Experimental Eimeria bovis infection in calves : cellular changes in peripheral blood and lymphoid tissues by Alwi Muhammad Shatry A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Veterinary Science Montana State University © Copyright by Alwi Muhammad Shatry (1991) Abstract: The frequencies of peripheral blood (PBL) and lymphoid tissue T and B cell subpopulations in normal and Eimeria bovis-challenged calves were determined by flow cytometry and the tissue distribution of these cells studied by immunoperoxidase staining. Increases in cells of the BoT4 phenotype were observed in both circulating and mesenteric lymph node (MLN) cells to account for total T cell increases. IgM and IgG1 cells were increased in PBL but no significant increases were observed for any antibody isotype in the tissues examined. Reduced expression of surface BoT8 in PBL of infected calves paralleled similar changes in the MLN spleen and gut lymphoid tissues. Results of in vitro PBL activation with mitogens indicated differences in responses between BoT4 and BoTB. Expression of the latter was markedly reduced after incubation with Con A, PHA and PMA, the magnitude of decreased expression being higher in cultures from non-infected calves. Together, these results emphasize differential lymphocyte subset alterations in the different lymphoid compartments and suggest that the BoT8 subset surface alterations may constitute a significant part of this subpopulation's response to E. bovis infection. Elevated sporozoite specific serum isotype levels, in particular IgM and IgG1, were consistent with increased frequencies of PBL's bearing these isotypes, suggesting that the increases may be related to higher frequencies of antigen-specific cell-surface isotypes. Analysis of sporozoite specific antibody isotype activity in culture supernatant fluid of lymphocytes of different tissue origins identified the MLN as the most active site of specific antibody synthesis. In addition, the differential antigen recognition profiles of serum isotypes suggested the preferential generation of antibody clonotypes, and may, in turn, have implications for the identification of immunodominant antigens. The ex vivo binding of sporozoites to various tissue sections revealed preferential attachment to gut tissue and that gut epithelial binding was higher than in the subepithelial microenvironment. These findings were consistent with observations of a modified blotting procedure in which biotinylated enterocyte protein extracts exhibited preferential binding to sporozoites and merozoite antigens over thymic extracts.  EXPERIMENTAL EIMERIA BOVIS INFECTION IN CALVES: CELLULAR CHANGES IN PERIPHERAL BLOOD AND LYMPHOID TISSUES by Alwi Muhammad Shatry A thesis submitted in partial fulfillment of the requirements for the degree Doctor of Philosophy in Veterinary Science MONTANA STATE UNIVERSITY Bozeman, Montana May, 1991 ii D 3 1 & APPROVAL of a thesis submitted by Alwi Muhammad Shatry This thesis has been read by each member of the thesis committee and has been found to be satisfactory regarding content, English usage, format, citations, bibliographic style, and consistency, and is ready for submission to the College of Graduate Studies. /aay /4 mi Date LLChairper^bn, Graduate Committee Approved for the Major Department / W /V; W / Date Head, Magbr Department Approved for the College of Graduate Studies Zr ̂ Graduate Dean iii STATEMENT OF PERMISSION TO USE In presenting this thesis in partial fulfillment of the requirements for a doctoral degree at Montana State University, I .agree that the Library shall make it available to borrowers under the.rules of the Library. I further agree that copying of this thesis is allowable only for scholarly purposes, consistent with "fair use" as prescribed in the U . S . Copyright Law. Requests for extensive copying or reproduction of this Thesis should be referred to University Microfilms International, 300 North Zeeb Road, Ann Arbor, Michigan 48106, to whom I have granted the "exclusive right to reproduce and distribute copies of the dissertation in and from microfilm and the right to reproduce and distribute by abstract in any format." Signature AAgj /V-'" /99/Date iv ' ACKNOWLEDGEMENTS I wish to express my sincere appreciation and thanks to my major advisor. Dr. C. A. Speer for accepting my candidacy under difficult circumstances, for his support, encouragement and guidance during the course of my studies. I am also grateful to the members of my Graduate Committee, Dr. D. E . Burgess, Dr. M. White, Dr. J. Cory, Dr. J . Berardinelli and Dr. J. Conant, for their encouragement and helpful suggestions. To Dr. M. A. Jutila, I am grateful for his guidance and encouragement, especially in flow cytometry, ex vivo parasite binding and immunohistology. I am grateful to Dr. I . Morrison of ILRAD, Nairobi, Kenya for T cell-specific monoclonal antibodies used in this study. The World Health Organization is gratefully acknowledged for providing the initial two year Fellowship award. The excellent technical assistance and friendship of Diane Welty, Andy Blixt and Sandy Kurk are deeply appreciated. I am also grateful to Gayle Callis1s assistance with tissue sections for the immunohistochemistry studies. Tim Clark1s assistance with computers was timely, given my own computer semi-literacy. I am grateful for the patience and help of Joan Haynes, Mary Herman, Linda Rees, and Charlotte McMilin. I dedicate this dissertation to my children, Ageel, Nafisa, Aisha, and my mother, Khadija. V TABLE OF CONTENTS Bage 1. LITERATURE REVIEW AND BACKGROUND Background..................... Surface Host-Parasite Interactions...... — ...... Phenotypic Characteristics of Bovine Lymphocytes.. Immunity— General................................... Antibody Responses.................................. T-Cell Mediated Immunity. ........................... Rationale............................ ........ ..... . Specific Aims....................................... 2. THE FREQUENCY OF CIRCULATING T AND B CELL SUBPOPULATIONS IN CALVES CHALLENGED WITH EIMERIA BOVIS.............. .......... .............. Introduqtion. . ................ ....................... Materials and Methods...................... . — Experimental Animals.......................... Soluble Sporozoite Antigen Preparation...... Aptibqdiee.... ........ • • ................ ; ' 'Fluorescence Activated Cell Sorter Analysis.. In Vitro PBM Activation.. .................... . Enzyme Treatment............................... Results...... ........ .............................. Frequency of T-Cell Subpopulations........... Cell-Surface Immunoglobulin Isotypes......... Expression of BoT4 and BoTS in Activated Cells............................ Discussiqn.......................................... 3. THE FREQUENCY AND DISTRIBUTION OF B AND T LYMPHOCYTE SUBPOPULATIONS IN LYMPHQID TISSUES OF EIMERIA BOVIS-CHALLENGED CALVES..... ........ I 1 2 4 8 9 15 19 21 23 23 25 25 26 27 28 3 0 31 31 31 39 45 47 61 Introduction. 61 Materials and Methods............... . Experimental Animals and Infection.. Lymphocyte Isolation from the Spleen Flow Cytometry....................... Immunohistochemistfy................. 65 65 66 68 69 TABLE OF CONTENTS— -CONTINUED Results. ...... ...................... .......... ...... 70 The Frequency of T-Cells in the Spleen....... 70 Ixnmunohistochemical Localization of T-Cells.. 74 The Frequency of Ig Isotype-Bearing Cells... 79 Localization of Ig Isotype- Bearing Lymphocytes...... 79 Discussion......... 89 4. ANTIGEN-SPECIFIC ANTIBODY ISOTYPES IN CALVES CHALLENGED WITH EIMERIA BOVIS. . .................... 93 Introduction............... 93 Materials and Methods......................... 94 Experimental Infection..................... . . . 94 Cell Culture................................... 95 Enzyme-Linked Immunosorbent Assay............ 95 Immunoblotting of Parasite Antigens.......... 97 Results.............................................. 99 Sporozoite-Specific Ig Isotypes.......... 99 Immunoblotting........................... 103 Discussion....... 106 5. INTERACTION BETWEEN EIMERIA BOVIS AND BOVINE TISSUES........... H O Introduction........................ H o Materials and Methods........................ 112 Ex Vivo Parasite Binding Assay.............. 112. Immunoblotting Procedure.......... 113 Results..................... lie Discussion........ 124 REFERENCES CITED. .................. 126 vi 0 LIST OF TABLES Table Ibge 1. The frequency of circulating T cells, BoT4 and BoTS cells in calves.......................... 3 2 2. Comparative levels of surface expression of BoTS on circulating and tissue lymphocytes....... 33 3. The frequency of Ig-bearing cells in calves...... 40 4. The frequency of T cell populations in the ileum, mesenteric lymph node and spleen............ 75 5. The frequency of Ig-bearing cells in the ileum, ■ mesenteric lymph node and spleen............ SI 6. Sporozoite-specific serum antibody isotypes in calves challenged repeatedly with Ei bovis... 102 7. Sporozoite-specific antibody isotypes in supernatant fluid of pokeweed mitogen-activated cells.... 102 8 . Binding of Eimeria bovis sporozoites to different bovine tissues. ................................ 117 vii viii Figure EGcp I. Sequential alterations in the frequency of T cells calves challenged with E . bovis oocysts..... 34 LIST OF FIGURES 2. Sequential alterations in the frequency of circulating BoT4+ cells in calves challenged. 35 3. Sequential changes in the frequency of circulating BoTS cells in calves challenged.............. 36 4. Sequential cell-surface expression of BoTS in calves challenged...................... 37 A. Comparative fluorescence levels of BoTS+ cells in 3 control and infected calves...... 38 5. Sequential alterations in the frequency of circulating IttIgM+ cells in calves inoculated.. 41 6. Sequential alterations in the frequency of circulating mlgGl cells in calves inoculated.. 42 I . Sequential changes in the frequency of circulating mIgG2 cells in calves inoculated............. 43 8 . Sequential alterations in the frequency of circulating mlgA in calves inoculated........ 44 9. Cell-surface BoTS expression on circulating lymphocytes activated with mitogens. . ........ 48 10. Cell-surface BoTS expression on circulating lymphocytes activated with mitogens.......... 49 II. Cell-surface expression of BoTS on circulating lymphocytes after in vitro activation........ 50 12. Cell-surface expression of BoT4 on circulating lymphocytes after in vitro activation...... 51 13. Cell-surface expression of BoT4 on circulating lymphocytes after in vitro activation........ 52 14. Cell-surface expression of BoT4 on circulating lymphocytes after in vitro activation........ 53 15. Cell-surface expression of BoTS (A) and BoT4 (B) in control and infected calves........... 54 ix 16. Effect of chymotrypsin (A) and neutral protease (B) on cell-surface BoT4 and BoTS............ 55 17. Histogram profile overlays for the frequency of peripheral blood T and B cells. . ......... . . . 60 18. Histogram profiles for cell-surface expression of TS in different tissues......................... 73 v19. The distribution of BoT2+ cells in the terminal ileum.................. 76 20. BoT4+ cells in the terminal ileum are mostly located in the interfollicular region........ 77 21. BoTS+ cells in interfollicular areas.............. 78 22. A. , B. Localization of IgM+ cells in the dome and lamina propria................ 82 C. The distribution of IgM+ cells in the ileal mucosa........................ 8 3 23. IgGl+ cells in the dome and corona regions of the Peyer1 s patch.................................. 84 24. IgG2+ cells in the corona region and dome of a Peyer1 s patch................... 85 25. Localization of IgA+ cells in a dome and the lamina propria............................ 86 26. IgM+ cells in the mesenteric lymph node........... 87 27. IgGl+ cells in a secondary follicle in the deep cortex of the MLN.............................. 88 28. A.,B. Immunodetection of sporozoite and merozoite 104 antigens by isotype-specific serum antibodies. 105 29. Sporozoite binding to ileal mucosa................ 