Biased Immunoglobulin G1 Isotype Responses Induced in Cattle with DNA Expressing msp1a of Anaplasma marginale

Abstract

Immunization with the native major surface protein 1 (MSP1) (a heterodimer containing disulfide and noncovalently bonded polypeptides designated MSP1a and MSP1b) of the erythrocytic stage of Anaplasma marginale conferred protection against homologous challenge (G. H. Palmer, A. F. Barbet, W. C. Davis, and T. C. McGuire, Science 231:1299–1302, 1986). The MSP1a polypeptide possesses a conserved neutralization-sensitive epitope. In the present study, the immune response to DNA-mediated immunization using msp1a was studied. The plasmid pVCL/MSP1a, which encodes the complete msp1a gene of A. marginale under the control of human cytomegalovirus immediate-early enhancer/promoter and intron A, was constructed. The immune responses elicited by immunization with pVCL/MSP1a into cardiotoxin-induced regenerating muscle were evaluated in mice and cattle. Antibody reactive with native MSP1a was detected in pooled sera of immunized BALB/c mice 3 weeks following primary immunization. Two calves seronegative for A. marginale were immunized four times, at weeks 0, 3, 7, and 13, with pVCL/MSP1a. By 8 weeks, both calves responded to MSP1a with an antibody titer of 1:100, which peaked at 1:1,600 and 1:800 by 16 weeks after the initial immunization. Interestingly, immunoblotting with anti-immunoglobulin G1 (anti-IgG1) and anti-IgG2 specific monoclonal antibodies revealed a restricted IgG1 anti-MSP1a response in both animals. T-lymphocyte lines, established after the fourth immunization, proliferated specifically against A. marginale homogenate and purified MSP1 in a dose-dependent manner. These data provide a basis for an immunization strategy to direct bovine immune responses by using DNA vaccine vectors containing single or multiple genes encoding major surface proteins of A. marginale.

Anaplasmosis, an economically important hemoparasitic disease of livestock, occurs in tropical, subtropical, and many temperate regions of the world, including the United States (50). Disease is caused by intraerythrocytic infection with the rickettsia Anaplasma marginale. This rickettsia, a member of the ehrlichial genogroup II, is transmitted either biologically, by Ixodid ticks, or mechanically, by blood-contaminated needles or biting flies (49). Following an incubation period of 20 to 40 days, there is an increase in rickettsemia, resulting in anemia, weight loss, abortion, and death (14, 49). Cattle that recover from acute disease become persistently infected and are life-long carriers, serving as a reservoir for the transmission of A. marginale (20). Persistently infected cattle are protected from challenge infection with homologous strains and are partially protected from challenge with heterologous strains (25). Control measures include chemotherapy, tick vector control, and immunization with either attenuated or killed organisms (6, 25, 28, 35, 53). However, vaccines presently available are associated with risks including neonatal isoerythrolysis and transmission of other blood-borne pathogens (51), underscoring the need for an improved immunization strategy for anaplasmosis.

Outer membrane proteins of the erythrocytic stage of A. marginale have been the focus of research directed toward an improved vaccine against anaplasmosis. The rationale for this approach is that outer membrane proteins are surface exposed, readily accessible to the immune system, and likely essential for the survival of the parasite in the host. These proteins may function in nutrient transport and in attachment to and invasion of host erythrocytes (30). Immunization of cattle with outer membranes of the erythrocyte stage of A. marginale induced protection against challenge with virulent A. marginale (9, 54). This finding indicates the potential for use of defined outer membrane proteins as components of recombinant protein or nucleic acid vaccines for anaplasmosis.

