INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Anaplasma marginale, a member of the ehrlichial genogroup II in the order Rickettsiales (17, 56), infects and replicates within bovine erythrocytes. Like other ehrlichial pathogens, A. marginale is principally transmitted either biologically by ixodid ticks or mechanically by biting flies, and it causes disease typified by an acute onset and rapid progression. High-level A. marginale rickettsemia results in severe anemia and, in 34% of clinical cases, death (2). Protective immunity can be induced by inoculation with live, attenuated A. marginale, but the use of these blood-based vaccines is limited by the difficulty of standardization and the risk of transmission of contaminating known or emergent pathogens. Although less efficacious than live vaccines, immunization using killed whole-organism vaccines, as well as purified outer membrane preparations, induced partial protection against high-level rickettsemia and severe disease (7, 36, 52). The ability of outer membrane proteins to induce protection supports a role for these as immune targets and vaccine candidates.

Protective immunity against A. marginale has been hypothesized to require induction of opsonizing immunoglobulin G2 (IgG2) antibody (34) against outer membrane protein epitopes and coordinated activation of macrophages for enhanced phagocytosis and killing (45). Both antibody and cell-mediated immune responses are associated with protection following immunization with live or killed whole A. marginale (14, 15, 21, 23, 52). Although antibody against outer membrane proteins can block the binding of A. marginale to erythrocytes (32) and, following in vitro incubation, its infectivity for cattle (46), passive transfer of antibody alone is insufficient to protect against experimental challenge (25). A requirement for concurrent cell-mediated effector mechanisms is supported by the recrudescence of rickettsemia in persistently infected cattle within 10 days following splenectomy or immunosuppression induced by dexamethasone or cyclophosphamide treatment (16, 26, 29). This prompt recrudescence occurs prior to a significant decrease in antibody titer. In cattle, gamma interferon (IFN-) is responsible for enhancing IgG2 production (19) and activating macrophages to produce nitric oxide (NO) (1, 51). Furthermore, bovine CD4⁺ T cells expressing IFN- have been shown to provide cognate help to B cells for IgG2 production (11). The efficacy of targeting IFN--mediated immunity is suggested by the enhancement of protection in cattle against A. marginale following inoculation with mycobacteria (50), known to induce interleukin 12 production and augment IFN- expression (18).

The outer membrane fraction of A. marginale is composed of at least six major surface polypeptides, which include major surface proteins (MSPs) MSP-1a, MSP-1b, MSP-2, MSP-3, MSP-4, and MSP-5 (46, 52, 55). Protection against homologous challenge in cattle immunized with purified A. marginale outer membranes correlates with the titer of antibody against these MSP epitopes exposed in the outer membrane (52). Immunization with either purified native MSP-1 or purified native MSP-2 induces protection against experimental challenge, as shown by the significant reduction in rickettsemia and anemia compared to sham-immunized control cattle (41, 42, 47). In addition, MSP-3, which shares conserved peptide blocks with MSP-2 resulting in 55% identity in the amino-terminal half (4), induces partial protection against challenge, as indicated by a significant delay in the onset of rickettsemia (45). Importantly however, A. marginale outer membrane proteins, including MSP-1, MSP-2, and MSP-3, vary both structurally and antigenically among strains (4, 5, 38, 43, 44). Consequently, identification of epitopes recognized by the protective immune response and conserved among otherwise antigenically distinct strains is a major challenge for the development of improved vaccines against A. marginale.