118 30. Adherent sporozoites on glandular and mucosal epithelial surfaces........................... 119 31. Sporozoite binding to spleen and thymus........... 12 0 32. E bovis sporozoites binding to renal tissue...... 121 33. Protein binding assay............................ 123 LIST OF FIGURES— CONTINUED. X ABSTRACT The frequencies of peripheral blood (PBL) and lymphoid tissue T and B cell subpopulations in normal and Eimeria boyis-challenged calves were determined by flow cytometry and the tissue distribution of these cells studied by immunoperoxidase staining. Increases in cells of the BoT4 phenotype were observed in both circulating and mesenteric lymph node (MLN) cells to account for total T cell increases. IgM and IgGl cells were increased in PBL but no significant increases were observed for any antibody isotype in the tissues examined. Reduced expression of surface B0T8 in PBL of infected calves paralleled similar changes in the MLN spleen and gut lymphoid tissues. Results of in vitro PBL activation with mitogens indicated differences in responses between BoT4 and B0T8 . Expression of the latter was markedly reduced after incubation with Con A, PHA and PMA, the magnitude of decreased expression being higher in cultures from non-infected calves. Together, these results emphasize differential lymphocyte subset alterations in the different lymphoid compartments and suggest that the BoTS subset surface alterations may constitute a significant part of this subpopulation's response to E. bovis infection. Elevated sporozoite specific serum isotype levels, in particular IgM and IgGl, were, consistent with increased frequencies of PBL1s bearing these isotypes, suggesting that the increases may be related to higher frequencies of antigen- specific cell-surface isotypes. Analysis of sporozoite specific antibody isotype activity in culture supernatant fluid of lymphocytes of different tissue origins identified the MLN as the most active site of specific ■ antibody synthesis. In addition, the differential antigen recognition profiles of serum isotypes suggested the preferential generation of antibody clonotypes, and may, in turn, have implications for the identification of immunodominant antigens. The ex vivo binding of sporozoites to various tissue sections revealed preferential attachment to gut tissue and that gut epithelial binding was higher than in the subepithelial microenvironment. These findings were consistent with observations of a modified blotting procedure in which biotinylated enterocyte protein extracts exhibited preferential binding to sporozoites and merozoite antigens over thymic extracts. I CHAPTER I LITERATURE REVIEW AND BACKGROUND Background Coccidiosis is a disease of various animals, including mammalian and avian hosts such as cattle, sheep, rabbits, turkeys and chickens. Coccidial infections generally result in hemorrhagic enteritis leading to diarrhea, dehydration, anemia, weight loss and in some cases death (I). The causative agents of bovine coccidiosis, which belong to the genus Eimeria. exhibit strong host specificity (2) . Eimeria bovis is regarded as the most frequent cause of bovine coccidiosis (3). In 1972, global economic losses were estimated to exceed $400 million annually (4) . To date, no vaccines or satisfactory prophylactic measures against bovine coccidiosis are available besides chemoprophylaxis. Eimeria bovis infections are initiated in susceptible hosts by ingestion of speculated oocysts. Oocysts each contain four sporocysts, each with two sporozoitqs. In the intestinal tract, the oocysts encounter carbon dioxide, trypsin and bile, causing sporozoite exeystation (5) . The sporozoites penetrate the intestinal epithelium and endothelial cells of the central lacteals (6). The sporozoites then undergo asexual reproduction (merogony, schizogony) to form first-generation merozoites. Fourteen to 2 fifteen days following the ingestion of sporulated oocystsz meronts reach maturity. First-generation merozoites then reach the large intestine and cecum where they penetrate glandular epithelial cells and develop to second-generation meronts with merozoites. Adjacent epithelial cells are invaded by the second-generation merozoites which differentiate into male and female gametocytes called micro- and macrogamonts respectively. Microgametes subsequently penetrate adjacent cells harboring macrogamonts (6, 7), where fertilization presumably occurs. Each zygote develops into an oocyst by forming an oocyst wall around itself resulting in destruction of the host cell and oocysts are discharged into the lumena of the cecal and large intestine and excretion in the feces as unsporulated oocysts (8), Oocyst sporulation occurs upon exposure to atmospheric oxygen. Sporulated oocysts are infective to a new susceptible host. Lysis of epithelial cells in the large intestine is caused by oocysts, resulting in hemorrhagic enteritis (9). Surface Host-Parasite Interactions A monoclonal antibody specific for a 20 kilodalton (kDa) sporozoite surface protein (p20) inhibits sporozoite penetration of Madin-Darby bovine kidney cells and a monocyte cell line (13). P20 has been shown to be a immunodominant surface antigen (14). These findings indicate a potential role of p20 in sporozoite penetration, and may be involved in 3 initial surface interactions with monocytes and vascular endothelial cells. The immunodominance of this molecule suggests its high immunogenic potential. The binding of p20 by circulating IgG from immune calves may be indicative of a possible mechanism of the expression of acquired humoral resistance, in blocking early surface interactions between sporozoites and host cells in challenge infections. The binding of parasite-specific IgA and IgG on the surface of sporozoites and merozoites has been demonstrated by immunoelectron microscopy (15). The association of the former isotype with structural damage on the sporozoite surface (16) lends support for sporozoite-targeted, antibody-mediated effector mechanisms resulting from host-parasite surface interactions. In other protozoal systems, immunoblotting techniques have been adapted to the study of molecules involved in interactions between parasites and mammalian host cells (17). This approach has led to the simultaneous identification of Trypanosoma cruzi and host cell molecules involved in the binding interactions. In Leishmania spp., a major glycoprotein (gp63) on the surface of promastigotes (18) and a lipophosphoglycan (LPG; 19) have been shown to mediate parasite binding to and uptake by macrophages. Both.molecules serve as ligands for various macrophage receptors among which the complement receptors CRl and CR3 appear to be most important (20, 21). Immunization against promastigote LPG and 4 gp63 peptides confer protection to mice challenged with infective promastigotes (22, 23). The identification of molecules that participate in parasite attachment to and uptake by host cells has, therefore, led to a better understanding of parasite evasion strategies (24) and novel molecular approaches to parasite vaccine design. Phenotypic Characteristics of Bovine Peripheral Blood and Tissue Lymphocytes . The phenotypic and functional charateristics of bovine peripheral blood mononuclear cells (PBM) have been extensively studied. Typically, monocytes comprise 5-20% of PBM isolated by density gradient centrifugation. The detection of monocytes is based on staining with cytoplasmic a-naphthyl esterase or monocyte specific antibodies (25). Lymphocytes constitute the bulk of remaining PBM cells. The generation of monoclonal antibodies (MAb) specific for bovine lymphocyte/ subpopulation determinants, in conjunction with flow cytometry, immunofluorescence and immunohistochemical techniques, have facilitated phenotypic and functional analysis of PBM and tissue lymphocyte subsets. Variable estimates on the frequency of lymphocyte subpopulations in PBL and tissues of healthy and infected cattle have been obtained. The frequency of T cells hes been estimated as ranging between 20 and 70% of PBM (25-29) based 5 on detection by MAb specific for T cell markers equivalent to the human CD2 (28), CD3 or CDS (25). T cells, similarly, constitute 60-70% of peripheral lymph nodes (25, 26). The frequency of. IgM+ cells appears to be the principal one reported in bovine PBM and peripheral lymphoid tissues, the frequency of which ranges between 4 and 30 -s (27, 31). This closely approximates the relative frequency of circulating B cells, as revealed by cells positive for surface immunoglobulin (25, 26, 29). The percentage of B cells expressing other (IgGl and IgG2) surface isotypes is reported to be "low" (25, 31) but supporting data are not available. The availability of B cell isotype-specific MAb have not thus far been applied to the study of lymphocyte differentiation. Available data indicate considerable variation in the frequency of given lymphocyte subsets within and between the different lymphoid compartments. Dramatic differences in subset distribution have been documented in afferent and efferent lymphatics of sheep, a phenomenon attributed to differential migration patterns among lymphocyte subpopulations (32, 115). In contrast to PBM and peripheral lymphoid organs, quantitative and immunohistochemical studies on lymphocyte subsets in the bovine intestinal tissues have received less attention. A recent study determined the frequency of B cells, T cells, T helper (BoT4) and T cytotoxic (BoT8) subsets (33). Percentages of the various subpopulations were 6 determined in intraepithelial, lamina proprial and Payer's patch lymphoid cell suspensions using flow" cytometry (33) . The data revealed that the sum of percentages of BoT4 and B0T8 subpopulations were in considerable excess over those obtained for total T cell populations. This would suggest the T cell- specific MAb may not have recognized target molecules on the T cell subsets. Conversely, the subsets may not all be expressing markers recognized by the putative T cell MAb, or the T4 and/or T8 reagents may recognize some non-T cells. No studies on tissue localization of T cell subsets in gut-associated lymphoid tissue (GALT) of the bovine have been conducted. Earlier studies focused attention on the distribution of B cells bearing the surface immunoglobulins, IgM, IgGl, IgG2 and IgA (34-37). Isotype-specific polyclonal antisera and immunofluorescence (34, 35, 37) were mostly used to determine the distribution of surface Ig isotypes. In one study (36), immunoperoxidase staining was applied but the results of all these studies were equivocal. The predominant surface isotypes reported in young calves were IgA and IgM, localized in the lamina propria and the intercryptal region (34). In calves 4 days to 24 months old, IgG2-bearing cells, which were less frequent than the former two isotypes, exceeded the relative frequency of IgGl+ cells. In the same Study, many cells reported to exhibit membrane IgGl and IgG2- staining were excluded from the count in favor of cells with intense cytoplasmic staining, which may reflect terminally 7 differentiated plasma cells. Further, the possible influence of elevated levels of circulating IgGl in colostrum-fed calves (38) on the frequency of surface IgGl+ cells in lymphoid tissues is not known. In contrast, IgGl was reported to be the predominant surface isotype expressed on GALT lymphoid cells in heifers 12 to 30 months old (35, 36) , the cells exhibiting similar tissue distribution patterns to those described for IgA and IgM. In view of the crucial importance of ileal Peyer1 s patch in the generation and export of B cells in sheep (39) and its complete involution with age (40) the effect of age on the frequency of surface isotypes remains to be determined. Additionally, the possibility of nonspecific fluorescence or binding of first- and second-step reagents via the Fe receptor (33) was not addressed. Consequently, the contribution of these phenomena to the frequencies observed in these studies remains unclear. Much of our present state of knowledge on ruminant B cell development comes from studies on lymphocyte migration in sheep. The investigations have led to consideration of the Peyer's patch