Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Retrospective Theses and Dissertations 1996 Placental pathology, immune responses, bacteriologic findings and clinical signs in pregnant cattle vaccinated with Brucella abortus strain RB51 Mitchell Van Palmer Iowa State University Follow this and additional works at: http://lib.dr.iastate.edu/rtd Part of the Veterinary Pathology and Pathobiology Commons Recommended Citation Palmer, Mitchell Van, "Placental pathology, immune responses, bacteriologic findings and clinical signs in pregnant cattle vaccinated with Brucella abortus strain RB51 " (1996). Retrospective Theses and Dissertations. Paper 11557. This Dissertation is brought to you for free and open access by Digital Repository @ Iowa State University. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Digital Repository @ Iowa State University. For more information, please contact [email protected]. INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter &ce, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely aSect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. Kgher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information G)nipany 300 North Zed) Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600 Placental pathology, immune responses, bacteriologic findings and clinical signs in pregnant cattle vaccinated with Brucella abortus strain RB51 by Mitchell Van Pahner A dissertation submitted to the graduate faculty in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Major: Veterinary Pathology Major Professor: Norman F. Cheville Iowa State University Ames, Iowa 1996 DMI Nximber: 9712585 UMI Microform 9712585 Copyright 1997, by UMI Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. UMI 300 North Zeeb Road Ann Arbor, MI 48103 ii Graduate College Iowa State University This is to certify that the Doctoral dissertation of Mitchell Van Palmer has met the dissertation requirements of Iowa State University Signature was redacted for privacy. Major Professor Signature was redacted for privacy. For the Major Program Signature was redacted for privacy. For the Graduate College iii TABLE OF CONTENTS ABSTRACT CHAPTER L GENERAL INTRODUCTION Introduction Literature Review Thesis organization References I I 2 24 25 CHAPTER 2. EXPERIMENTAL INFECTION OF PREGNANT CATTLE WITH THE VACCINE BRUCELLA ABORTUS STRAIN RB5I: PATHOLOGIC, BACTERIOLOGIC, AND SEROLOGIC FINDINGS 40 Abstract 40 Introduction 41 Materials and Methods 42 Results 46 Discussion 60 References 63 CHAPTER 3. SAFETY AND IMMUNOGENICITY OF THE VACCINE BRUCELLA ABORTUS STRAIN RB51 IN PREGNANT CATTLE Materials and Methods Results Discussion References 70 73 77 84 86 CHAPTER 4. TUMOR NECROSIS FACTOR-ALPHA IN PREGNANT CATTLE FOLLOWING VACCINATION WITH BRUCELLA ABORTUS STRAIN RB51 Abstract Introduction Materials and Methods Results Discussion References 89 89 90 93 95 99 104 CHAPTER5. GENERAL CONCLUSIONS General Discussion References 111 111 113 ACKNOWLEDGMENTS 116 iv ABSTRACT To determine the placental tropism, abortigenicity, immunogenicity and effect on TNF-a levels, pregnant cattle were vaccinated IV (n=10) or SC (n=5) with the vaccine Brucella abortus strain RB51 (SRB51), SC with strain 19 (SI9) (n=5), or saline (n=2) at 6 months gestation. Eight of ten IV vaccinated heifers developed placentitis and fetal infection. Strain RB51 was cultured from all tissues in which lesions were seen. No lesions were seen in SC vaccinated cattle and SRB51 was isolated from 2/5 superficial cervical lymph nodes in the area draining the site of SC vaccination. One premature calf bom to an rv vaccinated heifer had mild interstitial pneumonia and disseminated SRB51 infection. No lesions were seen in other calves from IV vaccinated heifers, heifers vaccinated SC with SRB51 or SI9, and neither SRB51 nor S19 could be recovered from calves of SC vaccinated cattle. Maternal PBMC from SRB51 vaccinates and S19 vaccinates showed proliferative responses to both y-irradiated SRB51 and S19 which were greater than controls. Strain RB51 vaccinates did not develop antibodies detected by the standard tube agglutination test, but did develop antibodies to SRB51 that reacted in a dot ELISA test with irradiated SRB51. Radioimmunoassay for bovine plasma TNF- a revealed insignificant differences (P>0.05) between SRB51 vaccinates, S19 vaccinates and controls. Similarly, TNF-a levels in amniotic or allantoic fluids from vaccinated cattle were not different from placental fluids of controls (P>0.05). Immunohistochemistry for TNF-a revealed increased immunoreactivity within trophoblastic epithelial cells which was most extensive in FV vaccinated cattle with vaccine- induced placentitis. These results indicate that SRB51 is less abortifacient than previously published reports vvith S19; however, SRB51 can infect the bovine placenta and fetus, can induce placentitis, and in some cases, lead to preterm expulsion of the fetus. Undesirable effects of SRB51 are dependent on dose and route of administration as SC vaccination with a lower dose does not result in placental or fetal infection. I CHAPTER 1. GENERAL INTRODUCTION Introduction Brucellosis is an economically important abortifacient disease of cattle. Since 1940, the United States has been actively engaged in a brucellosis eradication effort. A cornerstone of this effort has been the vaccination of young female calves with the live brucellosis vaccine Brucella abortus strain 19 (SI9). Strain 19 has been effective in decreasing the incidence of bovine brucellosis; consequently, the goal of the United States Department of Agriculture (USDA) is the eradication of bovine brucellosis from the United States by the end of 1998. Although effective in the prevention of brucellosis, S19 is not without limitations. An obstacle to the eradication effort is the inability to serologically distinguish S19 vaccinates from naturally infected cattle. Additionally, infrequent but problematic sequelae of SI9 vaccination include abortion in pregnant cattle, "yi "J and persistent infection. arthritis,^^ anaphylaxis,^^ gg In April, 1996, for the first time in half a century a new brucellosis vaccine was approved for use in cattle. The vaccine. Brucella abortus strain RB51, is as effective as S19 in the prevention of brucellosis," but allows the serologic distinction of vaccinates from naturally infected animals.^ ^ When brucellosis infected herds have been detected, the USDA has implemented the practice of whole herd vaccination, where all female cattle, regardless of age or pregnancy status are vaccinated. In such situations, safety of SRB51 in pregnant cattle would be necessary to avoid widespread abortion. The studies presented in this dissertation were 2 designed to determine; 1) the tropism of SRB51 for the bovine placenta, 2) the ability of SRB51 to colonize the placenta, induce placentitis and cause abortion, 3) the immimogenicity of SRB51 vaccination in pregnant cattle, measuring both humoral and cell mediated responses, and 4) the effect of SRB51 vaccination on levels of the potent proinflammatory cytokine, TNF-a, in plasma and placental tissues. Literature Review Historical Background Human brucellosis was known possibly as early as the time of Hipocrates who described the symptoms of prolonged, intermittent fever; however, the earliest accurate description of the disease in man was made in 1861 when Marston designated it as Mediterranean fever and described the disease as " a preliminary stage of dyspespsia, anorexia, nausea, headache, feeling of weakness, lassitude, and inaptitude of exertion, mental or physical; chills, muscular pains, and lastly a fever having a long course 3 to 5 or 10 weeks, marked by irregular exacerbations and remissions, great derangement of the assimilative organs, tenderness in the epigastric region, and splenic enlargement. It is prone to relapses, has a protracted convalescence, and is frequently marked by rheumatism." The name "imdulant fever" was given to the disease in humans by Hughes in 1897.38 In 1887 Colonel David Bruce discovered a microorganism isolated from the spleen of fatal cases of Malta fever, also known as Mediterranean fever or undulant fever, in soldiers on the island of Malta. The causative organism, now known as Brucella melitensis, was named J Streptococcus melitensis by Hughes in 1892 38 and Micrococcus melitensis in 1893 by Bruce. Further confirmation of Bruce's conclusions were made in 1897 when it was shown that the organism that Bruce identified was agglutinated by sera from afflicted patients.'" The association of the disease in man with that seen in goats was recognized in 1905 when local goats on the island of Malta, which were going to be used as experimental animals, were found to have preexposure agglutinating antibodies to the organism. Subsequent culture of milk from the goats yielded the causative organism of Malta fever.'"* As early as 1864 epizootics of abortion were known to occur in regions of Louisiana and the Mississippi River. In 1864 in Cattle and their Diseases, Richard Jennings describes epizootics of abortion which he hypothesized were of a sympathetic or empathetic nature; that is, when one cow observed another cow aborting she would also abort in a few days or a few weeks. As erroneous as his etiologic diagnosis was, his recommended treatment of isolation of cows during calving so other cows could not observe an abortion, did help in controlling the disease. In 1862, Nocard, a French bacteriologist, first recognized the presence of bacteria between the fetal membranes and the wall of the uterus of a pregnant cow following abortion. Similarly, in 1897, Bang, a Danish veterinarian, assisted by Stribolt, noted a thick yellow exudate between fetal membranes and the uterus from an abortion. From this exudate they isolated a microorganism which when injected into pregnant cows caused abortion. The organism was reisolated from the abortion materials and injected into pregnant cows, reproducing abortion. Bang and colleagues were also able to produce abortion in pregnant sheep injected with the organism. Bang designated the organism. 4 Bacillus abortus and the disease came to be known as Bang's disease.^' The diseases of Malta fever in man and contagious abortion in cattle were studied independently and no connection between the two diseases was made until 1918. Evans made this connection by demonstrating the close morphological, cultural and serological relationship between the two bacteria causing the disease in man and sheep, B. melitensis and B. abortus, respectively, which are now known to be very closely related.28 In Rhodesia, Duncan was the first to recognize and report brucellosis in man due to B. abortus}^ The first case of brucellosis in America that was reported as being caused by other than B. melitensis was recognized in 1924;^° however, it was later found to be caused by B. suis rather than B. abortus. Description of the genus. Brucella Brucella abortus is of the Order Eubacteriales, Family Brucellaceae. The genus Brucella is unrelated to other pathogens but is closely related to the agrobacteriumrhizobium complex. Seven species of Brucella exist with preferential host relationships: B. abortus (cattle), B. melitensis (goats), B. suis (swine), B. neotomae (desert wood rat), B. canis (dogs), B. ovis (sheep) and a poorly characterized Brucella sp. isolated from dolphins."^ All 88 Brucella are closely related and share greater than 90% DNA homology. Some authors have suggested that only one species be recognized. Brucella melitensis, and that other specific epithets previously used as species names be used to designate biovars (e.g. Brucella melitensis biovar Aborms). 88 5 Brucella organisms are Gram-negative, cocco-bacilli, arranged singly, or occasionally in pairs or small groups. True capsules are not detected and Brucella do not form spores and are nonmotile. Brucella are aerobic and many strains require supplemental CO2 for optimal growth. Growth can be stimulated by the addition of 5 to 10% normal serum. Colonies generally become visible after 3 days incubation; however, growth may require up to 10 days in some cases. Brucella colonies after 3-5 days incubation are round, 1-2 mm in diameter with smooth margins, translucent and a pale honey color when viewed from through a transparent medium using a low power microscope and obliquely transmitted light. Viewed from above, colonies are convex and pearly white.^ Description of B. abortus strain RB51 Brucella abortus strain RB51 (SRB51) is a rifampin-resistant strain of B. abortus derived by repeated passage of strain 2308 on trypticase soy (TS) broth supplemented with varying concentrations of rifampin.'^ SRB51 is highly deficient in the 0-side chain component of LPS when compared to the parental 2308, but may contain small amounts of the perosamine homopolymer which is the 0-chain component of B. abortusRough morphology is characterized by absorption of crystal violet, autoagglutination and lack of the 0-chain component in LPS extracted from SRB51 on SDS - PAGE gels. Furthermore, Western blot analysis with the monoclonal antibody BRU 38, specific for the perosamine homopolymer 0-chain of smooth Brucella LPS indicates that SRB51 is highly deficient in the 0-chain compared to 2308.^' Biochemically, SRB51 resembles 2308 in the ability to 6 metabolize erythritol/^ This ability is linked to the presence of the eri gene which is present in both SRB51 and 2308, but absent in B. abortus strain 19 (SI9). B. abortus strain RB51 as a vaccine Vaccination v^th SRB51 is protective in mice and cattle which are challenged after vaccination with virulent B. abortus Cattle vaccinated with SRB51 remain serologically negative when tested by conventional brucellosis surveillance tests including tube agglutination, particle concentration fluorescence immunoassay (PCFA), card test and complement fixation (CF).81 These tests measure antibody responses to the LPS 0-side chain of B. abortus and, therefore, do not measure anti-SRB51 antibodies. SRB51 is cleared more rapidly than S19 or S2308 in both murine spleens and bovine lymph nodes.In mouse spleens, SRB51 was not cultured beyond 4 weeks after infection while S19 and S2308 were present at 10 and 20 weeks after infection, respectively.^^ In bovine lymph nodes, SRB51 was not recovered from lymph node biopsy specimens at 6 weeks after vaccination while S19 and S2308 were cultured at 8 weeks after vaccination.'' Strain 19 induced abortion The eradication of bovine brucellosis has centered on the use of Brucella abortus strain 19 (SI9), an attenuated, live vaccine introduced into the federal brucellosis program in 1940. Although S19 has been an effective tool in the control of bovine brucellosis for over 7 50 years it has significant limitations. These limitations include not only the production of antibodies indistinguishable from those resulting from natural infections, but also the ability 32.61 8 58 to induce abortion in pregnant cattle. " ' As early as 1937 it was noted that experimental subcutaneous injection in pregnant cattle of a dose of 2 x lO" S19 organisms may produce typical clinical signs and lesions of brucellosis with the discharge of large numbers of the organisms in uterine material at the time of parturition.^^ Cattle in which S19 vaccination induced weak or dead calves were in late gestation when vaccinated. However, one cow vaccinated at 4 months gestational age and challenged with a virulent strain 32 days later gave birth to a weak calf with S19 being isolated from placental and fetal tissues.^^ Experimental intravenous inoculation of S19 leads to infection of the placenta and abortion. Two separate studies in the United States and United Kingdom utilized experimental intravenous inoculation of S19 in pregnant cattle in larger than recommended doses. ' S19 induced abortion in 100% of the cattle tested in both experiments. Abortions occurred between 16 days and 42 days after inoculation. In pregnant cattle inoculated by the traditional subcutaneous route, the incidence of abortion is less. In a Canadian smdy, the effects of subcutaneous S19 vaccination on pregnant cattle were investigated.58 Of 112 cattle vaccinated between one and nine months of gestation, 16 (14.29%) aborted. However, it was noted that of 30 cows between six and nine months of gestation, four (13.33%) aborted and S19 could not be recovered from colostrum or uterine exudate in any of the four.58 8 Not only is route of vaccination important but also vaccine dose. In cattle vaccinated during pregnancy with l/20th the reconunended dose of 1.12 x 10" and challenged with a virulent strain 10 weeks later, S19 was recovered in 4/9 (44%) following parturition with one abortion and one premature birth.^ Vaccination with l/400th the recommended dose caused no abortions and isolation of S19 in colostrum from 1/9 (11 %) cows." Some researchers have associated S19 colonization of the bovine placenta and abortion with erythritol utilizing mutants.^' In 4/9 (44%) pregnant cattle, vaccinated with 5.8 X lO' S19 organisms, S19 was isolated from cotyledonary tissues and in 2 cows colonization led to abortion or premature calving. These two isolates were determined to be erythritol utilizing mutants of SI 9.^^ Similarly, 307 head of cattle, three to five months pregnant were o vaccinated with a dose of 3 x 10 S19 organisms and abortion occurred in 15 (4.9%) of the cows with four of the isolates being erythritol utilizing variants of S19. In spite of these findings which suggest that erythritol utilization is responsible for the placental tropism of S19 and other Brucella strains there is evidence to the contrary. Species which do not synthesize erythritol in placental tissues, i.e. humans, rats, mice, rabbits and guinea pigs can become infected with Brucella sp. and develop placentitis.^^'"'^ Moreover, rate of erythritol metabolism has not been correlated with increased virulence in Brucella sp.^^'^'* Therefore, it is likely that although erythritol may enhance Brucella growth, it is not essential for growth or placental infection. Large numbers of pregnant cattle can be vaccinated with S19 with minimal concern for large numbers of vaccine-induced abortions. In a Florida field trial, cows in late gestation 9 were vaccinated with S19 at a dose of lO" organisms / animal. Vaccination with S19 induced abortion in less than 1% of over 10,000 cows.^' A small percentage of cows vaccinated with S19 as calves may retain S19 infection into adulthood and in some instances this persistent infection may be responsible for o/r abortion. A study in the United Kingdom showed that of 34 isolates of bovine origin determined to be SI9, 11 were from abortion material. 86 Calfhood vaccination was the only known exposure to S19 in all cases. Virulence Factors Brucella with smooth colony morphology are more virulent than Brucella with rough TX colony morphology, and this suggests a significant role for LPS in the virulence of B. abortus. Both smooth and rough Brucella are able to enter host cells; however, rough Brucella may enter host cells more effectively.'^ Smooth Brucella survive and even multiply while the rough bacteria are eventually eliminated.^^ Survival of Brucella has been related to various factors including their ability to inhibit phagosome-Iysosome fusion,^" '® resist destructive effects of lysosomal enzymes after fusion has occurred,^" and inhibit TNF-a secretion by phagocytic cells. An important property of Brucella sp. is the ability to survive and multiply intracellularly in host phagocytes. Extracts from the smooth virulent B. abortus strain 2308 are able to suppress phagosome-Iysosome fusion in murine peritoneal macrophages.^^ In bovine neutrophils, extracts of B. abortus strain 2308 show a concentration dependent 10 inhibition of primary granule release.This inhibition may be due to the production of 5' guanosine monophosphate (GMP) and/or adenine which are found in the culture supematant of live Brucella but not S. epidermidis or E. coli}^ Adenine and GMP have been shown to inhibit the ability of neutrophils to iodmate proteins.'^ Macrophage lysosomal release has also been shown to be inhibited by adenosine and other nucleosides and nucleotides most likely by inhibition of methylation of membrane proteins or phospholipids important in membrane fusion.^' Protein iodination and lysosomal release are important antibacterial mechanisms employed by phagocytes. Persistence of Brucella in phagolysosomes of nonactivated macrophages is thought to provide safety from host antibacterial factors. However, activated macrophages are able to phagocytose and kill Brucella. B. abortus has a predilection for the placental cotyledons, specifically the trophoblast.'