Because of the importance of T cells in protective immunity, the overall goal of our study was to characterize the helper T-cell response to MSPs present in outer membranes of A. marginale in calves immunized with whole outer membranes. Furthermore, we wished to determine whether the immunodominant proteins recognized by helper T cells from cattle protectively immunized with whole outer membranes were conserved among A. marginale strains. A. marginale-specific proliferation of peripheral blood mononuclear cells (PBMC) was detected 1 month following the first immunization with Florida strain outer membranes and persisted for 6 months following the final immunization. The PBMC responded to native MSP-1, MSP-2, and MSP-3 purified from the homologous Florida strain. Importantly, the PBMC also responded to all five strains of A. marginale tested, which include those with structurally distinct MSPs, and to Anaplasma ovis. Antigen-specific CD4⁺ T-cell lines from each of the immunized and protected cattle produced high titers of IFN-. Interestingly, complete protection against rickettsemia, as confirmed by nested PCR, was associated with development of an IgG2-specific response prior to challenge. These experiments provide the basis for detailed studies of the Th cell response against the protective antigens MSP-1, MSP-2, and MSP-3 (related to MSP-2), and for identifying conserved Th cell epitopes that are capable of eliciting both cellular and humoral anamnestic immune responses and that could be incorporated into subunit vaccines delivered by nucleic acid or other vectors.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Anaplasma strains and preparation of homogenates and membrane antigens. The A. marginale strains used in this study are designated by original location of isolation and include the Florida, South Idaho, Washington C, Washington O, and Virginia strains. These have been described or referenced previously (35). A strain of A. ovis originating from Idaho was also used (37). All Anaplasma strains were maintained as liquid-nitrogen-cryopreserved stabilates of infected bovine erythrocytes in dimethyl sulfoxide-phosphate-buffered saline (PBS). Anaplasma organisms were isolated from thawed, infected bovine erythrocytes by sonication and differential ultracentrifugation as previously described (46). To prepare antigen for in vitro assays, the organisms were resuspended in PBS containing the protease inhibitors antipain and E-64 (Boehringer Mannheim, Indianapolis, Ind.) at 25 µg/ml and phenylmethylsulfonyl fluoride (Sigma Chemical Co., St. Louis, Mo.) at 300 µg/ml and were homogenized by two passages though a French pressure cell (SLM Instruments, Inc., Urbana, Ill.) at 1,500 lb/in². Membranes were prepared from the Florida strain of A. marginale by sucrose density gradient centrifugation and were separated into two fractions (at 1.15 and 1.22 g of sucrose/cm³) as described elsewhere (52). Briefly, purified organisms were suspended in a 20% sucrose solution in 10 mM HEPES buffer containing DNase and RNase at 50 µg/ml each. The suspension was sonicated for 6 min at 250 W and centrifuged at 1,000 × g for 15 min to pellet residual organisms. The supernatant was layered on a sucrose step gradient consisting of equal volumes of 52, 48, 44, 38, and 32% sucrose and was centrifuged at 82,000 × g for 20 h at 4°C. After centrifugation, two visible bands at 1.15 and 1.22 g/cm³ were visualized and individually collected by inserting a 12-g cannula attached to a 10-ml syringe into the top of the gradient. Fractions were individually suspended in cold 10 mM HEPES buffer (pH 7.4) and centrifuged at 177,000 × g at 4°C for 1 h. Pellets were collected and washed twice in cold 10 mM HEPES buffer by centrifugation at 177,000 × g for 1 h. Protein concentrations were determined by the Bradford assay (Bio-Rad, Hercules, Calif.).

Calf immunization and challenge. Three male, neutered Holstein calves aged 3 months and designated 96BO5, 96BO6, and 96BO9 were immunized four times with outer membranes, consisting of the fraction banding at 1.22 g of sucrose/cm³, prepared by sucrose density gradient centrifugation from the Florida strain of A. marginale as described above (52). Briefly, each calf received subcutaneous inoculations of 67 µg of total protein resuspended in PBS containing 6 mg of saponin at 2-, 4-, and 4-week intervals. Three age-matched control calves designated 96B19, 96B20, and 96B21 received saponin alone. Six months following the last antigen inoculation, the calves were challenged by intravenous inoculation of approximately 10² erythrocytes infected with the A. marginale Florida strain in 2 ml of Hanks balanced salt solution free of calcium and magnesium (HBSS). To obtain live A. marginale-infected erythrocytes, freshly collected blood from a splenectomized donor calf with 2.0% rickettsemia was diluted in HBSS and immediately inoculated. Animals were observed for 46 days postinoculation. Blood was examined daily for the presence of A. marginale on Giemsa-stained blood smears, and packed cell volume (PCV) was determined. Data are presented as the day postchallenge when rickettsiae were first observed, the maximal level of rickettsemia and the day following challenge on which this level was reached, and the maximal percentage of decrease in PCV and the day following challenge on which this was observed. The Student one-tailed paired t test was used to determine the statistical significance of rickettsemia and reduction in PCV.

Serological determination. A competitive inhibition (CI) enzyme-linked immunosorbent assay (ELISA) was performed to determine seroconversion following immunization (27). Prior to immunization, sera from calves 96BO5, 96BO6, and 96BO9 were serologically negative for A. marginale. Sera obtained from calves 96BO5, 96BO6, and 96BO9 were all serologically positive at 2 weeks following the second inoculation of antigen and inhibited the binding of anti-MSP-5 monoclonal antibody (MAb) by 84, 75, and 79%, respectively, when used at a 1:2 dilution in this assay. Positive-control sera from two persistently infected cattle inhibited the binding by 75%.