as a central lymphoid organ for B cell development in sheep (36). The extracorporeal perfusion of isolated segments of the ileum, and the inclusion of fluorescein isothiocyanate (FITC) in the perfusate facilitates labeling of Peyer's patch lymphocytes. The presence of labeled cells in distant lymphoid organs revealed by 8 fluorescence microscopy," suggested substantial seeding of peripheral lymphoid organs by cells from the Peyef1s patch (39, 40). The high turnover of cells in the Peyer's patch (41), the high death rate of Peyer's patch lymphocytes (42) and the severe B cell deficiency in ileectomised prenatal and neonatal lambs (43) are taken as further evidence for the central role of Peyer's patch in the generation of B cells in ruminants. Immunity-General Immune responses to parasitic organisms are fundamentally polyclonal in nature leading to the generation of diverse populations of clonotypes, populations of effector cells and molecules with different consequences for host-parasite inter­ actions. While the aberrant immunological phenomena (immuno- depression, autoimmunity) associated with chronic infections with continuously replicating protozoal agents fLeishmania, Trypanosoma cruzi) do not appear significant in mammalian eimerid infections, at least not in E . bovis (44), host re­ lated factors (45), including immune status (46), are important in determining the nature of immune responses and disease pathogenesis. Better understanding of the immunobiology of host- parasite relationships requires analysis of functional inter­ actions between homogeneous populations of antigen-reactive cells and parasite populations. Non-availability of the 9 latter for any stage in the life cycle of coccidian parasites constitutes a major constraint in this regard. This problem should be circumvented by using molecular approaches in identifying antigens involved in the relevant interactions and cloning genes encoding them. Functional studies utilizing cloned populations of antigen-sensitive T-Iymphocytes, although limited in coccidian infections (47, 48), have yielded valuable information on T cell subset function. Antibody Responses Most Eimeria species appear, to be immunogenic and capable of inducing varying degrees of resistance to reinfection (10) . Immunity to challenge infection is manifested as reduced clinical severity and oocyst production; it is generally species-specific (11). Intraspecies immunological variation in the coccidia has been documented (12) and appears to induce partial cross-protection. A parasite-specific IgG response against first-generation merozoites has been demonstrated using indirect fluorescent antibody (49). The response is *first detectable 2 weeks following oral inoculation with 10 oocysts, peaks at about 21 days (44, 49) and is sustained for up to 98 days (49). In vitro complement-mediated lysis of Eimeria sporozoites arid merozoites occurs in the presence of parasite-specific IgG (50). Similarly, a sporozoite-specific IgG antibody was capable of in vitro sporozoite agglutination, 10 complement-mediated lysis and passive protection against challenge infection with E. tenella (51). The protective role of circulating antibody remains unresolved but appears to be of relatively minor importance. Past studies on the protective role of antibodies, largely relying on the passive transfer of serum or globulin fractions in mice and rats gave conflicting results (I) . Specific antibody titers could not be correlated with reduced oocyst production in immune L3T4 cell-reconstituted mice infected with E. vermiformis (52). Enrichment for antigen-specific antibody and evaluation of different antibody isotypes have not received adequate attention. Quantitative disparities in specific antibody may partially explain the reported variability of passively transferred immune sera in their protective effects. Failure of antigen-specific circulating antibody to accumulate at the sites of parasite development may further account for the apparent ineffectiveness of the passively transferred sera in conferring protection (53). The increased susceptibility to primary E . vermiformis infection in mouse strains with lowered or defective antibody production is not accompanied by failure to develop immunity to reinfection (54). The demonstration of antigen-specific secretory IgA in mice (16) and rats (55) and the ability of gut contents from immunized chicks to confer some measure of passive protection in avian coccidiosis (56), lends support for a protective role 11 fox secretory antibody. Further, secretory antibody-related ultrastructural damage and reduced motility of E . falciformis sporozoites was observed after incubation with enterocyte- associated mucus from immunized mice (16). A sporozoite-specific mouse monoclonal antibody (MAb) which recognizes an immunodominant 20 kDa protein (14) on the surface of E . bovis. inhibited sporozoite penetration of cultured bovine monocytes (13), suggesting a potential role for humoral antibody in modulating surface hostcell-parasite interactions, similar to a mechanism that had been proposed earlier for parasite-specific IgA (11). Similarly, sporozoites of E . tenella treated with a specific MAb failed to infect naive chickens, and ammonium sulphate-precipitated asciteg fluid injected intraperitonealIy protected recipient chicks against challenge infection (57). Of particular interest is the efficacy of a parenterally administered parasite-specific antibody of an IgG subclass, which might not be expected to reach high concentrations at mucosal sites. Inflammation-related leakage of circulating antibody has been proposed as a possible means of achieving high concentrations at sites of parasite infection (9, 58). Additionally, the tissue distribution of a passively transferred MAb in an unrelated recipient host may be altered. The marked increase in IgA-containing lymphocytes in the lamina propria but not mesenteric lymph nodes (MEN) of E . falciformis-immune mice suggests increased probability of 12 contact between specific secretory IgA and parasites at their development sites (59). In adoptive transfer studies (60)> MLN cells from immunized mice were superior to spleen cells in their ability to confer protection. It is tempting to speculate on the potential contribution of a higher frequency of secretory IgA precursors in MLN than in spleen cell suspensions. The precursors would, presumably, subsequently lodge in the lamina propria of challenged recipients and terminally differentiate into IgA-secreting plasma cells. In this regard, actively dividing lymphocytes conferred protection against E . yermiformis in mice while resting cells or those treated with the mitotic inhibitor, vinblastine, did not (60). This observation raises the possibility of the presence of IgA precursors among the MLN donor cell population, which, upon encountering antigen in recipient mice, are induced to recirculate and preferentially localize at sites of terminal differentiation in the lamina propria (61) . Adoptive transfer studies aimed at the critical evaluation of the protective role for E-cells should, ideally, control for surface isotype and frequency of antigen- specific cells in donor tissues. In addition, the potential contribution of antigen presentation by specific B-cells in the activation of T-cells (62) merits attention in the analysis of humoral protective responses to eimerid i n f e c t i o n s ’ 13 Analysis of E . bovis sporozoite and merozoite antigens (63, 64) revealed subtle differences: I) between merozoites obtained from infected animals and those obtained in vitro from sporozoiter-infected cultured bovine monocytes (ref. 63), 2) in the sporozoite antigen recognition profiles of specific serum IgG (obtained by differing immunizing protocols) in immunoblots, 3) recognition patterns of sera from different calves (14, . 65) . Differences in serum binding profiles reflect diverse repertoires of antibody clonotype specificities^ and the identification of immunodominant antigens require screening of larger numbers of immune sera that would be more representative of the genetic pool. Studies with polyclonal sera and MAbs revealed antigens unique to sporozoites and merozoites. Identification of stage-specific molecules should be valuable in the purification of potentially protective epitopes especially in view of the inhibitory effects of a 20 kDa specific MAb on sporozoite penetration of cultured bovine monocytes (14). In other coccidia, such as infection with Cryptosporidium spp. , as in the case with Eimeria spp, studies on passive transfer of immunity yielded mixed results. Immune bovine serum (65, 66), colostrum and MAbs (66) neutralized the infectivity of C. oarvum sporozoites for neonatal mice and partially protected them against infection with C. parvum oocysts. The oral administration of MAbs to mice had no effect on susceptibility to infection but significantly 14 lowered parasite burdens (66, 68). The MAbs, belonging to IgG3 and IgM isotypes, recognized nonprotein and protein epitopes, respectively (67). Neonatal calves fed hyperimmune bovine colostrum, with high concentrations of parasite- specific IgG, IgM and IgA were partially protected against Cryptosporidium infection (69). While the high concentrations of parasite-specific antibody isotypes may, in part, explain the partial protective effect, the contribution of other biologically active subtances in colostrum (i.e. cytokines, complement) had not been determined. Parasitological and clinical cure of cryptosporidiosis has also been reported in immunodeficient patients treated with bovine hyperimmune colostrum (70, 71) . In contrast, passive colostral protection could not be demonstrated in suckling mice (72) or calves (73). Similarly, the administration, of bovine colostrum failed to influence the course of cryptosporidiosis in immunocompromised patients (74). The problem of evaluating the protective role of humoral immunity is further compounded by the demonstration of parasite-specific serum IgM and IgG in immunocompromised subjects (75), suggesting that the eliciting of a humoral response may be insufficient in the expression of protective anti-cryptosporidial effects. Increased IgA and IgE responses to infection have also been documented (76), the former being considered important in clearing the infection (77). 15 The roles of circulating versus secretory antibodies in the immune responses to coccidial infections may be difficult to delineate. While both humoral components appear to possess in vivo and in vitro parasite modulatory activities, concentration of secretory immunoglobulins at the site of infection in the intestinal tract favors a more important role for them. It is conceivable, however, that the lytic, opsonic or cytophilie effects of circulating antibody may have a greater impact on the development of infection in circumstances which permit contact between such antibodies and the extracellular, invasive stages, such as changes in vascular permeability that occur within a short time of parasite challenge (102). Additionally, the ability of other isotypes (IgM) to complex with the secretory component in the bovine (36) suggests a potential protective role of this class of antibody on mucous surfaces. T-cell Mediated Immunity Several lines of evidence have led to the general conclusion that cell-mediated immunity may be more important than antibody in immune responses to Eimeria species: the induction of delayed hypersensitivity has been demonstrated using different Eimeria antigens in rabbits (103), chickens (104), and calves (78). Similarly, studies with E . bovis have demonstrated antigen-specific blastogenesis and the protective effects of dialyzable transfer factor (TE) from immune calf 16 lymph nodes (78) . Studies in athymic mice (79, 80) . and rats (81, 82) subsequently established the critical role of T-cells in acquired immunity to experimental infections with eimerid parasites. Athymic (nu/nu) rats passed 3 times more E . nieschulzi oocysts than did heterozygous (nu/+) controls. In contrast to mouse strains with B-cell-related defects, T-cell deficient mice are completely susceptible to reinfection (80). Furthermore, differences in responses between susceptible (C57BL/6) and resistant (BALB/c) strains were evident during primary but not subsequent infections (80). Findings in recombinant inbred strains (BALB/c X C57BL/6), leading to the conclusion that resistance to E . vermiformis was not associated exclusively with the H-2 locus (80), were later confirmed in experiments utilizing congenic strains infected with E. falciformis (83), suggesting a potential rolg in acquired resistance, o f hitherto undefined non H-2 genes. Effectors of DH in murine eimerid infection belong to the L3T4 (CD4+) subset in mice (84). Transfer of DH by spleen cells from E . falciformis-recovered donor mice was abrogated by depletion of CD4+ T-cells. Spleen cells from acutely infected mice suppressed the DH mediated by immune cells, suggesting, the transient generation of a suppressor subset, probably of the CD4+ phenotype, also observed in leishmaniasis (85) . Abrogation of protective effects was noted in this (84) and another study (86) in experimental infection, with E . 