unit time interval are given in Appendix A. The analysis of
variance of these slopes gave a calculated F value of 2.53. The tab-
ulated F values at the 0.01 and 0.05 levels of significance were
given as 6.37 and 3.50 respectively. No significant differences were
found between the test and control slopes (see Appendix A). The or-
ganisms maintained their typical curved morphology throughout the ex-
periment and no apparent change in the size of the cells was observed.
Experiment 4. Estrone and progesterone as combined growth
factors. The recorded optical densities and mean changes in OD per
unit time of the V. fetus cultures are given in Table 4. The slopes
of the curves derived during the unit time interval are given in
Appendix A. The analysis of variance of these slopes gave a cal-
culated F. value of 8.76. The tabulated F values at the 0.01 and
0.05 levels of significance were given as 5.06 and 3.11 respectively.
Significant differences were found between the slopes. Duncan's
Multiple Range Test when applied to the data at the 5% level of sig-
nificance indicated that the slopes of test concentrations of 0.005/
0.005, 0.007/0.003 and 0.009/0.001% estrone/progesterone were sig-
nificantly greater than the slopes of the hormone-free controls but
were not significantly different from each other. The slopes of test
concentrations of 0.005/0.005 and 0.009/0.001% estrone/progestrone
were significantly greater than the slopes of test concentrations of
0.001/0.009 and 0.003/0.007% estrone/progesterone (see Appendix A).
The organisms underwent no apparent changes in size or morphology
Progesterone concentrations (%)
0.01 0.009 0.007
A B A B A B
.220 .220 .210 .210 .210 .210
.280 .280 .270 .270 .280 .290
.320 .320 .350 .340 .320 .330
.365 .360 .370 .350 .370 .360
.410 .400 .410 .390 .400 .420
.430 .425 .450 .420 .430 .430
.500 .450 .510 .440 .440 .450
0.005 0.003 0.001
B A B A B A B
.210 .220 .215 .180 .200 .220 .220
.250 .230 .225 .240 .245 .240 .240]
.275 .270 .280 .280 .270 .320 .280
.300 .310 .315 .315 .300 .350 .310
.345 .360 .360 .340 .330 .390 .370
.365 .400 .415 .385 .380 .425 .390
.410 .400 .410 .410 .400 .460 .420
* .125 .130 .125 .095 .092 .132 .092 .140
Estrone and progesterone concentrations (%)
0.003/0.007 0.005/0.005 0.007/0.003
A B C A B C A B C
0 .310 .240 .260 .230 .220 .230 .160 .170 .160
0 .350 .330 .340 .310 .300 .300 .250 .260 .260
0 .420 .400 .400 .390 .370 .370 .310 .340 .340
0 .470 .450 .460 .460 .440 .440 .380 .400 .380
0 .490 .470 .480 .450 .440 .430 .400 .400 .400
0 .510 .490 .490 .500 .470 .470 .430 .440 .450
A B C
.190 .240 .32
.290 .360 .29
.350 .410 .36
.390 .460 .42
.380 .470 .38
.420 .520 .47
A B C
.210 .200 .210
.290 .290 .290
.360 .350 .370
.440 .420 .450
.450 .420 .440
.480 .480 .510
A B C
.120 .130 .130
.190 .200 .2101
.250 .260 .270
.300 .300 .310J
.330 .320 .330
.350 .370 .370
* .110 .120 .143 .130 .146 .103
during the experiment.
Experiment 5. Follicle stimulating hormone as a growth fac-
tor. The recorded optical densities and mean changes in OD per unit
time of the V. fetus cultures are given in Table 5. The slopes of
the curves derived during the unit time interval are given in Appendix
A. The analysis of variance of these slopes gave a calculated F value
of 2.42. The tabulated F values at the 0.01 and 0.05 levels of sig-
nificance were given as 5.06 and 3.11 respectively. No significant
differences were found between the test and control slopes (see Ap-
pendix A). There was no apparent change in the morphology or size of
the cells during the experiment.
The effect of chloroform treated medium on the growth of V. fetus.
The means of the slopes of growth curves derived from V. fetus cultures
grown in Brucella broth were compared to the means of the slopes of
growth curves of V. fetus grown in chloroform treated Brucella broth.
The Student's "t" test gave a calculated value of less than 1. No sig-
nificant difference was found between the two growth curve means (see
In vivo studies of the non-gravid guinea pig
Experiment 6. Intravaginal inoculaLion during the normal fol-
licular and luteal phases of the reproductive cycle. All the test and
control animals were free of V. fetus at the beginning of the experi-
ment. The test animals showed no apparent physiological response to
the inoculum. Vaginal discharge was absent and the animals remained
alert and active. No anorexia was observed.
Microscopic examination of Gram and Wright-Giemsa stained vaginal
Table 5. Optical density readings of V. fetus cultures with FSH added.
FSH concentrations (%)
Time 0.01 0.005 0.003 0.001 0.0005 Controls
hours A B C A B C A B C A B C A B C A B C
0 .160 .170 .200 .170 .170 .170 .170 .170 .180 .170 .160 .170 .170 .160 .170 .180 .180 .180
[2 .240 .250 .270 .240 .240 .260 .250 .250 .250 .240 .250 .250 .240 .220 .240 .250 .250 .2501
14 .340 .340 .350 .330 .330 .320 .330 .340 .340 .330 .340 .340 .330 .310 .330 .350 .340 .340J
6 .370 .390 .420 .370 .380 .400 .390 .380 .380 .380 .400 .380 .370 .350 .380 .410 .400 .390
.090 .080 .080 .090 .090 .094
A, B and C are replicates of each FSH concentration.
*M;ean change in OD in two hours determined between [2 and 4] hours.
OD readings taken at 525 m .
smears prepared during the inoculation period showed some neutrophils
together with degenerating epithelial cells and cell debris. The
vaginal flora was comprised of Gram positive and Gram negative cocci
and bacilli. Organisms morphologically similar to V. fetus were ob-
served in two of six animals during the inoculation period but at no
time during the post inoculation period or at necropsy five days later
There was apparently a slight increase in the number of leukocytes
seen on the third day post inoculation but the increase did not become
On each day of the three-day inoculation period V. fetus was
isolated from the vaginal cultures of the test animals. Isolation of
V. fetus was made from vaginal cultures taken from all the test ani-
mals 24 hours after the final inoculation was given. Positive vaginal
cultures were obtained from four of six test animals 48 hours after
the final inoculation was given and from two of six after 72 hours.
No isolations of V. fetus were made from the vaginal cultures of the
test animals at necropsy five days after the final inoculation was
given (see Table 6).
The uterus, cervix, blood, liver and spleen of each test animal
was cultured at necropsy and V. fetus was not isolated. Cultures of
the left and right uterine horns were free from secondary or contami-
nating flora as were those of the liver and spleen. Pseudomonas
aeruginosa, E. coli, Streptococcus sp. and Neisseria sp. were isolated
from the vaginas of the test and control animals. All the cultures
and smears taken from the control animals were negative for V. fetus.
No gross lesions were observed at necropsy. Iistological
studies of the uterine tissues revealed a slight inflammatory reaction
Animal number During inoculation*
24 hrs. 48 hrs.
24 hrs. 4B hrs. 72 hrs.
Necropsy 5 days post
3/3 2/3 1/3
and congestion of the lamina propria with evidence of edema between
the cells. No real differences were observed between the follicular
or luteal phase animals which could be attributed to the inoculation
of V. fetus.