Analysis of A. marginale-specific IgG1 and IgG2 responses. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblotting were performed with MAbs specific for bovine IgG1 and IgG2 to determine the subclass of the specific IgG response. A. marginale Florida homogenate (100 µg of protein) was applied in a single lane to a 7.5- to 17.5% polyacrylamide gradient gel and electrophoresed at 250 V for 3 h. Molecular weight Rainbow markers (10 µl) purchased from Amersham (Arlington Heights, Ill.) were added to a separate lane. The proteins were transferred at 4°C to nitrocellulose membranes by using an electrotransfer unit (Hoefer Scientific Instruments, San Francisco, Calif.) in transfer buffer (0.025 M Tris, 0.19 M glycine, 20% [vol/vol] methanol) for 2 h at 70 V and then overnight at 30 V. The membranes were air dried and immersed for 2 h in blocking solution consisting of PBS (pH 7.4) with 0.02% sodium azide, 0.1% Tween 20, and 10% normal horse serum (Vector Laboratories, Inc., Burlingame, Calif.) with gentle agitation. Following extensive washing in PBS-Tween (PBS [pH 7.4]-0.02% sodium azide-0.1% Tween 20), the membranes were placed in a Miniblotter 25 apparatus (Immunetics, Cambridge, Mass.), and 250-µl volumes of bovine sera serially diluted (1:10 to 1:32,000) in blocking solution were added per well according to the manufacturer's protocol. The membranes were incubated for 2 h at room temperature on a rocking platform. Following successive washes in PBS-Tween with 0.1% Nonidet P-40 (NP-40) and in PBS-Tween alone, the membranes were incubated overnight at 4°C and for an additional 1.5 h at room temperature on a rocker platform with murine anti-bovine IgG1 or IgG2 MAb (Serotec Ltd., Oxford, United Kingdom) diluted 1:100 in PBS-Tween with 2% horse serum. The membranes were washed extensively with PBS-Tween-NP-40 and then with PBS-Tween without sodium azide; then they were washed three times with TNT buffer (0.01 M Tris, 0.067 M NaCl, 0.05% Tween 20 [pH 8.0]) to completely remove sodium azide and phosphate from the filter. The reactions were then labeled for 1 h at room temperature with peroxidase-conjugated affinity-purified donkey anti-mouse IgG (heavy and light chains) (Jackson Immunoresearch Laboratories, West Grove, Pa.) diluted 1:5,000 in TNT buffer containing 1% horse serum. The membranes were washed repeatedly with TNT buffer, and the chemiluminescence was developed with a Renaissance Western blot chemiluminescence reagent (NEN Life Science Products, Boston, Mass.) according to the manufacturer's instructions.

Analysis of persistent rickettsemia by nested PCR. A nested PCR procedure was used with primers specific for the conserved msp-5 gene to detect persistent rickettsemias in immunized and control calves following challenge with viable A. marginale. Blood samples were obtained from the calves five times from 60 to 128 days after challenge, and the nested PCR was performed with DNA isolated according to the manufacturer's recommendations (Purogene; Gentra Systems, Inc., Minneapolis, Minn.) as described elsewhere (53). Blood from a persistently infected animal (bovine 95B808) was used as a positive control, and distilled water was used as a negative control, for the PCR.

A. marginale-specific T-cell lines. Short-term T-cell lines were established from PBMC of A. marginale membrane-immunized animals at various times following immunization but prior to challenge infection. Briefly, 4 × 10⁶ PBMC were cultured per well in 24-well plates (Costar, Cambridge, Mass.) in a volume of 1.5 ml of complete RPMI 1640 medium (10) with 1 to 15 µg of A. marginale antigen prepared from the Florida strain/ml. After 7 days, cells were subcultured to a density of 5 × 10⁵ per well and cultured with antigen and 2 × 10⁶ fresh irradiated (3,000 rads) autologous PBMC as a source of antigen-presenting cells (APC). T-cell lines were maintained for up to 11 weeks by weekly stimulation with antigen and APC, and cells were assayed periodically for antigen-dependent proliferation 7 days following the last antigenic stimulation.