17 vermiformis following depletion of adoptively transferred CD4+ cells. The latter study also demonstrated that lowered resistance to primary infection was greater in in vivo CD4+- than in CD8+-depleted mice. Supernatants of Con A- and antigen-activated peripheral blood lymphocytes are capable of inhibiting the development of E . bovls sporozoites in cultured bovine monocytes (87, 88) suggesting the participation of nonspecific and specific mechanisms of intracellular parasite elimination. The production of gamma interferon (IFN7.) by a T-cell clone. (47) suggests the generation by antigen of a functional subset probably equivalent to the T-helperl (89) . Phenotypic identity of the clone was not, however, established. IFN7. has been proposed as the soluble mediator inducing the inhibitory effects on sporozoite development (90). In vivo treatment of BALB/c mice with an IFN7-Specific MAb resulted in enhanced E. vermiformis infection, the effects of which waned when the MAb was administered between 4 and 7 days postinfection (91), indicating involvement of IFN7 in the control of primary infection but not the expression of immunity to reinfection. This observation would predict that failure of athymic nude mice to control primary infection may be related, at least in part, to deficiency of this mediator. Further support for the role of IFN7 in infection with Eimeria spp. comes from a study 't }. r •(92) in whiqh IFN7 titers in calves were elevated after primary but not challenge infection with E. bovis. This is 18 not surprising given the increased severity of infection in mice depleted of CD4+ cells, a subset of which synthesizes IFNr (89). Taken together, these observations suggest potential roles for both humoral and cell-mediated mechanisms in immunity to Eimeria species, in a manner that need not be mutually exclusive. The predominance of IgA precursor cells in Peyer1s patches and other gut-associated lymphoid tissues (GALT, 102) and the presence of other surface immunoglobulin­ bearing lymphocytes may have crucial antigen presenting functions for GALT T-helper cells (62). Activation of a subset of these T cells by the antigen-presenting B cells could lead to their elaborating lymphokines which enhance microbicidal and parasite modulatory capabilities of macrophages. Further, there may be temporal and transient patterns to the responses unique to each lymphocyte subpopulation at the infection site during progression of infection. In conclusion, the biological complexity of coccidian parasites, the host responses they elicit and the effector mechanisms involved, can not be adequately explained based on our present state of knowledge of host-parasite relationships. The role of factors not directly linked to immune response genes, which may also have a bearing on the disease outcome, and how the products of these genes subsequently interact with the immune regulatory networks leading to protective immune 19 expression is not known. The basis of age-related immunity, a feature of coccidian and other unrelated parasites, remains largely undefined. Similarly, host immune responses elicited by complex parasite molecules of varying biochemical compositions, each with widely differing immunogenic potential will, in all probability, within the available repertoire of responses, include many that are redundant having little to no direct impact on parasite elimination (93). Rationale . Differential patterns of tissue distribution among different subsets of lymphocytes is a well recognized phenomenon and may have implications for lymphocyte function (94) . B cells of blood, splenic and lymph node origin were found to have different requirements for proliferation and differentiation into Ig-secreting cells (95). Cowan I Staphylococcus aureus-stimulated blood and splenic B cells secreted immunoglobulin in the presence of recombinant IL-2 alone, while drainage lymph node B cells failed to elicit plaque-forming cells in response to the subcutaneous administration of the thymus-independent antigen TNP-Ficoll (96) . Antigen-presenting cells and con A-stimulated T cells from the spleen induced the secretion of only IgM whereas identical cell populations from Payer's patches induced high ■levels of TgA secretion and intermediate levels' of IgM and IgG 20 (97) . This implied diverse differentiation pathways of accessory cell populations of different tissue origin. Both qualitative and quantitative differences have been described for antigens capable of eliciting systemic and local mucosal response after oral administration (98). Oral immunization required amounts of antigen far in excess of those required for parenteral administration of systemic immunity. Intestinal and systemic responses were elicited by the oral administration of small quantities of Escherichia coli pili but not by identical amounts of bovine serum albumin. Proteins capable of oral immunization possess "lectin or lectin^like" binding activities while proteins that are unable to elicit oral immunization do not (98). These observations indicate that the functional diversity of lymphocyte populations may have a site or tissue-related basis, further implying site-related phenotypic diversity. Further, the differential distribution of different immunoglobulin isotypes in B cells of normal BALB/c mice and the subsequent quantitative changes in these cells in the lamina propria and mesenteric lymph nodes following E.. falciformis infection (59), reflects potential similarities tp the above experimental models. The concentration of IgA+ cells at the apical part of the lamina propria could facilitate contact between parasite-specific secretory IgA with parasites during their extracellular and intracellular 21 phases, thereby inhibiting their penetration and/or development. Studies focusing on qualitative and quantitative characteristics of lymphoid cells.in Peyer1s patches and MLNs have not been conducted in experimental bovine coccidiosis. The present study examined the frequencies and distribution of subsets of T and B cells in the circulation, Peyer1 s patches, MLN and SPL in calves challenged with E . bpvis, using flow cytometric and immunohistochemical techniques. Such information may provide insight on the comparative development of local responses versus responses distal to the parasite development site and the extent to which changes in peripheral circulating lymphocytes reflect the pattern of events at the local tissue level. Elucidation of these responses in Peyer1s patches and MLNs may partially explain the failure of parenteral immunizations using oocyst, sporozoite or merozoite antigens to confer protection against challenge infection (99, 100), or the partial to no protection afforded by the passive transfer of immune serum (101). Specific Aims I. Determine the frequency of T and B cell subsets in the peripheral blood of normal and E . bpyis-chailenged calves using flow cytometry. . 5; 22 2. Determine the frequency of T cell subsets and B cells bearing surface immunoglobulin isotypes (IgM, IgGl, IgG2 and IgA) in the ileum, MLN, and SPL of E. bqvis-challenged calves using flow cytometry. 3. Study the distribution of T and B cell subpopulations in the Peyer1s patch and in MLN using immunoperoxidase staining. 4. Compare the levels of sporozoite specific antibody isotypes in serum and ,PWM-activated culture supernatants and to determine antigen recognition profiles of serum antibody isotypes. 5. Determine the comparative tissue-binding characteristics of sporozoites using an ex vivo binding assay. 23 CHAPTER 2 THE FREQUENCY OF CIRCULATING T AND B CELL - SUBPOPULATIONS IN CALVES CHALLENGED WITH EIMERIA BOVIS Introduction Infection with Eimeria spp, a gut-associated coccidian parasite, results in partial to complete protection against homologous challenge (I) . Experimental infection with E . bovis, the principal causative agent of bovine coccidiosis, induces humoral (44, 49) and cell-mediated responses (44, 47, 78). Previous studies in this regard have focused on measurements of in vitro and in vivo correlates of these responses, namely, antigen-specific blast transformation (47, 78) , delayed type hypersensitivity (78) and serum antibody titers using indirect immunofluorescence (44, 49) or enzyme- linked immunosorbent assay (ELISA, ref. 106). Serum antibody assays have, in addition, primarily relied on the demonstration of parasite-specific total IgG during experimental infection. Antigen-specific responses by other antibody isotypes have not been characterized in bovine coccidiosis. Earlier studies on the identification of bovine lymphoid cells largely relied on the demonstration of their rosette­ forming or lectin-binding properties. For instance, T cells were identified on the basis of their capacity to bind 24 fluorochrome-labeled peanut agglutinin (PNA; 107). However, in addition to binding T lymphocytes, PNA also binds non­ lymphoid cells (25). Availability of monoclonal antibodies specific for bovine T lymphocyte subsets (5, 25, 29) and immunoglobulin isotypes (108, 109) has facilitated studies on the frequencies of lymphocyte subpopulations in the peripheral blood and tissues of normal cows and should permit studies on lymphocyte differentiation. Studies on alterations in circulating lymphocyte subpopulations in cattle experimentally infected with bluetongue ( H O ) , trypanosomiasis (111) and in mastitic cows have been made possible by subset-specific MAb. Investigations aimed at characterizing changes in circulating lymphocyte subsets should provide additional insights into cellular responses to infectious agents and should complement functional studies attempting to elucidate immune mechanisms associated with such infections. In vitro antigenic and antigen-independent stimulation of human T cells revealed the down regulation of CD3 and CD4 on human T cell clones (112). Similarly, blood forms of Trypanosoma cruzi cocultured with human peripheral blood mononuclear cells led to a marked decrease in the surface expression of CD3, CD4 and CDS on PHA-activatied cells (113) . Failure to demonstrate cytotoxic T lymphocyte activity in diabetes-prone Biobreeding rats has been associated, in part, to markedly reduced expression of cell-surface CDS. These 25 results suggest that the density of accessory molecules on T I cell surfaces are subject to antigen-specific and nonspecific modulation with functional and regulatory implications. This study was undertaken to characterize, sequentially, the frequency of circulating T and B lymphocyte subpopulations in calves receiving multiple inocula of E. bovis oocysts. In addition, the mode fluorescence obtained from a fluorescence activated cell sorter (FACS) was used as a measure of in vivo surface expression (114) of these molecules in E . bovis- challenged calves. The study also examined the in vitro effects of lymphocyte activation on the expression of CD4 and CDS. Materials and Methods Experimental Animals One week-old Holstein bull calves were purchased from the Bozeman Livestock Auction. The calves were confined to calf pens with