Characterization of A. marginale membrane proteins has revealed at least six major surface proteins (MSPs), which include MSP1a, MSP1b, MSP2, MSP3, MSP4 and MSP5 (3, 39, 40, 42, 55, 58). Whereas MSP1b is encoded by a polymorphic gene family (4), MSP1a is encoded by a single-copy gene and contains a neutralization-sensitive epitope defined by the monoclonal antibody (MAb) Ana 22B1 (1, 45). Despite size polymorphisms of MSP1a among isolates, the neutralization-sensitive epitope is conserved (34, 40, 41). Immunization of cattle with affinity-purified native MSP1 complex (a heterodimer containing MSP1a and a 100-kDa protein designated MSP1b) induced protective immunity against challenge with homologous and heterologous strains of A. marginale (43, 44). Peripheral blood mononuclear cells (PBMC) obtained from cattle protected against homologous A. marginale challenge proliferated in response to the MSP1 complex, indicating the immunogenicity of these proteins for helper T lymphocytes (9). Furthermore, it was shown that MSP1a and MSP1b localized to the surfaces of recombinant Escherichia coli bacteria and directed the adherence of these E. coli bacteria to bovine erythrocytes (30). For these reasons, we are exploring the use of MSP1a in a DNA-based vaccine for anaplasmosis.

DNA vaccines consist of an eukaryotic expression vector containing a gene of interest (19, 60). A mammalian promoter drives gene expression, and transcription is terminated by a mammalian polyadenylation signal in mammalian cells. Intramuscular or intradermal inoculation of DNA vaccines into animals transfects cells, which express the vector-encoded protein in vivo (17). The endogenously expressed antigens are processed and presented in the context of major histocompatibility complex (MHC) class I and class II molecules, thereby inducing specific cellular (both CD4⁺ and CD8⁺ T-lymphocyte) and antibody responses in immunized hosts (26, 56).

The majority of studies using DNA vaccines have been conducted on mice, and relatively few studies have been performed with large animals. The objective of the present study was to use MSP1a of A. marginale as a model antigen to evaluate the potential of DNA vaccination for bovine anaplasmosis. The plasmid pVCL/MSP1a was constructed, expressed in vitro in COS7 cells, and injected intramuscularly into mice and cattle. T-lymphocyte lines from immunized cattle proliferated in response to A. marginale homogenate and purified MSP1 in a dose-dependent manner, and the MSP1a antibody response in these cattle was shown to be restricted to the immunoglobulin G1 (IgG1) isotype.

MATERIALS AND METHODS

Experimental animals.

Eight 6- to 8-week-old female BALB/c mice and two A. marginale-negative male holstein calves (numbers C749 and C751) were used in this study. Two weeks prior to immunization, the calves were shown to be negative for antibody to A. marginale by Western immunoblotting and by a competitive enzyme-linked immunosorbent assay (ELISA) using MSP5 (55). During the course of the immunizations, both calves were monitored for antibody to other A. marginale MSPs by Western immunoblotting. Also, at 2 weeks prior to the initial immunization and 2 weeks after the final immunization, blood smears from both calves were examined by Giemsa staining.

Construction of the DNA vector containing msp1a.

Plasmid pAM420, containing the gene encoding MSP1a, was described previously (1). For insertion into the DNA expression plasmid, pVCL1010 (Vical Inc., San Diego, Calif.), the MSP1a gene was amplified from pAM420 by PCR using the primers 5′-CGTGTCTGCAGATGTCAGCAGAGTATGTGTCCAC3′ and 5′-AGCATCTGCAGGACTCTATCAAAGACCGGAA 3′, each of which contains a PstI site (underlined). The amplicons were purified twice by phenol-chloroform extraction followed by ethanol precipitation. The amplicons were then digested with PstI and ligated into the PstI-digested, alkaline phosphatase-treated eukaryotic expression vector pVCL1010. The ligation product was transformed into the competent E. coli strain DH5α and selected by kanamycin resistance by using a standard protocol (52). Kanamycin-resistant colonies were screened for the presence of the appropriate recombinant plasmid. Restriction enzyme digestions with PstI and BamHI identified a construct containing the correct orientation, and this plasmid was designated pVCL/MSP1a.

Purification of pVCL/MSP1a for vaccination.