Lymphocyte proliferation assays. Proliferation assays were carried out in replicate wells of round-bottom 96-well plates (Costar) for 6 days when PBMC were used and for 3 to 4 days when T-cell lines were used, essentially as described previously (9, 10). PBMC (2 × 10⁵) were cultured in triplicate wells with antigen in a total volume of 100 µl of complete medium. T-cell lines were assayed 7 days after the last stimulation with antigen and APC. T-cell lines (3 × 10⁴ cells) were cultured in duplicate wells in a total volume of 100 µl of complete medium containing 2 × 10⁵ APC and antigen. Antigens consisted of 0.016 to 25.0 µg of the following (per milliliter): membranes prepared from uninfected bovine erythrocytes (URBC) (from the same donor used to culture Babesia bovis), the Mexico strain of B. bovis (10), or the Florida strain of A. marginale; homogenates prepared from A. ovis or the different strains of A. marginale; and purified native MSPs. Native MSPs were immunoaffinity purified from the Florida strain of A. marginale by using specific MAbs as described in previous publications and include the MSP-1 heterodimer, consisting of MSP-1a and MSP-1b (42), MSP-2 (47), and MSP-3 (33). Protein concentrations in all antigen preparations were determined by the Bradford assay. To determine proliferation, cells were radiolabeled for the last 6 to 18 h of culture with 0.25 µCi of [³H]thymidine (Dupont, New England Nuclear, Boston, Mass.), radiolabeled nucleic acids were harvested onto glass filters, and radionucleotide incorporation was measured with a Betaplate 1205 liquid scintillation counter (Wallac, Gaithersburg, Md.). Results are presented as the mean counts per minute of replicate cultures ± 1 standard error of the mean (SEM) or as the stimulation index (SI), which represents the mean counts per minute of replicate cultures of cells plus antigen divided by the mean counts per minute of replicate cultures of cells plus medium. A SI of 2.0 was considered statistically significant.

Cell surface phenotypic analysis. Differentiation markers on PBMC and T-cell lines were analyzed by indirect immunofluorescence and flow cytometry as previously described (12). The MAbs used were specific for bovine CD2 (MAb MUC2A), CD3 (MAb MM1A), CD4 (MAb CACT 138A), CD8 (MAbs CACT 80C and BAT 82B), and the  chain of the T-cell receptor (MAb CACT 61A). These MAbs were kindly provided by William C. Davis, Washington State University, Pullman, Wash. MAb IL-A29, which recognizes the WC1 molecule on a subset of T cells, was obtained from the International Laboratory for Research on Animal Diseases, Nairobi, Kenya.

Detection of IFN- in supernatants of Anaplasma-specific T cells. T-cell lines were cultured for 24 to 72 h at densities of 0.7 × 10⁶ to 2.0 × 10⁶ cells per ml with 2.0 × 10⁶ APC per ml and 1 to 15 µg of A. marginale homogenate or membranes prepared from the Florida strain of A. marginale per ml. Supernatants were harvested by centrifugation and stored frozen at 70°C. The bovine IFN- assay was performed by using a commercial ELISA kit (IDEXX Laboratories, Westbrook, Maine) according to the manufacturer's protocol. The IFN- activity in culture supernatants diluted 1:5 to 1:500 was determined by comparison with a standard curve obtained with a supernatant from a Mycobacterium bovis purified protein derivative-specific Th cell clone that contained 440 U of IFN- per ml (previously determined by the neutralization of vesicular stomatitis virus [12]).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Immunization and challenge. All three calves immunized with outer membranes prepared from the Florida strain of A. marginale seroconverted, as determined by CI ELISA. Six months following the last antigen inoculation, immunized and adjuvant-inoculated control calves were challenged with the homologous strain and monitored for rickettsemia and clinical signs of anaplasmosis (Table 1). All three control animals developed acute signs of infection, characterized by rickettsemias ranging from 3.5 to 7.4% infected erythrocytes and a 35 to 47% reduction in PCV between days 35 to 37 after challenge. In contrast, rickettsiae were never detected in two immunized calves, and only calf 96BO5 developed detectable rickettsemia, which was significantly lower than those of control calves. Although the three immunized animals had somewhat decreased PCVs (13, 17, and 33%), the mean percentage of reduction in PCV in the immunized animals was significantly lower than that in the control animals. The 13 and 17% drops in PCV in the completely protected vaccinees is within the normal range of variation in healthy calves (52). To verify the lack of infection in two of the three immunized calves, persistent rickettsemia was measured at five time points (60 to 128 days) postchallenge by nested PCR using primers specific for the conserved msp-5 gene. As shown in Fig. 1, an msp-5-specific PCR product was amplified from blood taken on day 60 postchallenge from all three control calves and from immunized calf 96BO5, whereas no product was amplified from blood taken from immunized calves 96BO6 and 96BO9. Identical results were obtained with blood sampled at the later time points (data not shown). Thus, the results of PCR confirmed the absence of rickettsemia in blood smears from two of three immunized calves. These results are consistent with previous studies which showed that immunization with outer membranes in saponin conferred protection, as measured by a lack of detectable rickettsiae in 7 of 10 calves (52). Importantly, the T cells used to characterize antigen-specific responses are derived from cattle demonstrated to be protected.