slatted floors in isolation facilities, Department of Veterinary Molecular Biology, for 3 to 4 weeks prior to parasite inoculation. The holding facilities were thoroughly cleaned and disinfected prior to the arrival of newly purchased calves. The diet comprised primarily of a commercial milk replacer fed twice daily. The milk replacer was withheld and oral electrolytes administered to calves developing scours prior to infection with Eimeria bovis. At O 26 termination of the experiment, the calves were sacrificed by stunning and bleeding and then prepared for the aseptic collection of appropriate tissues. The strain of E . bovis used in this study was originally isolated by Dr. D. M. Hammond (Utah, Dr. Speer, personal communication) and subsequently maintained by serial passage in outbred Holstein-Freisian calves. Primary infection was established by the oral administration of 4 X IO4 sporulated oocysts of E. bovis in 5 ml physiological saline. To ensure patency of the infection, fecal samples were processed for oocyst collection at 18-22 days after inoculation. Age and sex-matched control calves not receiving the inoculum were confined to the isolation facilities during the entire experimental period. Eight weeks following the establishment of the primary infection, a challenge inoculum of IO5 oocysts was administered orally at 14-day intervals on three occasions. Venous blood was collected weekly in tubes containing heparin to a final concentration of 10 IU sodium heparin per ml blood. The anticoagulated blood was processed for flow cytometry. Soluble Sporozoite Antigen Preparation Eimeria bovis oocysts were separated from calf feces by sugar flotation, sporulated, and then stored in aqueous 2.5% K2Cr2O7 at 40C until further use. The oocysts were further purified from contaminating debris by repeated washing in 27 Hank's Balanced Salt Solution (HBSS), pH 7.2 to remove the K2Gr2O7, resuspending them twice in sodium hypochlorite (Clorox) for 30-60 min and then harvesting the oocyst-rich supernatant. The Clorox was then removed by several washes in HBSS, after which the oocysts were resuspended in HBSS and broken by grinding in a motor-driven Teflon-coated tissue grinder. The resultant sporocyst suspension was then pelleted and sporozoites excysted by resuspension in RPMI 1640 containing 0.25% w/.v trypsin (Gibco Laboratories) and 0.75% w/v sodium taurocholate (Sigma Chemical Co) and incubated in a 37°C water bath for 60-90 minutes (87, 117). Sporozoites were purified by passage over a nylon wool column, enumerated and pelleted. The pelleted sporozoites were then lysed in IOOjul sterile distilled water, subjected to 5 cycles of freeze-thawing and resuspended in CRPMI to a concentration equivalent to 2 X IO6 sporozoites ml'1. The supernatant was filter sterilized and used as soluble sporozoite antigen. Antibodies Monoclonal antibodies (MAb) specific for the bovine equivalents of human CD2 (BoT2, IL-A42), CD4 (BoT4, IL-A12) and CD8 (BoT8 , IL-A 51) were the kindly provided by Dr, W. I. Morrison, ILRAD, Nairobi, Kenya. The MAb, used as mouse ascites, belong to the IgG2a (IL-A12, IL-Al2) and IgGl 28 isotypes; they were all used at a final concentration of 1/4000 for flow cytometry (Morrison, personal communication). Bovine immunoglobulin isotype-specific MAb (108) were purchased from Ultimate Conceptions (Millers Falls, MA) . These antibodies are specific for heavy chains of bovine IgM (DAS 6, mouse ascites), IgG1 (DAS 17, culture supernatant), IgG2 (DAS 2) and IgA (DAS 7, mouse ascites). All the isotype- specific MAb belong to the IgG1 subclass. The MAb DREG 55, specific for the human homing receptor (116), was the kind gift of Dr. M. Jutila; this was used as an isotype control. Fluorescence Activated Cell Sorter (FACS) Analysis Peripheral blood mononuclear (PBM) cells were isolated by density gradient centrifugation described by Julius et al. (115). Briefly, the heparinized blood was mixed with an equal volume of cold calcium and magnesium-free Hanks' Balanced Salt Solution (HBSS, pH 7.2) containing 5mM sodium EDTA (Sigma Chemical Co., St. Louis, Mo). Twenty five ml of the mixture were layered over 15ml Ficoll-Hypaque (Histopaque 1077; Sigma) and spun at 3 50 X g for 45 min at 4°C. The PBM-rich interphase was removed, pelleted and treated with 0.16M NH4Cl, 0.17M tris, pH 7.65 to lyse contaminating erythrocytes (rbc lysis buffer). The cells were then washed thrice at 100 X g for 10 min to help remove platelets that separated at the PBM- rich interphase. They were then enumerated, assessed for 29 viability by staining with trypan blue (Sigma Chemical Co.)- Viability as determined by trypan blue exclusion, always exceeded 90%. Cell suspensions were subsequently prepared in cold GKN buffer (Sg NaCl, 0.4g KCl, I.Ilq Na2HPCM . 2H20, 0.69g NaH2PO4.H20 , 2g glucose, per liter deionised water; pH 7.2) containing 0.1 % sodium azide and, heat-inactivated 2% gamma globulin-free horse serum (GGF-HS; Gibco Laboratories, Grand Island, NY) . Aliquots of IO6 cells were pelleted in 12 X 75mm Falcon tubes (No. 2052, Becton Dickinson Labware, Lincoln Park, NJ) . All incubation and washing steps were subsequently carried out on ice in these tubes. In order to block non-specific and Fc-receptor binding sites (33) , the cells were resuspended in 100/il GKN containing 5% normal, rabbit serum, 5% GGF-HS, 2% goat serum 0.1% sodium azide and incubated for 15 minutes, washed once in 4ml GKN and incubated for a further 3 0 min with SOjLtl of the appropriate primary step MAb dilution in CGKN. For FACS analysis, the isotype-specific MAb were used at the following dilutions; DAS 6, 1/1000; DAS 17, 1/100; DAS 2, 1/500; DAS 7, 1/500). Phycoerythrin-Iabeled goat anti-mouse IgG (Fab12, SOjLil, 1/100; TAGO Inc. , Burlingame, CA.) was added to the cells and incubated for 30 minutes. After the final wash, the cells were resuspended in I ml CGKN. Control tubes received the secondary reagent only or were treated with the MAb DREG 55, a mouse IgGl, specific for the human homing receptor (116) as an antibody isotype control. Untreated cells were used for 30 generating forward-scatter profiles and selection of the cell population to be analyzed. Single-parameter flow cytometric analysis was performed using the FACSCAN (Becton Dickinson, Immunocytometry Systems, Mountain View, CA) . Data acquisition and analysis was obtained using the Consort 30 software. Data was acquired in list mode on 10,000 events. Histogram profiles were based on relative cell numbers and fluorescence intensity. The cell population examined was gated on a scatter profile and excluded most non-lymphoid and dead cells. In Vitro PBM Activation To study the effects of in vitro activation on the cell- surface expression of BoT4 and BoTS, aliquots' of I X IO6 PBM were resuspended in 0.5ml RPMI 1640 medium (Gibco), supplemented with 2OmM L-glutamine, 100 U ml'1 penicillin G , IOOjLtg ml"1 streptomycin, 10% fetal bovine serum and 5 X 10"5M 2-mercaptoethanol (CRPMI). The cells were incubated with or without concanavalin A (Con A, 4jug ml"1 final concentration) , phytohemagglutinin (PHA, Ijug ml'1; Calbiochem Corp., La Jolla, CA), 3 OnM phorbol, 12-myfistate, 13-acetate (PMA, Calbiochem) , IjLtM ionomycin (calcium salt; Calbiochem) or soluble sporozoite antigen. The cultures were incubated in 24-well plates (Corning Glass Works, Corning, NY) at 37°C in 5% C02-95% air. Untreated cultures received 0.5 ml CRPMI. After 18 hours, cells were harvested by gentle pipetting, 31 washed twice with cold CGKN and processed for flow cytometry as described above. Enzyme Treatment PBM cell cultures were incubated with or without varying doses of proteolytic enzymes to determine the relative sensitivities of surface BoT4 and BoTS. Suspensions of PBM cells at a final concentration of I X IO6 ml’1 were incubated with a-chymotrypsin (ICN Nutritional Biochemicals, Cleveland, OH) or neutral protease (Boehringer-Mannheim Biochemicals, Indianapolis, IN) at 37°C for I hour. The cells were harvested and processed for FACS analysis as described above. Results Frequency of T-Cell Subpopulations In one experiment in which 2 infected and one healthy calves were monitored sequentially, two weeks following the first challenge inoculum, the frequency of T cells in the gated population had nearly doubled, relative to the non- inf ected control (Fig. I). Considerable fluctuation in the frequency of T cells was observed during the course of the study. However, significant differences (p=.01) were observed between pooled data obtained from additional calves similarly infected, in a separate study (Table I). The transient increases appear to reflect responses to infection as they were detectable within 14 days after the administration of each challenge inoculum- Although the levels eventually dropped to near baseline, the relative increases were sustained through most of the experimental period. The frequency of circulating BoT4+ cells roughly paralleled alterations in T cells, the most dramatic increases occurring at 14, 35 and 42 days after the initial challenge inoculum (Fig. 2). Differences (p=0.I) were also observed in the pooled data (Table I). In contrast, relative increases in the frequency of BoTS lymphocytes did not parallel those of T cells, except on day 35 (Fig. 3) . No differences were observed in the pooled data. These results suggest increases in the frequency of T cells are largely due to both absolute and relative increases in BoT4+ cells and not due to decreases in BoTS+ cells. Table I. The frequency of circulating T cells, BoT4 and BoTS cells in calves 49 days post-challenge with Eimeria bovis oocysts. 3 2 Phenotype Infected3 % positive (mean + SD) Controls6 % positive BoT2 CO H i+ H O O O 28.0 ± 6.0 (30.5 ± 18.4) BoT4 23.8 ± 5.6d 13.8 ± 4.9 (17.7 ± 12.9) BoTS 13.1 ± 3.8 9.1 ± 1.4 (13.1 ± 6.9) an=5 calves. bn=4 calves. ^significantly different from controls (p=0.01). d(p=0.I). Parentheses: PBL values from 9 healthy calves. 33 Sequential mode fluorescence values of BoTS were lower in infected than in non-infected controls for most of the sampling intervals (Fig. 4). Surface expression of BoTS as indicated by mode fluorescence, showed considerable fluctuation in both sets of calves although control levels at some intervals were 4 to 5-fold higher than in the infected and challenged calves. Similar trends were not observed for BoT2 and BoT4. In 9 calves surveyed, the mean TS mode fluorescence values were 10 to 20-fold higher than in the infected calves at day 42 (Table 2). Although mode fluorescence values were not obtainable in a separate experiment, histogram profiles revealed lowered surface BoTS expression in PBL of infected calves (Fig. 4A) Table 2: Comparative levels of surface expression of BoTS on circulating and tissue lymphocytes of infected and non- infected calves. Mode Fluorescence PBL MLN Ileum SPLN Naive 312 (1039 + 399*) 889 417 1187 Naive ND 691 416 242 Infected 51 53 44 49 Infected 53 48 61 43 ^Mean ± SD Mode fluorescence values obtained from PBL of 9 healthy calves. (ND= not done) ce nt 34 D-----D Control A— -A inf A ° — °ln f B 14 21 28 35 42 49 56 Days post challenge Fig. I. Sequential alterations in the frequency of circulating T cells in calves challenged with ICr E . bovis oocysts. 35 Fig. D-----DControI A— -A |nf A ° — °ln f B s $ 14 21 28 35 42 49 56 Days post challenge 2. Sequential alterations in the frequency circulating BoT4+ cells in calves challenged IO5 E . bovis oocysts. of with ce nt po si tiv e 36 D— DCcntroI a— A|nf A o— oinf B 14 21 28 35 42 49 56 Days post challenge Fig. 3. Sequential changes in the frequency of circulating BoTS+ cells in calves challenged with IO5 E . bovis oocysts. 37 Fig. 4. D-----EC trI A— -AInf A O— OInf B 4000 3000 2000 1000 - □— — 14 21 28 35 42 49 56 Days post challenge Sequential cell-surface expression of BoTS in calves challenged with IO5 E . bovis oocysts. 38 Fig. 4 Continued. Comparative fluorescence levels of circulating BoTS+ cells in 3 control (left histograms) and 3 E . bovis - infected calves (right) at 49 days post­ challenge . 