Experiment 7. Intravaginal inoculation of the normal estrous
guinea pig with subsequent mating. The statistical analysis of the re-
sults of this experiment showed a significant difference (p = 0.05) be-
tween the average number of young which survived in the control popu-
lation when compared with that of young which survived in the test
population. The average for the controls was 2.B and that for the
test animals was 1.3 (see Appendix A). The average number of young
carried by the control group was 3.3. The average of the test group
was 2.3. This difference was significant (p = 0.05). Although the
litter size of the control animals was greater than that of the test
group, the number of stillbirths in the control group was higher. One
animal (I) in the control population and two in the test population
(9 and 10) failed to show definite signs of pregnancy. These results
are presented in Table 7.
A single animal (2) in the test group died in late gestation
and four mummified fetuses were found in the uterus at necropsy
(see Appendix C). Bacteriological examinations of the fetal and
maternal tissues were negative for V. fetus. Histopathological ex-
amination of the tissues from the reproductive tract of the dam re-
vealed thrombosis and extreme congestion of the endometrial vessels
and some sloughing of the endometrium. Extreme necrosis of the
placenta was noted especially around the area of attachment to the
Number Number surviving
stillborn more than 30 hrs.
Died; 4 mummified fetuses
V. fetus isolated from vagina
Total 33 5 28
uterine wall. No nuclear staining was present. The liver showed
some evidence of fatty change.
Three weeks after intravaginal inoculation one animal (8)
in the test population was observed to have a bloody vaginal dis-
charge which was characterized by increased numbers of neutrophils,
numerous erythrocytes and contained many vibrio-like organisms.
Subsequent isolation of V. fetus was made from a vaginal culture.
No apparent abortion was observed; however, the animal failed to
show any obvious signs of pregnancy and when euthanized six days
after the expected date of delivery was foundnon-pregnant. No gross
lesions were observed in the reproductive tract of this animal and
no isolations of V. fetus were made from the reproductive tract,
liver, spleen, blood or peritoneum.
The offspring of test animals No. 4, 5, 6 and 7 were apparent-
ly normal. The offspring of test animals 1 and 3 showed signs of
extreme weakness immediately after birth and did not attempt to
nurse the dams. Their gait was characterized by muscular twitching
and staggering. All died within 30 hours post-partum.
Experiment 8. Intravaginal inoculation during induced estrus.
All the test and control animals were free of V. fetus at the begin-
ning of the experiment. The animals tolerated the estrogen injec-
tions well and a physiological response to the hormone was observed.
The vagina dilated and there was an increase in the amount of vaginal
mucus, which was profuse in one of the test animals. The inoculum
of bacteria was tolerated well by both test and control animals. They
remained active and alert throughout the test period. No anorexia was
At the time of inoculation the vaginal cytology of the test
and control groups was characterized by cornified epithelial cells
and some cellular debris. The results of vaginal cultures taken
during the test period are given in Table 8. The vaginal cultures
of the test and control animals were positive for V. fetus during
the inoculation period but became negative within 24 hours after the
final inoculation. No isolations of V. fetus were made from vaginal
cultures taken from the test or control guinea pigs at intervals of
48 and 72 hours after the final inoculation was given.
Cultures for V. fetus taken from the uterus, cervix, blood,
liver and spleen of the test and control animals at necropsy five
days after the end of the inoculation period were negative.
Gram stained preparations of vaginal smears of test and con-
trol animals taken during the inoculation period showed organisms
morphologically similar to V. fetus in only one out of six animals.
A mixed vaginal flora of Gram positive and Gram negative bacilli
and cocci was observed. These secondary organisms when cultured in-
cluded Proteus sp., E. coli, Streptococcus sp., Staphylococcus sp.,
Neisseria sp. and Diphtheroid organisms. The uterine horns of all
the animals were apparently free of secondary microorganisms.
Cytological studies of the vagina of the estrogen treated
animals made on the second day after the final inoculation showed
a marked increase in the number of neutrophils accompanied by a de-
crease in the total number of microorganisms observed.
Histological examination showed no abnormalities in the uterine
tissue of any of the animals. The condition of the endometrium and
uterine glands was compatible with that of the uterus at estrus.
Animal number During inoculation* Post inoculation
24 hrs. 48 hrs. 24 hrs. 48 hrs. 72 hrs.
Necropsy 5 days post
Experiment 9. Intravaginal inoculation of progesterone treaL-
ed female guinea pigs. The pretrial vaginal cultures of the test and
control animals were negative for V. fetus. The animals tolerated the
inoculum well and remained alert and active. Some trauma was observed
at the site of hormone injection, notably stiffness of the muscle and
some tenderness. The vaginal cytology of the hormone treated test
animals was characterized by a marked eosinophilia not observed in
the vaginal smears of the control animals. The results of vaginal
cultures taken during the study are given in Table 9. The vaginal
cultures of the test and control animals were positive for V. fetus
during the inoculation period. The organism was not isolated from
the vagina of four of six test animals or from the vagina of one con-
trol animal 24 hours after the final inoculation was given. No iso-
lations of V. fetus were made from any vaginal cultures taken at 48
and 72 hours post inoculation or from the vaginal cultures at necropsy.
The vaginal smears of the test animals taken in the post inoc-
ulation period showed a moderate to heavy cellular debris with numerous
neutrophils and the eosinophilia noted at the start of the test period
persisted. The vaginal flora comprised Gram positive and Gram negative
cocci and bacilli. No organisms which morphologically resembled V.
fetus were seen in Gram stained vaginal smears at any time. The cytology
and flora of the vagina of the control animals were similar to those of
the test group, however, there was no indication of eosinophilia.
No isolations of V. fetus were made from cultures of the uterus,
cervix, blood, liver or spleen of the test or control animals when these
organs were cultured at necropsy five days after the final intravaginal
inoculation. The predominant microorganisms isolated from the vagina
Animal number During inoculation* Post inoculation
24 hrs. 48 hrs 24 hrs. 48 hrs. 72 hrs.
Necropsy 5 days post
6/6 2/6 0/6 0/6
2/2 1/2 /2 0/2
2/2 1/2 0/2 0/2
of the test and control animals were L. coli, Pseudomonas aeruginosa,
Proteus sp., Neisseria sp., Streptococcus sp and diphtheroid or-
No abnormalities were detected in the histological studies of
uterine tissue from any animal. The condition of the endometrium and
uterine glands was compatible with the hormonal state of the uterus.
Experiment 10. Intrauterine inoculation with V. fetus during
exogenous estrogen administration. All the test animals and sham op-
erated controls tolerated the surgical procedure and anesthesia very
well and showed no adverse after effects. The exogenous estrogen
caused enlargement of the vulva and increased secretion of cervical-
vaginal mucus. Hyperemia of the uterus was observed at laparotomy.
The results of the cultures taken at necropsy from the animals eu-
thanized at 24, 48 and 72 hours after inoculation are given in Table
10. Isolation of V. fetus was made from blood from the heart of three
of four animals examined at each time period. The left and right
uterine horns of all the animals examined 24 hours after inoculation
were positive for V. fetus. Isolation of V. fetus was made from the
saline washings of the uterine horns of these animals. Cultures taken
from the left and right uterine horns of three of four animals ex-
amined 48 hours after inoculation were positive for V. fetus. The
uterus of one animal (F) in this group was heavily contaminated with
Proteus sp., and isolation of V. fetus was made only from the saline
washings of the horns. The saline washings, from the left and right
uterine horns of one animal (I), and from Lhe left uterine horn of
another (K), were negative for V. fetus 72 hours after inoculation.