^ Pathogenesis studies have shown that in placental infection, B. abortus is first seen in phagosomes of erythrophagocytic trophoblastic epithelial cells of the placentome. Following replication within erythrophagocytic trophoblastic cells, B. abortus next replicate within the rough endoplasmic reticulum cistemae and nuclear envelope of adjacent chorioallantoic cells of the trophoblast."* Hematogenous spread to chorionic villi and the fetus follow necrosis of trophoblast and ulceration of the chorioallantoic membrane. It remains unclear if Brucella are passively phagocytosed by the trophoblastic cells or if there is a more selective receptor-ligand interaction between the bacteria and trophoblastic cells. However, indiscriminate phagocytosis of Brucella would not account for the remarkable numbers of 11 bacteria within trophoblastic cells when few bacteria are seen free in the extracellular space, suggesting a more specific tropism for the placental trophoblast. Brucella sp. synthesize several proteins which have been determined to be important in protecting bacteria against the damaging effects of host cell enzymes including superoxide dismutase/'* catalase, and high temperature resistant stress proteins.^ Construction and testing of genetically altered B. abortus mutants deficient in the Cu-Zn superoxide dismutase gene, htrA heat shock gene, catalase gene, urease gene, recA gene or a gene encoding a 31kDa protein of unknown fimction have not resulted in mutants of decreased . , 83,I9,46J4.64 virulence. Iron chelating siderophore synthesis is a virulence mechanism utilized by various pathogens to survive in the low iron environment of the host. In animal cells and body fluids free iron levels are low. Rather, iron is bound to host proteins such as transferrin and lactoferrin. Bacteria have developed proteins (siderophores) which scavenge iron from host 48 iron binding proteins in low iron envirorunents. Brucella abortus has been shown to produce the siderophore 2,3-dihydrobenzoic acid when grown under iron-limiting situations in vitro.^^ However, the importance of siderophores in vivo has not been established. Enzymes such as nitric oxide are of limited importance in host defense against brucellosis. Addition of I-arginine, an inhibitor of nitric oxide synthesis, to Brucellainfected macrophages in vitro has only minor affects on killing of Brucella?^ Nitric oxide has been shown to be important in resistance to infection with the intracellular parasite Leishmania.^^ 12 Placental lesions in cattle with brucellosis Early descriptions of gross lesions of brucellosis consisted of an abundant yellow exudate with a slimy, somewhat lumpy character found between the uterine mucosa and the allantochorion. This exudate contained numerous short bacilli.® Later microscopic descriptions reported necrosis of trophoblast cells and large numbers of intracellular bacilli within trophoblast cells. Uterine mucosal epithelial cells as well as amniotic epithelial cells were reported to be unaffected.'^ More recent and complete descriptions describe placental cotyledons with marked inflammation characterized by cell debris, intact and degenerate neutrophils and bacteria lying between the maternal caruncle and the allantochorion.®^ These changes in the placental cotyledon were limited to the periphery of the cotyledon early in infection (3-4 weeks after infection) and did not extend to the interior of the cotyledon until 5 weeks after infection. Close examination of the trophoblast layer showed masses of bacteria within the trophoblast. Inflammation extended to the connective tissues of the allantochorion which was edematous and contained numerous neutrophils. Endothelial cells lining fetal blood vessels contained large numbers of intracellular bacilli. Maternal tissues of the placentome were largely unaffected.®® Additionally, a mild interstitial lymphoplasmacytic mastitis with acinar lumina variably filled with neutrophils was described in experimentally infected cattle.®® 13 Development and anatomy of the placenta Placental development begins early during the morula stage as the outer layer of ectodermal cells differentiate into trophoblastic cells. Trophoblastic epithelimn is differentiated to physically cormect the developing embryo with the uterus. Differentiation includes the redistribution of intercellular adhesion molecules such as E-cadherin and Na^/K^ ATPases to the basolateral aspect of the trophoblastic epithelial cell. Cadherins and ATPases are present on all cells in the morula, but redistribution occurs only in trophoblastic epithelial cells. The Na^/K^ ATPase may fimction in ion and fluid fluxes necessary to create the cavity of the blastocyst known as the blastocoel. Cytokeratins characteristic of epithelial cells are detectable in differentiating trophoblastic epithelial cells at this early morula stage. Remaining cells of the morula differentiate at one pole into the inner cell mass. Endodermal and mesodermal components of the placenta later develop from portions of the inner cell mass. The continued development of the remaining inner cell mass into the embryo does not proceed until the first structures of the placenta have been formed. Early in placental development and coincident with implantation, maternal recognition of pregnancy is required. In a broad sense, maternal recognition is the sequence of events required to ensure that the conceptus is not rejected by the maternal immune system. It is hypothesized that release of cytokines selectively modulates the immune cells at the implantation site. In some species, the release of factors from the placenta act in a paracrine or endocrine fashion to 14 ensure continued pregnancy. In ruminants the secretion of type I interferons from trophoblastic epithelial cells plays a key role in maternal recognition of pregnancy. Progesterone is required during pregnancy to maintain uterine endometrium in a secretory and receptive state which is optimal for embryonal growth and development. Therefore, it is vital during pregnancy to ensure the continued secretion of progesterone. The major source of progesterone is the postovulation ovarian structure the corpus luteum (CL). Since the gestational period for most species is longer than the functional period of the normal corpus luteum, during pregnancy it becomes necessary to extend the functional life of the CL. Different mammals utilize different strategies to extend CL function. Primates and horses produce a luteotrophic substance, chorionic gonadotropin, which supports the CL by supplementing the action of luteinizing hormone from the anterior pituitary. In ruminant species, a type I interferon, IFN-x, is produced by mononucleate trophoblastic epithelial cells. IFN-t inhibits prostaglandinF2a (PGF2a)-^ Prostaglandin F2a is produced by the uterus and serves to lyse the CL in nonpregnant animals. The expression of IFN-t mRNA occurs as early as postconception day 7, is maximal around days 16-19, and is limited to the mononucleate trophoblastic epithelial cells and has not been detected in binucleate frophoblastic cells, embryonic disc cells or embryonic endodermal cells.^ Implantation of the conceptus proceeds differently among different species. In rodents and primates, implantation occurs soon after the blastocyst hatches from the zona pellucida. In ruminants, swine and horses the conceptus remains free in the uterine lumen for several days before implantation. In rodents, the study of implantation has shown that 15 trophoblast - uterine attachment may be initiated by carbohydrate - lectin interactions. Mouse blastocysts have been shown to express several carbohydrate structures including the lectin ligand, sialyl lewis X which is also important in lymphocyte - high endothelial cell interaction in lymph node postcapillary venules. The maternal response to the implanting embryo includes vascular changes such as • • 89 angiogenesis, vasodilation and increased permeability. Hypercellularity occurs as inflammatory cells (large granular lymphocytes, macrophages and 78+ T cells) are recruited to the implantation site.^°'^^ Several proinflammatory cytokines are produced in the uterus and increased angiogenesis is noted. In primates and rodents the uterine epithelium is lost, and poorly characterized stromal cells, decidual cells, undergo transformation and proliferation. This decidua contains numerous inflammatory cells such as macrophages, lymphocytes, natural killer cells and lymphocytes of the y/S subtype, y/6 T-cells have been shown to play a protective role on some mucosal surfaces and may be especially important in infections due to intracellular bacterial pathogens. 78 Early steps in placental development and cell differentiation may proceed similarly between species; however, the degree of invasion of uterine tissues and the final form the placenta takes are quite variable. In primates and rodents the trophoblast invades maternal vasculature in the submucosa creating a hemochorial placenta. In the pig, the trophoblast is noninvasive and forms a diffiise apposition to the uterine epithelium resulting in a diffuse epitheliochorial placenta. In ruminants, the trophoblastic epithelium is noninvasive; however, some binucleate trophoblastic cells fuse with uterine epithelial cells forming 16 syncytia which then release granules of placental lactogen into the maternal circulation. Syncytia formation is more pronounced in small ruminants such as sheep and goats than in cattle. Intimate contact between chorionic trophoblast and uterine epithelium occurs in ruminants in specialized structures termed placentomes. The maternal contribution, the uterine caruncles, occur in four rows more or less evenly spaced in the uterus and are present in both uterine homs.''"^^"^' The bovine uterus contains approximately 100-160 caruncles.^^ The size of the placentome is dependent on stage of pregnancy and position in the uterus.'^ Nonpregnant, multiparous females and virgin heifers have well-defined caruncles composed of small cells with small, dense oval or spindle-shaped nuclei. Ultimately, these cells are the source of growing matemal connective tissue during the development of the placentome.^^ The expanding chorioallantoic membrane of the early embryo is apposed closely to the uterine wall, and in those areas apposing caruncles, the chorioallantoic membrane enters narrow spaces within the caruncle. The chorioallantoic membrane associated with caruncles forms thickenings described as "milky patches". These fragile attachments between caruncle and chorioallantois occur as early as 33 days after ovulation." The mesoderm of these patches contains thin walled allantoic blood vessels with primitive nucleated erythrocytes. The caruncular tissue forms a crypt system composed of numerous deep and extensively branched crypts which open toward the chorioallantoic membrane. Initially the crypts are filled with cell debris and erythrocytes while being lined by uterine epithelial cells. The corresponding chorioallantoic membrane forms villi with loose mesenchymal cores which are well vascularized by allantoic blood vessels. Villi are covered by a simple layer of 17 trophoblastic epithelium.'^" Trophoblastic epithelium comprises a heterogeneous population of erythrophagocytic, placentomal and chorioallantoic trophoblastic cells that differ in location and fimction. Chorioallantoic villi penetrate into a network of caruncular crypts. Villi advance through the caruncle until their tips reach the base of the caruncle. The trophoblastic and cryptal cells have surface microvilli which interdigitate with each other, thus increasing the sxirface area of contact between fetal and maternal tissues. " Primary chorionic villi run deep and straight to the base of the placentome, decreasing slightly in diameter toward the base. Branches of secondary and tertiary villi form, and interdigitate with maternal crypts of primary, secondary, and higher order. The bases of the primary villi form arcades around club-shaped endings of maternal septa." Further grovrth of the placentome comes from enlargement of the diameter and length of existing villi and/or from 37 extensive villus branching and increased surface corrugation. In ruminants, the placentomal arcade zone is the area between the base of primary chorionic villi and the tips of maternal septa. Within this space there is often extravasated blood, autolyzed cells and uterine secretions. This material is often resorbed by the adjacent trophoblast cells." In sheep, this zone has been studied extensively where extravasation of blood is significant and can be seen grosslyIn the bovine placenta the amount of extravasation is not as great; however, small deposits of blood are first noted in the third month of gestation and increase in volume during gestation. Erythrophagocytic chorionic trophoblastic epithelial cells in this zone contain erythrocytes in various degrees of degeneration as well as hemoglobin derived pigment.^"'" Erythrophagocytosis is thought to 18 be a major pathway of iron acquisition for the fetus as well as an important factor in the pathogenesis of placentitis and fetal infection in brucellosis infection in goats'* and sheep/^ The intercotyledonary area is situated between placentomes and is comprised of a maternal side of glandular uterus lined on the surface by simple epitheliiun of varying height. The fetal side is represented by the allantochorion which is sometimes folded. The surface facing the maternal side is lined by trophoblast. Trophoblastic epithelial cells opposite the surface openings of uterine glands are also highly phagocytic, ingesting uterine secretions. Pathogenesis of bovine abortion Both maintenance and termination of pregnancy involve a complex interplay between fetus, placenta, uterus and various fetal and maternal endocrine control mechanisms. Any of these factors must be considered a potential target of any abortifacient. Further complicating the scenario are species differences concerning known abortifacients. For example, the abortifacient activity of PGF2a in rats is largely due to the disturbance of luteal function while in guinea pigs, PGF2a causes abortion due to oxytocic effects on uterine contractility."^ The definitive mechanism of abortion in brucellosis is unknown. Fetal death and abortion may be due to interruption of nutrient and oxygen transfer by the placenta and subsequent fetal hypoxia;however, rats infected with Brucella abortus or Listeria monocytogenes showed large areas of placental necrosis but no fetal death. However, if placental destruction was extreme, survival was not possible.^^ These findings suggest that more than disruption of the fetal maternal interface is involved in brucellosis abortion. 19 Another hypothesis involves the toxic effects of endotoxin, including increased uterine sensitivity to oxytocin following exposure to Brucella endotoxin.^'* As early as 1943 endotoxin was found to be a potent abortifacient. Placental hemorrhage and abortion in mice and rabbits was demonstrated following intraperitoneal injections with Shigella paradysenteriaeP It was hypothesized that decidual and placental capillaries have a much lower threshold of damage due to endotoxin than vessels of other organs.'^ Investigators later described fibrinoid swelling in the placental interstitium in addition to placental hemorrhage and abortion in mice and rabbits following administration of endotoxin. These findings led to the suggestion that endotoxin caused a Shwartzmann-like reaction in the placenta resulting in abortion. This idea was later discounted when it was found, by other researchers, that there were areas of hemorrhage in the decidua basalis in untreated mice at term as well as endotoxin-treated mice.^^ This finding suggested that this histologic change represented a mechanism by which normal placental separation proceeded and not a change specific for endotoxin-induced abortion. A finding which was not seen in control mice; however, was vascular engorgement of the placental labyrinth. Cattle given LPS during gestation abort within 10 days; however, the fetus does not appear to play a significant role in endotoxin-induced abortion.^' Several parameters were evaluated in pregnant cattle following intravenous infusion of Salmonella typhimurium LPS using catheterized fetal and placental structures.^' Maternal responses included increases in PGF2a, TNFa, ACTH, Cortisol and decreases in leukocytes and progesterone with rapid clearance of endotoxin firom maternal plasma. Fetal changes included increases in ACTH 20 and Cortisol with delayed increases in PGE2 in amniotic and allantoic fluids and no elevations of fetal TNFa levels and no detection of LPS in fetal plasma. No gross or microscopic lesions were seen in the fetuses and fetal motion and blood O2 levels were normal up to the time of abortion. These findings do suggest; however, a possible role for increases in PGF2a, TNFa or Cortisol in endotoxin induced abortion. The luteolytic effect of PGF2a may play a more important role in early first trimester abortion rather than in abortions which occur during the second or third trimester, more typical of brucellosis.^^ In cattle, pregnancy during early gestation is maintained primarily by progesterone secreted by the CL. In advanced gestation, extraluteal sources of progesterone, such as the placenta, predominate.^' One report showed cows in the first trimester of pregnancy are more sensitive to the abortigenic effects of LPS than are cows in the second or third trimester.^^ The increased sensitivity of first trimester cows may be attributable to the lack of extraluteal sources of progesterone. The oxytocic effects of PGFia; however, could play a role in abortion at any stage. In humans, LPS inhibits prostaglandin dehydrogenase, the prostaglandin inactivating enzyme creating a net increase in prostaglandins.^'' Gram negative bacterial invasion of placental and uterine tissues in humans is characterized by the release of bacterial LPS and phospholipases from bacterial and placental sources. Phospholipases cause the release of arachidonic acid and subsequent elevations in prostaglandins as well as the induction of cytokines. 78 Metabolism of PGE2 from human amnion cells has also been shown to increase during term as well as pathogen -induced 21 labor.' The role of prostaglandins in the initiation of term and preterm parturition are supported by several observations: 1) prostaglandins cause cervical ripening and uterine contractions; both necessary preparatory changes for fetal expulsion, 2) the precursor of prostaglandins, arachidonic acid, is found in increased levels in the anmiotic fluid of laboring women, 3) injection of arachidonic acid into the amnion stimulates labor, and 4) inhibitors of prostaglandin synthesis inhibit parturition. In addition to the effects of LPS mediated through prostaglandins, maternal endotoxemia results in the stimulation of both the fetal and maternal hypothalamic - pituitary - adrenal axes.^^"^' This results in a sharp rise in fetal and maternal adrenocorticotrophic hormone (ACTH) and Cortisol. Activation of this response in normal term pregnancies in ruminants is thought to initiate the physiologic process of parturition.^" Increased Cortisol values have been noted in fetuses inoculated with virulent B. abortus as early as PID 6;27 therefore, this may play a role in premature labor as Brucella organisms multiply within trophoblastic cells. As trophoblastic cells are shed into the inflammatory exudate, high numbers of organisms can potentially be achieved. As local LPS and prostaglandin concentrations increase, parturition may be induced. A strong correlation between prostaglandin levels with TNF-a and IL-1 levels in amniotic fluid has been noted in women during preterm labor.'^ It is well known that LPS can cause the release of cytokines such as TNF-a and IL-1 from macrophages; furthermore, TNF-a stimulates prostaglandin synthesis by several cell types.23 The significance of uteroplacental TNF-a is not entirely clear, but it likely involves the following; regulation of 22 cell proliferation at the uteroplacental interface (TNF-a inhibits DNA synthesis of trophoblastic cells), regulation of expression of MHC-I antigens necessary for the binding of many hormones and growth factors, and elevated levels of TNF-a have been implicated in the initiation of preterm labor associated with infection in women/'^^ In humans epidemiological, bacteriologic and histologic evidence for the association of bacterial infection and preterm labor exists. Prenatal association of specific infectious agents and preterm labor are suggestive but inconsistent. These agents include, Neisseria gonorrhea, group B streptococci. Chlamydia trachomatis. Mycoplasma hominis, Ureplasma urealyticum, and Trichomonas vaginalisJ'^ Histologic evidence of an association of infection and inflammation with preterm labor has been reported.'