39 Cell-Surface Immunoglobulin Isotypes Circulating lymphocytes positive for DAS-6 (IgM-specific) ranged between 7 to 2 5% of the gated population in non- infected versus 13-53% in infected calves. Membrane IgM (mlgM) positive cells exhibited a steady increase in infected calves peaking at 35 days following the initial challenge inoculum (Fig. 5) . The maximum percentage increase of infected over naive values reached approximately 3-fold.at day 49 post challenge. Lymphocytes positive for the monoclonal DAS-17 (IgGl- specific), likewise exhibited steady increases during the experimental period (fig. 6), The increased frequency of these cells was more marked at days 35 and 49 post­ inoculation, being 10-fold higher than the frequency of control cells. The observed levels in both groups of calves were comparable at 0 - 14 days post-inoculation. Cells expressing both the IgM and IgG1 isotypes were significantly higher (p=.05) at day 49 post challenge than either the control group or 9 healthy calves surveyed (Table 3) . In contrast to the patterns observed for pan T and surface BoT4, the decline in the frequency of mlgGl during the intervals between the initial inocula and the challenge inocula was not observed. This suggests the generation of a stable memory HtIgG+ population which became expanded during subsequent oocyst inoculations. This may further indicate that these 40 cells are maintained as a stable circulating subpopulation in view of the sustained increases during the intervals between initial and challenge inocula. Alterations in mIgG2+ cell frequency were less uniform than those in InIgGl+ cells, but increases were more marked at 21 and 35 days post-challenge (Fig. 7). The marked fluctuation in both groups in circulating mIgA+ cells suggests transient increases may be random and reflect fluctuations in transiting circulating cells (Fig. 8) . Neither of the latter isotypes revealed significant differences from any of the control groups. Histogram profiles for surface T and B cell surface,phenotypes are appended (Fig. 17). Table 3. The frequency of circulating Ig-bearing cells in calves infected and challenged with IO5 E . bovis oocysts. Isotype Infected3 % positive (mean+SD) Controls6 IgM 23.4 ± *13.8, 7.7 ± 3.6 (10.8 ± 6 .0) IgGl 11.2 ± 11.9* 0.9 ± 0.9 (2.5 ± 2 .2) lgG2 2 .6 ± 2.2 0.7 ± 0.3 (0.4 ± 0.3) IgA 1.4 ± 0.6 0.6 ± 0.2 (0.3 ± 0.1) ^n=S; n=4; ‘significantly different from controls (p=.05). Figures in parentheses represent values obtained from 9 healthy calves. ce nt ce ll: 41 □— DControI A— A|nf A o— oinf B Days post challenge Sequential alterations in the frequency circulatinq mIgM+ cells in calves inoculated IO5 E . bovis oocysts. of withFig. 5. ce nt po si t! 42 D— n Control A— -Ainf A ° — Olnf B 14 21 28 35 42 49 56 Days post challenge Fig. 6 . Sequential alterations in the frequency of circulating mlgGl cells in calves inoculated with IO5 E . bovis oocysts. P er ce nt p os iti ve 43 a— D Control A— Ainf A o— 0 |nf B 14 21 2 8 3 5 4 2 4 9 5 6 Days post challenge Fig. 7. Sequential changes in the frequency of circulating mIgG2 cells in calves inoculated with IO5 E. bovis oocysts. P er ce nt po ai tiv 44 D— DControI A— Ainf A o— Olnf B 14 21 28 35 42 49 56 Days poet challenge . Sequential alterations in the frequency circulating mlgA cells in calves inoculated IO5 E . bovis oocysts. of with Fig. 8 45 Expression of BoT4 and B0T8 in Activated Cells Circulating lymphocytes cultured in the presence of mitogens, the Ca++ ionophore ionomycin or soluble sporozoite antigen exhibited differential responses in the surface expression of BoT4 and B0T8 . The expression of BoTS was markedly reduced in cells treated with Con A, PHA and PMA. The reductions, in non-infected cells represented 9 to 17 fold decreases in mode fluorescence values of treated versus untreated cultures (Fig. 9). The decreases in mode fluores­ cence values of treated cells from infected calves ranged between 3- to 5-fold in one calf (Fig. 10) and 10- to 16-fold in another (Fig. 11) . The initial levels of expression of Bo- TS in the untreated cells of the former calf may have been a factor in the reduced magnitude of the depressed values. These trends were reproducible in all mitogens tested in two independent experiments, except in one instance where an initial mode fluorescence value of 36 in untreated cells of one infected calf was increased 3-fold in the presence of Con A (data not shown). It should be noted that this low value was comparable to the baseline mode fluorescence values which ranged from 20 to 38 in cells treated with the second step PE- labeled antibody only. In contrast to BoTS, there was variable enhancement in the expression of BoT4 in Con A-treated cells (Figs. 12-14). The mitogens PHA and PMA caused reduced expression in both groups of calves with greater fold 46 reductions in infected over non-infected mode fluorescence values. Sporozoite antigen-dependent activation resulted in no alterations in the surface expression of either molecule in cells from a non-infected calf (Fig. 15 A,B) . In infected calves, antigen-induced modulation of surface BoTS resulted in a two-fold increase in the calf with baseline mode fluorescence values but not in the other calf (Fig. 15A). In both calves, however, BoT4 was moderately increased in one and two-fold in the other (Fig. 15B). These results may indicate that antigen activation in vivo renders the surface molecules, in particular BoTS, less susceptible to in vitro antigen stimulation. This may further imply that the in vivo antigen-induced alterations of surface expression resulting from repeated antigen challenge may represent stable and specific responses imposed by antigen, since surface expression exhibited greater sensitivity to antigen- independent (i.e. nonspecific) activation. Sensitivity of cell-surface BoT4 and BoTS to two pro­ teolytic enzymes in cells from a non-infected calf revealed dose-dependent reductions in mode fluorescence in the former with both a-chymotrypsin and neutral protease (Fig. 16). Re­ duction in BoTS expression occurred only at the higher dose of neutral protease (Fig. 16B) but not with the a-chymotrypsin (Fig. 16B). The differential sensitivity of the two surface 47 molecules to proteases may indicate differences in regulatory mechanisms of surface expression in the two cell phenotypes. Discussion Calves repeatedly challenged with E . bovis oocysts exhibit detectable alterations in the frequency of the majority of circulating lymphocyte subpopulations studied. Increases in T cells are more marked within a fourteen-day period following oral challenge than in the immediate post­ challenge period. The fluctuations in frequency in the intervening periods may be an indication of several events occurring in response to the infection: antigen-specific T cells may have selectively expanded upon encounter with antigen. This event, during the induction phase, will conceivably be more accentuated at sites of increased concentration, namely the GALT. As antigen-sensitive lymphoblasts are induced to recirculate, their numbers in the peripheral circulation will likely reflect this event. Since sporozoite release from oocysts and their invasion of the gut epithelial lining occurs within the first 24 hours (Speer, personal communication) , the changes in T cell frequencies seen in the peripheral circulation may be a reflection of clonally expanded, recirculating, antigen-sensitized lymhocytes. Alterations in circulating BoT4+ cells closely parallel T cell changes, suggesting that absolute increases in the 48 Fig. 9. Cell-surface B0T8 expression on circulating lymphocytes activated in vitro with mitogens, in a non-infected calf. M od e Fl uo re sc en ce 49 300 r Treatment None Con A PHA PMA 10. Cell-surface BoT8 expression on circulating lymphocytes activated in vitro with mitogens in a calf, 42 days after challenge with IO5 E . bovis oocysts. Fig. 50 400 r Treatment None Con A PHA PMA Fig. 11. Cell-surface expression of BoTS on circulating lymphocytes after in vitro activation with mitogens in a calf 42 days after challenge with E . bovis oocysts. 51 Fig. 12. Cell-surface expression of BoT4 on circulating lymphocytes in a non-infected calf after in vitro activation with mitogens. 52 2000 r Treatment None Con A PHA PMA Fig. 13. Cell-surface expression of BoT4 on circulating lymphocytes after in vitro activation with mitogens in a calf 42 days after challenge with E . bovis oocysts. 53 1500 r I 1000 - 500 Treatment None Con A PHA PMA Fig. 14. Cell-surface expression of BoT4 on circulating lymphocytes after activation with mitogens in a calf 42 days after challenge with E . bovis oocysts. 54 a) Antigen Antigen Fig. 15. Cell surface expression of B0T8 (A) and BoT4 (B) in control (open) and infected (hatched) calves. PBM (2 X IO5) were incubated for 18 h at 37°C with solubilized sporozoite antigen (2 X IO6 sporozoite/well) and processed for cytometry. 55 o— OBo T4 •— ego TB 2000 1000 Enzyme oono Uhlte/ml) o— OBo T4 •— OBo TB 2000 1000 Enzyme oono (Uhlts/mO Fig. 16. Effect of chymotrypsin treatment (A) and neutral protease (B) on cell-surface expression of BoT4 and BoTS on PBL from a normal calf. Cells were incubated with enzyme in serum-free RPMI 1640 for I h at 37°C prior to staining for cytometry. 56 former may be the major factor contributing to the T cell responses. Higher mean T4:T8 ratios in the infected group compared to control ratios for the experimental period concurs with these findings. Marginal changes in the T8+ cell populations support the importance of the contribution of T4+ cells to the overall changes in peripheral T cells. The relatively lower levels in the surface expression of B0T8 molecule in infected calves suggests, for this lymphocyte subset, that regulation of surface expression may have more important functional implications than altered frequency in response to the infection. In response to bluetopgue virus (HO) and trypanosomiasis (111), both absolute increases and decreases, respectively, in frequency have been documented for BoT8+ cells. These studies, therefore, suggest the cell frequency of BoTS+ is subject to quantitative alteration in response to viral and protozoal infection. The simultaneous increase of BoT4+ cell numbers and the decreased expression may reflect the predominance of T-helper-related functions over cytolytic/suppressor functions in parasite-specific, immune effector functions. Of the T-helper-induced mechanisms, the induction of bovine macrophage microbicidal and growth inhibitory activities against intracellular sporozoite by activated T cell supernatants (87, 88), provides a functional illustration in this regard. Further, the -importance of interferon-r infection (91) and the role of L3 57 T4+ cells (86) in murine resistance to experimental E. vermiformis have been demonstrated. Steady increases in the frequency IgM- and IgG- bearing cells in infected calves suggests the kinetics of clonal expansion and traffic regulation of antigen-specific peripheral B cells may be different from those affecting cells of the T lineage. Studies on lymphocyte recirculation suggests that differential traffic regulation of lymphocyte subsets, based on their tissue and vascular distribution, occurs in sheep (32, 118). Recirculating specific B cells may continue to expand in the peripheral circulation where they encounter soluble parasite antigen after their initial induction at GALT sites. This, in turn, may be a reflection of, possibly, a relatively long half-life of these antigens, thereby providing a continual source of stimulation for sensitized B cells. In vitro activation of PBM cells produced variable responses in the surface expression of T4 and T8 . Cell- surface expression of T8 was significantly decreased in both groups of calves in cultures treated with Con A, PHA, and PMA. Increased T4 cell-surface expression by Con A, in contrast, suggests that different activation signals may be delivered to the T cell receptor (TcR) on the two subsets by this mitogen. Conversely, this may imply differences in subset TcR, which has been identified as the ligand for Con A (119). PMA caused the down-regulation of T4 but not T8 in mouse 58 lymphocytes (120). In longer term cultures, however, the expression of surface T8 molecules was