Table 10. Cultures taken at necropsy following the intrauterine inoculation with V. fetus of
ovariectomized estrogen treated guinea pigs.
Uterus R. horn
cultures per animal
Hours after inoculation
A B C D E F G H
+ + -
+ + +
+ + +
+ + +
+ + + + +
+ + + + +
+ + + + +
+ + +
7 7 7 6 10 5 10 9
I J K L
Sham operated controls
(2 animals )
+ + +
+ + + +
+ + +
- + + +
+ + + +
- + +
- + +
5 10 4 10
- = V. fetus isolated
- no growth of V. fetus after 5 days at 37 C
Tissues of animal F were contaminated with Proteus sp.
Similar cultures made on the remainder of the test animals in this
group were positive for V. fetus.
Isolations of V. fetus were made from the cervical cultures
of all the animals in every test group. The cultures from the vagina
of animals in the 24- and 48-hour test groups were positive for V.
fetus. In the 72-hour group V. fetus isolations were made from
the vagina of two of the four animals.
No isolations of V. fetus were made from the liver, spleen and
peritoneum 24 hours after inoculation. Isolations of V. fetus were
made from these sites in three of four animals in the 48-hour test
group. At 72 hours post inoculation V. fetus was isolated from the
liver of three of four animals and from the spleen and peritoneum of
two of four animals. No isolations of V. fetus were made from the
bile of any animal examined.
The blood cultures of three of four animals in each time group
were positive for V. fetus. Two of the positive blood cultures re-
vealed non-motile organisms on dark-field examination which proved
to be V. fetus when the blood cultures were subcultured on selective
blood agar plates- The results of the viable cell counts of V. fetus
performed on the washings from the right uterine horns of the animals
are given in Table 11. This table also lists the secondary organisms
isolated from the tissues of the animals at necropsy. A decrease was
observed in the number of viable V. fetus cells in the right uterine
horns of the animals examined at necropsy.
The animals tolerated the inoculum and remained active and alert
throughout the entire test period. No anorexia was noted. No gross
Table 1L. Cie co : a .- -,it utsrinr- .orns an. r secondary oaiiaisms
!ro. the ti.s -. f st-ro-o'. .--_ated ov.,ric.to.jize. guinea pigs following
intrauter n- in(culati_- '.. Lt '. "-t :.
-- ~ -~-~--t
Animal of '. fe
5.r l -
(5.0 nl total
Pradorirant seccr"ary nicrcorganisms
and source of isolation
Total vt..ci' cjil of 7. fetus inoc.la-esd = 1.2 x 106
24 hours post inoculation
2 1,4 2
A -1.' x 10 Proteus sp. 14; Statococcus so.
B 6.1 s 102 Str--tococccus s'. ; Stah',loocccus so.
C 3-0 x 12 StrxFp-.ococcus so. ; Sta.'hvlococcus aureus2
D 9.8 x 103 Str-U.tocoQcus sp.
48 hours pcst ijnoculation
E les- taan 100 ::eiss-ria sp. 23; D. pneuzroniae
F* less than 10l Protu. sp.'2 ; Aerobac-or sp.
G less t -an 100 Eo pred)ninart organi'.sr
h less than 100 No predorina.:t organisms
72 hours post inoculation
I less than 100 Aercoacter sp. ; Proteus sp.
j less than 100 !-o predominant orgarn.ss
K less than 130 Stre-?tcocc"s 2.,,iri -s
L less than 100 'N predominant cr?--isms
Key to superscripos given to predominant organis-s: 1 cervix; 2 left
uterinB horn; 3 ri ht uterine horn; 4 vagina; 5 blood from the heartt
*Proteus so. iJolated fro- ths liver, spleen and c-:irtonpal cavity
lesions were observed at necropsy and ovarian tissue was absent.
The sham operated control animals were negative for V. fetus and
showed no apparent gross changes in the reproductive system or
other abdominal viscera.
Histological studies of the uterine horns from animals in
the 24-hour test group showed neutrophilic infiltration of the
uterine glands with some neutrophils in the lamina propria. Uterine
tissues taken from animals in the 72-hour test group showed mild in-
flammation with some neutrophilic infiltration between the endometrial
Experiment 11. Intrauterine inoculation with V. fetus during
exogenous progesterone administration. The test and sham operated
control animals tolerated the surgical procedure and anesthesia with-
out any adverse aftereffects. The progesterone injections produced
some muscle stiffness but this was not as pronounced as that described
previously in the intact animals (Experiment 9). The vaginal cytology
was typical of that produced by progesterone and the uterus of each
animal was observed to be smaller and lacked the hyperemia present in
the estrogen treated animals.
The results of cultures for V. fetus taken at necropsy 24, 48
and 72 hours after intrauterine inoculation are given in Table 12.
Isolation of V. fetus was made from the blood of two of four animals
24 hours after inoculation and from all the animals in the 48- and
72-hour time intervals.
Isolations of V. fetus were made from the left and right uterine
horns, the saline washings from these horns, the cervix, vagina, liver
Hours after inoculation
Uterus R. horn
cultures per animal
M N O P
- -+ +
+ ++ +
Q R S T
+ + +
+ + +
+ + *4
9 8 9 10 9 9 9 9
U V w X
Sham operated controls
+ ++ +
+ + ++
+ ++ +
7 8 6 10
and spleen of all animals examined 24 and 4B hours after inocula-
tion. The spleen of one animal (S) contained an abscess from which
E. coli was isolated. Twenty-four hours after inoculation V. fetus
was isolated from the bile of two animals. Cultures made of the
bile of the other two animals in the 24-hour group and from animals
in the 40- and 72-hour groups did not result in the isolation of V.
Cultures of the left and right uterine horns, cervix, and
vagina of all animals in the 72-hour test group were positive for
V. fetus. The saline washings from the left and right uterine horns
of two of four animals in the 72-hour group also were positive. Iso-
lations of V. fetus were made from the liver and spleen of two of four
animals in the 72-hour test group. The peritoneal cultures of animals
examined at 24 and 48 hours post inoculation were negative for V. fetus,
but V. fetus was isolated 72 hours post inoculation from the peritoneal
cultures of three of four animals. No isolations of V. fetus were made
from cultures taken from the sham operated control animals.
The results of viable cell counts of V. fetus made on the right
uterine horn washings are given in Table 13. This table also lists the
secondary microorganisms isolated from the reproductive systems and
other organs cultured during the study. A reduction in the number of
viable cells of V. fetus isolated from the right uterine horns was ob-
served. Secondary organisms were confined to the vagina and distal
aspect of the cervix of the animals examined. No secondary organisms
were isolated from the uterine horns, blood, liver or spleen of any
Table 13. Cell counts of the right uterine horns and secondary organisms
from the tissues of progesterone treated ovariectumized guinea pigs follow-
ing intrauterine inoculation with V. fetus.
Animal Viable count
of V. fetus
(5.0 ml total)
Predominant secondary microorganisms
and source of isolation
Total viable cells of V. fetus inoculated = 6.6 x 106
24 hours post inoculation
M 1.5 x 103 Coliform organisms '4
N 2.6 x 102 E. coli1'4
0 3.0 x 102 E. colil,4; Streptococcus sp. ,4
P less than 100 Pseudomonas sp.4; E. coli4
48 hours post inoculation
S 1.8 x 102 No predominant organisms
R 6.9 x 10 No predominant organisms
S* less than 100 Coliform organisms' 4; Staphylococcus sp.'