View larger version (34K): [in this window] [in a new window]   FIG. 1.   Nested PCR analysis of the msp-5 gene in A. marginale membrane-immunized calves and nonimmunized control calves after challenge with A. marginale. Nested PCR was performed with primers specific for msp-5 by using distilled water as a negative control (lane 1) and DNA prepared from the following animals: a bovine persistently infected with the Florida strain (positive control; lane 2), immunized calves 96BO5 (lane 3), 96BO6 (lane 4), and 96BO9 (lane 5), and control calves 96B19 (lane 6), 96B20 (lane 7), and 96B21 (lane 8). Nested PCR products (345 bp) were visualized in a 2% agarose gel following electrophoresis. Molecular weight markers (M) consisting of a 100-bp DNA ladder and prominently displaying the 600-bp fragment were included in the gel.

IgG1 and IgG2 responses in immunized calves before and after challenge. Titers of bovine IgG1 and IgG2 specific for A. marginale Florida were determined by immunoblotting with serially diluted sera obtained from calves at 2 weeks following the second immunization and at 1 and 2 months postchallenge. MAbs specific for bovine IgG1 and IgG2 were used to determine the subclasses of the specific antibodies. The results, summarized in Table 2, show differences in the IgG subclass response in the different cattle. Calf 96BO5, which was not completely protected and developed rickettsemia following challenge, developed an early IgG1 and IgG2 response directed against both MSP-2 and MSP-4. In contrast, calves 96BO6 and 96BO9, which were completely protected against challenge infection, developed only transient and weak IgG1 responses and strong IgG2 responses directed against MSP-2. The IgG2 titers measured before challenge in the completely protected animals were two- to fourfold higher than that in calf 96BO5. The IgG2 titers did not increase following challenge of calves 96BO6 and 96BO9, an observation consistent with complete protection from infection. Two of the three adjuvant control calves produced IgG1, but not IgG2, 2 months following challenge infection, but they were not examined thereafter.

Proliferative responses of PBMC to homogenates and outer membranes. PBMC obtained from the three immunized calves before and at several time points following immunization, ranging from 4 weeks to 6 months after the first antigen inoculation, were tested in vitro to determine Anaplasma-specific proliferation (Table 3). Before immunization, either no detectable responses or weak but insignificant proliferative responses to A. marginale were observed. Although these calves had no possible exposure to B. bovis, the proliferative response to this parasite antigen was somewhat higher, and the SI ranged from 1.2 to 13.8. In contrast, for two of the three calves, PBMC obtained at 4 weeks following the first antigen inoculation had strong responses to A. marginale in comparison with B. bovis, and by 7 weeks after the first antigen inoculation, PBMC proliferated strongly and significantly in response to A. marginale. At this time, there was little difference between the response to A. marginale homogenate and the response to A. marginale purified outer membranes (Fig. 2). In comparison with the control antigen, URBC, both A. marginale antigen preparations induced strong, dose-dependent proliferation of PBMC from all three calves. In most cases, stimulation with 25 µg of antigen/ml was less effective than stimulation with a lower concentration of antigen, as particularly noted with the response of PBMC from calf 96BO5 to homogenate. Although the reasons for this are not known, a suboptimal response with high antigen concentrations is typical with crude antigens of other organisms, such as B. bovis (10). Proliferative responses in these calves continued to be very strong and specific up until the time of challenge at 6 months following the last antigen inoculation (data not shown). Anaplasma-specific proliferation was never detected in control calves immunized with adjuvant alone (data not shown).

View larger version (23K): [in this window] [in a new window]   FIG. 2.   Dose-dependent proliferation of PBMC from A. marginale-immunized calves to A. marginale homogenate and outer membrane antigens. PBMC were obtained 10 days following the third immunization with membranes prepared from the Florida strain of A. marginale and were assayed for proliferation against medium or 1 to 25 µg of URBC (open circles)/ml, 0.2 to 25 µg of Florida strain homogenate (solid circles)/ml, 0.2 to 25 µg of outer membranes purified from Florida strain organisms (open triangles)/ml. PBMC were cultured for 6 days in triplicate with antigen, radiolabeled, and harvested. Results are presented as mean counts per minute from replicate cultures ± 1 SEM.