also decreased (121). The differences observed in T8 and T4 responses to Con A suggest that the cell-surface expression levels of the two molecules may be regulated by distinct intracellular mechanisms. Treatment with antigen resulted in increased surface T4 expression resembling those induced by Con A, suggesting that mitogenic activation via the TcR by Con A may mimic the effects of antigen-specific activation on surface T4+ but not T8+ lymphocytes. Although lymphocyte activation causes receptor phosphorylation and internalization and could partially explain the down-regulatory effects, the modulation of receptors on activated cells by surface proteases could be an additional mechanism responsible for this phenomenon. PMA- and chymotrypsin-induced down-regulation of MEL-14 antigen oh neutrophils suggested the potential regulatory role of surface proteases on this cell-surface receptor (122). In this regard, the differential effects of chymotrypsin and neutral protease on surface BpT4 and BoTS may, in addition to structural differences, reflect differential susceptibility to surface regulatory events. In summary, increases in the frequency of T cells in calves was largely due to altered BoT4 cells. There were also increases in the frequency of B cells bearing IgM. and IgGl isotypes. Alterations in BoTS in vivo were principally 59 qualitative, as reflected in the decreased surface expression of this molecule. Activation with Con A, PHA and PMA mimicked this effect on T8 in vitro. The patterns of altered frequencies of T and B cells suggest differences in dynamics of responses of the two subsets to repeated exposure to E . bovis could be an additional mechanism responsible for this phenomenon. PMA- and chymotrypsin-induced down-regulation of MEL-14 antigen on neutrophils suggested the potential regulatory role of surface proteases on this cell-surface receptor (122). In this regard, the differential effects of chymotfypsin and neutral protease on surface BoT4 and BoTS may, in addition to structural differences, reflect differential suceptibility to different surface regulatory events. 60 Bo T 4Pan T Bo TB F l u o r e s c e n c e Fig. 17. Histogram profile overlays for the frequency of peripheral blood T and B cell surface markers in normal (solid line) and infected calves (dashed). 61 CHAPTER 3 THE FREQUENCY AND DISTRIBUTION OF B AND T LYMPHOCYTE SUBPOPULATIONS IN LYMPHOID TISSUES OF EIMERIA BOVIS-CHALLENGED CALVES Introduction Studieg on the relative frequencies of B and T cell subpopulations in GALT and other lymphoid tissues in response to experimental infection with E . bovis should provide useful information on parasite-induced alterations at the various tissue level relative to the primary infection sites. Tissue localization pf subset-specific surface phenotypes should also shed light on likely differentiation events induced by the coppidia. The predominance of IgA-containing cells in the gut and mesenteric lymph nodes in mice experimentally infected with E. falciformis suggests the relative importance of these cells at primary infection or drainage sites and their preferential generation in response to infection. In non-ruminants, the Peyer1s patch has been found to be an enriched source of IgA precursor cells in the gut lamina propria (105, 123). In sheep, however, the Peyer1s patch is regarded as a primary lymphoid organ that exports immunoglobulin-bearing lymphocytes to all other peripheral lymphoid tissues (39). This may imply possible differences in the frequencies of IgA precursors in ruminants compared to other mammals. 62 In "normal" cattle, variable estimates of T and B lymphocyte subpopulation frequencies have been documented in lymphoid tissues. In peripheral lymph nodes, T cells constitute 60-70% of the lymphocytes (25, 26). The BoT4 and BpTS phenotypes constitute 40 and 25%, respectively, of lymphocyte suspensions in the same tissue (25). Lower (20- 24%) frequencies have been reported for B in cells peripheral lymph nodes and the spleen (26, 124, 125). In bovine GALT, the frequency of T cell subsets and B cells havp been determined for intraepithelial (IE), lamina prpprial (LP) and Peyer1s patch of the ileum by flow cytometry (29, 33) and immunofluorescence (33b). T cells constituted 44% of Peyer's patch lymphocytes compared to 26% and 38% of IE and LP lymphocytes, respectively. Similarly, the frequencies of B cells and the T-helper phenotype in the LP and IE were twice those observed in IEL. These observations suggest preferential accumulation of lymphoid subpopulations in distinct microenvironments of the bovine gut. These findings were not, however, supplemented with immunphisto- chefliical localization. Certain discrepancies in the frequency of T cells and T cell subpopulations are apparent. In all tissue preparations from the different micro-anatomical sites, the sum of frequencies of BoT4 and TS cells was in considerable excess over those obtained for total T cell populations, suggesting possible cross-contamination of lymphoid cells from the different sites, the target MAb 63 epitopes present on PBL T cells may be different from those on IE lymphocytes, the absence of pan T markers on cells of the BoTS phenotype (33) or recognition of non-T cells by the T cell-specific MAb. To the best of our knowledge, immunohistochemical localization of domestic ruminant GALT T cell subsets has been demonstrated only in sheep (126). The relative micro- anatomical distribution of B and T cells varied according to the origin of the gut tissue. In the jejunum, the majority of T cells localized in the interfollicular areas, while this site in the ileocecal tissue contained mainly B cells. Some T cells were also present in the dome and the corona regions in tissues of both origins. In the calf, the interfollj,cular region has been identified as a T cell-dependent area based on the accumulation of labeled cells following the infusion of 3H-thymidine into the thymic arteries (127). Earlier studies (34-37) on histochemical identification of immunoglobulin (Ig)-bearing GALT cells yielded mixed results. The tissue distribution of B cells expressing surface IgM, IgGl, IgG2 and IgA has been studied in cattle using polyclonal antisera and immunofluorescence (34, 35* 37) or immunoperoxidase staining (36). The predominant surface Ig isotypes reported in young Calves were IgA and IgM located in the lamina propria and intercryptal regions (34), The relative frequency of IgG2+ cells in 4 day-old calves, although considerably lower than the former two isotypes, 64 exceeded that of IgGl-bearing cells. In the same study, many cells reported to exhibit membrane IgGl and IgG2 fluorescence were not enumerated; only cells with intense cytoplasmic staining were included. These populations may reflect the more mature, terminally differentiated plasma cells, or, in the case of IgM-containing cells, they may also reflect a less differentiated cytoplasmic ^-containing pre-B cell. The latter is less likely in view of the location of positively surface staining cells in the lamina propria or intercryptal regions. Further, the influence of elevated levels of circulating IgGl in colostrum-fed calves (38) on the frequency and distribution of lymphocytes expressing this isotype is not known. Other studies reported IgGl as the predominant cell- surface isotype on GALT lymphocytes (35, 36), exhibiting tissue distribution patterns similar to those described for IgA- and IgM-bearing cells. Factors contributing to the differences observed have not been identified, .but, in view of the crucial role of ileal Payer's patch in the generation and export of B cells in another ruminant (ovine) model (39) and it's complete involution with age (40), age-related changes in isotype tissue distribution and frequency may partially contribute to this phenomenon. Furthermore, the possibility of nonspecific fluorescence or the binding of ■ • ' • . . . . . . .first and second step reagents via the Fe receptor (33) were not addressed in these earlier studies. Consequently the 65 contribution of these phenomena to the tissue distribution of cell-surface immunoglobulin isotypes described in these studies remains undetermined. The present study was undertaken to determine the frequencies of T and B lymphocyte subpopulations in the terminal ileum, mesenteric lymph node (MLN) and spleen using flow cytometry in calves repeatedly challenged with E . boyis. Immunohistochemical methods were, in addition, used to study the tissue distribution of surface phenotypes of both cell types in GALT and MLN. These two approaches should be complementary in obtaining quantitative estimates on the relative frequencies and the qualitative immuno-histochemical localization of lymphocytes bearing the various B and T cell markers. The use of MAb specific for T cells, their subsets and Ig isotypes should enhance the specificity of the reactions. Information obtained there-from should improve our understanding of cellular composition and events occurring at primary infection, drainage and distant sites in tissues from infected calves and may indicate likely parasite-induced B cell differentiation events. Materials and Methods Experimental Animals and Infection Calves used in the study and the infection regime used are described in Chapter 2. At 63 days after the initial 66 challenge inoculum, calves were killed with the captive bolt gun and bled. The skin on both flanks was thoroughly cleaned with disinfectant soap and 70% ethanol prior to the removal of ileal, mesenteric lymph node and splenic tissues. Lymphocyte Isolation from the Spleen, Mesenteric Lvmoh Node and Gut A portion of the spleen was aseptically removed into cold, calcium and magnesium-free HBSS containing 50,000 lU/ml penicillin, 50,000 mcg/ml streptomycin and 125 mcg/ml fungizone (CSHBSS Flow Laboratories, McLean, VA). After the splenic capsule was removed and the tissue gently teased with, a thumb tissue forceps onto a petri dish containing cold HBSS, the suspension was transferred to 50 ml centrifuge tubes. Large tissue debris were allowed to sediment by standing the tubes in ice for 5-10 min., the cell-rich supernatant pelleted and the erythrocytes (rbc) lysed with 0.16M NH4Cl (rbc lysis buffer). After two washes, the cells were resuspended in HBSS containing 5mM EDTA and IO7 splenocytes/ml were layered on Ficoll-Hypaque (Histopaque 1077, Sigma) in 15 ml centrifuge tubes and spun at 1000X g for 20 min at 40C to remove dead cells. Viability was determined by trypan blue exclusion. For FACS analysis, IO6 cells were used per aliquot- Mesenteric lymph node cells were obtained in a similar manner, from the mesenteric lymph nodes draining the terminal ileum. 67 For intestinal tissue, a segment of the ileum terminating at the ileocecal junction was identified and freed from the mesentery. Removal of the ingesta was accomplished by flushing the gut lumen with PBS several times using a 60 ml syringe fitted with a 14g needle. This was followed by 3 flushes with PBS containing penicillin, streptomycin and fungizone at the above concentrations. A 6 inch segment of the terminal ileum was then removed and placed in a 300ml bottle containg cold C-HBSS. Further cleaning was accomplished by vigorous shaking the tissue, decanting the C- HBSS and replacing it with fresh C-HBSS. In order to remove epithelial cells, the gut segment was incubated in warm HBSS/5mM EDTA in a 37°C water bath and monitored frequently for the presence of epithelial cells in the buffer. The HBSS/EDTA was replaced for further incubation and the solution monitored for the presence of epithelial cells. This process was terminated when epithelial cells were no longer detectable in the buffer. The EDTA was then removed by several washes of cold HBSS and epithelial cell removal generally required 60-90 min. The peritoneum and residual mesenteric fat was then removed from the gut segment which was then incubated in RPMI containing the mucolytic agent dithiothreitol (IOmM; DTT, Sigma) for 30 min at room temperature (128) the tissue minced in cold HBSS to release lymphocytes and incubated for Ih at 37°C in RPMI 1640 containing 45 units/ml collagenase, 60 units/ml hyaluronidase (Worthington Biochemical Corp., 68 Freehold NJ). The cells were then washed thrice in cold HBSS and large tissue debris removed in a manner identical to the spleen and MLN. The SPL and MLN cells were also incubated in the enzyme solution. Residual epithelial cells were removed by the rapid passage of the cell suspension over a 0.25g nylon wool column equilibrated with cold HBSS. Non-viable cells were removed as described for the SPL above. Few to no epithelial cells were present in the cell suspensions treated in this manner. Flow Cytometry Tissue cell suspensions were processed for cytometry as described above (Chapter 2, Materials and Methods). Briefly, IO6 cells were incubated in GKN containing 5% GGF-HS, 2% goat serum and 0.1% sodium azide in ice for 15 min to block non­ specific and Fc-receptor binding sites. After one wash in GKN, first step antibodies were added to the cells in the concentrations indicated for PBL cells, the cells washed and phycoerythrin-labelled goat anti-mouse Ig (Fab12, TAGO, Inc., Burlingame, CA) added and the cells washed and resuspended in 0:5 ml GKN for cytometry. Control cells were treated with the second step reagent only or with the irrelevant MAb, DREG 55 (ref. 116). Untreated cells were used to generate forward scatter profiles and select the population for analysis. 