*^ Histologic findings such as these, as well as bacteriologic findings can be debated on the argument that bacteriologic and tissue changes may have occurred after labor. However, studies in which measurements of the onset of uterine contractions and membrane integrity were controlled, show that histologic and microbiologic changes occur prior to the initiation of parturition.''^ A potential model is hypothesized in which bacterial infection of uterine and/or placental tissues occurs. Bacterial factors such as LPS activate the host monocyte macrophage system to produce TNF-a as well as other proinflammatory cytokines. TNF-a and other cytokines stimulate the release of prostaglandins from intrauterine tissues which signal the initiation of parturition.^' This model is supported by the detection of measurable TNF-a in >90% of women with preterm labor and amniotic fluid from which bacteria were 23 cultured. TNF-a was not detected in amniotic fluid from women during the second and third 71 72 This model is not without opponents. Other trimesters who were not in labor. ' researchers have found that amniotic fluid TNF-a levels increase during pregnancy and normal parturition as well as preterm labor.'^^ This same study showed no detectable changes in plasma TNF-a of pregnant women at any time during gestation. The cellular origin of uteroplacental TNF-a is also not completely understood. In vitro studies have shown the release of TNF-a by decidual and chorionic villi explants, cytotrophoblast cells in culture, dispersed amniochorion cells and macrophages isolated from chorionic and decidual tissues."*^ In ruminants, TNF-a levels rise following the intravenous infusion of endotoxin. Chronic administration of endotoxin to cattle results in depression, cachexia, and diarrhea with renal hemorrhage and interstitial nephritis, thymic atrophy, myocardial atrophy and subendocardial hemorrhage, skeletal muscle atrophy, serous atrophy of fat, lymphoid depletion, hepatic necrosis and vacuolar change, pancreatic necrosis and ileal villus atrophy In periparturient cattle, isolated mononuclear cells are more responsive to the effects of endotoxin, producing more TNF-a, than those of cows in mid to late lactation.^' Furthermore, supramammary lymph node mononuclear cells were more responsive to endotoxin than those from the peripheral blood.'' The role of increased mononuclear cell responsiveness to the effects of endotoxin in periparturient diseases such as late term abortion, endometritis and mastitis is unclear; however, it may provide insight into the high prevalence of these periparturient diseases of cattle. 24 In addition to the uterine synthesis of prostaglandins stimulated by proinflammatory cytokines, the luteolytic action of PGF2a may also be mediated through TNF-a. Using a procedure known as continuous flow microdialysis, it has been shown in vivo, that TNF-a is increased in luteal tissue coincident with a decline in progesterone/^ Immunohistochemical studies have localized the source of luteal TNF-a to macrophages within the CL. These findings suggest that the luteolytic action of PGF^a may, in part, be mediated through TNFa. Tumor necrosis factor is increased in the milk of dairy cattle near parturition.68 Milk TNF-a levels peak 4 to 6 weeks prior to parturition and decrease to undetectable levels at parturition. Following parturition, TNF-a levels increase rapidly and maintain a midlevel concentration throughout lactation.68 It is not noted if plasma TNF-a levels follow a similar pattern as those seen in the milk; however, it is assumed that the source of the TNF-a is the increased number of macrophages in milk which parallel the increase in TNF-a.68 Dissertation Organization This dissertation is prepared in the alternate format and includes manuscripts prepared and submitted to refereed scientific journals. The format used in the individual chapters is consistent with the scientific journal to which the manuscript was submitted. The general introduction and general conclusion chapters are prepared in a format consistent v^th Veterinary Pathology. Literature cited in the general introduction and general conclusion sections appears at the end of the respective section. 25 References 1. Adams JL, Semrad SD, Czuprynski CJ: Administration of bacterial lipopolysacharide elicits circulating tumor necrosis factor-alpha in neonatal calves. J Clin Micro 28:9981001,1990 2. Alton GG, Comer LA, Plackett P: Vaccination of pregnant cows with low doses of Brucella abortus strain 19 vaccine. Austral Vet J 56:369-372,1980 3. Alton GG, Jones LM, Angus RD, Verger JM. Techniques for the brucellosis laboratory. Bacteriologic Methods. Paris France: Institut National de la Recherche Agronomique; 1988:34-60. 4. Anderson TD, Meador VP, Cheville NF: Pathogenesis of placentitis in the goat inoculated with Brucella abortus. I. Gross and histologic lesions. Vet Pathol 23:219-226,1986 5. Austgulen R, Espevik T, Mecsei R, Scott H: Expression of receptors for tumor necrosis factor in human placenta at term. Acta Obstet Gynecol Scand 71:417-424,1992 6. Bang B: Infectious abortion in cattle. J Comp Path Therapeut 19:191-202,1906 7. Bazer FW, Spencer TE, Ott TL: Placental interferons. Amer J Reprod Immunol 35:297308,1996 26 8. Beckett FW, MacDiarmid S.C.: The effect of reduced dose Brucella abortus strain 19 vaccination in accredited dairy herds. Br Vet J 141:507-514,1985 9. Bennett PR, Rose MP, Myatt L, Elder MG: Preterm labor; Stimulation of arachidonic acid metabolism in human amnion cells by bacterial products. Obstet Gynecol 156:649-655,1987 10. Bertram TA, Carming PC, Roth JA: Preferential inhibition of primary granule release from bovine neutrophils by a Brucella abortus extract. Infect Immun 52:285-292,1986 11. Bjorkman N: Morphologic and histochemical studies on the bovine placenta. Acta Anat (Basel) 22 Suppl. 22-2:5-26,1954 12. Bjorkman N, Sollen P: Morphology of the bovine placenta at normal delivery. Acta Vet Scand 1:347-362,1960 13. Braun AI, Pamales-Lebron A, Stinebring WR: Interaction between mononuclear and phagocytes and Brucella abortus strains of different virulence. Proc Soc Exp Biol Med 97:393-397,1958 14. Bricker BJ, Tabatabai LB, Judge BA, Deyoe BL, Mayfield JE: Cloning, expression and occurrence of the Brucella Cu-Zn superoxide dismutase. Infect Immim 58:2935-2939,1990 27 15. Canning PC, Roth JA, Deyoe BL; Release of 5'-guanosine monophosphate and adenine by Brucella abortus and their role in the intracellular survival of the bacteria. J Infect Dis 154:464-470,1986 16. Canning PC, Roth JA, Tabatabai LB, Deyoe BL: Isolation of components of Brucella abortus responsible for inhibition of function in bovine neutrophils. J Infect Dis 152:913921,1985 17. Caron E, Gross A, Liautard J-P, Domand J: Brucella species release a specific, proteasesensitive, inhibitor of TNF-a expression, active on human macrophage-like cells. J Immunol 156:2885-2893,1996 18. Caron E, Peyrard T, Kohler S, Cabane S, Liautard JP, Domand J: Live Brucella spp. fail to induce tumor necrosis factor alpha excretion upon infection of U937- derived phagocytes. Infect Immun 62:5267-5274,1994 19. Cheville NF, Jensen AE, Hailing SM, Tatum FM, Morfitt DC, Hennager SG, Frerichs WM, et al.: Bacterial survival, lymph node changes and immunologic responses of cattle vaccinated with standard and mutant strains of Brucella abortus. Am J Vet Res 53:18811888,1992 28 20. Cheville NF, Stevens MG, Jensen AE, Tatum FM, Hailing SM: Immvine responses and protection against infection and abortion in cattle experimentally vaccinated with mutant strains of Brucella abortus. Am J Vet Res 54:1591-1597,1993 21. Comer LA, Alton GG: Persistence of Brucella abortus strain 19 infection in adult cattle vaccinated with reduced dosages. Res Vet Sci 31:342-344,1981 22. Dale HE, English JE: Effect of bmcellosis vaccination on rectal temperature and feed consumption of beef and dairy calves. Vet Med 52:477-480,1957 23. Dayer JM, Beutler B, Cerami A: Cachectin/tumor necrosis factor stimulates collagenase and prostaglandin E2 production by human synovial cells and dermal fibroblasts. J Exp Med 162:2163-2168,1985 24. Dennis SM: The effect of bacterial endotoxin on pregnancy. Vet Bull 36:123-127,1966 25. Duncan JT: Contagious abortion and undulant fever. Br Med J 1:554,1925 26. Elger W: Pharmacology of parturition and abortion. Animal Reproduction Science 2:133148,1979 27. Enright FM, Walker JV, Jeffers G, Deyoe BL: Cellular and humoral responses of Brucella aiorrw^-infected bovine fetuses. Am J Vet Res 45:424-430,1984 29 28. Evans AC: Further studies on Bacterium abortus and related bacteria. II. A comparison of Bacterium abortus with Bacterium bronchisepticus and with the organism which causes Malta fever. J Infect Dis 22 :580,1918 29. Ewalt DR, Payeur JB, Martin BM, Cummins DR. Miller WG: Characteristics of a Brucella species from a bottlenose dolphin {Tursiops truncatus). J Vet Diagn Invest 6:448452,1994 30. Diagnosis of abortion. Veterinary Clinics of America: Food Animal Practice. Foley GL, Schlafer DH. Bacterial Endotoxemia and Reproductive Effects in Ruminants. Philadelphia, PA: W.B. Saunders Co.; 1994;I0:3:491-501. 31. Foley GL, Schlafer DH, Elsasser TH, Mitchell M: Endotoxemia in pregnant cows: Comparisons of maternal and fetal effects utilizing the chronically catheterized fetus. Theriogenology 39:739-762,1993 32. Frenchick PJ, Markham F, Cochrane AH: Inhibition of phagosome-lysosome fusion in macrophages by soluble extracts of virulent Brucella abortus. Am J Vet Res 46:332335,1985 30 33. Giri SN, Emau P, Cullor JS, Stabenfeldt GH, Bruss ML, Bondurant RH, Osbum BI: Effects of endotoxin infusion on circulating levels of eicosanoids, progesterone, Cortisol, glucose and lactic acid, and abortion in pregnant cows. Vet Microbiol 21:211-231,1990 34. Hailing SM, Detilleux PG, Tatum FM, Judge B, Mayfield JE: Deletion of the BCSP31 gene of Brucella abortus by replacement. Infect Immun 59:3863-3868,1991 35. Haring CM, Traum J: Observations of pathogenic and antigenic effects of Brucella abortus. United States Bureau of Animal Industry strain 19. J Agric Res 55:117-128,1937 36. Harmon BG, Adams LG, Frey M: Survival of rough and smooth strains of Brucella abortus in bovine manamary gland macrophages. Am J Vet Res 49:1092-1097,1988 37. Hradecky P, Mossman HW, Stott GG: Comparative development of ruminant placenta. Theriogenology 29:715-729,1988 38. Hughes ML: The natural history of certain fevers occurring in the Mediterranean. The Mediterranean Naturalist 2:325,1892 39. Jiang X, Leonard B, Benson R, Baldwin CL: Macrophage control of Brucella abortus: Role of reactive oxygen intermediates and nitric oxide. Cell Immunol 151:309-319,1993 31 40. Keefer CS: Report of a case of Malta fever originating in Baltimore, Maryland. Bull Johns Hopkins Hosp 35:6,1924 41. Kenison DC, Elsasser TH, Payer R: Tumor necrosis factor as a potential mediator of acute metabolic and hormonal responses to endotoxemia in calves. Am J Vet Res 52:13201326,1991 42. Kniazeff AJ, Elberg SS, Traum J: Pathogenesis of experimental brucellosis of rabbits in pseudopregnancy. J Comp Path 74:141-150,1964 43. Knox IC, Hoemer JK: The role of infection in preterm rupture of membranes. Amer J Obstet Gynecol 59:190-194,1950 44. Kohn MA, Hillier SL, Martius J, Eschenbach DA, Kiviat NB: Bacterial vaginosis is related to prematurity. Abstract 1216. 26th Interscience Conference on Antimicrobial Agents and Chemotherapy. New Orleans, LA ,1986 45. Laham N, Brermecke S, Bendtzen K, Rice G: Tumour necrosis factor a during human pregnancy and labour; matemal plasma and amniotic fluid concentrations and release from intrauterine tissues. Eur J Endocrinol 131:607-614,1994 32 46. Latimer E, Simmers J, Sriranganatlian N, Roop RM, Schurig GG, Boyle SM: Brucella abortus deficient in copper/zinc superoxide dismutase is virulent in BALB/c mice. Microb Pathog 12:105-112,1992 47. Liew FY, Millot S, Parkinson C, Palmer RMJ, Moncada S: Macrophage killing of Leishmania parasite in vivo is mediated by nitric oxide from L-arginine. J Immunol 144:4794-4797,1990 48. Litwin CM, Calderwood SB: Role of iron in regulation of virulence genes. Clin Micro Rev 6:137-149,1993 49. Lopez-Goni 1, Moriyon I, Neilands JB: Identification of 2,3 dihydroxybenzoic acid as a Brucella abortus siderophore. Infect Immun 60:4496-4503,1992 50. Marston JA: Report on fever (Malta). Army Medical Report 3:486,1861 51. McDonald LE, McNutt SH, Nichols RE: On the essentiality of the bovine corpus luteum of pregnancy. Am J Vet Res 14:539-541, 1953 52. McGregor JA, French JI, Lawellin DJK: Preterm birth and infection: Pathogenic possibilities. Amer J Reprod Immunol 16:123-132,1988 33 53. Meyer ME: Metabolic characteristics of the genus Brucella V. Relationship of strain oxidation rate of /-erythritol to strain virulence for guinea pigs. J Bacteriol 92:584-588,1966 54. Meyer ME: Metabolic characteristics of the genus Brucella VI. Growth stimulation by /erythritol compared with strain virulence for guinea pigs. J Bacteriol 93:996-1000,1967 55. Mincheva-Nilsson L, Hammarstrom SHM-L: Human decidual leukocytes from early pregnancy contain high numbers of y5+ cells and show selective down regulation of alloreactivity. J Immunol 149:2203-2211,1992 56. Mingle CK, Manthei CA, Jasmin AM: The stability of reduced virulence exhibited by Brucella abortus strain 19. J Am Vet Med Assoc 99:203-204,1941 57. Mollelo JA, Jensen El, Collier JR, Flint JC: Placental pathology. III. Placental lesions of sheep experimentally infected with Brucella abortus. Am J Vet Res 24:915-921,1963 58. Moore T, Mitchell CA: Vaccination of sexually mature cows wdth Brucella abortus strain 19 vaccine. Can J Comp Med 14:209-213,1950 59. Mossman HW. Vertebrate Fetal Membranes. New Brunswick, NJ: Rutgers Univ. Press; 1987:290. 34 60. Murai TYS: Erythrophagocytosis by the trophoblast in a bovine placentome. Jpn J Vet Sci 48:75-88,1986 61. Nicoletti P: A preliminary report on the efficacy of adult cattle vaccination using strain 19 in selected dairy herds in Florida. Proc. 80th Annu. Meet. U.S. Animal Health Assoc. 91100,1977 62. Nielsen K, Duncan JR. Animal Brucellosis. Nicoletti P. Vaccination. Boca Raton, FL: CRC Press; 1990:283-300. 63. Ohmann HB, Campos M, Snider M, Rapin N, Beskorwayne T, Popowych Y, Lawman MJP, et al.: Effect of chronic administration of recombinant bovine tumor necrosis factor to cattle. Vet Pathol 26:462-472,1989 64. Palmer MV, Cheville NF, Tatum FM: Morphometric and histopathologic analysis of lymphoid depletion in murine spleens following infection Mdth Brucella abortus strains 2308, RB51 or an htrA deletion mutant. Vet Pathol 33:282-289,1996 65. Payne JM: Changes in the rat placenta and foetus following experimental infection with various species of bacteria. J Pathol Bacteriol 75:367-385,1958 35 66. Payne JM: The pathogenesis of experimental brucellosis in the pregnant cow. J Pathol Bacteriol 78:447-459,1959 67. Reider RF, Thomas L: Studies on the mechanisms involved in the production of abortion by endotoxin. J Immunol 84:189-193,1960 68. Rewinski MJ, Yang TJ: Lactation stage-dependent changes in levels of tumor necrosis factor/cachectin in milk. Amer J Reprod Immunol 31:170-176,1994 69. Riches DWH, Watkins JL, Henson PM, Stanworth DR: Regulation of macrophage lysosomal secretion by adenosine, adenosine phosphate esters and related structural analogues of adenosine. J Leuk Biol 37:545-557,1985 70. Riley LK, Robertson DC: Ingestion and intracellular survival of Brucella abortus in human and bovine polymorphonuclear leukocytes. Infect Immim 46:224-230,1984 71. Romero R, Brody DT, Oyarzun E, Mazor M, Wu YK, Hobbins JC, Durum SK: Infection and labor: Interleukin-l: A signal for the onset of parturition. Amer J Obstet Gynecol 160:1117-1123,1989 72. Romero R, Mazor M, Sepulveda W, Avila C, Copeland D, Williams J: Tumor necrosis factor in preterm and term labor. Amer J Obstet Gynecol 166:1576-1587,1992 36 73. Roop RM, Jeffers G, Bagchi T, Walker J, Enright FM, Schurig GG: Experimental infection of goat fetuses in utero with a stable, rough mutant of Brucella abortus. Res Vet Sci 51:123-127,1991 74. Sangha RK, Walton JC, Ensor CM, Tat H-H, CahlUs JRG: Immunohistochemical localization, messenger ribonucleic acid abundance, and activity of 15hydroxyprostaglandin dehydrogenase in placenta and fetal membranes during term and preterm labor. J Clin Endocrinol Metabol 78:982-989,1994 75. Schurig GG, Roop RM, Bagchi T, Boyle S, Burhman D, Sriranganathan N: Biologic properties of RB51; a stable rough strain of Brucella abortus. Vet Microbiol 28:171188,1991 76. Shaw DW, Britt JH: Concentrations of tumor necrosis factor a and progesterone within the bovine corpus luteum sampled by continuous-flow microdialysis during luteolysis in vivo. Biol Repro 53:847-854,1995 77. Smith T: A characteristic localization of bacillus abortus in the bovine fetal membranes. J Exp Med 29:451-456,1919 37 78. Snow EC. Handbook of B and T lymphocytes. Bom W, Kelly BCA, O'Brien RL. y/5 TCells. San Diego, CA: Academic Press Inc.; 1994:179-214. 79. Sordillo LM, Pighetti GM, Davis MR: Enhanced production of bovine tumor necrosis factor-a during the periparturient period. Vet Immunol Immunopathol 49:263-270,1995 80. Starkey PM, Sargent IL, Redman CWG: Cell populations in human early pregnancy decidua: characterization and isolation of large granular lymphocytes by flow cytometry. Immunology 65:129-134,1988 81. Stevens MG, Hennager SG, Olsen SC, Cheville NF: Serologic responses in diagnostic tests for brucellosis in cattle vaccinated with Brucella abortus strain 19 or RB51. J Clin Micro 32:1065-1066,1994 82. Stevens MG, Olsen SC, Pugh GW, Palmer MV: Immune and pathologic responses in mice infected with Brucella abortus strain 19, RB51 or 2308. Infect Immun 62:32063212,1994 83. Tatum FM, Morfitt DC, Hailing SM: Construction of a Brucella abortus RecA mutant and its survival in mice. Microb Pathog 14:177-185,1993 38 84. Tatum FM, Cheville NF, Morfitt DC: Cloning, characterization and construction of htrA and htrA-X^k& mutants of Brucella abortus and their survival in BALB/c mice. Microb Pathog 17:23-36,1994 85. Taylor AW, McDiarmid A: The stability of the avirulent characteristics of Brucella abortus strain 19 and strain 45/20 in lactating and pregnant cows. Vet Rec 61:317-318,1949 86. Thomas EL, Bracewell CD, Corbel MJ: Characterization of Brucella abortus strain 19 cultures isolated from vaccinated cattle. Vet Rec 108:90-93,1981 87. Tobias L, Schurig GG, Cordes DO: Comparative behavior of Brucella abortus strains 19 and RB51 in the pregnant mouse. Res Vet Sci 53:179-183,1992 88. Verger J-M, Grimont F, Grimont PAD, Grayon M: Brucella, a monospecific genus as shown by deoxyribonucleic acid hybridization. Int J Syst Bacteriol 35:292-295,1985 89. Welsh AO, Enders AC: Chorioallantoic placenta formation in the rat: IL Angiogenesis and maternal blood circulation in the mesometrial region of the implantation chamber prior to placenta formation. Amer J Anat 192:347-365,1991 90. Wright AE, Semple D: On the employment of dead bacteria in the serum diagnosis of typhoid and Malta fever. Br Med J 1:1214,1897 39 91. Wyn-Jones G, Baker JR, Johnson PM: A clmical and immunopathological study of Brucella abortus strain 19-induced arthritis in cattle. Vet Rec 107:5-9,1980 92. Yelavarthi K, Chen H-L, Yang Y, Cowley B, Fishback J, Hunt J: Tumor necrosis factora mRNA and protein in rat uterine and placental cells. J Immunol 146:3840-3848,1991 93. Zahl PA, Bjerknes C: Effect of the endotoxin of Shigella paradysenteriae on pregnancy in rabbits. Proc Soc Exp Biol Med 56:153-155,1944 94. Zammit T: A preliminary note on the examination of the blood of goats suffering from Mediterranean fever. Reports of the Commission on Mediterranean Fever 3:83,1905 40 CHAPTER 2. EXPERIMENTAL INFECTION OF PREGNANT CATTLE WITH THE VACCINE CANDIDATE BRUCELLA ABORTUS STRAIN RB51: PATHOLOGIC, BACTERIOLOGIC AND SEROLOGIC FINDINGS A paper published in Veterinary Pathololgy 1996;33:682-691 Mitchell V. Palmer, Norman F. Cheville, and Allen E. Jensen AbstractTo determine the placental tropism and abortigenicity of the vaccine candidate, B. abortus strain RB51 (SRB51), a rough mutant of the virulent strain 2308, 10 Polled Hereford heifers were inoculated intravenously in the sixth month of gestation. Heifers were euthanatized and examined at 8 weeks p.i.