T 6.2 x 102 Proteus sp.1,4; E. coli4
72 hours post inoculation
U less than 100 No predominant organisms
V less than 100 Proteus sp.1
W No growth No predominant organisms
X less than 100 E. colil'4; Pseudomonas aeruginosal'4
Key to superscripts given to predominant organisms: 1 cervix; 2 left
uterine horn; 3 right uterine horn; 4 vagina
*Splenic abscess; E. coli isolated
Apart from the splenic abscess previously described no other
gross lesions were found at necropsy. Ovarian tissue was absent in
all the animals examined.
Histological examination of the uterine tissues from all the
test animals showed congestion of the lamina propria with slight
lymphocytic infiltration. These changes were more pronounced in the
right uterine horn.
Experiment 12. Intrauterine inoculation of pregnant guinea
pigs. All the test animals and sham operated controls were vibrio-
free at the start of the test. The surgical procedure and anesthesia
were well tolerated and no adverse aftereffects were noted. The non-
infected controls maintained normal pregnancy.
The test animals appeared listless and became inactive within
12 hours after inoculation. The hair coat became ruffled and harsh
in appearance. Some anorexia was noted together with an increase in
water consumption. There was no rise in body temperature. Six of the
11 test animals (1, 2, 3, 4, 9 and 11) aborted within 12 to 36 hours
after inoculation and one (7) aborted between 36 and 48 hours post in-
oculation. Four animals (5, 6, 8 and 10) did not abort. Two of these
animals (8 and 10) were allowed to go to term.
The abortions were preceded by the appearance of a bloody vaginal
discharge which was initially reddish-brown in color. Later the dis-
charge became bright red. Rapidly motile organisms were present in the
discharge which when plated on selective blood agar plates proved to
be V. fetus.
Two animals (1 and 2) which aborted were euthanized 24 hours
after inoculation. A single animal (7) was euthanized following
abortion 48 hours after inoculation. The animals which survived
abortion and were not euthanized (3, 4, 9 and 11) together with those
allowed to term (8 and 10) developed a vaginal V. fetus infection
which persisted for 72 hours. This infection was characterized by
the isolation of decreasing numbers of V. fetus organisms toward the
end of the 72-hour test period. No isolations of V. fetus were made
from the vagina of these animals after four days. The vaginal cytology
during this period comprised numerous neutrophils and erythrocytes with
heavy cellular debris and some intact epithelial cells. Numerous
vibrio-like organisms were seen in Gram stained vaginal smears taken
24 hours after inoculation. After 72 hours vibrio-like organisms were
no longer seen in the vaginal smears although isolations of V. fetus
were made from the vagina of test animals examined.
The results of cultures taken at necropsy on the euthanized
dams are given in Table 14. Isolations of V. fetus were made from
the uterus, cervix and vagina of each animal examined 24, 48 and 72
hours after inoculation. The bile and ovaries of each animal were
negative for V. fetus. Culture of the spleen resulted in the isola-
tion of V. fetus from one animal (1) in the 24-hour examination group
and from another (6) in the 72-hour group. V. fetus was isolated from
the blood of two animals (1 and 2) 24 hours after inoculation and from
another (6) 72 hours after inoculation. The peritoneal cavity of one
animal (1) was positive for V. fetus when cultured 24 hours after inocu-
Gross lesions observed at necropsy of the dams were confined to
Table 14. Cultures taken at necropsy from qravid guinea pigs euthanized
following intrauterine inoculation with V. frtus.
Date of inoculation: 6-2-70, 8.00 to 10.00 a.m.
Time and date 8.00 a.m. 11.00 a.m. 7.00 a.m.
abortion complete 6-3-70 6-3-70 6-4-70
Time and date
L. uterine horn
R. uterine horn
cultures per animal
+ + +
6/11 5/11 8/11
+ = growth of V. fetus
- = no growth of V. fetus in
* Further results of fetal
5 days at 37 C
cultures are given in Table 15
the uterus, cervix and vagina. In those animals which aborted be-
fore euthanasia the uterus was generally enlarged, flaccid and ex-
tremely hyperemic. The uterine cavity was filled with a mucoid
blood-stained fluid. The cervix was relaxed and lacked tone. Tena-
cious blood-stained mucus occluded the cervical os. The vagina was
filled with a similar blood-stained mucus. Gram stained preparations
of this mucus revealed numerous neutrophils and erythrocytes with
large masses of vibrio-like organisms clumped together (see Appendix
C). The organisms were unevenly distributed throughout the smears
and many high power fields were devoid of them.
The vaginal smears of the animals which aborted but were not
euthanized showed some intact epithelial cells with moderate numbers
of neutrophils and erythrocytes. Vibrio-like organisms were seen in
only two vaginal smears.
The reproductive tracts of the two animals (5 and 6) which
were euthanized without history of abortion differed in gross ap-
pearances. The right uterine horn of animal number 5 was enlarged,
and hyperemic. The left uterine horn was enlarged hyperemic and
congested about a small area of the endometrium. The uterine glands
at this point appeared tortuous and the endometrial surface showed
evidence of previous nidation, although no embryo was found. The
uterus of animal number 6 contained three small embryos of approx-
imately 0.5 cm crown rump length, one in the right uterine horn and
two in the left uterine horn. The placental tissue was not well
differentiated and was of a pulpy consistency. Petechial hemorrhages
were seen on the uncut surface of the placenta. The uterine washings
from both animals were grossly blood-stained and contained many
neutrophils. Occasional vibrio-like organisms were seen in Gram
stained preparations of the uterine washings.
The results of the cultures taken from the aborted fetuses
and the fetuses in utero are given in Table 15. The number of
fetuses aborted by the dams ranged from one to six with a mode of
three. The crown rump lengths of the fetuses ranged from 0.5 cm to
5.0 cm, the length of the fetuses from any one dam being somewhat
uniform. The gross appearances of the fetuses varied from normal to
macerated. Some aborted fetuses showed evidence of cannibalism by the
Some of the fetuses had petechial hemorrhages of the skin sur-
faces. Other fetuses were reddish-brown in color with softer tissues
than those of control guinea pig fetuses at the same state of gesta-
tion. Impression smears of the skin surfaces of the macerated fetuses
showed numerous highly motile vibrio forms when examined by dark-field
technique. These organisms were not found on impression smears of the
skin surfaces of non-macerated fetuses. Some of the aborted fetuses
showed signs of post-mortem lividity.
Isolation of V. fetus was made from the stomach contents of six
aborted fetuses (from dams 6 and 7). Isolation of V. fetus was also
made from the fetal thoracic and abdominal viscera and from the placentae
of fetuses from seven of the dams. Isolation of V. fetus was made in
every instance in which amnionic fluids, amnionic membranes or fetal
blood were available for culture.
Accurate viable cell counts for V. fetus were not possible on
the uterine fluids due to their mucoid nature and relatively small
volumes. Direct culture of the fluids and infected uterine surfaces
Table 15. Bacterial flora of
of the dam with V. fetus.
aborted fetuses and
fetuses in utero following intrauterine inoculation
1 2 3 4 6 7 9 11
Number of fetuses
aborted or in utero
3* 3* 6* 3*
3* 1* i*
3.3 4.5 4.5 5.0 0.5 5.0 2.5 3.0
NA + NA NA EC
::A = not available for culture
+ = isolation of V. fetus from all fetuses cultured
EC = entire fetus cultured; V. fetus isolated
plated on selective blood agar resulted in a heavy, confluent and
pure growth of V. fetus.