69 Immunohistochemistrv Sections from the ileum and MLN were collected in aluminum foil cups containing embedding medium (O.C.T., Tissue-Tek, Elkart, IN) , rapidly transferred to a beaker containg 2-Methyl butane (Aldrich Chemical Co. , Milwaukee, WIS), and snap frozen in liquid nitrogen. The frozen tissue blocks were stored at -80°C. Cryostat sections (4/um thick) were fixed in acetone for 5 min, air-dried and stored in slide boxes, under moisture-free conditions at -SO0C. Prior to staining, tissue sections were kept at room temperature for one hour. The tissue area was demarcated with a wax pencil to prevent reagent diffusion. All incubation steps were carried out in a humid chamber, and the slides carefully blot-dried around the sections, between incubations. Tris-buffered saline (TBS, 0.15M, pH 7.60 containing .05% equine serum) was used as the diluent for the primary MAb at the following dilutions: ILA-42, 1/400; ILA-12, 1/400, I LA- 51, 1/200; DAS 6, 1/600; DAS 17, 1/200; DAS 2, 1/400; DAS 7, 1/400) . The same buffer was also used for rinsing the sections between incubations. Immunoperoxidase staining was performed with a staining kit (Histo-probe, Immunohistological Staining Kit, TAGO, Inc., Burlingame, CA) according to the manufacturer's instructions. Briefly, after 100/xl of primary antibody was placed on the tissue for 10 min at 37°C, the slides were rinsed and incubated with 50/zl of biotinylated 70 goat anti-mouse IgG. After a TBS rinse, the sections were similarly incubated with streptavidin peroxidase, rinsed, incubated with substrate (5 min, 37°C) and counterstained with hematoxylin (3 min, room temperature). After rinsing in TBS, the sections were treated with SOjLtl ammonia water, rinsed in distilled water and one drop of mounting fluid added, followed by the application of a glass covers!ip. Control sections received the irrelevant MAb DREG-55 (1/100 dilution) or TBS/equine serum only. The slides were examined with a Nikon Labophot light microscope. Results The Frequency of T cells in the Spleen. Mesenteric Lvmoh Node and Ileum The frequencies of tissue T cells and their subsets are summarized in Table 4. Mesenteric lymph node cell suspensions from E . bovis infected calves had significantly (p=0.05) higher percentages of T cells than MLN from non-infected calves. No significant alterations in this population were evident in the ileum or spleen from infected calves. In addition, the frequency of BoT4 cells was higher in MLN of infected than naive calves. This suggests that alterations in cells bearing the BoT 2 phenotype are preferentially increased at this site, and that this is due to increased BoT4 cells. Similarly, in both sets of calves, the frequency of T cells in the SPL and MLN is significantly higher (p=0.05) 71 than in the ileum. This finding is an indication of the differential distribution of T cells in diverse lymphoid compartments of normal and infected calves. Furthermore, although the frequency of B0T8 in identical tissues of infected compared to naive calves was not different, compartmental differences were similarly observed, with the highest frequency of this subset occurring in the spleen; A three-fold increase in the T4/T8 ratio was observed in the MLN of infected compared to naive calves which further indicates that both absolute and relative increases of the BoT4 subset contributed to the T cell increases (Table 4) . Similarly, compartmental differences were observed for this subset, the spleen in both calf groups had higher frequencies of B0T8 relative to BoT4 cells, as indicated by lower T4/T8 values. Although a marginal shift in the ratio also occurred i n ,the spleen in infected calves, the infection did not have a similar impact in the MLN and ileum, indicating the likelihood of a more stable quantitative relationship of the two subsets in this lymphoid organ. A comparison of the expression of BoTS in infected and naive calf tissues is summarized in Table 2 (see Chapter 2, Results). In all tissues examined, decreases in surface mode fluorescence values in infected calves ranged between 7-fold in ileal cells to 28-fold in the spleen, with intermediate values observed in the MLN. Similar trends were observed for surface BoT8 expression in peripheral blood lymphocytes (Table 72 2). Because quantitative estimates for the mode fluorescence were unobtainable, histogram profiles were appended to demostrate this trend in calf tissues (Fig. 18). These findings indicate that the decreases in surface T8 expression in peripheral circulation are a reflection of similar trends in the tissues. 73 C o n t r o l Infecte d P B L SPL u o r e s c e n c e Fig. 18. Relative expression of BoTS in peripheral blood (PBL), mesenteric lymph node (MLN), spleen and Payer's patch lymphocytes in control and Ei. bovis- infected calves. 74 Immunohistochemical Localization of T cells In the ileum T cells were characterized by random distribution of these cells in the dome areas of the Peyer1s patches, and larger, discrete clusters in the interfollicular tissue. . The rest of the follicular Payer's patch contained scanty numbers of BoT2+ . cells, in some instances giving the impression of being almost devoid of T cells, depending on the level of sectioning (Fig. 19A,B). Considerable numbers of T cells were diffusely distributed throughout the lamina propria and intercryptal regions. The intraepithelial region, especially the glandular crypts were also infiltrated with T cells. Subsets belonging to the BoT4 and BoTS phenotypes closely parallelled those of BoT2 in their tissue distribution in the ileum (Figs. 20, 21) . Intraepithelial localization appeared more prominent with TS cells (Fig. 2IB) which were also less concentrated in the dome and interfollicular areas than T4 cells. These findings suggest the existence of discrete T cell-dependent interfollicular and dome areas in the ileum, similar to sheep (126) and the mouse (127) . About 40% of sheep ileal Payer's patch dome cells were positive for surface IgM (126). Visually discernible differences in the localization of T cells were not apparent between tissues from infected and non-infacted calves. In the MLN, distribution of T cell populations conformed to those described elsewhere (25) namely, the vast majority 75 of these cells being localized in the cortical and paracortical areas (not shown). Fewer cells expressing T cell and subset markers were located in the deeper cortical or medullary areas. The concentration of cells in these areas varied between lymph nodes of different calve? and between different cortical regions of the same node, but in general, the distribution patterns remained largely confined to the cortex and paracortex. Table 4. The frequency of T cell populations in the ileum, mesenteric lymph node and spleen of E. bovis-infected and naive calves. Percent Ileum positive, cells (MeaniSD) MLN SPL Infection + - + + - Bo T2 9.9±2.I 10.7±1.5 44.119.4" 23.6113.5 46.017.6 40.818.0 Bo T4 9.4±4.9 9.111.4 25.317.7* 9.91 3.6 10,613.2 8.412.0 Bo T8 6.2+5.1 7.314.2 8.713.8 13.5110.3 15.015.6 17.615.7 T4:T8 3.4±3.4 1.710.9 3.211.1* 0.91 0.4 0.910.8 0.510.2 n=5, infected; 4, controls. ‘Significantly different (p=0.05) from control values. I 76 Fig. 19. A. The distribution of Bo T2+ cells in the terminal ileum. A) Positive cells are localized in inter follicular (IF) and dome (D) areas of the peyer's patch and intercryptal areas of the lamina propria (L) . X160. Fig. 19 B. Note the high density of positive cells in the IF areas and fewer cells in the peyer's patch follicle (P). XlOO. 77 Fig. 20 78 Fig. 21 A. BoTS+ cells in interfollicular (IF) areas of the Peyer's patch. Note absence of these cells in the Peyer's patch (P). XlOO. Fig. 21 B . In the ileal mucosa, positive cells are located in the intraepithelium (arrows), or randomly distributed in the lamina propria. X200. 79 The Frequency of Iq Isotype- Bearinq Cells in the Ileum, SPL and MLN The frequencies of B cell subpopulations are summarized in Table 5. Although no significant differences (p=0.05) were apparent between identical tissues from infected and naive calves, the frequency of IgM+ cells in the ileum was significantly higher than in the spleen and MLN from infected calves. Similarly, the ileum had higher IgGl- and IgA­ positive cells than the spleen and MLN, indicating the differential distribution of numbers of Ig-bearing cells in the different lymphoid tissue compartments. The reasons for the disparity between alterations in circulating IgM+- and IgG,,-positive and those in tissues are not clear. A possible explanation may be related to the fact that recirculating cells of GALT origin may not necessarily return to sites where tissues were sampled. Relocation of these cells to other sites in the intestinal tract or other mucosal sites (131) may have a dilution effect on the concentration of cells in the GALT tissues sampled. Localization of Iq Isotvpe-Bearinq Lymphocytes In the ileum, surface IgM+ cells in the submucosa are located in Peyer1s patch domes, in intrafollicular spaces and randomly distributed throughout the lamina propria between glandular crypts and fewer numbers of positive cells could be 80 observed in the intraepithelial spaces (Fig. 22A-C). In addition, the stain was more intense on the glandular epithelial luminal surfaces. The periphery of Peyer1s patch follicles contained the highest numbers of positively staining cells, especially in the quiescent follicles; however, the intensity of staining was considerably decreased compared to positive cells in the submucosal areas. This phenomenon has been described in sheep (126). Similarly, more intensely staining IgM+ cells associated with germinal centers were peripheral to this structure. IgGl-, IgG2- and IgA-bearing cells were similarly located in the dome areas in larger numbers (Figs 23-25) and the former two isotypes more commonly detected in the intraepithelial spaces. Although intraepithelial localization was not observed for IgA+ cells, the luminal epithelial surfaces stained most intensely in sections incubated with the MAb DAS 7 (anti IgA) compared to the other isotypes. In Peyer's patch follicles, localization of the IgG- and IgA- bearing cells were localized on the peripheral zones of the germinal centers (Figs. 23B, 25B). The observations described here suggest subtle differences in the tissue localization of Ig-isotype bearing cells in calves, with fewer cells of the IgG and IgA isotypes localized in interfollicular T-dependent areas. As in the case of T cells, there were no discernible differences in distribution patterns between tissue sections from infected or naive calves. Further, germinal center 81 activity was also evident in some Peyer1s patch sections from naive calves, making it difficult to distinguish between E . bovis- or environmental antigen-induced changes. In the