(n=5) or at full term (n=5). Four of five infected heifers sampled at 8 weeks p.i. and 3/4 infected heifers at term had placentitis while reproductive tissues of 3 normal cows used for comparison had no placentitis. Numerous macrophages, immunoreactive for SRB51 antigen, as well as neutrophils, fibrin and cell debris filled the arcade zone between chorion and maternal septae. Trophoblastic epithelium of the placentomal arcade zone had intracellular bacteria that were immunoreactive for SRB51 antigen. The tips of matemal septae had a lymphoplasmacytic infiltrate with small multifocal erosions and ulcerations of matemal epithelium. SRB51 was cultured fi-om all tissues in which lesions were seen. Placentae of one cow from each group had no placentitis and contained no SRB51. In mammae, interstitial lymphoplasmacytic infiltrates and suppurative infiltrates within alveoli and intralobular ducmles were seen in 2/5 heifers at 8 weeks p.i. SRB51 was cultured firom liver, spleen, lung and bronchial lymph nodes in 4/5 calves at 8 wks. p.i. and 3/4 full term calves, but no lesions were seen. One term heifer had disseminated infection, placentitis, lymphoplasmacytic endometritis, and delivered a 41 premature weak calf. These results establish that SRB51 is less abortifacient than previously published reports with strain 19, in that only 1/4 heifers delivered prematurely following intravenous inoculation with SRB51 while intravenous inoculation with strain 19 leads to 100% abortion. However, it also shows that SRB51 can infect the bovine placenta, mammary gland and fetus; can induce placentitis and in some cases, lead to preterm expulsion of the fetus. IntroductionThe eradication of bovine brucellosis has centered on the use of Brucella abortus strain 19 (SI9), an attenuated, live vaccine introduced into the federal brucellosis eradication program in 1940. Although effective and important in the control of bovine brucellosis, S19 has significant limitations including virulence for humans, ability to induce abortion in 6,11_30 pregnant cattle, and the development of agglutinating antibodies indistinguishable from 29 those seen in naturally infected animals. Experimental intravenous infection of pregnant cattle using S19 typically leads to placentitis and abortion; e.g. in two separate studies abortions occured in 100% of the cows.^^"''^ Abortion occured as early as 16 days and as late as 42 days postinoculation."*^ Additionally, in one of the preceeding studies, B. abortus strain 45/20, given intravenously caused abortion in 100% of the cows between days 9 and 47 after inoculation. A separate study utilizing B. abortus strain 544 resulted in abortion in 100% of the cows in 3-4 weeks when given intravenously.38 B. abortus S19 has also been shown to induce abortion in a small percentage of pregnant cattle vaccinated subcutaneously with the normal vaccine dose.^-'^-^° B. abortus strain RB51 (SRB51) is a stable rough mutant of B. abortus strain 2308 37 (S2308), which lacks much of the LPS 0-side chain, and has been investigated as an g alternative vaccine to S19. A major concern m evaluating SRB51 is the possibility of 42 tropism for the bovine placenta and subsequent abortion. In mice, SRB51 has been shown to 46 produce a mild to minimal placentitis which is not associated with fetal death. Although SRB51 did localize in the giant trophoblast cells of the mouse placenta similar to S19 and the virulent S2308, placental damage and colonization by SRB51 were markedly less than 46 with either S19 or S2308. Further testing has shown that SRB51 did not induce abortion in pregnant goats when fetuses were inoculated intramuscularly with SRB51.^^ SRB51 was found to colonize bovine chorioallantoic membrane explants in a manner similar to S19 and S2308; however, the degree of cytotoxicity associated with SRB51 infection was much less than that meastired for S2308 infection and not significantly different firom that seen with 36 S19. The abortifacient potential of SRB51 in pregnant cattle has not been determined. It is essential that SRB51 or any other vaccine candidate be tested in cattle to determine the ability to not only cause abortion but any other pathologic changes which may be considered detrimental. The purpose of this study is to determine the potential for abortion, placental infection, placentitis and fetal infection in pregnant cattle infected with SRB51. Materials and MethodsTen virgin Polled Hereford heifers used in the experiment were obtained from nonvaccinated, brucellosis-free herds. Six heifers were 27-months old while the remaining 4 were 16 months old at breeding. All cattle were serologically negative for brucellosis by standard tube agglutination."* Upon arrival, cattle were kept on pasture and bred by natural service using a Polled Hereford bull which was also serologically negative for brucellosis by standard tube agglutination. During challenge, animals were housed individually as described^ in concrete isolation rooms with positive air flow from a clean corridor and negative air flow to a contaminated corridor. 43 Normal tissue for comparison was obtained from a meat packing plant. Three reproductive tracts containing fetuses estimated by fetal crown-rump length to be approximately 240 days gestation were collected, processed for light microscopy and examined. At 6 months of gestation, heifers were injected intravenously in the right jugular vein 10 with 10 CFU B. abortus stram RB51 suspended in 10 ml sterile saline (0.85% NaCl). At postinfection week 8 (PIW), 5 heifers (3 from older group and 2 from younger group) were randomly selected, euthanatized and examined. The remaining cattle were examined at term (n=3) or following premature delivery (n=l). Samples of maternal uterus, placentomes, intercotyledonary placenta, mammary gland, supramammary lymph node, liver, spleen, and fetal liver, spleen, lung and bronchial lymph node were obtained for bacteriologic culture, histopathology and immunohistochemical staining for SRB51 antigen. Matemal blood, milk, vaginal swab, allantoic and arrmiotic fluids and fetal blood, cerebrospinal fluid, abomasal contents and rectal swab, were also collected for bacterial culture. All cattle were euthanatized by intravenous administration of sodium pentobarbital. Following a premature delivery, heifer (#9) was treated with oxytetracycline (Liquamycin LA-200, Pfizer, New York, N.Y.), 1 Img/kg, SID for 5 days prior to necropsy. Treatment of cows aborting or delivering premature calves has been standard protocol in our laboratory as a matter of personnel safety due to the zoonotic potential of B. abortus and the unknown virulence of SRB51 for humans. Histologic and Immunohistochemical analyses- Tissue specimens were placed in 10% neutral buffered formalin. Tissues were processed by routine paraffin embedment techniques, cut 4-6 [am thick and stained with hematoxylin and eosin. Unstained sections were used for immunohistochemical staining of SRB51 antigen using an avidin-biotinalkaline phosphatase complex staining techmque as described 25.32 and a commercially 44 available kit (Histomark, BCirkegaard and Perry Laboratories, Inc., Gaithersburg, MD). Primary antibodies were prepared by hyperinununization of rabbits with whole cell, irradiated SRB51. Anti-SRB51 antisera was used at a dilution of 1:1000. Optimal dilutions for the primary antibody had previously been determined using formalin-fixed, paraffinembedded sections of SRB51-infected murine spleens or bovine lymph nodes. Control slides were stained with serum from clinically normal rabbits. Sections of SRB51-infected BALB/c mouse spleen and liver were used as positive control slides. Staining of bovine macrophages using the monoclonal antibody EBM 11 (anti-CD68) (Dako, Carpenteria, CA) was done as described.' Samples of placentome and intercotyledonary placenta were also collected, embedded in O.C.T. compound (Tissue Tek, Miles Scientific, Naperville, IL) and immediately frozen by immersion in liquid nitrogen. Frozen sections of intercotyledonary placenta and placentome were processed and stained with monoclonal antibodies to CD4, CDS and y/5 lymphocyte surface markers as described."' Bacteriologic analysis- B. abortus strain RB51 was used in this study.SRB51 was prepared from master seed stock of B. abortus RB51 (ARS-1) at the Brucellosis Research Unit, National Animal Disease Center, Ames, lA. SRB5I was propagated from working seed stock on tryptose agar and incubated at 37°C for 48 hours. The surface of the agar was flooded with saline and the suspended bacteria were collected by aspiration, twice washed by centrifiigation and resuspended in saline. Bacterial suspensions were subsequently diluted to the appropriate concentration prior to inoculation. 8 Tissue samples to be tested for SRB51 were processed as described using tryptose agar with 5% bovine serum, cyclohexamide (30|ig/ml), bacitracin (7.5 U/ml), polymyxin B sulfate (1.8 U/ml) and ethylviolet (1:300,000). Brucella colonies were identified as described previously.^ 45 Collection and handling of bloodfor culture- At time periods corresponding to postinoculation day (PID) 1-6 and postinoculation week (PIW) 1-12, 100 ml of blood was collected aseptically for culture from jugular veins on alternating sides beginning with the side contralateral to the initial injection. The samples consisted of one 125 ml serum vial containing 50 ml of blood and 50 ml of blood culture media,^ and 5 - 20 ml tubes each containing 10 ml of blood and 10 ml of blood culture media. Upon receipt, 5 ml of blood was drawn aseptically from the 125 ml serum bottle and 1 ml was inoculated on to each of 5 plates containing tryptose agar with 5% bovine serum. These plates were incubated at 11^C and checked at 24 hrs., 3, 7,10 and 14 days for bacterial growth. Any suspect growth was subcultured and processed for SRB51 identification. The remaining sample in the 125 ml serum vial was incubated at 37°C and subcultured as above on days 3,10, 20, 30 and 40. The 5 serum tubes were immediately placed at -20°C and kept until placed in biphasic flasks.^ Frozen blood was thawed at 37°C and transferred to biphasic culture flasks 2 containing tryptose agar and processed as described. These flasks were checked for bacterial growth at 24 hrs., 3, 7, 10,20, 30 and 40 days. Suspect growth was subcultured and processed for SRB51 identification. Calf blood was examined similarly at necropsy. Serological analysis- Twenty milliliters of blood was collected by venipuncture from each heifer prior to inoculation and PIW 1, 2,4, 6, 8, 10 and 12. Blood was collected into sterile 10-ml tubes, allowed to clot for 6-24 hrs. at 4°C, and centrifiiged. Serum was divided into 2 ml aliquots, frozen, and stored at -70°C. Blood was similarly collected from the calves at necropsy. Samples were later thawed and tested. The standard tube agglutination test was performed on all samples. Antibodies to SRB51 were measured by use of an antibody dot blot assay ^ using anti-bovine IgG, heavy chain specific and anti-bovine IgM 46 antibodies (Kirkegaard & Perry Laboratories, Gaithersburg, Md.) at dilutions of 1:1,000 and 1:100 respectively. Results Clinical Signs - Eight of ten cattle were febrile during the first 24-48 hours after inoculation. Elevated temperatures ranged from 39.1°C to 41.1°C (normal range 38.5±0.5 °C). Six of eight heifers returned to normal body temperature by 24 -48 hours after inoculation; however, two heifers (#4 and #9) had elevated body temperatures which persisted for 72 hrs., being normal by 96 hours after inoculation. Febrile responses were not accompanied by other notable clinical signs. All cattle remained on feed and no obvious signs of depression were noted. One heifer (#9) delivered a premature, small, weak calf, one month prior to the expected calving date (PID 59). The calf was unable to stand or nurse and had a weak suckling reflex. The dam was afebrile at the time of delivery. Bacterial Culture of Blood- SRB51 was detected in blood in one or more sampling times up to 6 days after inoculation in 8/10 cattle, however, no organisms were cultured beyond that time (Table 1). Two of ten heifers showed no SRB51 in the blood at any sampling time. Calf blood culture revealed 2/10 calves contained SRB51 in the blood (Table 2). Bacterial Culture of Tissues and Fluids- Four of five heifers (80%) sampled at PIW 8 had SRB51 in placentomes, intercotyledonary placenta, uterus and spleen (Table 3). Mammary tissue and supramammary lymph nodes from 2/5 (40%) yielded SRB51. One 47 Table 1. Isolation of B. abortus strain RB51 from blood' of pregnant cattle inoculated i.v. at six months of gestation. PID ID 2 3 I + + - ND" ND 2 - - ND ND + - ND ND - +•' + ND - ND 3 + 4 5 6 1 - 4 + - 5 - - 6 + - - ND ND - 7 + + + ND ND - 8 . 9 - _ ND ND 10 ' Unless otherwise stated, B. abortus RB51 isolation was successful only after culture enhancememt (see text for details). '' ND = cow not tested on specified day. ° B. abortus RB51 isolated from direct plating, 4.16 x 10" CFU/ml blood. •* B. abortus RB51 isolated from direct plating, one colony. 48 heifer was free of SRB51 by PIW 8. The calf from this heifer was also free of SRB51 in all tissues and fluids cultured. Bacteriologic culture of 4/5 (80%) calves from heifers euthanatized at PIW 8 showed SRB5I in liver, spleen, lung and bronchial lymph node (Table 2). Three of four heifers (75%) which went to full term had SRB51 in samples of uterus, placenta and placentome (Table 3). One heifer which went to frill term was negative for SRB51 in all tissues and fluids cultured. The calf from this heifer was similarly free of SRB51. All other full term calves had SRB51 in spleen and bronchial lymph nodes. Two of four calves (50%) also had SRB51 in limg (Table 2). After the premature delivery of a weak calf, SRJB51 was found in milk samples from all four mammary quarters of heifer #9. A vaginal swab and samples of expelled placenta also contained SRB51. After oxytetracycline therapy for 5 days, SRB51 was still recovered from intercaruncular uterus and caruncle. No SRB5I was found in milk or mammary tissue following antibiotic therapy (Table 3). The premature weak calf was euthanatized within 8 hours of delivery and examined. All samples cultured from the calf contained SRB51 except cerebrospinal fluid and a rectal swab specimen (Table 2). Serology- Ail heifers and calves had no antibody titers as determined by standard tube agglutination at all time points sampled. Serologic results measured by dot blot assay are reported in Table 4. No calf from either group had an antibody titer, as measured by the dot blot assay, for anti-SRB51 IgG antibody. The premature calf (#9) had an IgM titer to SRB51 of 1:160 at necropsy. Other calves had no anti-SRB51 IgM antibody titer. Anti-lgM titers were not evaluated on the heifers. 49 Table 2. Summary of culture results to isolate B. abortus RB51 from calves of pregnant heifers inoculated i.v. with RB51 at six months of gestation. PIW 8 Full term Animal Number Tissue 1 2 J 4 5 Total 6 7 Liver - + + + ND' 3/4 + Spleen + + + + - 4/5 Lung + + + + - Bronchial LN*" + + + + Blood - - - CSF' - - Abomasal - - 8 9 10 Total - + + 3/5 + - + + 3/5 4/5 + - + - 2/5 - 4/5 + + + + 4/5 - - 0/5 + - + - 2/5 - - - 0/5 - - - ND - 0/4 + - + - 2/5 - - - NA" + 1/4 - + - - 1/5 + - 4- + 3/5 Contents Rectal Swab '' Tracheobronchial lymph node. Cerebrospinal fluid. Not available for culture. 50 Table 3. Summary of culture results to isolate B. abortus RB51 from pregnant heifers inoculated i.v. at six months of gestation and examined at PIW 8 or at fiill term. Animal Number PIW 8 Tissue 1 2 3 4 Full Term 5 Total 6 7 8 9' 9 Pre Post 10 Total Liver + + - + -3/5 + +- na** ND" + 3/4 Spleen + + + +.4/5 + +. na - + 3/5 Uterus + + + +-4/5 + +- NA + + 4/5 Placenta + + + +-4/5 + +- + na + 4/5 Placentome + + + + -4/5 + +- + na + 4/5 Mamm-'-RF - - + - - 1/5 - - - NA - - 0/5 Mamm-RR - - + - - 1/5 - - - NA - + 1/5 Mamm-LF + - -- .1/5..- nA - - 0/5 Mamm-LR + - - - - LN'-R LN-L Milk-RF + - + - - 1/5 - - - NA - - 0/5 0/5 - - - NA - + 1/5 2/5 + - - NA - + 2/5 0/5 - - - + - - 1/5 51 Table 3 (continued). Summary of culture results to isolate B. abortus RB51 from pregnant heifers inoculated i.v. at six months of gestation and examined at PIW 8 or at full term. Animal Number PIW 8 Tissue Full Term Total 6 10 Pre Total Post Milk-RR 0/5 Milk-LF 0/5 1/5 Milk-LR 0/5 1/5 Blood 0/5 0/5 Allantoic*^ ND Amniotic® VS" + + 2/4 0/5 ND 1/5 4- + NA NA NA + 2/3 NA NA NA + 1/3 0/4 ' Pre and post samples refer to oxytetracycline treatment given after premature delivery of a weak calf. '• NA = not available for culture. " ND = not done. Mamm = mammary gland tissue. ' LN = supramammary lymph node. ^Allantoic fluid. ® Amniotic fluid. '* Vaginal swab 1/5 52 Gross Lesions - Mild placentomal edema was present in 4 of 5 cattie at PIW 8. The chorioallantois covering caruncles was mildly thickened, but translucent. Placentomes of 4/5 cattle at PIW 8 contained a small amount of tan, creamy exudate found between the chorioallantois and maternal tissues. This exudate was most prominent over caruncles although it was found less frequently in intercotyledonary areas. Gross lesions were not present in other organs. Amniotic and allantoic fluids appeared normal. Placentae from 3/4 cows exammed at full term were more edematous than those examined at PIW 8 with thick gelatinous edema of the intercotyledonary placenta which was slightly opaque but translucent. Extensive areas of creamy, tan exudate were between the chorioallantois and uterine epithelium within placentomes and intercotyledonary areas. Allantoic and amniotic fluids appeared normal. The cow delivering a premature calf had cotyledons covered with a brown creamy exudate while the intercotyledonary areas were normal. Postparturient lochia contained small amounts of brown, creamy exudate. Microscopic LesionsPIW 8 - Placentitis was present in 4/5 cattle examined at PIW 8 (Figures 1,2). Infiltrates of numerous macrophages, neutrophils, necrotic cell debris and fibrinchemorrhagic exudate was seen in the placentomal arcade zone. Inflammatory infiltrate separated the tips of matemal septa from the trophoblast layer. Epithelial cells of the trophoblast contained abundant intracytoplasmic bacteria which were immunoreactive with anti-RB51 antibody (Figure 3). Many trophoblastic epithelial cells with intracellular bacteria were detached from the chorion and present in the inflammatory infiltrate. In some areas of intense inflammation, the trophoblast layer of the chorioallantois was flattened with cuboidal to squamous rather than columnar epithelial cells. Immunostaining for SRB51 showed intense staining within trophoblastic epithelial cells as well as within macrophages as a Figure 1. Section of placentomal arcade zone from a heifer examined at PIW 8. Note extensive inflammatory infiltrate separating chorioallantois (short arrow) from maternal septa (long arrow). HE. Bar= 100 |im. Figure 2. Section of placentome from a heifer examined PIW 8. Note chorionic villus surrounded by inflammatory cells and necrotic cell debris. Trophoblastic epithelial cells are enlarged and granular in appearance due to numerous intracellular bacteria (arrows). HE. Bar = 50 fim. 55 component of the inflammatory infiltrate. SRB51 immunoreactive macrophages were most prominent adjacent to the tips of the maternal septa and were rarely seen adjacent to the trophoblast layer (Figure 4). Neutrophils were seen in the tunicae intima and muscularis of arterioles of the chorioallantois with fewer located in the stroma of the chorioallantois. Endothelial cells of chorioallantoic arterioles were rounded with some containing clear, smoothly contoured cytoplasmic vacuoles. Bacteria immunoreactive for SRB51 were not seen within maternal septal epithelial cells of the placentome. Tips of maternal septa contained areas of lymphoplasmacytic infiltration with mild to moderate fibrosis. Small ulcerations of maternal septa were seen with loss of maternal epithelial cells, infiltrates of moderate numbers of neutrophils and small numbers of bacteria which were immunoreactive for SRB51 antigen. Intercotyledonary placenta had similar placentitis to that seen in placentomes with inflammatory infiltrate which separated maternal and fetal tissues. Foci of endometritis subjacent to areas of placentitis contained infiltrates of numerous lymphocytes, plasma cells, macrophages and neutrophils. These infiltrates were most intense in the strata compactum and spongiosimi but also extended deep to surround uterine submucosal glands. Immunostaining of T-lymphocytes in the endometrial inflammation revealed most to be of the CD4 type, with few present of the CDS or y/S (CD4- CD8-) subtype (Figures 5a, 5b). Infiltrates of neutrophils and fibrin were in submucosal glands of the uterus (Figure 6). Low numbers of submucosal gland epithelial cells, neutrophils and macrophages vvath cytoplasmic staining for SRB51 antigen were seen within and surrounding submucosal glands. Mammary glands had multifocal interstitial lymphoplasmacytic infiltrates and intraluminal infiltrates of large numbers of neutrophils within some alveoli and intralobular ductules. Supramammary lymph nodes had infiltrates of large numbers of macrophages and neutrophils in medullary sinuses which extended superficially to the deep cortex. Cortical regions contained prominent follicles with germinal centers. 