Recombinant E. coli DH5α was grown in Terrific broth (12 g of Bacto Tryptone, 24 g of Bacto yeast extract, 4 ml of glycerol, 2.31 g of KH₂PO₄, and 12.54 g of K₂HPO₄ per liter) in the presence of 50 μg of kanamycin/ml. To enhance plasmid recovery, the cultures were treated with 170 μg of chloramphenicol/ml for 16 to 18 h. Plasmid pVCL/MSP1a was purified by alkaline lysis of cultures followed by CsCl density gradient ultracentrifugation (52). The resultant pVCL/MSP1a was further purified by RNAse and proteinase K digestion followed by phenol-chloroform extraction and ethanol precipitation. The purity of the DNA was confirmed by measuring the optical density at 260 and 280 nm, followed by agarose gel electrophoresis and ethidium bromide staining. The purified DNA was resuspended in sterile saline and maintained at −20°C.

Transfection of COS7 cells with pVCL/MSP1a.

Expression of MSP1a in vitro was evaluated in monkey COS7 cells. Cells were maintained in Dulbecco modified Eagle medium (DMEM) high-glucose medium (Gibco BRL, Gaithersburg, Md.) and supplemented with 2 mM l-glutamine, 20 mM HEPES, 22 mM NaHCO₃, 10% fetal bovine serum, and 1,000 IU of penicillin-streptomycin/ml in 100-mm plates (Becton Dickinson, Franklin Lakes, N.J.). The cells were held in a humidified incubator with 5% CO₂ at 37°C. Cells were released by trypsinization and split 24 to 48 h prior to transfection. Lipofectamine (Gibco BRL)-mediated transfection was performed when cells were at 95% confluence. For each plate, 8 μg of plasmid DNA and 40 μl of Lipofectamine were mixed in 1.6 ml of serum and antibiotic-free medium (OPTI-MEM; Gibco BRL) and kept at room temperature for 20 min. After 20 min, 6.4 ml of OPTI-MEM was added to the above mixture and overlaid onto cells. After incubation for 6 h at 37°C with 5% CO₂, 8 ml of OPTI-MEM containing 6% fetal calf serum was added to the transfected plate. Twenty-four hours following transfection, the medium was replaced with fresh OPTI-MEM supplemented with 3% fetal calf serum. At 72 h following transfection, the cells were rinsed twice with ice-cold sterile phosphate-buffered saline (PBS) and lysed with 250 μl of protease inhibition buffer (50 mM Tris [pH 8], 5 mM EDTA, 5 mM iodoacetamide, 0.1 mM Nα-p-tosyl-l-lysine chloromethyl ketone [TLCK], and 1 mM phenylmethylsulfonyl fluoride [PMSF]) containing 1% Nonidet P-40 (NP-40) and 0.1% sodium dodecyl sulfate (SDS). The lysate was centrifuged at 12,000 × g for 10 min, and the supernatant was collected and stored at −20°C. Expression of MSP1a was demonstrated by Western immunoblotting as described below.

Immunization of mice with pVCL/MSP1a.

Eight 6- to 8-week-old female BALB/c mice were used in this study. To enhance the cellular uptake of DNA, all mice were anesthetized by intraperitoneal injection of ketamine-xylazine, and their quadriceps were injected with 60 μl of 0.01 mM cardiotoxin (Sigma, St. Louis, Mo.) (17) by using an insulin syringe and a 27-gauge needle (Becton Dickinson). Five days later, mice were anesthetized as described above and injected with 100 μg of pVCL/MSP1a (at a concentration of 1 μg/μl in PBS) in the cardiotoxin-treated regions by using a 27-gauge needle. Control mice were injected with 100 μg of the empty vector pVCL1010 in PBS at a concentration of 1 μg/μl. Mice were immunized at days 0 and 22. Pre- and postimmune serum samples were collected from the tail vein and stored at −20°C until they were tested for MSP1a-specific antibody by immunoblotting.

Immunization of calves with pVCL/MSP1a.

Two male holstein calves (no. C749 and C751) were used in this experiment. Five days prior to each immunization, inoculation areas were shaved and marked with a permanent marker pen. Each calf was inoculated intramuscularly with 400 μl of cardiotoxin (Sigma) (16) by using a 22-gauge needle. Five days later, each calf was immunized with 2 mg of pVCL/MSP1a at a concentration of 500 μg/ml in PBS in the cardiotoxin-treated sites by using a 22-gauge needle. Calves were injected at weeks 0, 3, 7, and 13. The first three immunizations were given in the gluteal muscles (1 mg of DNA/gluteal site), and the final dose was given in the rectus femoris. Calves were bled at 2-week intervals, and anti-MSP1a antibody titers were determined by Western immunoblotting.