Histopathological examination of the tissues taken from
euthanized dams examined following abortion showed superficial de-
nudation of the endometrium with a severe inflammatory reaction
characterized by polymorphonuclear and mononuclear leucocytic in-
filtration into the endometrium. Thrombi were observed in some of
the uterine vessels and hemorrhage was present in the lamina propria.
An acute patchy inflammatory reaction with edema was noted in the
The placental tissues revealed extensive gross areas of necrosis
and hemorrhage which varied in intensity. Fetal membranes were con-
gested and necrotic with evidence of hemorrhage present.
Histological examination of euthanized animals which did not
abort showed areas of inflammatory cell infiltration into the uterine
glands and small hemorrhages into the lamina propria and some thrombosis.
The placental tissues showed an acute inflammatory reaction with hemorrhage
and areas of necrosis.
The two animals (8 and 10) in which gestation was to be main-
tained to term although pregnant at the time of inoculation showed no
signs of a progressive pregnancy and were found non-pregnant when a
necropsy was performed on the expected date of parturition. Cultures
taken from the reproductive tract, liver, spleen, blood and the peritoneal
cavity of these animals were negative for V. fetus.
The four animals which aborted but were not euthanized at the
time of abortion (3, 4, 9 and 11) were subsequently euthanized 26 days
after intrauterine inoculation. No isolations of V. fetus were made
from any of the tissues cultured. No gross lesions were found at
necropsy in any of these animals, or the animals maintained to term.
No changes were seen in tissues from the reproductive tracts of these
animals when histopathological examination was made following necropsy.
The in vitro studies reported in the present work were at-
tempted in an effort to determine the possible effect of erythritol,
follicle stimulating hormone (FSH), estrone and progesterone on the
growth of V. fetus in an artificial medium. These effects were
assessed in the exponential phase of cell growth. A curve of change
in optical density (OD) against time was substituted for the standard
The standard bacterial growth curve described in textbooks
of bacteriology (Burrows,1959; Stanier, Douderoff and Adelberg, 1957)
is based on the characteristic growth pattern derived by counting the
number of viable cells in a growing culture over a period of time-
The logarithm of the number of viable cells is then plotted against
time. The use of OD to replace the need for repeated viable cell
counts in studies of bacterial growth was found to be of value by
Lyon, Hall and Costas-Martinez (1970). A standard of cell dry weight
was used in the study of Mycobacterium tuberculosis. The growth curves
derived in this study are recognized as not being absolute measures
of bacterial cell numbers since neither the actual dry weight of the
cells, the cell size nor the total cell count was determined. The re-
liability of the OD method of growth study was reported to improve with
the use of synchronized cultures (Wilson and Miles, 1964). Synchronized
cultures allow the subsequent increase in OD of the culture to be
proportional to the increase in cell numbers. This proportionality
ceases to be a linear function when the OD reading exceeds 0.4 (Anon.,
1957). The slopes of the linear plots obtained in Experiments 1 through
5 were calculated from the points derived during the interval when
the growth of the cells was apparently logarithmic.
The OD of a culture is affected by changes in cell size. Most
of these changes occur in the lag phase of growth and are accompanied
by increased cellular metabolism. The actual lag refers to cell
division rather than cell metabolism (Thimann, 1963). The size of the
cell increases in the lag phase and usually reverts back as the actual
exponential or log phase of growth begins.
Turbidometric methods were used in studies of possible growth
factors for V. fetus by Trueblood and Tucker (1957) and Reich, Dunne,
Bortree and Hokanson (1957). In all cases these authors compared the
test growth curves with those of controls treated in an identical
manner but without the growth factor or substrate added.
The choice of Brucella broth as a growth medium was made on the
basis of its wide acceptance and value in the study of V. fetus (Reich,
Morse and Wilson, 1956). A true minimal medium that will support the
growth of all V. fetus isolants has not been defined although Alexander
(1957) described a medium of brain liver heart, sodium chloride, phos-
phate buffer and 0.05% agar for the growth of ovine isolants of V. fetus.
The erratic growth of V. fetus in a chemically defined medium was de-
scribed by Fletcher and Plastridge (1963).
The metabolism of V. fetus was determined to be microaerophilic by
May (1953). The organism was found to lack fermentative ability when
studies of possible energy sources were made by Alexander (1957).
The inability of V. fetus to metabolize carbohydrates, especially
dextrose, is noted in Bergey's Manual of Determinative Bacteriology
The possibility of erythritol as a growth factor for V. fetus
was suggested by Herzberg (1966). The presence of erythritol in
bovine fetal fluids was determined by Pearce et al. (1962). The
position of erythritol in the metabolic pathway is not well under-
stood. Hers (1958) proposed an intermediary function between D-
erythrose and D-erythrulose in the pentose pathway. The possibility
existed that if erythritol was a growth factor for V. fetus it could
be injected into the guinea pig prior to inoculation and enhance the
virulence of V. fetus. Keppie et al. (1965) used this technique to
enhance the virulence of Brucella abortus in the guinea pig.
The results of Experiment 1 clearly indicated that, under the
experimental conditions employed, erythritol was not a growth factor
for V. fetus.
Osborne and Bourdeau (1955) reported the stimulatory effects
of steroid hormones on the growth of V. fetus. Their study mentioned
neither the method of addition of the steroid hormones to the culture
medium nor an assay of the purity of the hormone preparations used.
The extent of growth increase was measured by the depth of the sub-
surface layer of cells obtained in semi-solid thiol medium. Zemjanis
and Iloyt (1960) reported an increase in the depth of the subsurface
growth layer of V. fetus cultures to which a concentration of 0.005%
17, beta-cstradiol was added. Concentrations of 0.1 to 0.2% 17,
beta-estradiol were inhibitory to V. fetus. In the present study no
statistically significant change (p = 0.05) in the growth rate of
V. fetus was obtained in media which contained concentrations of es-
trone or progesterone ranging from 0.0005 to 0.01%. Under the ex-
perimental conditions employed neither estrone nor progesterone were
growth factors for V. fetus. The analysis of the results of Experiment
4 indicated a statistically significant increase (p = 0.05) in the
growth rate of V. fetus when combinations of estrone and progesterone
were incorporated into the medium. The mechanism by which this increase
in growth rate occurred was not determined. However, there was the
possibility of a synergistic effect when the two hormones were combined.
In view of the differences in methodology it was not possible to make
a critical comparison of the results of Osborne and Bourdeau (1955) and
Zemjanis and Hoyt (1960) with those obtained in this study.
One of the prime factors to be considered in such in vitro
growth experiments as those reported in the present work was the ex-
treme insolubility of the steroid hormones in an aqueous medium. Both
estrone and progesterone are almost insoluble in water (Anon.,1952).
This factor alone must influence the outcome of the experiments since
no matter how great the quantity of hormone physically present in the
medium the quantity that actually goes into solution remains extremely
small and may not be increased by the mere addition of a greater quan-
tity of hormone to the medium. The results shown in Tables 2, 3 and 4
are best considered from a qualitative rather than from a strictly quan-
titative point of view.