Proliferative responses of PBMC to geographically different strains of A. marginale and A. ovis. The cross-reactive proliferative response to homogenized organisms was then determined. Homogenates were prepared from different strains of A. marginale, including the immunizing Florida strain and the Idaho, Washington O, Washington C, and Virginia strains, as well as from an Idaho strain of A. ovis. All strains of A. marginale and A. ovis induced levels of proliferation by PBMC isolated at 5 months following the last antigen inoculation that were comparable to the level induced by the Florida strain of A. marginale (Fig. 3). The responses of calves 96BO5 and 96BO9 were significant. The high background proliferation of PBMC resulted in relatively low SIs for calf 96BO6 (1.3 to 2.0), but the results for all three calves indicate the presence of immunogenic epitopes shared by A. marginale strains and between A. marginale and A. ovis.

View larger version (43K): [in this window] [in a new window]   FIG. 3.   Proliferative responses of PBMC from A. marginale-immunized calves to homogenates prepared from homologous Florida (FL) and heterologous Virginia (VA), Washington O (WA O), Washington C (WA C), and Idaho (ID) strains of A. marginale and from A. ovis. PBMC were obtained approximately 5 months following the last immunization and were assayed for proliferation against URBC membrane antigen or homogenates of the indicated A. marginale strains or A. ovis. PBMC were cultured for 6 days in triplicate with antigen, radiolabeled, and harvested, and the results for stimulation with the optimal concentration of protein (0.4 µg/ml for calf 96BO6 and 10.0 µg/ml for calves 96BO5 and 96BO9) or medium are presented as the mean counts per minute of replicate cultures ± 1 SEM.

Proliferative responses of PBMC to native MSPs isolated from the Florida strain of A. marginale. The MSPs of A. marginale found in the outer membrane fraction include MSP-1, MSP-2, and MSP-3. Previous studies have shown that these proteins all induce strong antibody responses (3, 33, 41, 42, 47), and native MSP-1 and MSP-2 have been shown to induce partial protective immunity in cattle (41, 42, 47). To determine the recognition of these MSPs by T cells obtained from calves immunized with outer membranes, proliferation was assessed. In three independent experiments, performed at approximately 2 and 5 months after the last antigen inoculation, PBMC from each animal responded to all three MSPs tested (Fig. 4). Optimal stimulation was frequently achieved with 4.4 µg of antigen/ml, and in most cases 8.8 µg/ml was less stimulatory, which likely reflects residual detergent present after extensive dialysis in the samples that were immunoaffinity purified from detergent-disrupted organisms. Animal 96BO9 had stronger responses (as determined by the SI) than animal 96BO5 or 96BO6. With the exception of the response of calf 96BO5 to MSP-1, the proliferative responses to individual MSPs were significant.

View larger version (42K): [in this window] [in a new window]   FIG. 4.   Proliferative responses of PBMC from A. marginale-immunized calves to purified native MSP-1, MSP-2, and MSP-3. PBMC were obtained at approximately 2 (calf 96BO5) or 5 (calves 96BO6 and 96BO9) months following the last immunization and were assayed for proliferation against the different MSPs isolated from the Florida strain of A. marginale. PBMC were cultured for 6 days in triplicate with antigen, radiolabeled, and harvested. Results for stimulation with 4.4 µg of MSP antigen/ml, 2.0 µg of URBC antigen/ml, or medium are presented as the mean counts per minute of replicate cultures ± 1 SEM.

Specificity and IFN- production of A. marginale-stimulated T-cell lines. T-cell lines established from animals 96BO5, 96BO6, and 96BO9 at different times after immunization were tested for proliferation and IFN- production between 1 and 11 weeks of continuous culture. Examples of T-cell proliferation are shown in Fig. 5. T-cell lines established 4 weeks following the initiation of immunization and cultured with A. marginale membrane antigen for 2 to 4 weeks responded in a dose-dependent manner to antigen prepared from the Florida strain of A. marginale but had little or no response to control URBC (data not shown) or B. bovis membrane antigen. Cell lines from animal 96BO9 consistently had the highest responses to A. marginale antigen (Fig. 5C). Lines 96BO5 and 96BO9 were tested with membrane antigen and APC from an allogeneic animal, and proliferation was not detected, indicating that the response is major histocompatibility complex restricted (data not shown). Levels of IFN- secreted by 3-week-old cell lines stimulated for 3 days with antigen and APC were 244 to 264 U/ml. When tested after 8 to 10 weeks of culture, Th cell lines derived from animals 96BO5, 96BO6, and 96BO9 produced 101, 46, and 38 U of IFN- per ml, respectively, whereas control cultures of APC plus antigen produced between 0 and 3 U of IFN- per ml. Thus, the recall response to A. marginale by peripheral blood lymphocytes from immunized calves is characterized by vigorous proliferation and high levels of IFN- production, which are maintained for several weeks by T-cell lines.