Western immunoblotting.

Antigen preparations for immunoblotting consisted of solubilized A. marginale homogenates (Florida strain), or COS7 cell lysates following transfection with pVCL/MSP1a. A. marginale lysate was prepared by differential centrifugation and sonication as described previously (44, 57). Protein concentrations were determined by bicinchoninic acid (BCA) assay, with bovine serum albumin (BSA) used as a standard (Pierce Chemical Co., Rockford, Ill.).

Antigen was suspended in electrophoresis sample buffer (0.025 M Tris-hydrochloride, 2% SDS, 15% glycerol, 2.5% 2-mercaptoethanol), boiled for 10 min, and cooled on ice. Antigen preparations were then loaded on a 7.5 to 17.5% gradient gel with a 5% polyacrylamide stacking gel (60). After separation, proteins were transferred to a 0.45-μm-pore-size nitrocellulose filter (70 V for 3 h) in transblot buffer (25 mM Tris, 190 mM glycine, and 20% methanol).

To detect the expression of MSP1a in COS7 cells, approximately 50 μg of transfected COS7 cell lysate per lane was separated and transferred to nitrocellulose. Approximately 50 μg of untransfected COS7 cell lysate (same amount of protein as transfected cell lysate) was used as a negative control, and approximately 7.5 μg of A. marginale homogenate was included as a positive control (57). The membrane was probed with MAb Ana22B (57). Briefly, the membrane was blocked overnight in buffer A (10 mM Tris HCl [pH 8], 150 mM NaCl, 0.05% Tween 20) with 10% skim milk, washed three times in buffer A, and incubated with MAb Ana22b1 (2 μg/ml) in buffer A. Bound antibody was detected with horseradish peroxidase-conjugated goat anti-mouse IgG (Kirkegaard & Perry Laboratories, Gaithersburg, Md.) at a 10⁻⁴ dilution, followed by enhanced chemiluminescence by use of a kit according to the instructions of the manufacturer (NEN Life Science Products, Boston, Mass.). An isotype control (IgG3) MAb (2 μg/ml), followed by a second antibody (peroxidase-conjugated goat anti-mouse IgG), was also tested.

To detect MSP1a antibodies in cattle, A. marginale homogenate was used as the antigen. The membrane was blocked overnight in buffer B containing 10 mM Tris HCl (pH 8.0), 150 mM NaCl, 0.2% Tween, and 1% (wt/vol) polyvinyl pyrrolidone (Sigma). The membrane was then incubated with test sera diluted in buffer C (10 mM Tris HCl [pH 8.0], 150 mM NaCl, 0.05% Tween 20, and 1.0% [wt/vol] polyvinyl pyrrolidone) for 4 h at room temperature in a miniblotter. After a wash with buffer C, the bound antibody was detected by incubation for 1 h with peroxidase-conjugated protein G (Zymed, San Francisco, Calif.) diluted 1:20,000, followed by enhanced chemiluminescence.

Isotype analysis of MSP1a antibodies from immunized calves.

To analyze bovine immunoglobulin isotypes IgG1 and IgG2 in immunized calves, sera from calves obtained at 8, 14, and 18 weeks postimmunization were analyzed by immunoblotting. A. marginale (Florida strain) lysate (7.5 μg/lane) was used as the antigen. After transfer of proteins, the membrane was rinsed in transblot buffer (Tris glycine buffer with 20% methanol) and dried at room temperature. The membrane was rehydrated in distilled water for 5 min and blocked in PBS-T (PBS [pH 7.4] with 0.1% Tween 20) containing 0.02% sodium azide and 10% equine serum. The membrane was washed in PBS-T and then incubated with test sera diluted in PBS-T containing 0.02% sodium azide and 3% equine serum for 6 h in a miniblotter at room temperature. The membrane was washed once with PBS-T containing 0.1% NP-40 and once with PBS-T and was incubated for 2.5 h with anti-bovine IgG1 (mouse IgG2b isotype) and IgG2 (mouse IgG1 isotype) MAbs (Serotec, Raleigh, N.C.) diluted 1:100 in PBS-T containing 0.02% sodium azide. The membrane was then washed with PBS-T containing 0.1% NP-40, followed by one wash with buffer A. The bound MAbs were detected by incubating the blot for 1 h with horseradish peroxidase-conjugated donkey anti-mouse IgG (Jackson Immunoresearch Laboratories, West Grove, Pa.) diluted 1:5,000 in buffer A containing 1% equine serum, followed by enhanced chemiluminescence.