Earlier preliminary studies made by the author indicated that in
order to maintain the total quantities of the steroid hormones used
in a completely soluble form, a level of organic solvent was required
which was toxic to the test organism. For this reason a minimal
volume of chloroform was used to add the required quantity of steroid
hormone to the medium. The subsequent distillation of the solvent from
the aqueous medium removed the possibility of toxicity to the test or-
ganism. As the chloroform was distilled off the bulk of the hormone
The precise level of soluble steroid hormones achieved in Ex-
periments 2, 3 and 4 was not determined and the results must therefore
be carefully interpreted. Any change in cell growth rate was due
either to the amount of insoluble or soluble hormone present in the
medium. There was also the possibility that both the soluble and in-
soluble factors could influence the growth of V. fetus. The nature of
hormonal action in vivo is characterized by the extremely small quan-
tities of hormone necessary to exert an effect on the target tissues
(Harper, 1963). A similar situation may influence the growth of
microorganisms in vitro when hormones are added to culture media.
Studies of the effect of the hormones at a cellular level were not
made; however, quantitative estimates of increased protein, DNA or RNA
synthesis would be of value.
This study did not give any indication of actual breakdown of
the hormones by microbial action or any indication of physical changes
induced in the culture medium by the presence of the hormones. These
factors would need to be determined by biological, physical and chemical
assay. The in vitro aspects of this study were directed toward the
possibility that estrogen or progesterone either individually or in
combination could influence the growth of V. fetus in vivo. It was
interesting to note that the only significant growth response oc-
curred in the estrone-progesterone combinations (Experiment 4) in
which there was the possibility of a synergistic effect.
The relationship between the in vitro hormone levels achieved
in this study and those which occurred naturally or were induced by
the exogenous hormone administered in vivo can only be presumed
(Nalbandov, 1964). Evidence of the direct effect of estrogen and
progesterone upon cell components and growing tissues was given by
Brooks, Leitheiser, DeLoecker and De Wever (1969); Hahn, Church,
Gorbman and Wilmot (1968); and O'Malley, Aronow, Peacock and Dingman
(1968). Studies made by these authors indicated significant increases
in protein synthesis in uterine microsomal supernatants in vitro (Brooks
et al., 1969) induced by estrone sulphate and progesterone; and estrogen
induced increase of t-RNA formation and the production of new types of
RNA in the chick oviduct (Hahn et al., 1968; O'Malley et al., 1968).
The statistical analysis of the results of Experiment 5 indicated
that the addition of FSH to the culture medium in concentrations ranging
from0.0005 to 0.01% did not affect the growth rate of V. fetus.
Although certain combinations of estrone and progesterone sig-
nificantly increased the growth rate of V. fetus in vitro, the possi-
bility that estrogen and progesterone, either singly or in combination,
had the same effect in the guinea pig was not evident in the present
study. It is known that estrogen can induce an antibacterial state in
the uterus and progesterone produces changes in which the uterus is
more susceptible to bacterial infection (Hawk, 1958). Whether thcse
hormonal influences apply to V. fetus infection requires further re-
search. Black et al. (1953),in studies of the response of the rabbit
uterus to V. fetus infection, failed to demonstrate any of the changes
induced by pyogenic organisms. Although V. fetus was isolated from
the inoculated uterine horns of estrous and pseudopregnant rabbit uteri
the pathological response of the tissues was minimal.
The in vivo studies reported here made use of the vaginal route
of inoculation although Ristic and Morse (1953) reported this route to
be the least effective. Justification for the use of the intravaginal
route of inoculation was based on the normal venereal route of infec-
tion in cattle. An attempt was made to duplicate this venereal route
of infection as closely as possible in the guinea pig. Table 6 shows
a relative lack of persistence of V. fetus in the vagina of the guinea
pig even when the organism was introduced in large numbers. Vaginal
cultures taken 72 hours after inoculation were negative for V. fetus
in four of six animals regardless of the hormonal condition of the
reproductive tract. Five days post-inoculation all the vaginal cultures
were negative. This was in direct contrast with the infection in the
heifer in which V. fetus was reported to persist in the vagina for as
long as eight months (VanDeplassche et al., 1956).
The bacteriological and histological findings made after intra-
vaginal inoculation of the intact guinea pig (Experiment 6) gave no
indication that V. fetus entered the uterus. The possibility of uterine
invasion of shorter duration, less than five days, was not ruled out.
An experiment in which larger numbers of animals are used and examined
bacteriologically at closer intervals would be of value. The absence
of a contaminating microflora in the uterus at necropsy contrasted
with a vaginal flora of Pseudomonas sp. and E. coli apparently in-
dicated an effective uterine defense mechanism in which the cervix
acted as a barrier to direct infection of the uterus.
The use of exogenous estrogen in an attempt to increase the
susceptibility of the guinea pig to V. fetus was reported by Adler
(1953) and by Power (1954). Experiment 8 was an attempt to repeat
the work of Adler (1953). The results of bacteriological studies of
the estrus induced guinea pigs reported in Table B were very similar
to the results of Experiment 6 in which guinea pigs with normal cycles
were used. A lack of persistence of the organism in the vagina was
again observed with some indication of an even more rapid rate of
clearance. These results were in direct contrast with those of Adler
(1953) who reported a 50% efficiency in the rates of isolation of V.
fetus from guinea pigs inoculated intravaginally with infected bull
semen and from animals inoculated with pure V. fetus cultures. Adler
(1953) relied on phase contrast microscopy to establish the identity
of V. fetus from the uterine tissues of guinea pigs. The presence of
motile organisms, other than V. fetus, in the cultured tissues was a
possibility not discussed by the author. Power (1954) also reported
data in direct contrast with that of the present study and successfully
infected three of six guinea pigs with infected bull semen inoculated
into the vagina. Two of the negative animals were found to be pregnant.
The results of Experiment B were supported by the data of Robinson
et al. (1956) who reported failure in attempts to infect two groups of
estrus induced guinea pigs with intravaginal inoculations of V. fetus.
The apparent dichotomy between the results of Adler (1953)
and Power (1954) and those reported by Robinson ct al. (1956) to-
gether with the data reported in this study shows an obvious need
for further work in the study of v. fetus infection in the non-gravid
female guinea pig. These studies should include an increase in the
numbers of animals used, accurate identification of the V. fetus
isolants and due regard for the possible loss of virulence in lab-
oratory maintained cultures of V. fetus.
Exogenous progesterone apparently had little effect to increase
the susceptibility of the female guinea pig to intravaginal inoculation
with V. fetus. The organism failed to persist in the vagina (Table 9)
although V. fetus was isolated from three animals 24 hours after the
final intravaginal inoculation was given. Histological evidence of
uterine invasion was absent and cultures taken from the uterus at
necropsy were negative for V. fetus.
It is of interest to note that Manclark and Pickett (1965a)
were unable to detect the formation of either mucosal or humoral
antibody following the intravaginal inoculation of the virgin female
guinea pig with live cells of V. fetus. The pattern of the immune
response of the guinea pig to V. fetus infection was influenced by the
route of inoculation, the type of antigen used and the hormonal con-
dition of the reproductive tract. The highest antibody titers re-
ported by Manclark and Pickett (1965a) were obtained when live V. fetus
cells were inoculated intraperitoneally into pseudopregnant guinea pigs.
In this case both humoral and mucosal antibodies were detected. An
explanation for the absence of mucosal or humoral antibody in the virgin
guinea pig, following intravaginal inoculation with live V. fetus
cells, may be a lack of antigen persistence as evidenced by the
results of Experiments 6, 8 and 9.
The use of ovariectomized guinea pigs in the study of the
intrauterine route of inoculation was made to permit the appraisal
of the effect of estrogen and progesterone in vivo. The contribution
of estrogen from the adrenals to compensate for the estrogen normally
produced by the ovary was not considered significant in a system in
which exogenous estrogen was administered to ovariectomized guinea
pigs. Although the quantitative levels of estrogen induced by the
hormone injections were not determined, the gross appearance of the
reproductive tract, the vaginal cytology and the histological ap-
pearance of the uterus were compatible with estrogen stimulation.