56 Figure 3. Placentome from full tenn heifer, chorionic villus. Note extensive intracellular staining for SRB51 antigen. Avidin-biotin-alkaline phosphatase method. Gill's hematoxylin counterstain. Bar = 35 |im. Figure 4. Placentome from fiill term heifer, matemal septa (arrow). Note extensive inflammatory cells occupying space between chorioallantoic membrane and matemal septa which stain positive for SRB51 antigen. Avidin-biotin-alkaline phosphatase method. Gill's hematoxylin counterstain. Bar = 50 jim. Figure 5. Section of uterus from heifer at PIW 8. Periglandular lymphocytic infiltrate is composed primarily of CD4+ T-cells (a), with rare y/S T-cells (arrow)(b). Avidin-biotinalkaline phosphatase method. Gill's hematoxylin counterstain. Bar = 35 [im. Figure 6. Section of uterus from full term heifer. Note periglandular inflammation and intraluminal neutrophils. HE. Bar = 50 |im. \ 59 Maternal spleens had large numbers of macrophages with abimdant cytoplasm which separated white pulp areas and obliterated red pulp. These cells stained strongly for the macrophage marker CD68. Areas of lymphoid follicle formation in white pulp were normal. Calves examined at PIW 8 contained no significant microscopic lesions. In spleens, red pulp was well developed, periarteriolar lymphoid sheaths were small, and germinal centers were absent. Full term - Placentae from 4/5 full term cows had placentitis resembling that seen at PIW 8. However, placentitis was more widespread and extended deeper into the placentome involving not only the arcade zone but primary chorionic vilU and maternal epithelium and stroma. Bacterial colonies, immunoreactive for SRB5I antigen, were seen among the inflammatory infiltrate separating trophoblast from maternal septa. Immunostaining for SRB51 antigen showed results similar to those seen at PIW 8 with abundant staining of trophoblast epithelial cell cytoplasm. Similar microscopic lesions were observed in other organs as those seen in cows examined at PIW 8. As with the calves examined at PIW 8, full term calves showed no splenic germinal center formation and small periarteriolar lymphoid sheaths. No other gross or microscopic lesions were seen in full term calves. Extensive necrosis was seen in placentomes from heifer # 9 which delivered a premature weak calf. Necrosis of chorioallantois and trophoblastic layers extended deep to the base of the placentome. Aggregates of basophilic coccobacilli between trophoblastic and matpmfll septal epithelium were immunoreactive for SRB51 antigen. Endometritis characterized by infiltrates of large numbers of lymphocytes, plasma cells and neutrophils was present in the superficial layers of the uterus as well as surrounding submucosal uterine glands. Multifocal small erosions and ulcerations of uterine mucosa with numerous neutrophils were also seen. 60 Mammary tissue from heifer #9 had alveoli containing moderate numbers of intraluminal neutrophils. Low numbers of lymphocytes were seen within the interstitium surrounding such alveoli. Supramammary lymph node changes in heifer #9 were consistent with those observed in cattle examined at PIW 8. Lung from the calf from heifer #9 had very mild multifocal interstitial thickening due to infiltrates of moderate numbers of macrophages and neutrophils. These areas were often found but not limited to peribronchiolar regions. Increased numbers of cells immimoreactive for CD68 were seen within the interstitium. Bronchi and bronchioles were normal with no intralimiinal or intraepithelial inflammation. Placentomes from noninfected, normal cows did not contain areas of necrosis or inflammation. In some sections, small areas of hemorrhage were present within the arcade zone. Foci of hemorrhage were not associated with necrosis of ttophoblastic epithelial cells or with infiltrates of inflammatory cells. DiscussionThis study shows that SRB51 has a tropism for the bovine placental trophoblast and induces placentitis which can result in premature birth. Abortion is defined as the premature expulsion from the uterus of the products of conception, of the embryo, or of a nonviable fetus.Although the calf from heifer #9 was premature and weak, the judgement of potential viability is subjective. To avoid confusion we have classified this delivery as a premature, weak calf and not an abortion. The infravenous inoculation of heifers in the sixth month of gestation represents a stringent test of virulence for any brucellosis vaccine candidate. It is unlikely that SRB51 would be given infravenously in field situations. However, despite this stringent test, lack of abortion shows that SRB51 is less virulent than other brucellosis vaccine strains, including SI9, when administered intravenously."^"^^'^ This decreased virulence in pregnant cattie contributes to the attractiveness of SRB51 as a vaccine for the 61 prevention of bovine brucellosis. Moreover, lack of antibodies which are detected by brucellosis surveillence tests, e.g. standard tube test, allows differentiation of SRB51vaccinated animals from naturally infected animals.'^^ Such differentiation is not possible following S19 vaccination. SRB51 - induced placental and uterine lesions in our cattle are similar to those 27 reported in cattle and other species infected with virulent B. abortus, and are consistent with the role of the trophoblast epithelium as the primary site of infection.^'^^ Placental inflammation and tropohoblast epithelial cell necrosis could allow passage of bacteria into the allantoic sac. Such a mechanism could explain the presence of SRJB51 in allantoic fluid of 40% of our heifers while the amniotic fluid was infected in only 10%. The amniotic membrane is not in intimate association with maternal tissues making extension of infection from the trophoblast layer less likely. The disseminated infection of many of the fetuses as well as the chorioallantoic vascuUtis suggest that hematogenous spread of infection also plays a role in fetal infection. T-lymphocytes of the CD4 type are known to be important in infections with intracellular bacteria." The activation of CD4+ T-cells and resultant secretion of IFN-y have been shown to be important in the protection of mice infected with other intracellular pathogens such as Listeria monocytogenes and Mycobacteria spp.'^ The presence of numerous CD4+ T-cells in the endometritis seen in the present study is consistent with a role for CD4+ T-cells in the response to SRB51. Although IFN-y levels were not measured in our cattle, IFN-y has been shown to be important in the control of intracellular growth of SRB51 in murine macrophages 18 as well as the infection of murine spleens with S2308.43 y/S T-cells do not appear to play a major role in SRB51-induced endometritis. The rare identification of y/6-T-cells in our study differs from the proposed mechanism of y/5-Tcells in protection against intracellular pathogens in other species.^ In mice, y/5 T-cells colonize various tissues, including lung, tongue, skin, intestine, uterus, vagina and placenta.^ 62 In M. tuberculosis-m£ec\.Q& mice deficient in y/5 T-cell receptors, pulmonary infection is disseminated and not confined in granulomata as it is in genetically normal control mice or mice deficient in only the a/p T-cell receptor. This suggests a role for y/S T-cell in the containment of M. tuberculosis infection. A similar protective role has been seen in mice ''S infected with L. monocytogenes.' Extrapolation of results regarding lymphocyte subsets in other species to cattle may not be appropriate. In contrast to mice, y/5 T-cells comprise a significantly higher proportion of peripheral blood mononuclear cells in ruminants.Although the precise role and significance of y/5 T-cells in ruminants is not clear, one study has shown that numbers of y/6 T-cells are not altered in draining lymph nodes following subcutaneous inoculation of SRB51 It may be possible that y/5 T-cells are not major effector cells in SRB51-induced inflammation; however, in the present study, the lack of periodic analysis of y/5 T-cell involvement over the course of infection does not rule out the early involvement and/or possible destruction of y/5 T-cells in SRB51-induced endometritis. The incidence of fetal lung lesions in 5n/ce//a-infected, aborted fetuses or premature weak calves is highly variable, and is characterized as bronchitis or bronchopneumonia with predominantly mononuclear infiltrates."°'''° Interstitial lesions similar to those seen here have been reported.''^ Persistence of SRB51 infection to the termination of our study at 14 weeks differs from previous studies with SRB51 in which SRB51 was cleared by 6-weeks after inoculation in subcutaneously inoculated heifers Q Prolonged persistence may be due to differences in inoculation route, dosage or pregnancy status. Route of inoculation has been shown to influence the outcome of infection."^ Subcutaneous inoculation with B. abortus strain 45 is inocuous while intravenous inoculation results in abortion.^"* Additionally, our inoculum 10 9 dose, 10 CPU, is slightly higher than the dose of 5 to 7 x 10 used in previous vaccination g studies. The pregnancy status of cattle at inoculation also plays an important role in the 63 development of infection as other Brucella abortus strains have been shown to have a tropism for the placental trophoblast.^"^' In contrast to the persistence in tissue, lack of persistence of SRB5I in blood samples in the current study is possibly due to increased phagocytosis and killing of rough strains as compared to smooth strains.Previous studies have isolated virulent smooth B. abortus from pregnant cows 52 - 62 days following intravenous inoculation.'^ SRB51 is more sensitive to the bactericidal effects of complement when compared to smooth strains.'" 10,13 These differences have been attributed to the lack of 0-side chain on SRB51. Weak or negative serologic responses to SRB51 by perinatal calves are consistent with previous findings in SRB51 vaccinated cattle.^ Morphologic evidence for this lack of response was evident in the present study as splenic germinal center formation was absent in all calves examined. Although anatomically the components of the fetal immune system may be present by day 175 of gestation/^ fetal changes responsible for immunologic maturity may not occur at the same time for all antigens.^' In a previous study, bovine fetuses experimentally infected with the smooth S2308 had increased levels of both IgM and IgG.'"* Being a rough derivative of S2308, it is possible that the rough nature of SRB51 may contribute to decreased immunogenicity in fetuses. References 1. Ackermaim MR, DeBey BM, Stabel TJ, Gold JH, Register KB, Meehan JT: Distribution of anti-CD68 (EBMl 1) immunoreactivity in formalin fixed, paraffin-embedded bovine tissues. Vet Pathol 1994;31:340-348. 2. Alton GO, Jones LM, Angus RD, Verger JM: Techniques for the brucellosis laboratory, pp. 18-30. Institut National de la Recherche Agronomique, Paris, 1988. 64 3. Alton GG, Jones LM, Angns RD, Verger JM: Techniques for the brucellosis laboratory, pp. 34-60. Institut National de la Recherche Agronomique, Paris, 1988. 4. Alton GG, Jones LM, Angtis RD, Verger JM: Techniques for the brucellosis laboratory, pp. 123-126. Institut National de la Recherche Agronomique, Paris, 1988. 5. Anderson TD, Meador VP, Cheville NF: Pathogenesis of placentitis in the goat inoculated with Brucella abortus. I. Gross and histologic lesions. Vet Pathol 1986; 23:219-226. 6. Beckett FW, MacDiarmid SC: The effect of reduced dose Brucella abortus strain 19 vaccination in accredited dairy herds. Br Vet J 1985;141:507-514. 7. Bom W, Kelly KA, O'Brien RL. y/5 T cells. In\ Handbook of B and T lymphocytes. Ed. Snow EC. 1994. Academic Press Inc., San Diego, CA. pp. 179-214. 8. Cheville, NF, Jensen AE, Hailing SM, Tatum FM, Morfitt DC, Heimager SG, Frerichs WM, Schurig GG: Bacterial survival, lymph node changes and immunologic responses of cattle vaccinated with standard and mutant strains of Brucella abortus. Am J Vet Res 1992;S3:1881-1888. 9. Cheville NF, Stevens MG, Jensen AE, Tatum FM, Hailing SM: Immune responses and protection against infection and abortion in cattie experimentally vaccinated with mutant strains of Brucella abortus. Am J Vet Res 1993;54:1591-1597. 10. Corbeil LB, Blau K, Inzana TJ, Nielsen KH, Jacobson RH, Corbeil RR, Winter AJ: Killing of Brucella abortus by bovine senmi. Infect Immun 1988;56:3251-3261. 65 11. Comer LA, Alton GG: Persistence of Brucella abortus strain 19 infection in adult cattle vaccinated with reduced doses. Res Vet Sci 1981; 31:342-344. 12. Cotton, WE. New research and developments on contagious abortion (Bang's disease). Proc. 12th Int Vet Cong. 1934, p. 283. 13. Detilleux PG, Deyoe BL, Cheville NF: Penetration and intracellular growth of Brucella abortus in nonphagocytic cells in vitro. Infect Immun 1990;58:2320-2328. 14. Enright FM, Walker JV, Jeffers G, Deyoe BL: Cellular and humoral responses of Brucella abortus-m^QcX&d bovine fetuses. Am J Vet Res 1984;45:424-430. 15. Harmon BG, Adams LG, Frey M: Survival of rough and smooth strains of Brucella abortus in bovine mammary gland macrophages. Am J Vet Res 1988;49:1092-1097. 16. Hein WR, Mackay CR. Prominence of y/5 T cells in the ruminant immune system. Immunol Today 1991,12:30-34. 17. Hubbert WT: Recommendations for standardizing bovine reproductive terms. Cornell Veterinarian 1972;62:216-237. 18. Jiang X, Baldwin CL. Effects of cytokines on intracellular growth of Brucella abortus. Infect Immun 1993,61:124-134. 66 19. Kaufinann SHE, Reddehase MJ. T cell subsets and defense against bacteria and viruses. In: Handbook of B and T lymphocytes, ed. Snow EC, pp. 245-248. Academic Press, San Diego, CA., 1994. 20. Kennedy PC, Miller RB: The female genital system. In: Pathology of domestic animals, ed. Jubb KVF, Kennedy PC, Palmer N, 4th ed., vol. 3, pp. 396-399. Academic Press, San Diego, CA., 1993. 21. Kunkle RA, Steadham EM, Cheville NF: Morphometric analysis of CD4+, CD8+ and 7/8 T-lymphocytes in lymph nodes of cattle vaccinated with Brucella abortus strains RB51 and 19. Vet Immunol Immimopathol 1995;49:271-279. 22. Ladel CH, Blum C, Dreher A, Reifenberg K, Kaufinann SHE. Protective role of y/5 T cells and a/p T cells in tuberculosis. Eur J Immunol 1995, 25:2877-2881. 23. McClure SJ, Hein WR, Yamaguchi K, Dudler L, Beya M-F, Miyasaka M; Ontogeny, morphology and tissue distribution of a unique subset of CD4- CD8- sheep T-lymphocytes. Immunol Cell Biol 1989;67:215-221. 24. McEwen AD: Experiments on contagious abortion. Vet Rec 1937;49:1585-1596. 25. Meador VP, Tabatabai LB, Hagemoser WA, Deyoe BL: Identification of Brucella abortus in formalin-fixed paraffin-embedded tissues of cows, goats and mice with an avidinbiotin-peroxidase complex immunoenzymatic staining technique. Am J Vet Res 1986;47:2147-2150. 67 26. Mingle CK, Manthei CA, Jasmin AM: The stability of reduced virulence exhibited by Brucella abortus strain 19. J Am Vet Med Assoc 1941;99:203-204. 27. Mollelo JA, Jensen R, Collier JR, Flint JC: Placental pathology. HI. Placental lesions of sheep experimentally infected with Brucella abortus. Am J Vet Res 1963;24:915-921. 28. Mombaerts P, Amoldi J, Russ F, Tonegawa S, Kaufinann SHE: Different roles of a/p and y/5 T cells in immimity against an intracellular pathogen. Nature 1993,365:53-56. 29. Nicoletti P: Vaccination. In: Animal Brucellosis, ed. Nielsen K, Duncan JR, pp.283-300. CRC Press, Inc. Boca Raton, Fla., 1990. 30. Nicoletti PL: A preliminary report on the efficacy of adult cattle vaccination using strain 19 in selected dairy herds in Florida. Proc. 80th Annu. Meet. U.S. Animal Health Assoc., 1977:91-100. 31. Osbum BI: Ontogeny of immunity and its relationship to diagnosis of congenital infections. In: Current Therapy in Theriogenology. Ed. by Morrow D.A., pp.203-204. W.B. Saunders Co., Philadelphia, P.A.,1986. 32. Pahner MV, Cheville NF, Tatum FM: Morphometric and histopathologic analysis of lymphoid depletion in murine spleens following infection with Brucella abortus strains 2308, RB51 or an htrA deletion mutant. Vet Pathol 1996;33:282-289. 33. Payne JM: The pathogenesis of experimental brucellosis in the pregnant cow. J Pathol Bacteriol 1959;78:447-459. 68 34. Price RE, Templeton JW, Adams LG: Survival of smooth, rough and transposon mutant strains of Brucella abortus in bovine mammary macrophages. Vet Immunol Immunopathol 1990;26:353-365. 35. Roop RM, Jeffers G, Bagchi T, Walker J, Enright FM, Schurig GG: Experimental infection of goat fetuses in utero with a stable, rough mutant of Brucella abortus. Res Vet Science 1991;51:123-127. 36. Samartino LE, Enright FE: Interaction of bovine chorioallantoic membrane explants with three strains of Brucella abortus. Am J Vet Res 1992;53:359-363. 37. Schurig GG, Roop RM, Bagchi T, Boyle S, Buhrman D, Sriranganathan N: Biologic properties of RB51; a stable rough strain of Brucella abortus. Vet Microbiol 1991;28:171188. 38. Smith H, Keppie J, Pearce JH, Fuller R, Williams AE: The chemical basis of the virulence of Brucella abortus. I. Isolation of B. abortus form bovine foetal tissues. Brit J Exp Pathol 1961;42:631. 39. Smith T: A characteristic localization of bacillus abortus in the bovine fetal membranes. J Exp Med 1919;29:451-456. 40. Smith T: Pneumonia associated with Bacillus abortus (BANG) in fetuses and newborn calves. J Exp Med 1925; 41:639-647. 69 41. Spink WW, Hall EH JW, Finstad J, Malle E: Immunization with viable Brucella organisms. Result of a safety test in humans. Bull. W.H.0.26:409-419,1962. 42. Stevens MG, Hennager SG, Olsen SC, Che\'ille NF: Serologic responses in diagnostic tests for brucellosis in cattle vaccinated with Brucella abortus 19 or RB51. J Clin Microbiol 1994;32:1065-1066. 43. Stevens MG, Pugh GW, Tabatabai LB: Effects of gamma interferon and indomethacin in preventing Brucella abortus infections in mice. Infect Immun 1992,60:4407-4409. 44. Taylor AW, McDiarmid A: The stability of the avirulent characters of Brucella abortus. strain 19 and strain 45/20 in lactating and pregnant cows. Vet Rec 1949;61:317-318. 45. Tizard I: Immimity in the fetus and newborn. In: Veterinary Immunology, 3rd ed., pp. 171-184. W.B. Saunders Co. Philadelphia, PA., 1987. 46. Tobias L, Schurig GG, Cordes DO: Comparative behavior of Brucella abortus strains 19 and RB51 in the pregnant mouse. Res Vet Science 1992;53:179-183. 70 CHAPTER 3. SAFETY AND IMMUNOGENICITY OF THE VACCINE BRUCELLA ABORTUS STRAIN RB51 IN PREGNANT CATTLE A paper submitted to the American Journal of Veterinary Research (In press, accepted for publication 11/19/96) Mitchell V. Palmer DVM, Steven C. Olsen DVM, PhD, and Norman F. Cheville DVM, PhD Objective- Determine the safety and immunogenicity of Brucella abortus strain RB51 as a vaccine in pregnant cattle. Animals- Twelve Polled Hereford heifers obtained from a brucellosis-free herd and bred on site at 16 months of age with a brucellosis-free bull. Procedure - Pregnant heifers were vaccinated SC at 6 months' gestation with 10^ CFU Brucella abortus strain RB51 (n=5), 3x10^ CFU B abortus strain 19 (n=5), or sterile pyrogen free saline (n=2). Periodic blood samples were collected for serology and lymphocyte blastogenesis assays. At full gestation, heifers were euthanatized and maternal and calf tissues and fluid specimens were collected for bacteriologic culture, histologic analysis, and lymphocyte blastogenesis using various antigenic stimuli (y-irradiated strain RB51, y-irradiated strain 19, strain RB51 LPS or strain 2308 LPS). 