Establishment of T-lymphocyte lines and proliferation assays.

Following the final immunization, T-cell lines were established several times from both calves as described previously (9, 10). Briefly, 4 × 10⁶ PBMC were cultured in the presence of 5 μg of A. marginale (Florida strain) homogenate/ml in a final volume of 1.5 ml in complete RPMI 1640 medium in 24-well plates (Costar, Cambridge, Mass.). The cells were incubated at 37°C in a humidified incubator with 5% CO₂. Proliferating T lymphocytes were stimulated weekly with 5 μg of A. marginale homogenate/ml, 7.5 × 10⁵ T lymphocytes, and 2 × 10⁶ irradiated (3 kilorads) autologous PBMC as a source of antigen-presenting cells (APC).

T-lymphocyte proliferation assays were performed in duplicate wells of 96-well U-bottom plates (Costar) at 37°C in a humidified incubator containing 5% CO₂ for 4 days as described previously (9, 10). Briefly, each well (final volume, 100 μl) contained 3 × 10⁴ responder cells, 2 × 10⁵ APC, complete RPMI 1640 medium, and 0.2 to 5 μg of antigen/ml. The antigens used in this assay were A. marginale homogenate and affinity-purified native MSP1 complex (3, 44). Babesia bovis (crude membrane fraction) and membranes from uninfected bovine erythrocytes (URBC) were used as negative-control antigens (9). The lymphocytes were radiolabeled with 0.25 μCi of [³H]thymidine during the final 16 h of culture, harvested onto glass filters by using an automatic cell harvester, and counted with a β scintillation counter. Results are presented as mean counts per minute and as the stimulation index (SI), defined as the mean counts per minute of cells cultured with antigen divided by the mean counts per minute of cells cultured with medium. An SI of ≥3.0 was considered significant (5).

T-lymphocyte surface phenotypic analysis.

Differentiation markers on T-lymphocyte lines were analyzed by indirect immunofluorescence and flow cytometry as previously described (9). The MAbs used were specific for bovine CD3 (MAb MM1A), CD4 (MAb CACT 138 A), CD8 (MAbs CACT 80C and BAT 82B), and the δ chain of the γ/δ T-cell receptor (TCR) (MAb CACT 61A). William C. Davis, Washington State University, Pullman, kindly provided these MAbs.

Detection of IFN-γ in supernatants of Anaplasma-specific T lymphocytes.

T-lymphocyte lines were cultured for 24 h at a density of 2.0 × 10⁶ cells per ml with 2.0 × 10⁶ APC per ml and 5.0 μg of A. marginale homogenates prepared from the Florida strain of A. marginale/ml. Supernatants were harvested by centrifugation and stored frozen at −70°C. The bovine gamma interferon (IFN-γ) assay was performed by using a commercial ELISA kit (IDEXX Laboratories, Westbrook, Maine) according to the manufacturer’s protocol. The IFN-γ activity in culture supernatants diluted 1:4 and 1:20 was determined by comparison with a standard curve obtained with a supernatant from a Mycobacterium bovis purified protein derivative (PPD)-specific Th cell clone that contained 440 U of IFN-γ per ml (previously determined by the neutralization of vesicular stomatitis virus [9]).

RESULTS

Expression of pVCL/MSP1a in COS7 cells.