Previous studies of V. fetus infection in which an intrauterine route
of inoculation was used include the works of Black et al. (1953) in
the estrous and pseudopregnant rabbit and the work of Osburn et al.
(1969) in the cow. The present work is the first reported use of the
intrauterine route of inoculation of V. fetus in the guinea pig and the
first study to define the V. fetus isolant used as V. fetus var. vene-
The results of the bacteriological examinations of the non-pregnant
animals (Tables 8 and 10) inoculated by the intrauterine route were in
contrast to those previously described in Experiments 6, 8 and 9 in which
the intravaginal route of inoculation was used. After introduction into
the lumen of the uterus V. fetus spread throughout the reproductive system
within the first 24 hours post inoculation. The organism was also isolated
from the blood of the animals. Within 48 hours the liver and spleen
were involved and the organism still persisted in the reproductive
tract. Cultures taken 72 hours after inoculation showed an apparent
clearance of the organism from the reproductive system and the spleen.
Isolations of V. fetus from the liver and spleen of non-gravid guinea
pigs inoculated intraperitoneally were reported by Smith (1923). The
isolation of V. fetus var.venerealis from the blood of a non-gravid
guinea pig has not been described previously and constitutes an im-
portant finding in the present work since V. fetus var. venerealis
was presumed to localize in the reproductive tract. The possibility
of bacteremia as a means for the systemic spread of V. fetus in the
guinea pig was suggested by Smith (1919) and Ristic et al. (1954a)
but was apparently not followed up by these workers in their later
studies of V. fetus infection in the guinea pig. V. fetus var. vene-
realis infection in the cow was considered to be a localized infection
centered in the reproductive tract (Manclark and Pickett, 1965a). No
data are presently available on studies designed to determine if the
organism ever enters the circulatory system of the cow.
The cultural data obtained from the progesterone treated ovari-
ectomized guinea pigs (Table 12) after intrauterine inoculation were
similar to that described from the estrogen treated animals. A notable
difference was the isolation of V. fetus from the gallbladder of two
animals examined 24 hours after intrauterine inoculation. Previous iso-
lation of V. fetus from the gallbladder of the guinea pig was reported
by Ristic ct al. (1954). This followed the inoculation of a mixed cul-
ture of ovine and bovine isolants of V. fetus. The ovine isolant
designated strain K was possibly of the intestinalis variety. The
isolation of V. fetus var. intestinalis from the gallbladder of the
ewe is not uncommon and formed part of a study by Firehammer, Love-
lace and Hawkins (1962).
The viable cell counts of V. fetus obtained from the saline
washings of the right uterine horns of the estrogen and progesterone
treated animals gave no indication that the organisms reproduced in
the uterus. Failure to ligate the left uterine horn prior to inocula-
tion allowed the organisms to gain direct access to the left uterine
horn via the cervix. This precluded the use of the left horn of the
uterus as a control tissue. An increase in V. fetus cell numbers
could only be proven by count of the total number of cells throughout
the body of the host. This was not attempted. In Experiments 10 and
11 the reduction in the viable cell count from the right uterine horn
was possibly due to the spread of the organisms throughout the body.
The organisms appeared to spread more rapidly in the estrogen treated
animals than in the progestrone treated animals. It is speculative
at this time to consider this as an effect of hormonal mediation.
The possibility of a uterine clearance mechanism under the control of
the endocrine system cannot be ruled out; however, the present study
lends support to the opinion of Black et al. (1953) that V. fetus is
not affected by such a mechanism.
The failure of V. fetus to persist in the non-gravid guinea
pig uterus and vagina under the experimental conditions employed in
this series of tests was probably due to an absence of host specificity
and lends support to the earlier findings of Smith et al. (1920) in
which V. fetus infection in the guinea pig was described as transitory.
The absence of any naturally occurring deaths and the lack of sig-
nificant uterine lesions among the test populations (Experiments 6, 8,
9, 10 and 11) further supported the opinion of Smith et al. (1920)
that V. fetus was not pathogenic for the non-gravid guinea pig. This
opinion was reiterated by Manclark and Pickett (1965a).
The isolation of bacterial species, other than V. fetus, from
the uterus of the test animals examined after intrauterine inoculation
(Experiments 10 and 11) contrasted with the negative uterine cultures
obtained when the intravaginal route of inoculation was used (Experi-
ments 6, 8 and 9). The presence of these secondary isolants (Tables 11
and 13) could be due to the trauma of surgery and the possible post
surgical debilitation of the animals; however, no secondary bacterial
isolations were made from the non-infected sham operated control animals.
The experimental evidence (Experiments 10 and 11) indicated the
passage of V. fetus from the uterus to the cervix and then into the
vagina. In previous Experiments (6, 8 and 9) and in the report of
Ristic and Morse (1953) V. fetus was unable to survive in the vagina
or to pass through the cervix when inoculated intravaginally into the
non-gravid guinea pig. It is possible that when placed directly into
the bacteria-free non-gravid uterus V. fetus is able to survive and pass
from the uterus through the cervix into the vagina. The ability of the
organism to survive may be due to the absence of microbial competition
for available substrates in the bacteria-free uterus. The differences
in the chemicophysical environment of the uterus and that of the vagina
may also be determining factors in the survival of V. fetus in the
reproductive tract. The continued passage of V. fetus from the uterus
provides a continual reinfection of the vagina which persists until
the uterine infection is cleared. The open cervix must provide the
portal of entry by which the vaginal flora gain access to the uterine
horns. The present study did not include the intrauterine inoculation
of intact non-pregnant guinea pigs.
The investigations of Smith (1918; 1919 and 1923) did not in-
clude studies with pregnant guinea pigs. The absence of pathological
change and the failure to isolate V. fetus four days after intra-
peritoneal inoculation led Smith (1918) to conclude that the guinea
pig was refractory to V. fetus infection. Ristic and Morse (1953) re-
ported abortions in guinea pigs one to 12 days after oral, intraperitoneal
and intravaginal inoculation but, as reported earlier, these workers
used a mixed inoculum of bovine and ovine isolants. In the present
study abortion occurred in seven of 11 animals within 36 hours after
intrauterine inoculation with V. fetus var. venerealis. This interval
is shorter than that observed by Robinson et al. (1956) in which preg-
nant guinea pigs aborted four days after intraperitoneal inoculation
with infected bovine semen. Osburn et al. (1969) noted abortion five
days following the intrauterine inoculation of V. fetus in the cow
during the second trimester of pregnancy. Cows in the third trimester
of pregnancy aborted between nine and 20 days after similar inoculation.
Whether this apparent resistance of the bovine fetus to V. fetus in-
fection applies equally to the guinea pig is not known.
It is possible that V. fetus must first gain access to the uterus
in order for abortion to occur in which case the route of infection is
an important factor. The isolation of the organisms from the blood of
the animals which aborted (Experiment 12) and from the other non-
pregnant animals previously described emphasizes the importance of
the circulatory system in the spread of the infection from the uterus.
The role of the circulatory system in the establishment of the natu-
rally acquired infection in the cow, in which the organisms are in-
oculated into the vagina or into the cervix at artificial insemination,
The gross appearance of the aborted fetuses (Experiment 12)
closely resembled that described by Ristic and Morse (1953). The
bacterial examination of the aborted fetuses (Experiment 12) indicated
that V. fetus had gained access to the stomach, viscera and blood re-
gardless of the normal appearance of some specimens (see Appendix C).