71 Results- None of the strain RB51- or strain 19-vaccinates aborted or had gross or microscopic lesions at necropsy consistent with brucellosis. Two of five strain RB51vaccinates had strain RB51 in superficial cervical lymph nodes draining the site of vaccination. No other maternal or calf tissues contained strain RB51. Maternal PBMC from strain RB51-vaccinates and strain 19-vaccinates showed proliferative responses to both yirradiated strain RB51 and strain 19 which were greater than nonvaccinated controls. In contrast, maternal superficial cervical lymph node cells from strain 19-vaccinates, but not strain RB51-vaccinates, demonstrated proliferative responses to y-irradiated strain RB51 or strain 19 bacteria that were greater than that seen in nonvaccinated controls. PBMC from calves of strain RB51- and strain 19-vaccinates showed proliferative responses greater than nonvaccinated controls to irradiated strain RB51, strain RB51 LPS, and strain 2308 LPS. None of the heifers vaccinated with strain RB51 developed antibodies detected by the standard tube agglutination test, but all developed antibodies to strain RJB51 that reacted in a dot ELISA test using irradiated strain RB51 as antigen. Conclusions- Pregnant cattle can be safely vaccinated with 10 CPU strain RB51 without subsequent abortion or placentitis. Furthermore, strain RB51 is immunogenic in pregnant cattle and induces humoral and cell mediated responses. 72 Clinical relevance- Safety and immunogenicity of B abortus strain RB51 in pregnant cattle, with the absence of antibodies that interfere with serologic diagnosis of field infections, make it a desirable vaccine to be used in situations in which whole herd vaccination is necessary. Brucellosis is an economically important abortifacient disease of cattle caused by the bacterium Brucella abortus. Vaccination of heifers with the vaccine Brucella abortus strain 19 (SI 9) has been integral to the decreased prevalence of bovine brucellosis in the United States since the inception of the brucellosis eradication effort in 1940. Although effective as a vaccine, S19 has several limitations including the ability to induce abortion in pregnant cattle.' Another limitation which continues to interfere with brucellosis eradication efforts is the production of antibodies following S19 vaccination which are indistinguishable on brucellosis siurveillance tests from those produced by natural infection with field strains." Limitations of the current brucellosis serologic tests makes the distinction of naturally infected animals from S19-vaccinates problematic. Brucella abortus strain RB51 (SRB51), a rough mutant of the virulent Brucella abortus strain 2308,^ has recently been approved in the United States for use as a brucellosis vaccine for calves. SRB51 has been shown to be protective in cattle and does not produce antibodies which cross react on serologic surveillance tests,'*'^ enabling the distinction between vaccinated and naturally infected animals. The possibility of placental and fetal infection with SRJB51 was established by the experimental FV inoculation of pregnant cattle with very high numbers of SRB51, however, abortion was not observed.^ The question of 73 placental tropism when SRB51 is given in lower numbers and by the recommended route, i.e., SC inoculation, has not been established. This study was designed to determine the safety and immunogenicity of SRB51 in pregnant cattle when the vaccine is administered by the recommended route, using a dosage which has been used in previous field investigations.^ Both humoral and cell mediated immune responses were analyzed as a measure of immune responsiveness. Materials and Methods Polled Hereford heifers used in the experiment were obtained firom a nonvaccinated, brucellosis-free herd. Heifers were 16 months old at breeding. Prior to shipment and again prior to vaccination, all cattle were serologically negative for brucellosis by the standard tube g agglutination test (STAT). Upon arrival, and throughout the experiment, cattle were kept on pasture. Breeding was by natural service using a Polled Hereford bull which was serologically negative for brucellosis as determined by STAT. At six mondis' gestation, animals were randomly divided into 3 groups and injected with 9.4 X 10^ CFU SRB51 (n=5), 3x10^ CPU S19 (n=5), or 2 ml sterile pyrogen free saline, (n=2). All animals were injected SC in the right prescapular region. The dosage of SRB51 used previously in adult vaccination studies is 10^ CFU,' while the recommended adult dose of strain 19 is 3-10 x 10^ CFU. ^ 74 Body temperatures were recorded daily for the first four days after vaccination and animals were observed twice daily for the duration of the experiment for adverse reactions or clinical signs. Necropsy- All cattle were euthanatized at full gestation by IV administration of sodium pentobarbital/ Samples of maternal uterus, placentome, intercotyledonary placenta, mammary gland, supramammary lymph node, liver, spleen, and calf liver, spleen, lung and bronchial lymph node were obtained for bacteriologic culture and histologic analysis. Immunohistochemical staining for SRB51 antigen was done on tissues from SRB51-infected heifers and calves. Maternal blood, milk, vaginal swab specimens, allantoic and amniotic fluids and calf blood, cerebrospinal fluid, abomasal contents and rectal swab specimens, were also collected for bacterial culture Histologic and Immunocytochemical analyses- Tissue specimens were placed in neutral buffered 10% formalin. Tissues were processed by routine paraffin embedment techniques, cut 4 to 6 |im thick and stained with H&E. Unstained sections were used for immunohistochemical staining of SRB51 antigen using an avidin-biotin-alkaline phosphatase complex staining technique as described and a commercially available kit.'' Polyclonal anti-SRB51 antibodies were prepared by hyperimmunization of rabbits with whole cell, yirradiated SRB51. 'Sleepaway, Fort Dodge Co., Fort Dodge, lA. ''Histomark, Kirkegaard and Perry Laboratories, Inc., Gaithersburg, MD. 75 Anti-SRB51 antiserum was used at a dilution of 1:1000 based on previous work using sections of SRB51-infected murine spleens or bovine lymph nodes.'' Control slides were stained with serum from nonimmunized, clinically normal rabbits. Sections of SRB51-infected BALB/c mouse spleen and liver were used as positive control slides. Bacteriologic analysis - B abortus strain RB51 and strain 19 were used in this study. SRB51 vaccine was obtained from a commercial vaccine supplier*^ and diluted appropriately using the supplied diluent. Strain 19 vaccine was obtained and prepared as directed for the vaccination of adult animals. Tissue and fluid samples to be tested for SRB51 or S19 were processed as described using tryptose agar (TSA) and TSA with 5% bovine serum, cyclohexamide (30 |ag/ml), bacitracin (7.5 U/ml), polymyxin B sulfate (1.8 U/ml) and ethylviolet (1:300,000).'^ Brucella colonies were identified by colony morphology and growth characteristics. g Collection and handling of blood for culture - At time periods corresponding to postvaccination day (PVD) 1-4 and postvaccination week (PVW) 1-12, 100 ml of blood was collected aseptically for culture and processed as described.^ Calf blood was processed similarly at necropsy. "^Colorado Serum Co., Denver, CO. '' National Veterinary Services Laboratory, Ames, lA. 76 Serologic analysis - Blood was collected by venipuncture from each heifer prior to vaccination and at PVW 1-12, and from calves at necropsy. Maternal and calf blood was collected into sterile 10-ml tubes, allowed to clot for 6-24 hrs at 4°C, and centrifuged. Serum was divided into 2 ml aliquots, frozen, and stored at -70''C. The STAT, and a previously described dot ELISA for antibodies to SRB51 were used to evaluate antibody responses/^ Lymphocyte proliferation assay - Assays were done on peripheral blood mononuclear cells (PBMC) collected prior to vaccination (PVD 0), on PVW 2,4, 6, 8, 10 and 12, and from calves at necropsy. Fifty milliliters of blood was collected in 5 ml acid citrate buffer as anticoagulant. Following centrifugation, plasma was decanted and cells were placed on a Ficoll gradient^ (s.g. 1.083 g/dl) for separation. PBMC were processed and resuspended to a 7 concentration of 10 viable cells/ml in RPMI 1640 medium. 14 A 50 |j.l aliquot (5x105 cells) of the PBMC suspension was added in duplicate to separate flat-bottom wells of a 96-well microtiter plate containing 100 jj.1 of various concentrations of y-irradiated B abortus SRB51 (10^ to 10^ bacteria/well), S19 (10^ to 10^ bacteria/well), SRB51 LPS (20 to .02 |ig/well), or S2308 LPS (2.5 to .025 |ig/well) in a solution of RPMI 1640 medium containing Lglutamine, 25mM Hepes, 10% fetal bovine serum, lOOU/ml penicillin, lOOfxg/ml streptomycin and 5 x 10'^ M 2-mercaptoethanol. Lymphocytes were collected for blastogenesis assays at necropsy from maternal sites (superficial cervical lymph nodes, supramammary lymph nodes), and calf sites (parotid lymph nodes, mesenteric lymph nodes and spleen) and prepared as described.'^ A 50 |il aliquot (5 x 10^ cells) of lymph node (LN) cell suspension was added in duplicate to separate flat-bottomed wells of a 96-well microtiter plate as described.'^ Lymph node and PBMC cell cultures were incubated for 7 ® Histopaque 1083, Sigma Daignostics, St. Louis, MO. 77 days at 37°C in 5% CO2, pulsed with l.O fiCi per well of [^H]-thynudine for 18 hours, harvested on to glass filters and counted for radioactivity in a liquid scintillation counter. Cell proliferation results were expressed as the mean counts per minute (CPM) ± the standard error of the mean. Statistical analysis - Differences in responses between treatments were compared by a repeated measures, general linear models procedure using the SAS system.^ For statistical comparisons, [^H] thymidine incorporation and antibody titers were also evaluated as the logarithm of the CPM or logarithm of the antibody titer. Means for individual treatments were separated by use of a least significant difference procedure (P < 0.05). Results Clinical Signs - All animals remained clinically normal throughout the experiment. Febrile responses were not detected in any animal during the first 4 days after vaccination. Abortions or premature births did not occur. Evidence of arthritis or anaphylactic shock was not seen in animals in either vaccine group. Postmortem examination - Gross or microscopic lesions were not seen in cattle vaccinated with either SRB51 or SI9. Similarly, lesions were not seen in calves fi-om vaccinated cattle in either group. Immunohistochemical staining of tissue sections fi:om SRB51-vaccinates revealed no SRB51 antigen. Bacteriology results - Strain RB51 was isolated from superficial cervical lymph nodes on the side of initial vaccination in 2/5 cases. No other maternal tissues, including placentomes ^ SAS Institute Inc., Cary, NC. 78 or intercotyledonary placenta contained SRB51 or SI9. Similarly, no maternal blood samples taken at any time point from either vaccine group contained SRB51 or SI9. No tissues or fluid specimens examined from calves from either group contained SRB51 or SI9. Serology - All SRB51-vaccinates remained serologically negative by STAT throughout the experiment (Fig 1). SI9 vaccinates showed peak STAT titers at PVD 14 which steadily declined to the termination of the smdy. Both SRB51- and S19-vaccinates showed dot ELISA titers to y-irradiated SRB51 (Fig 2). However, peak dot ELISA titers, were greater (P<0.05) in SRB51-vaccinates at PVD 42, 56 and 70. Nonvaccinated control heifers and calves from all groups showed no titers by either STAT or dot ELISA (data not shown). 600 SRB51 470 O H 340 t3 0 -0-- S19 I / V b / / 1 210 < 0. •0.^ 80 .0 -50 I 2 4 6 8 10 12 Postvaccination Week Figvire 1- Serum antibody responses of heifers vaccinated during pregnancy with Brucella abortus SRB5I or S19 as measured by the standard tube agglutination test. Results are expressed as mean titer ±S.E. 79 200 - SRB51-0- S19 (U 13 100 O c < ,0 ^ --O 0 1 2 4 6 8 10 Postvaccination Week Figure 2- Serum antibody responses of heifers vaccinated during pregnancy with Brucella abortus SRB51 or S19 as measured by a dot ELISA test for antibodies to SRB51. Results are expressed as mean titer ± S.E. Titers of SRB51-vaccinates at 6, 8 and 10 weeks after vacciation are significantly (P<0.05) greater than titers of SI9-vaccinates. Maternal cell mediated response - Blastogenesis assays of maternal PBMC revealed a greater (P<0.05) proliferative response in both vaccine groups than in nonvaccinated controls when PBMC were incubated with either y-irradiated-SRB51 or S19 (Fig 3). Although proliferative responses to y-irradiated S19 were greater (P<0.05) in S19-vaccinates than in SRB51-vaccinates on PVD 28 and 70, proliferative responses to y-irradiated SRB51 were 80 significantly different between S19 and SRB51-vaccinates at PVD 56 only. Similarly^ differences were not seen (P>0.05) between vaccinated and nonvaccinated cattle when maternal PBMC were incubated with LPS derived from SRB51 or S2308 (data not shown). Response of maternal superficial cervical node cells obtained at necropsy from SRB51-vaccinates did not differ (P>0.05) from that seen in nonvaccinated controls to any of the antigens tested (Fig 4). However, superficial cervical lymph node cells from S19vaccinates showed greater (P<0.05) proliferative responses to y-irradiated SRB51, and yuxadiated S19 than that seen in nonvaccinated controls and SRB51-vaccinates. Cells from maternal supramammary lymph node of SRB51- or SI9-vaccinates did not show proliferative responses that were significantly different (P>0.05) from controls (data not shown). Calf cell mediated response - PBMC from calves of SRB51-vaccinates showed greater (P<0.05) proliferation than nonvaccinated controls in response to y-irradiated SRB51, RB51 LPS and 2308 LPS ( Fig 5). PBMC from calves of S19-vaccinates also showed significantly higher (P<0.05) responses to y-irradiated SRB51, SRB51 LPS and S2308 LPS, as well as yirradiated S19 than did nonvaccinated controls (Fig 5). PBMC from calves of S19-vaccinates had greater responses (P<0.05) to y-irradiated SRB51, y-irradiated SI9, and SRB51 LPS than did PBMC from calves of cows vaccinated with SRB51. Cells from calf parotid lymph node, calf mesenteric lymph node and calf spleen of SRB51- or S19-vaccinates did not show proliferative responses that were significantly different (P>0.05) from controls (data not shown). 81 A. Control Postinfection Day 300 240 SRBSI c 180 £ 120 519 Control 60 14 28 42 56 70 85 Postinfection Day Figure 3- Proliferation of PBMC from SRB51-vaccinated, Sl9-vaccinated, or salineinoculated pregnant heifers to A)10^y-irradiated SRB51 or B) lO^y-irradiated S19. PBMC were incubated for 7 days with killed bacteria prior to being pulsed for 18 hrs. with [^H]thymidine. Results are presented as mean ± S.E. cpm of duplicate cultures; PVD = postvaccination day. 82 srb51 Control Control Figure 4- Proliferation of maternal superficial cervical lymph node cells from SRB51vaccinated, SI9- vaccinated, or saline-inoculated pregnant heifers to: A-10^ y-irradiated SRB51, B-10^ y-irradiated SI9, C- 2.5 |j.g SRB51 LPS, or D- 3.1 jig S2308 LPS. Lymph node cells were incubated for 7 days with antigen and then pulsed for 18 hrs. with [^H]thymidine. Results are expressed as mean ± S.E. cpm of duplicate cultures. Means with different superscripts differ significantly (P<0.05). 83 Control SRB5I Control Figure 5- Proliferation of calf PBMC from SRB51-vaccinated, SI 9- vaccinated, or salineinoculated pregnant heifers. See figure 4 for legend. 84 Discussion Results of this study support the use of SRB51 in pregnant cattle by showing that pregnant cattle can be safely vaccinated with SRB51 when inoculated SC using a dose of lO' CFU SRB51. Specifically, vaccination of pregnant cattle with SRB51 did not result in generalized infection of maternal or calf tissues but did induce cell mediated and humoral immune responses. However, small animal numbers decrease the likelihood of detecting rare vaccineinduced abortions. Likewise, no S19-induced abortions were seen in the present study. Although S19-induced abortions do occur, a large study involving the SC vaccination or >10,000 pregnant catde with S19, showed abortions occur in <1%.' Lymphocyte blastogenesis results of maternal PBMC in this study indicate that pregnant cattle develop a cell mediated immune (CMI) response to SRB5I by PVD 28. Heifer calves have been shown to develop a CMI response to SRB51 and S2308 protein fractions 10-12 wks after vaccination vnth a slightly higher dose of SRB51.''' Blastogenic responses to S2308 and SRB51 proteins were similar even though SRB51 fractions do not contain the 0-side chain of LPS.'"* Our studies flirther confirm that the LPS of S2308 is not blastogenic for PBMC or superficial cervical lymph node cells from SRB51-vaccinated cattle. The blastogenic and humoral responses of S19-vaccinates to y-irradiated SRB51 was not unexpected as the outer membrane proteins of Brucella species have been shown to be remarkably similar.Results of this study are also similar to data obtained from another study in which S19-vaccinates developed humoral responses to irradiated SRB51.'^ These studies suggest that S19 and SRB51 share common epitopes to which immune responses are directed. The development of a CMI response in the PBMC of calves from SRB51- or SI 9vaccinates suggests exposure of the calf to vaccine antigens although SRB51 or S19 were not isolated from calf tissues or placental membranes. It cannot be excluded that fetal infection occurred following vaccination, with bacteria cleared by immune responses prior to 85 parturition. Fetal immune responses to Brucella antigens are possible as the anatomic components of the fetal immime system are present by day 175 of gestation.18 Although pregnant heifers were vaccinated at 180 days of gestation, lack of CMI response by other calf lymphoid tissues and the lack of a calf humoral response suggest that the calf immune system may not be functionally mature in relation to response to Brucella antigens. A previous study using IV inoculation of pregnant cattle with SRB51 showed generalized fetal infection with SRB51 while only one of ten fetuses developed a humoral response.® Another previous study with bovine fetuses experimentally inoculated with S2308 showed increased levels of both IgG and IgM.'^ Differences in fetal response to S2308 and SRB51 may be due to the lack of the 0-side chain of the lipopolysaccharide of SRB51 or to differences in the amount of antigen to which the fetus was exposed. The results of this study further indicate that route of vaccination and dosage have significant effects on the outcome of vaccination with SRB51. Intravenous inoculation of 10'° CFU SRB51 to cattle at 6 months of gestation results in placentitis and fetal infection.® Conversely, lack of placentitis as well as absence of localization of SRB51 in other tissues, i.e. spleen, liver, etc., in the present study, suggest that systemic spread and colonization of SRB51 do not occur following SC vaccination with 10^ CFU SRB51. Route of inoculation has also been demonstrated to be critical with other brucellosis vaccines such as strain 45 or S19 which consistently cause abortion when administered IV but cause few or no abortions when administered SC.' Persistence of SRB51 in the present study in superficial cervical lymph nodes lymph nodes to 12 weeks after vaccination differs from previous findings.'^ In calves vaccinated SC with 10'° CFU SRB51 in the superficial cervical region, SRB51 was not isolated from lymph node biopsies beyond 6 weeks after vaccination.'^ Persistence in pregnant cattle may be related to age or pregnancy status at vaccination which may influence immune ftmction. Moreover, clearance of S19 from superficial cervical lymph nodes prior to the clearance of 86 SRB51 in the present study differs from that seen elsewhere.^^ However, the adult dose of S19 used in the current study is roughly 30-fold less than that used previously. t -J In agreement with previous findings, our SRB51-vaccinates did not develop antibody titers which were detected by STAT, thereby, facilitating the differentiation of SRB51vaccinates and naturally infected animals. Antibodies detected in the STAT are directed to the 0-side chain;" therefore, lack of STAT response in SRB51-vaccinates is thought to be due to the lack of 0-side chain on SRB51. ^ Lymphocyte blastogenesis and serologic reactions in the present study suggest an active immune response in pregnant animals without generalized infection of maternal reproductive or calf tissues. However, vaccination with SRB51 during pregnancy followed by challenge with a virulent strain of Brucella would be necessary to prove protective immune responses. Additionally, field tests with large numbers of pregnant cattle would also be needed to determine the true prevalence of SRB51-induced abortion. In outbreak situations it has been the policy of the USDA to "whole herd vaccinate", vaccinating both female adults and calves. The current findings suggest that vaccination of pregnant cattle with SRB51 would provide a safe means to enhance resistance to Brucella infection. References 1. Nicoletti P. A preliminary report on the efficacy of adult cattle vaccination using strain 19 in selected dairy herds in Florida. Proc US Animal Health Assoc 1977:91-100. 