View larger version (22K): [in this window] [in a new window]   FIG. 5.   Dose-dependent, antigen-specific proliferative responses of short-term T-cell lines from A. marginale-immunized calves. T-cell lines were established at 4 weeks after the primary antigen immunization (2 weeks after the second immunization) and cultured for either 2 (calf 96BO6) or 4 (calves 96BO5 and 96BO9) weeks prior to assay. T cells were cultured for 4 days in duplicate with medium or 0.04 to 5.0 µg of A. marginale Florida homogenate (solid circles) or B. bovis membrane antigen (open circles)/ml, radiolabeled, and harvested. Results are presented as the mean counts per minute of duplicate cultures ± 1 SEM.

Phenotype of the lymphocytes responding to A. marginale. T cells cultured with A. marginale antigen for 11 to 22 days were examined by single-color flow cytometry with MAbs specific for bovine lymphocyte differentiation antigens, and surface phenotypes were compared with those of PBMC obtained at the initiation of cell culture. After 11 to 13 days of culture, approximately 70% of the cells were CD4⁺ Th cells, and 25 to 30% of the cells were T cells, as observed previously for B. bovis- and Fasciola hepatica-stimulated cell lines (8, 9). However, by 22 days of culture, 96 to 99% of the T cells expressed CD4, verifying that CD4⁺ T helper cells are the subset responding to antigen in these cultures.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

This study demonstrates that calves immunized with outer membranes prepared from the A. marginale Florida strain were subsequently protected against homologous challenge. The immune response in these calves was characterized by vigorous antigen-specific recall responses of PBMC and cell lines composed of CD4⁺, IFN--producing T cells. These studies confirm and extend earlier experiments that demonstrated significant protection against homologous challenge in cattle immunized with outer membranes of the Zimbabwe Norton strain (52). In the present study, two of three outer-membrane-immunized calves were protected from clinical disease, as demonstrated by significantly lower decreases in PCV compared with values for saponin-inoculated calves. Rickettsemia levels were significantly lower in all three immunized calves compared with control calves, and we show for the first time that outer-membrane-immunized animals can be completely protected from the development of persistent infection following challenge. The two calves with microscopically undetectable peripheral rickettsemias were negative for the presence of the conserved msp-5 gene by nested PCR, a technique which can detect as few as 30 infected erythrocytes per ml of blood (53). The absence of an increase in antibody response after challenge is consistent with the absence of A. marginale organisms in these two calves. The ability to completely prevent infection by immunization is critical for preventing the formation of a reservoir of persistently infected cattle among vaccinated cattle.

The A. marginale-specific Th cell lines described in this study produced high titers of IFN- in response to antigenic stimulation ex vivo. The levels of IFN- secreted by antigen-stimulated T-cell lines cultured for 3 weeks (average, 254 U/ml) were relatively high in comparison with those secreted by B. bovis-stimulated T-cell lines from two B. bovis-immune cattle, which averaged 8 to 13 U per ml (54). The relatively strong IFN- response by CD4⁺ T-cell lines from calves protectively immunized against Anaplasma is consistent with the hypothesis that the protective immune response against rickettsial parasites is in part dependent on IFN-. IFN- production by A. marginale-stimulated PBMC taken from calves undergoing acute infection was also observed (23). IFN- is important for protection in mice infected with Rickettsia spp. (20, 28, 30) and activates macrophages to secrete oxygen and nitrogen intermediates, such as NO, that are inhibitory for many intracellular pathogens, including rickettsiae (31, 49). Soluble factors released by PBMC obtained from cattle during acute A. marginale infection and cultured with specific antigen were toxic for intraerythrocytic A. marginale (58). Although recombinant bovine IFN- used at 500 U per ml was not inhibitory for A. marginale in vitro (57), this cytotoxicity mediated by a soluble factor is consistent with inhibitory molecules being induced in mononuclear cells by A. marginale in combination with IFN- and/or tumor necrosis factor alpha produced by antigen-specific T cells. The role of NO in protection against acute Anaplasma rickettsemia has recently been questioned. In studies reported by Gale et al. (24), administration of a chemical inhibitor of inducible NO synthase (iNOS), aminoguanidine, to calves 9 days after infection with A. marginale failed to increase levels of rickettsemia and in fact resulted in significantly lower levels of rickettsemia and anemia. In the same study, a neutralizing anti-bovine IFN- MAb was given to calves 9 days postinfection, and although serum levels of the MAb remained elevated for 10 days, no effect on rickettsemia was observed. However, in these studies, in vivo inhibition of iNOS activity was not determined, and complete neutralization of IFN- was not verified. Additional experiments need to be performed before either IFN- or NO can be ruled out as important in the defense against acute anaplasmosis.