The macerated appearance of some of the fetuses was apparently related
either to the extent of bacterial invasion or was due to the actual
uterine position of the fetus in relation to the inoculum. Care was
taken to avoid the direct inoculation of any fetus in utero; however,
there was an initial concentration of the inoculum around the injection
site and this may have caused the death of the fetus nearestto the site
of injection. The mortality due to V. fetus infection among chicken
embryos was reported between 10.7% and 85.7% by White, Ristic and
Sanders (1958). The response was greatest when a human isolant of
V. fetus was tested. There was no correlation between embryonic mor-
tality and the number of viable cells in the inoculum.
The histopathologic changes observed in Experiment 12 were
comparable with those described in the pregnant guinea pig by Ristic
et al. (1954a) and to the histopathologic changes observed in the
heifer following V. fetus infection (Dozsa, Olson and Campbell, 1960).
The presence of edema, inflammatory cell reaction and vessel throm-
bosis in the uterus of the cow as a result of V. fetus infection was
described by Simon and McNutt (1957).
The work reported here does not eliminate the possibility of
endotoxic action as a cause for the observed abortions since whole
cells of V. fetus in the growth medium were used as the inoculum. Os-
borne (1965) reported the incidence of abortion induced by V. fetus
endotoxin in the goat, cow, ewe and rabbit. The histopathology of the
uterus of the above animals was similar to that found in naturally oc-
curring infections and in the pregnant animals described in Experiment
12. The gross findings noted by Osborne (1965) in the heart, lungs,
and liver of calves following the injection of V. fetus endotoxin
were not apparent in the guinea pigs which were examined after abor-
tion (Experiment 12). The possibility of differences in the responses
of various animal species to endotoxins must be considered. The role
of endotoxin as the cause of the infertility and abortion observed as
the result of V. fetus infection was discounted by Dennis (1967). This
author determined the toxicity of the endotoxins of pathogenic isolants
of V. fetus and saprophytic vibrios and found them to be similar in
potency. Dennis (1967) suggested a need to investigate causes other
than endotoxic activity to account for the pathogenicity of V. fetus.
In Experiment 7 the animals were mated after intravaginal in-
oculation during the proestrus period. There was a significant differ-
ence (p = 0.05) in the number of offspring which survived for longer
than 30 hours in the control population when compared with the nunmer
of survivors in the test population. (Table 7 and Appendix A.) Sig-
nificantly larger litters (p = 0.05) were produced by females in the
control population than those produced by the test population (Table 7
and Appendix A). The number of stillborn offspring was higher in the
control population than in the test population. This was probably a
result of increased litter size. The stillborn offspring of both
populations frequently weighed more than 75 grams, a birthweight which
increased the trauma to the young at birth.
The results shown in Table 7 did not indicate "infertility" in
the test or control groups; however, there was an indication that the
presence of V. fetus in the vagina at the time of mating could sub-
sequently affect the embryos either before or after nidation. At what
stage in the gestation period V. fetus affects the embryo or fetus is
not known. Ulberg and Burfening (1967) reported the effects of an
adverse uterine environment on the embryo. They showed a direct re-
lationship in the type and duration of stress to the ultimate survival
of the embryo or the fetus. The possibility that V. fetus produced
an adverse uterine environment in the guinea pig is worthy of consider-
ation. The change induced in the intrauterine environment could be
slight, in which case the fetus developed to term but died shortly
after birth. A more severe change in the intrauterine environment
could result in early abortion, resorption of the embryo or apparent
failure of nidation. The possibility of a severe change was reinforced
by the apparent complete resorption of the embryos of the two animals
selected to go to term in Experiment 12. The predilection of V. fetus
for the gravid uterus was clearly shown in Experiment 12 and the re-
sults of this experiment support the findings of Manclark and Pickett
(1965a) in this regard.
This work has attempted to describe the effects of V. fetus
var. venerealis in the gravid and non-gravid guinea pig. The transi-
tory nature of the infection and the absence of lesions in the non-
gravid guinea pig, whether treated with steroid hormones or not, ren-
dered the study of V. fetus infection in such a model difficult and
produced inconclusive results. There is an indication that the steroid
hormones (estrone and progesterone) may act as growth factors in vitro
but no definite conclusion can be made that this indication holds true
in vivo. The value of the non-gravid guinea pig as a diagnostic tool
for the presence of V. fetus infection in cattle was not substantiated
by this study. A diagnostic procedure which makes use of the intra-
peritoneal inoculation if the pregnant guinea pig seems feasible but
more work is needed in this area.
The abortions observed in this study and those noted by the other
workers cited in the text do not explain the "infertility" manifested
by V. fetus infection in the cow. The refractory response of the guinea
pig to infection by V- fetus is considered to be due to the strict host
specificity of V. fetus var. venerealis which allows the fullest ex-
pression of its virulence solely in the reproductive system of the cow
and renders an accurate study of such factors as pathogenicity, in-
fectivity and antigenicity difficult, if not impossible in other animal
This study was undertaken to determine the influences of
specific endocrine factors on the growth of V. fetus var. vene-
realis in vitro and to relate any effects to similar factors in
non-gravid and gravid guinea pig exposed to V. fetus var. venerealis
infection. The study also included an appraisal of routes of inoc-
ulation and the general features of V. fetus var. venerealis infec-
tion in the female guinea pig.
1. The addition of erythritol, follicule stimulating hormone,
estrone or progesterone to culture medium failed to give statistically
significant evidence of growth stimulation of V. fetus var. venerealis
2. The addition of estrone in combination with progesterone
to culture medium provided statistically significant evidence (p = 0.05)
of an increase in the growth rate of V. fetus var. venerealis in vitro.
3. Non-gravid guinea pigs with normal cycles and non-gravid
guinea pigs treated with exogenous estrogen or progesterone failed to
develop a prolonged vaginal or a systemic infection when inoculated
intravaginally with V. fetus var. venerealis. No histopathologic
lesions were seen in the reproductive tract.
4. Proestrus females inoculated with V. fetus var. vencrealis
intravaginally and then mated showed significant decreases (p = 0.05)
in litter size and in numbers of offspring surviving 30 hours post-
5. Ovariectomized guinea pigs treated with either exogenous
estrogen or progesterone and inoculated with V. fetus var. venerealis
by the intrauterine route developed a transitory but apparently not
progressive infection of the uterus, cervix and vagina. The infection
became systemic and involved the blood, liver, spleen, gallbladder and
peritoneum. A hormonal influence on the course of the infection was
not established. No histopathologic changes were seen in the repro-
6. Gravid guinea pigs inoculated by the intrauterine route
with V. fetus var. venerealis aborted within 24 to 30 hours post inocu-
lation. The aborted fetuses were infected and V. fetus var. venerealis
was isolated from the stomach fluid, viscera, blood and body surfaces.
Isolation of V. fetus var.venerealis was made from the fetal maternal
membranes and from the uterus, cervix, vagina, liver, spleen, blood
and peritoneum of the dams. The animals which did not abort showed
signs of apparent fetal resorption and only transitory infection of
the vagina following intrauterine inoculation with V. fetus var. vene-
realis. Histopathological examination of the uterus and placenta of
the guinea pigs which aborted showed evidence of necrosis, thrombosis,
hemorrhage, edema and inflammatory reaction.