2. Nielsen K, Duncan JR. Animal Brucellosis. Nicoletti P. Vaccination. Boca Raton, FL: CRC Press; 1990:283-300. 3. Schurig GG, Roop RM, Bagchi T, et al. Biologic properties of RB51; a stable rough siizxa. of Brucella abortus. Vet Microbiol 1991;28:171-188. 87 4. Cheville NF, Stevens MG, Jensen AE, et al. Immune responses and protection against infection and abortion in cattle experimentally vaccinated with mutant strains of Brucella abortus. Am J Vet Res 1993;54:1591-1597. 5. Stevens MG, Hennager SG, Olsen SC, et al. Serologic responses in diagnostic tests for brucellosis in cattle vaccinated with Brucella abortus strain 19 or RBSl.JClin Micro 1994;32:1065-1066. 6. Palmer MV, Cheville NF, Jensen AE. Experimental infection of pregnant cattle with the vaccine Brucella abortus strain RB51: Pathologic, bacteriologic and serologic findings. Vet Pathol 1996;33:682-691. 7. Olsen SC, Stevens MG, Palmer MV, et al. Efficacy of Brucella abortus strain RB51 to protect cattle against brucellosis. Proc US Animal Health Assoc 1995:108-110. 8. Alton GG, Jones LM, Angus RD, et al. Techniques for the brucellosis laboratory. Bacteriologic Methods. Paris France: Institut National de la Recherche Agronomique; 1988:34-60. 9. United States Department of Agriculture, Animal and Plant Health Inspection Service, Veterinary Service. Brucellosis eradication. Uniform methods and rules, pp 36-37. Washington, D.C.: U.S. Government Printing Office; 1986. 10. Meador VP, Tabatabai LB, Hagemoser WA, et al. Identification of Brucella abortus in formalin-fixed, parafRn-embedded tissues of cows, goats, and mice with an avidinbiotin-peroxidase complex immunoenzymatic staining technique. Am J Vet Res 1986;47:2147-2150. 11. Palmer MV, Cheville NF, Tatum FM. Morphometric and histopathologic analysis of l)anphoid depletion in murine spleens following infection with Brucella abortus strains 2308, RB51 or an htrA deletion mutant. Vet Pathol 1996;33(3):282-289. 88 12. Cheville NF, Jensen AE, Hailing SM, et al. Bacterial survival, lymph node changes and immunologic responses of cattle vaccinated with standard and mutant strains of Brucella abortus. Am J Vet Res 1992;53:1881-1888. 13. Olsen SC, Stevens MG, Cheville NF, et al. Experimental use of a dot-blot assay to measure serologic responses of cattle vaccinated with Brucella abortus strain RB51. J Vet Diag Lab Invest 1996;(submitted). 14. Stevens MG, Olsen SC, Cheville NF. Lymphocyte proliferation in response to Brucella abortus RB51 and 2308 in RB51-vaccinated or 2308-infected cattle. Infect Immun 1996;64(3):1007-1010. 15. Stevens MG, Olsen SC, Cheville NF. Comparative analysis of immune responses in cattle vaccinated with Brucella abortus strain 19 or strain RB51. Vet Immunol Immunopathol 1995;44:223-235. 16. Santos JM, Verstreate DR, Perrera VY, et al. Outer membrane proteins from rough strains of four Brucella species. Infect Immun 1984;46(1):188-194. 17. Brooks-Worrell BM, Splitter GA. Sodium dodecyl sulfate- and salt-extracted antigens from various Brucella species induce proliferation of bovine lymphocytes. Infect Immun 1992;60(5):2136-2138. 18. Tizard I. Veterinary Immunology. Immunity in the Fetus and Newborn. 3rd ed. Philadelphia, PA: W.B. Saunders Co.; 1987:171-184. 19. Enright FM, Walker JV, Jeffers G, et al. Cellular and humoral responses of Brucella a^or/w^-infected bovine fetuses. Am J Vet Res 1984;45:424-430. 20. Taylor AW, McDiarmid A. The stability of the avirulent characteristics of Brucella abortus strain 19 and strain 45/20 in lactating and pregnant cows. Vet Rec 1949;61:317318. 89 CHAPTER 4. TUMOR NECROSIS FACTOR-a PRODUCTION IN PREGNANT CATTLE AFTER INTRAVENOUS OR SUBCUTANEOUS VACCINATION WITH BRUCELLA ABORTUS STRAIN RB51 A paper prepared for submission to Veterinary Immunology and Immunopathology Mitchell V. Palmer DVM , Theodore H. Elsasser PhD, and Norman F. Cheville DVM, PhD Abstract TNF-a is a key cytokine in inflammatory processes and mediates many of the clinical signs associated with endotoxemia due to gram negative bacterial infection. In order to determine the influence of brucellosis vaccination on TNF-a levels in pregnant cattle and the possible role of the placenta in TNF-a production, pregnant cattle were vaccinated FV with Brucella abortus strain RB51 (n=IO), SC with B. abortus strain RB5I (n=5), or SC with B. abortus strain 19 (n=5); controls received pyrogen free saline SC (n=2). Radioimmunoassays showed no elevations in TNF-a levels in serum or plasma from IV or SC vaccinated cattle that differed from controls (P>0.05). Similarly, TNF-a levels in amniotic and allantoic fluids from SC vaccinated cattle were not different from controls (P>0.05). Immunohistochemistry for TNF-a revealed increased immunoreactivity within trophoblastic epithelial cells in SC and rV vaccinated cattle. Immunoreactivity was most extensive in FV vaccinated cattle that developed vaccine induced placentitis. These studies indicate that SC vaccination for the prevention of brucellosis using recommended adult dosages does not result in elevations of 90 TNF-a in plasma, serum or placental fluids; however, vaccination of pregnant cattle does stimulate trophoblastic epithelial cells to express TNF-a even when placentitis is not histologically evident. 1. Introduction Tumor necrosis factor-alpha (TNF-a) is a cytokine capable of altering immune and pathophysiologic responses. TNF-a is expressed in a variety of cell types; the most quantitatively important being the macrophage (Myers et al.,1995; Tracey et al.,1992). However, TNF-a is also produced by lymphocytes, mast cells, NK cells, neutrophils, B-cells, keratinoc5^es, microglial cells, smooth muscle cell and various tumor cells (Tracey et al.,1992). Various cell types also carry receptors for TNF-a (Le et al.,1987; Dayer et al.,1985). Consequently, the biological effects of TNF-a are numerous including: inhibition of lipoprotein lipase (Beutler et ai.,1985), pyretogenesis (Dinarello et al.,1986; Le et al.,1987), activation of neutrophils (Tsujimoto et al.,1986; Klebanoff et al.,1986), T-cell activation (Le et al.,1987), osteoclast activation (Bertolini et al.,1986), induction of endothelial cell procoagulant activity (Nawroth et al.,1986; Nawroth et al.,1986), mitogenic action on fibroblasts (Vilcek et al.,1986; Sugarman et al.,1985), induction of oncogenes, and induction of synthesis of acute phase proteins, prostaglandins (Dinarello et al.,1986; Dayer et al.,1985), collagenases (Dayer et al.,1985), granulocyte/macrophage-colony stimulating factor, IL-1 and TNF-a itself (Dinarello et al.,1986; Nawroth et al.,1986). 91 TNF-a levels are increased in response to a variety of stimuli, the most potent of which is endotoxin (Adams et al.,1990; Kenison et al.,1991)(Mannel et al.,1987), a lipopolysaccharide (LPS) in the cell wall of gram-negative bacteria. Endotoxin exposure results in numerous clinical signs, many of which are mediated through the production of TNF-a, including fever, anorexia, depression, hypotension, metabolic acidosis, transient hyperglycemia followed by hypoglycemia, gastrointestinal infarction, hemorrhage of the adrenal glands and pancreas, acute nephrosis, and interstitial pneimionitis (Mannel et al.,1987). Endotoxin exposure can also have significant effects on reproductive performance in cattle. Endotoxin has been shown to be luteolytic (Benyo et al.,1992), causing shortened interestrous intervals. Infusion of Salmonella typhimurium endotoxin in pregnant cows induces abortion within 4 to 10 days of infusion (Foley et al.,1993). Endotoxin infusion is followed by a rapid increase in maternal but not fetal TNF-a (Foley et al.,1993). Bovine brucellosis is an economically important abortifacient disease caused by the gram-negative bacterium Brucella abortus. Since 1940, an integral part of brucellosis prevention and eradication in the Untied States has been the vaccination of female calves with the live vaccine Brucella abortus strain 19 (S19). Although effective in decreasing the incidence of brucellosis, S19 has several disadvantages including the induction of abortion in some pregnant cattle (Nicoletti,1977). A new vaccine. Brucella abortus strain RB51 (SRB51), has recently been approved in the United States as an official calfhood vaccine. A study in pregnant cattle showed that IV injection of SRB51 results in placental infection, placentitis and fetal infection; although, abortion was not seen (Palmer et al.,1996). Further 92 study showed SC vaccination of SRB5I did not result in placental infection, placentitis or fetal infection (Palmer et al., (in press)). This work was part of a vaccine safety study involving vaccination of pregnant cattle with SRB51. Clinical signs and pathology associated with reproduction of all groups have been reported elsewhere (Palmer et al.,1996; Palmer et al., (in press)). Briefly, one heifer vaccinated IV with SRB51 delivered a weak calf one month prior to the predicted calving date. No other premature births or abortions were noted in either the IV or SC vaccinated groups (Palmer et al.,1996; Palmer et al., (in press)). Eight of ten IV vaccinated cattle and 0/10 SC vaccinated cattle developed placentitis which is described in detail elsewhere (Palmer et al.,1996; Palmer et al., (in press)). In the rV vaccinated group, 80% of cattle were febrile during the initial 24 to 48 hrs. after vaccination. All but two animals had normal body temperatures by 48 hrs. after vaccination and the remaining two were normal by 96 hrs. after vaccination. Cattle vaccinated SC with SRB51 or S19 showed no febrile response and did not differ from controls (Palmer et al.,1996; Palmer et al., (in press)). The current investigation was designed to determine the changes in TNF-a levels following vaccination with SRB51 or S19 as a measure of the toxicity of vaccine-associated LPS. Additionally, this investigation was designed to determine placental tissue for cellular sites of TNF-a production. 93 2. Materials and MethodsPolled Hereford heifers used in this experiment were obtained from a nonvaccinated, brucellosis-free herd. Heifers were 16-27 months old at breeding. All cattle were serologically negative for brucellosis by standard tube agglutination (Alton et al.,1988). Upon arrival, and throughout the experiment, cattle were kept on pasture. Breeding was by natural service using a Polled Hereford bull which was serologically negative for brucellosis by standard tube agglutination. IV vaccination - At 6 months of gestation heifers were injected IV in the right jugular 10 vein with 10 CPU SRB51 suspended in 10 ml sterile saline (0.85% NaCl). At postvaccination week 8 (PVW), 5 heifers were randomly selected, euthanatized, and examined. The remaining catde were euthanatized at term (n=3) or following premature delivery (n=l). SC vaccination - At six months of gestation animals were randomly divided into 3 groups and injected SC in the right prescapular region with 9.4 x 10® CPU SRB51, (n=5), 3 x 10^ CPU SI9, (n=5), or 2 ml sterile pyrogen free saline, (n=2). The exact inocula dose reported above was determined retrospectively. The dosage of SRB51 used in adult vaccination studies is 10^ CPU (Palmer et al.,1995), while the recommended adult dosages of 94 S19 is 3-10 X 10^ CFU (United States Department of Agriculture et al.,1986). All animals were euthanatized at full term of pregnancy. In both rV and SC vaccinated cattle, body temperatures were recorded daily for the first 4 days following vaccination and animals were observed twice daily during the experiment for clinical signs of abortion, illness or other adverse reactions. Radioimmunoassay- Blood was collected 3 times prior to vaccination on days -7, -5 and on the day of vaccination (day 0) and on postvaccination days (PVD) 1,2, 3,4, 7, 14, 21, 28, 35,42,49, 56, 63. Blood was collected from the calves at necropsy. Amniotic and allantoic fluids were collected at necropsy from cattle vaccinated SC with SRB51 or SI9. Blood and placental fluids from SC vaccinated cattle were immediately placed in sterile tubes with EDTA, and centrifuged. Samples were filtered through a 0.22 ^im low protein binding filter and aliquotted in 2 ml vials and frozen at -70 C. Similarly, serum was collected from IV vaccinated cattle in sterile tubes without anticoagulant and allowed to clot for 4 to 6 hrs at 4 C and centrifuged. Serum was removed, filtered as above, and aliquotted in 2 ml vials and frozen at -70 C. Plasma, serum, and placental fluid TNF-a levels were measured using a double-antibody radioimmimoassay (RIA) as described (Kenison et ai.,1990). Immunohistochemistry - Placental samples were placed in neutral 10% buffered formalin and processed by routine paraffin embedment techniques, cut 4-6 pim thick and left unstained. Staining for bovine TNF-a was done as described (Cirelli et al.,1995), with 95 modifications for deparaffinization of parafBn sections. An avidin-biotin-alkaline phosphatase complex staining technique was used as described (Pahner et al.,1996; Meador et al.,1986). Anti-bovine TNF-a antibody was prepared as described (Kenison et al.,1990), and used at a dilution of 1:500 - 1:1000. Blocking with 1% normal bovine serum for 15 min. prior to the application of the anti-TNF-a antibody was done to reduce background staining. Negative control slides included histologic sections stained with serum from clinically normal rabbits and slides stained with omission of the primary antibody. Macrophages within the individual tissue sections were highly immunoreactive for bovine TNF-a and were used as an intemal positive control. Statistical Analysis - Differences in TNF-a levels between groups were compared by a repeated measures, general linear models procedure using the SAS system (SAS Institute, Gary, NC). Differences between means were considered significant at a level of P<0.05. Prevaccination means on days -7, -5 and 0 were combined within groups to yield one prevaccination TNF-a value per group. 3. Results Plasma TNF-a levels - Tumor necrosis factor-a levels in plasma of cattle vaccinated SC are shown (Figure 1). There were not significant differences between cattle vaccinated SC with SRB51 or S19 or given saline (P>0.05). However, in these three groups a significant rise (P<0.05) was seen in TNF-a levels over the course of the experiment which 96 peaked at 42-56 days after vaccination. Similarly, levels of TNF-a in serum of IV vaccinated cattle did not show significant elevations as compared to controls. Levels of TNF-a in amniotic fluid and allantoic fluid from SC vaccinated cattle were either undetectable or not elevated above that seen in control cattle (Figure 2). Calf plasma or serum did not contain significantly elevated TNF-a as compared to controls (Figure 3). SRB51SC 90 - - S19SC 80 ^ A SRB51IV • 70 60 I ^ - 7 - Control - r 5 0 40 30 - - ^ y ^ := r 4 * " - ^ i-- i ^ - * - ^ - ^ 20 - 10 - 0 0 2 4 14 28 42 56 Postvaccination Day Figvire 1. Plasma TNF-a levels in cattle vaccinated SC with 10^ CFU SRB51 (n=5) or 3x10^ CFU S19 (n=5), IV v^ith 10'° CFU SRB51 (n=lO), or given saline (n=2). TNF-a levels were determined by radioimmunoassay and represent mean ± standard error. * Significantly greater in SC vaccinates than values at 0-14 days after vaccination (P<0.05). 97 A. SRB51 Controls 100 80 60 SRB51 S19 BO Q, 40 Controls 20 0 Figure 2. TNF-a levels in amniotic (A), and allantoic (B) fluids from cattle vaccinated SC with lO' CFU SRB51 (n=5), 3x10® CFU S19 (n=5) or sterile saline (n=2). Values are expressed as mean ± standard error. 98 Immunohistochemistry Results - Placental sections from SRB51 IV vaccinated catde with placentitis showed staining of large regions of the trophoblast layer. Strongly immunoreactive trophoblastic epithelial cells contained diffuse cytoplasmic staining for TNF-a (Figure 4). In contrast, placental sections from SC vaccinated cattle (SI9 and SRB5I), which did not have placentitis, showed a punctate staining of trophoblastic epithelial cells with few cells having diffiise cytoplasmic staining (Figure 5). Immunoreactive pimctate foci within trophoblastic epithelial cells were multiple, round, 1-3 (im in diameter and strongly immunoreactive for TNF-a. Staining was seen in multiple groups of adjacent cells bounded by groups of cells without TNF- a immunoreactivity. Macrophages which stained strongly for TNF-a were present within the chorion, endometrium, myometrium and within the submucosal region of the utems surrounding glands and vessels of both IV and SC vaccinated cattle (Figure 6). Saline-inoculated control catde showed immunoreactivity for TNF-a in low numbers of macrophages within the myometrium and subserosa. Macrophages between the outer longitudinal muscle layer and serosa, and macrophages adjacent to submucosal glands or vessels, stained strongly for TNF-a (Figiire 7). Low numbers of individual trophoblastic epithelial cells were also immunoreactive for bovine TNF-a with moderate difilise staining of the entire cytoplasm; however, the remainder of the trophoblast layer lacked immimoreactivity for TNF-a (Figure 8). 99 S19SC L_1 SRB51 IV Figure 3. TNF-a levels in plasma of calves from dams vaccinated SC with 10^ CFU SRB51 (n=5), 3x10^ CFU S19 (n=5), IV with 10'° CFU SRB51, or sterile saline (n=2). Values are expressed as mean ± standard error. 