The outer-membrane fraction of A. marginale is composed of at least six major surface polypeptides, which include MSP-1a, MSP-1b, MSP-2, MSP-3, MSP-4, and MSP-5 (46, 55). MSPs 1 through 4 were originally identified by surface radioiodination, and MSP-5 was originally described by Tebele et al. (52), who showed serological reactivity against this 19-kDa protein in calves immunized with the outer-membrane fraction. PBMC from calves immunized with outer membranes of the Florida strain proliferated specifically in response to purified outer membranes and to whole A. marginale homogenate from the immunizing strain as early as 4 weeks postimmunization. The specificity of the response to A. marginale is indicated by the relatively weak responses of PBMC from all three calves prior to immunization. Furthermore, Anaplasma-naïve control calves showed no proliferative response to A. marginale antigen either before (nine calves tested) or after (three calves tested) adjuvant administration (data not shown). The two calves completely protected from developing infection upon challenge had the strongest levels of proliferation at 4 weeks after the initial immunization. Interestingly, these same calves (96BO6 and 96BO9) developed a strong and biased IgG2 antibody response against A. marginale. Animal 96BO5, which did develop low-level rickettsemia following challenge, developed both IgG1 and IgG2 responses. Thus, our results suggest a role for A. marginale-specific CD4⁺ T cells as helper cells to induce immunoglobulin isotype switching to IgG1 and IgG2. These results, although based on a few animals, are also consistent with the hypothesis that a strong Th cell response characterized by IFN- and IgG2 production is important for protective immunity.

Outer membrane proteins (OMP) of human ehrlichial pathogens, including the Ehrlichia chaffeensis OMP-1, which is a homologue of A. marginale MSP-2, are candidate vaccine antigens (39). When native A. marginale OMP were used as immunogens, the MSP-1 complex and MSP-2 induced protection (41, 42, 47) and MSP-3 stimulated a delayed onset of infection (45). In the present study, PBMC from all immune calves responded to the native affinity-purified MSP-1a-MSP-1b complex, MSP-2, and MSP-3, demonstrating that these three proteins contain immunogenic T-cell epitopes. Furthermore, MSP-2 and MSP-3 proteins contain blocks of conserved amino acids that are of sufficient length to comprise T-cell epitopes (4), so that T-cell recognition of common epitopes on these MSPs is possible. The presence of helper T-cell epitopes on MSP-2 was clearly demonstrated by the strong anti-MSP-2 IgG response. MSP-1b, MSP-2, and MSP-3 are all encoded by multigene families with genetic polymorphism among strains (4, 6, 38, 44). However, cross-reactive proliferative responses to four additional strains of A. marginale and to A. ovis were observed, showing conservation of immunogenic T-cell epitopes within the crude homogenates of these different strains and species. It is known that MSP-1, MSP-2, and MSP-3 contain B-cell epitopes shared between A. marginale strains and Anaplasma centrale (33, 35, 43, 48, 52), and MSP-2 and MSP-5 B-cell epitopes are also conserved between A. marginale and A. ovis (37, 40). Thus, it is not surprising to find shared Th cell epitopes between Anaplasma strains or species. These experiments provide a rationale for identification of antigenically conserved helper T-cell epitopes on MSP-1a, MSP-1b, MSP-2, and MSP-3 that could be incorporated into protein or nucleic acid vaccines to stimulate memory T-cell responses cross-reactive for heterologous strains of A. marginale.

In summary, the studies reported in this paper provide a foundation for more detailed analyses of the helper T-cell response and its role in protective immunity to A. marginale. The immune response is notably complex, and although calves can recover from an acute infection, they remain persistently infected for years. If persistence does result from antigenic variation (22), then characterizing the variant and conserved epitopes recognized by immune T cells is critical for designing vaccines (45). Use of cloned T cells derived from immune calves will permit precise mapping of epitopes on the well-characterized MSP-1, MSP-2, and MSP-3 proteins, as well as identification of additional proteins immunogenic for Th cells. Functional studies performed in vitro with Anaplasma-specific T cells can also help define their roles in protection. These would include experiments designed to measure Th cell cytokine-induced macrophage activation and rickettsial inhibition, and induction of isotype switching by autologous B cells. Characterization of Anaplasma-specific cloned CD4⁺ T cells is ongoing and is the subject of a separate publication (13).