4. Discussion These studies indicate that bovine trophoblastic epithelial cells can produce TNF-a and that TNF-a is produced by more trophoblastic epithelial cells during SRB51-induced placentitis than in controls or vaccinated cows without placentitis. The diffuse cytoplasmic TNF-a immunoreactivity seen in cases of placentitis may reflect increased expression of TNF-a. Intense punctate immunoreactivity for TNF-a seen in SC vaccinated cows also suggests increased expression over that seen in controls. Figure 4. Section of placenta from heifer vaccinated IV at 6 months gestation with 10^° CFU B. abortus strain RB51 and examined 8 weeks later. Note diffuse cytoplasmic immunoreactivity for TNF-a in trophoblastic epithelial cells. Alkaline phosphatase method. Gill's hematoxylin counterstain. Bar = 30 jim. Figure 5. Section of placenta from heifer at full term which was vaccinated SC at 6 months gestation with 10^ CFU B. abortus strain RB51. Note punctate cytoplasmic immunoreactivity for TNF-a in trophoblastic epithelial cells. Alkaline phosphatase method. Gill's hematoxylin counterstain. Bar = 30 jim. Figure 6. Section of placenta from heifer at full term which was vaccinated SC at 6 months gestation with 10^ CFU B. abortus strain RB51. Note immunoreactivity for TNF-a in macrophages surrounding chorioallantoic arteriole. Alkaline phosphatase method. Gill's hematoxylin counterstain. Bar = 50 |im. Figure 7. Section of placenta from control heifer at full term which was injected SC with saline at 6 months gestation. Note lack of immunoreactivity for TNF-a in trophoblast layer. Alkaline phosphatase method. Gill's hematoxylin counterstain. Bar = 50 |im. Figure 8. Section of uterus from control heifer at full term which was injected SC with saline at 6 months gestation. Note immunoreactivity for TNF-a in macrophages in subserosa. Alkaline phosphatase method. Gill's hematoxylin counterstain. Bar = 30 i^m. 101 102 However, Brucella were not recovered from SC vaccinated cattle and histologic signs of inflammation were not seen. In mice, endotoxin treatment results in the appearance of TNF- a in secretory granules of granulocytes, monocytes and neutrophils in all tissues examined (Schmauder-Chock et al.,1994). The punctate TNF-a immunoreactivity in trophoblastic cells of SC vaccinated cattle may represent secretory granules containing TNF-a. The dose of endotoxin associated with our vaccinations may have been insufficient to induce changes seen in catde injected with endotoxin of other bacterial species. An experimental study in pregnant cattle with endotoxemia showed elevations in maternal plasma TNF-a following injection of Salmonella endotoxin; however, fetal plasma and amniotic and allantoic fluid TNF-a levels were not elevated above controls (Foley et al.,1993). The endotoxin of Brucella has been shown to be 10,000 fold less potent than that of Escherichia coli in eliciting fever in rabbits (Goldstein et al.,1992), and 1,000 fold less potent than Salmonella endotoxin in eliciting lysozyme release in neutrophils (Rasool et al.,1992). It should be noted that radioimmunoassay measures TNF-a regardless of bioactivity. Factors which inhibit TNF-a bioactivity, such as soluble TNF-a receptors, may interfere with bioactivity but not immunoreactivity of secreted TNF-a (Myers et al.,1995). It may be possible tliat SC vaccination induced changes within the placenta which were significant enough to increase TNF-a expression, but not significant enough to result in chemotaxis of inflammatory cells and disruption of cells or tissue. Immunoreactivity of TNF-a in bovine trophoblastic epithelial cells in cases of placentitis has not been demonstrated previously but is consistent with studies in women with preterm labor due to 103 infection. Elevated TNF-a has been measured in the amniotic fluid of women in preterm labor (Romero et al.,I989; Romero et al.,1989) (Laham et al.,1994) (Halgunset et al.,1994) with clinical as well as subclinical intrauterine infection. Furthermore, TNF-a production has been documented in human cytotrophoblasts and decidual cells (Chen et al.,1991; Stallmach et al.,l995). In response to TNF-a and/or EL-l, several cell types, including amnion and decidua synthesize prostaglandins (Dayer et al.,1985) (Riley et al.,1984)(Romero et al.,1989). Similarly, 10 |ag of recombinant human TNF-a causes abortion in mice (Parant,1987), while in a model of malaria-induced abortion, as little as 1.5 ^g will induce abortion in mice infected with Plasmodium vinckei (Clark et al.,1988). Therefore, it is hypothesized that bacterial products, such as LPS, stimulate maternal and/or fetal cells to produce cytokines, such as TNF-a, which in turn increase prostaglandin synthesis by intrauterine tissues resulting in the onset of labor (Romero et al.,1989). In cultured bovine luteal cells, TNF-a induces a dose dependent increase in luteal PGF2a and has an inhibitory effect on lutenizing hormone-induced progesterone production (Benyo et al.,l992). Both increased PGF2a and decreased progesterone levels are changes consistent with the onset of parturition in cattle. The reason is unknown for the increase in plasma TNF-a seen in SC vaccinated and control cattle which occurred 42-56 days after vaccination. This time period corresponds to approximately 7.5 to 8 months' gestation. Mononuclear cells from periparturient cattle have been shown to produce significantly more TNF-a than cows in mid to late lactation (Sordillo et al.,1995). However, all cattle in our study were housed together during the experiment; 104 therefore, the possibility of subclinical disease affecting all cattle which resulted in elevations of plasma TNF-a cannot be ruled out. Reasons for lack of postvaccination fever in our SC vaccinated cattle may include vaccine dose and age at vaccination. A previous report with S19 vaccination described fever and anorexia of 3-5 days duration in dairy and beef calves (Dale et al.,1957). However, the animals used were calves and the dose of S19 vaccine was unspecified. This study shows that SC vaccination vidth SRB51 or S19 and FV vaccination with SRB51 results in increased TNF-a immunoreactivity in histologic sections of placenta with and without placentitis. Although neither IV vaccination with SRB51 nor SC vaccination with SRB51 or S19 resulted in elevations in plasma TNF-a, the increased immunoreacitivity of TNF-a indicates the potential exists for a role in bovine abortion for TNF-a similar to that suspected in humans. References 1. Adams JL, Semrad SD, Czuprynski CJ. 1990. Administration of bacterial lipopolysacharide elicits circulationg tumor necrosis factor-alpha in neonatal calves. J Clin Micro 28:998-1001. 2. Alton GG, Jones LM, Angus RD, Verger JM. Techniques for the brucellosis laboratory. Bacteriologic Methods. Paris France: Institut National de la Recherche Agronomique; 1988:34-60. 105 3. Benyo DF, Pate JL. 1992. Tumor necrosis factor- a alters bovine luteal cell synthetic capacity and viability. Endocrinology 130:854-860. 4. Bertolini DR, Nedwin GE, Bringman TS, Smith DD, Mundy GR. 1986. Stimulation of bone resorption and inhibition of bone formation in vitro by human tumor necrosis factors. Nature 319:516-518. 5. Beutler BA, Cerami A. 1985. Recombinant interleukin 1 suppresses lipoprotein lipase activity in 3T3-L1 cells. J Immunol 135:3969-3971. 6. Chen H-L, Yang Y, Hu X-L, Yelavarthi K, Fishback J, Hunt J. 1991. Tumor necrosis factor alpha mRNA and protein are present in human placental and uterine cells at early and late stages of gestation. Amer J Pathol 139:327-335. 7. Cirelli RA, Carey LA, Fisher JK, et al. 1995. Endotoxin infusion in anesthetized sheep is associated with intrapulmonary sequestration of leukocytes that immunohistochemically express tumor necrosis factor-a. S Leuk Biol 57:820-826. 8. Clark lA, Chaudhri G. 1988. Tumor necrosis factor in malaria-induced abortion. Am J Trop Med Hyg 39:246-249. 9. Dale HE, English JE. 1957. Effect of brucellosis vaccination on rectal temperature and feed consumption of beef and dairy calves. Vet Med 0ctober:477-480. 106 10. Dayer JM, Beutler B, Cerami A. 1985. Cachectin/tumor necrosis factor stimulates collagenase and prostaglandin E2 production by human synovial cells and dermal fibroblsasts. J Exp Med 162:2163-2168. 11. Dinarello CA, Carmon JG, Wolff SM, et al. 1986. Tumor necrosis factor (cachectin) is an endogenous pyrogen and induces production of interleukin 1. J Exp Med 163:14331450. 12. Foley GL, Schlafer DH, Elsasser TH, Mitchell M. 1993. Endotexemia in pregnant cows: Comparisons of maternal and fetal effects utilizing the chronically catheterized fetus. Theriogenology 39:739-762. 13. Goldstein J, Hoffman T, Frasch C, et al. 1992. Lipopolysaccharide (LPS) from Brucella abortus is less toxic than that from Escherichia coli, suggesting the possible use of B. abortus or LPS from B. abortus as a carrier in vaccines. Infect Immun 60:1385-1389. 14. Halgunset J, Johnsen H, Kjollesdal A, Qvigstad E, Espevik T, Austgulen R. 1994. Cytokine levels in amniotic fluid and inflammatory changes in the placenta form normal deliveries at term. Eur J Obstet Gynecol 56:153-160. 15. Kenison DC, Elsasser TH, Fayer R. 1990. Radioimmunoassay for bovine tumor necrosis factor: Concentrations and circulating molecular forms in bovine plasma. J Immunoassay 11:177-198. 107 16. Kenison DC, Elsasser TH, Payer R. 1991. Tumor necrosis factor as a potential mediator of acute metabolic and hormonal responses to endotoxemia in calves. Am J Vet Res 52:1320-1326. 17. Klebanoff SJ, Vadas MA, Harlan JM, et al. 1986. Stimulation of neutrophils by tumor necrosis factor. J Immunol 136:4220-4225. 18. Laham N, Brennecke S, Bendtzen K, Rice G. 1994. Tumour necrosis factor a during human pregnancy and labour: maternal plasma and amniotic fluid concentrations and release from intrauterine tissues. Eur J Endocrinol 131:607-614. 19. Le J, Vilcek J. 1987. Tumor necrosis factor and interleukin-1: Cytokines with multiple overlapping biological activities. Lab Invest 56:234-248. 20. Mannel DN, Northhoff H, Bauss F, Falk W. 1987. Tumor necrosis factor: A cytokine involved in toxic effects of endotoxin. Rev Infect Dis 9:S602-S606. 21. Meador VP, Tabatabai LB, Hagemoser WA, Deyoe BL. 1986. Identification of Brucella abortus in formalin-fixed, paraffin-embedded tissues of cows, goats, and mice with an avidin-biotin-peroxidase complex immunoenzymatic staining technique. Am J Vet Res 47:2147-2150. 22. Myers MJ, Murtaugh MP. Biology of tumor necrosis factor. In. Cytokines in Animal Health and Disease. New York, NY: Dekker Publishing; 1995:121-151. 108 23. Nawroth PP, Bank I, Handley D, Cassimeris J, Chess L, Stem D. 1986. Tumor necrosis factor/cachectin interacts with endothelial cell receptors to induce release of interleukin 1. J Exp Med 163:1363-1375. 24. Nawroth PP, Stem DM. 1986. Modulation of endothelial cell hemostatic properties by tumor necrosis factor. J Exp Med 163:740-745. 25. Nicoletti P. 1977. A preliminary report on the efficacy of adult cattle vaccination using strain 19 in selected dairy herds in Florida. Proc. SOth Annu. Meet. U.S. Animal Health Assoc. 91-100. 26. Palmer MV, Cheville NF, Jensen AE. 1996. Experimental infection of pregnant cattle with the vaccine Brucella abortus strain RB51: Pathologic, bacteriologic and serologic findings. Vet Pathol 33:682-691. 27. Palmer MV, Cheville NF, Tatum FM. 1996. Morphometric and histopathologic analysis of lymphoid depletion in murine spleens following infection with Brucella abortus strains 2308, RB51 or an htrA deletion mutant. Vet Pathol 33:282-289. 28. Palmer MV, Olsen SC, Cheville NF. (in press). Safety and immunogenicity of the vaccine Brucella abortus strain RB51 in pregnant cattle. Am J Vet Res . 29. Pahner MV, Olsen SC, Stevens MG, Cheville NF. 1995. Placentitis induced by Brucella abortus strain RB51 in pregnant cattle. Proc US Animal Health Assoc :104-107. 109 30. Parant M. 1987. Role of TNF in nonspecific stimulation of the mouse resistance to infection. Immunobiology 175:26. 31. Rasool 0, Freer E, Moreno E, Jarstrand C. 1992. Effect of Brucella abortus lipopolysacharide on oxidative metabolism and lysozyme release by human neutrophils. Infect Immim 60:1699-1702. 32. Riley LIC, Robertson DC. 1984. Ingestion and intracellular survival of Brucella abortus in human and bovine polymorphonuclear leukocytes. Infect Immun 46:224-230. 33. Romero RJ, Durum S, Dinarello CA, Oyarzun E, Hobbins JC, Mitchell MD. 1989. Interleukin-1 stimulates prostaglandin biosynthesis by human amnion. Prostaglandins 37:1322. 34. Romero RJ, Manogue KR, Mitchell MD, et al. 1989. Infection and labor FV. Cachectin-tumor necrosis factor in the amniotic fluid of women with intraamniotic infection and preterm labor. Amer J Obstet Gynecol 161:336-341. 35. Romero R, Brody DT, Oyarzun E, et al. 1989. Infection and labor; Interleukin-1: A signal for the onset of parturition. Amer J Obstet Gynecol 160:1117-1123. 36. Schmauder-Chock EA, Chock SP, Patchen ML. 1994. Ultrastructural localization of tumour necrosis factor-alpha. Histochem J 26:142-151. 110 37. Sordillo LM, Pighetti GM, Davis MR. 1995. Enhanced production of bovine tumor necrosis factor-a during the periparturient period. Vet Immunol Immunopathol 49:263-270. 38. Stallmach T, Hebisch G, Joller-Jemelka HI, Orban P, Schwaller J, Engelmarm M. 1995. Cytokine production and visualized effects in the feto-matemal unit: Quantitative and topographic data on cytokines during intrauterine disease. Lab Invest 73:384-392. 39. Sugarman BJ, Aggarwal BB, Hass PE, Figari IS, Palladino MA, Shepard HM. 1985. Recombinant human tumor necrosis factor-a: Effects on proliferation of normal and transformed cells in vitro. Science 230:943-945. 40. Tracey ECJ, Cerami A. 1992. Tumor necrosis factor and regxilation of metabolism in infection: Role of systemic versus tissue levels. Proc Soc Exp Biol Med 200:233-239. 41. Tsujimoto M, Yokota S, Vilcek J, Weissmaim G. 1986. Tumor necrosis factor provokes superoxide anion generation from neutrophils. Biochem Biophys Res Comm 137:1094-1100. 42. United States Department of Agriculture, Animal and Plant Health Inspection Service, Veterinary Service. Brucellosis Eradication. Uniform Methods and Rules, pp. 26-37. Washington, D.C.: U.S. Goverrmient Printing Office; 1986. 43. Vilcek J, Palombella VJ, Henriksen-DeStefano D, et al. 1986. Fibroblast growth enhancing activity of tumor necrosis factor and its relationshop to other polypeptide growth factors. J Exp Med 163:632-643. Ill CHAPTER 5. GENERAL CONCLUSIONS General Discussion These studies show the vaccine Brucella abortus strain RB51 (SRB51) does have a tropism for the bovine placenta, similar to other strains of B. abortus\^^ however, even v^rith moderate placentitis and disseminated fetal infection, abortion was not seen. Placentitis could only be induced by higher than recommended dosages and IV inoculation. Subsequent studies showed that pregnant cattle can be safely vaccinated SC with lO' CPU SRB51 which induced both humoral and cell mediated immune responses without placentitis or fetal infection. When infected herds have been identified in the United States, action taken by the USDA may involve vaccination of the whole herd, including pregnant animals.The results presented here support the vaccination of pregnant cattle v^th SRB51. The protective value of SRB51 vaccination in pregnant cattle; however, can only be suggested from the data presented. Studies involving vaccination of pregnant cattie followed by challenge with the virulent strain 2308 would be required to determine true efficacy. These results further suggest that SRB5I induced abortion is a rare event in cattle. Vaccination of 10,000 head of pregnant cattle with S19 resulted in <1% abortion.® The number of cattle used in the present studies was small and the true prevalence of SRB51induced abortion will require the vaccination of large numbers of pregnant cattle. Field studies with SRB51 have vaccinated >800 pregnant cattle in various stages of gestation with 1 SRB51-induced abortion.' 112 The disseminated infection of full tenn calves following IV vaccination without immune response by the calf suggests that calf infection may occur very late in gestation prior to calving or calves may not be functionally capable of moimting a response to SRB51. The presence of SRB51 infection in calves as early as 8 weeks after infection and 4 weeks prior to calving discounts the possibility of calf infection immediately prior to calving. In contrast, calves infected in utero with the virulent strain 2308 developed increased IgG and IgM.^ Differences between responses reported in S2308 infected calves and those seen here may be due to the rough nature of SRB51. Smooth Brucella spp. survive in macrophages which are important antigen presenting cells as well as effector cells of the immune response/ Humoral and cell mediated responses in pregnant cattle are similar to those seen in nonpregnant catde.'^' Lack of humoral responses detected by routine brucellosis surveillance tests, such as the card test or standard tube agglutination test, allow the differentiation of SRJB51 vaccinated animals from naturally infected animals.'^ Such differentiation is not possible with S19 vaccinated animals and should accelerate the brucellosis eradication effort. Adverse effects due to LPS induced TNF-a elevations may be of minimal concern under the parameters described in these studies. Lack of significant increases in circulating TNF-a levels following IV or SC vaccination with SRB51 and SC vaccination with S19 may be due to the decreased toxicity of Brucella LPS as compared to other gram negative pathogens."*' However, immunohistochemical detection of TNF-a in trophoblastic epithelial 113 cells of vaccinated cattle but not control catde suggests that vaccination did result in increased TNF-a expression by trophoblastic epithelial cells. The proposed role of placental TNF-a in women with premature labor,'" '' and abortion in mice,"^ is a key one. Our findings suggest that vaccination does have an effect on bovine placental TNF-a; however, this role remains poorly defined and deserves further investigation. References 1. Cheville NF, Stevens MG, Jensen AE, Tatum FM, Hailing SM: Immune responses and protection against infection and abortion in cattle experimentally vaccinated with mutant strains of Brucella abortus. Am J Vet Res 54:1591-1597,1993 2. Clark lA, Chaudhri G: Tumor necrosis factor in malaria-induced abortion. Am J Trop Med Hyg 39:246-249,1988 3. Enright FM, Walker JV, Jeffers G, Deyoe BL: Cellular and humoral responses of Brucella a^or/wj-infected bovine fetuses. Am J Vet Res 45:424-430,1984 4. Goldstein J, Hoffinan T, Frasch C, Lizzio EF, Beining PR, Hochstein D, Lee YL, et al.: Lipopolysaccharide (LPS) from Brucella abortus is less toxic dian that from Escherichia coli, suggesting the possible use of B. abortus or LPS from B. abortus as a carrier in vaccines. Infect Immun 60:1385-1389,1992 114 5. Harmon BG, Adams LG, Frey M: Survival of rough and smoodi strains of Brucella abortus in bovine mammary gland macrophages. Am J Vet Res 49:1092-1097,1988 6. Nicoletti P: A preliminary report on the efficacy of adult cattle vaccination using strain 19 in selected dairy herds in Florida. Proc. 80th Annu. Meet. U.S. Animal Health Assoc. 91-100,1977 7. Olsen SC, Palmer MV, Stevens MG, Cheville NF: Brucella abortus strain RB51 as a calfhood vaccine in bison or adult pregnant cattle. Proceedings 100th Meeting US Animal Health Assoc (in press),1997 8. Parant M: Role of TNF in nonspecific stimulation of the mouse resistance to infection. Immunobiology 175:26,1987 9. Rasool O, Freer E, Moreno E, Jarstrand C: Effect of Brucella abortus lipopolysacharide on oxidative metabolism and lyso2yme release by human neutrophils. Infect Immun 60:1699-1702,1992 10. Romero RJ, Manogue KR, Mitchell MD, Wu YK, Oyarzun E, Hobbins JC, Cerami A: Infection and labor IV. Cachectin-tumor necrosis factor in the amniotic fluid of women with intraamniotic infection and preterm labor. Amer J Obstet Gynecol 161:336-341,1989 115 11. Romero R, Mazor M, Sepulveda W, Avila C, Copeland D, Williams J: Tumor necrosis factor in preterm and term labor. Amer J Obstet Gynecol 166:1576-1587,1992 12. Smith T: A characteristic localization of bacillus abortus in the bovine fetal membranes. J Exp Med 29:451-456,1919 13. Stevens MG, Hennager SG, Olsen SC, Cheville NF: Serologic responses in diagnostic tests for brucellosis in cattle vaccinated with Brucella abortus strain 19 or RB51. J Clin Micro 32:1065-1066,1994 14. United States Department of Agriculture, Animal and Plant Health Inspection Service, Veterinary Service. Brucellosis Eradication. Uniform Methods and Rules, pp. 36-37. Washington, D.C.: U.S. Government Printing Office; 1986. 116 ACKNOWLEDGMENTS The author wishes to thank the many people who have played a role in the completion of this thesis and the work it represents. My foremost appreciation is to my wife Janell and my daughters Rachel, Angela, Chelsea and Vanessa, for without the support and companionship of my wife I would be nothing, and without my children nothing would be worth doing. I thank Dr. Norman Cheville for serving as my major professor and mentor throughout these studies and for serving as an inspirational supervisor at NADC. He has served as a model of scientist, professor, pathologist and leader to be emmulated. I am also grateful for the moral support, advice and positive influence of Dr. Mark Ackermarm as a member of my committee, collegue and friend. I am appreciative of the other members of my committee; Drs. John Kluge, Lyle Miller, Joseph Haynes and Charles Thoen for their time and effort spent in review of my program of study, preparation of preliminary exams and review of this dissertation. I am grateful for those who have provided technical assistance to these studies, especially Aileen Duitt, Randy Capsel, Richard Homsby and Dr. Allen Jensen. I value their technical expertese, professionalism, advice and friendship. I also appreciate the excellent work done by Jean Donald, Martha Church and Judi Stasko in the the microscopic services lab at NADC. They have provided valuable assistance and advice in preparation and processing the many samples generated during these studies. 117 The animal caretakers at NADC; Terry Krausman, Howard Cox, Dennis Weuve and James Carlson deserve my boundless gratitude for providing safe and humane care for the animals used in this study and for spending many hours caring for animals and assisting in sample collection, often during unpleasant weather. I am appreciative of my collegues in the Zoonotic Diseases Research Unit for assistance, advice and positivity. I especially thank Drs. Steven Olsen and Ed Steadham for assistance with experimental design, methods, statistical analysis and interpretation. The quality photographic reproductions are due to the patient help of Gene Hedburg, Tom Glasson and Chuck Greiner. I am appreciative of there advice, professionalism and pride in their work. Lastly, I am grateful to all my collegues and associates at NADC and ISU who have morally supported me and positively influenced me throughout the course of these studies. A special thanks is extended to Drs. Dominique Breese, John Vahle, David Hutto, Tom Gidlewski, Amanda Fales-Williams, Lani Vincent, Janice Miller, David Alt, Susan Brockmier, Carole Bolin and to Diana Whipple, Rebecca Lyons and Nicolle HeallyLehman.