Figure 1A is an immunoblot demonstrating the binding of MAb Ana22B₁ to a protein at the expected molecular mass of 105 kDa and a slightly smaller protein in the pVCL/MSP1a-transfected cell lysate, but not to any protein in the untransfected cell lysate. The minor lower-molecular-mass proteins observed in Fig. 1A, lane 3 or in previous publications may represent processing of MSP1a (1, 33, 57). The IgG3 isotype control MAb (Fig. 1B) and the goat anti-mouse IgG alone (Fig. 1C) did not react with native or recombinant MSP1a.

FIG. 1.

Expression of MSP1a in COS7 cells by immunoblotting. Lanes: 1, 50 μg of COS7 cell lysate transfected with pVCL/MSP1a; 2, 50 μg of untransfected COS7 cell lysate; 3, 7.5 μg of A. marginale homogenate. (A) MAb Ana22b1; (B) IgG3 isotype control MAb; (C) no MAb; anti-mouse IgG only. Molecular mass markers (in kilodaltons) are shown on the left.

Antibody responses in BALB/c mice and calves immunized with pVCL/MSP1a.

Serum antibodies, collected at day 21, from the mice immunized with pVCL/MSP1a bound a 105-kDa protein in A. marginale homogenate, whereas there were no A. marginale-specific antibodies in preimmunization sera or sera from the four mice immunized with vector pVCL1010 (data not shown).

Six and eight weeks following initial immunization, serum antibody from both calves bound MSP1a in immunoblots (data not shown). Figure 2 shows the antibody reactivity of sera collected 16 weeks following the initial immunization. Only postimmunization sera had antibodies to the 105-kDa MSP1a. The bound lower-molecular-mass proteins resemble the stepwise pattern noted with both native and recombinant bacterial MSP1a (1, 33, 57). Also, it has been shown that only MSP1a possesses the neutralizing epitope defined by MAb 22B₁ and that MSP1a and MSP1b are antigenically distinct (3). Since A. marginale organisms were obtained from infected erythrocytes, the bands identified by preimmunization sera may be erythrocyte proteins recognized by antibodies in the sera of C749 and C751. As shown in Fig. 2, pre- and postimmunization serum antibodies did not bind other recognized MSPs, including the 19-kDa diagnostic antigen MSP5 (55, 58) of A. marginale. To further confirm the specificity of the immune responses of C749 and C751, antibodies in postimmunization sera but not preimmunization sera were shown to bind to affinity-purified recombinant MSP1a by immunoblotting (data not shown).

FIG. 2.

MSP1a antibody titers in sera from calves C749 and C751 at 16 weeks after the initial immunization. Each lane contains 7.5 μg of A. marginale homogenate. (A and B) Homogenate was reacted with C749 sera (A) or with C751 sera (B), consisting of preimmunization serum (1:50) (lanes 1), and postimmunization sera at dilutions of 1:200, 1:400, 1:800, and 1:1,600 (lanes 2 through 5, respectively). (C) Homogenate was reacted with control sera. Lanes: 1, positive-control bovine immune serum (B523, immunized with recombinant MSP1a) (1:100); 2, preimmunization serum (B523) (1:100); 3, protein G control. Molecular mass markers are indicated on the left in kilodaltons.

Additionally, the anti-MSP1a antibody titers of the calves at 16 weeks following initial immunization (shown in Fig. 2) were 1:1,600 and 1:800 for C749 and C751, respectively. The development of the antibody responses by the calves to MSP1a for weeks 0 through 18 post-initial immunization is presented in Table 1.

TABLE 1.

MSP1a serum antibody titers of calves immunized with pVCL/MSP1a

Wk postimmunization	Serum antibody titera of calf:
	C749	C751
0	NEGb	NEG
2	NEG	NEG
4	NEG	NEG
6	1:100	NEG
8	1:100	1:100
10	1:200	1:200
12	1:400	1:200
14	1:800	1:400
16	1:1,600	1:800
18	1:1,600	1:800

^aTiter presented as the last dilution showing binding to MSP1a in immunoblot. 

^bNEG, not detectable. 

Determination of the IgG isotypes of antibody to MSP1a in immunized calves C749 and C751.

To determine the IgG isotypes of antibody to MSP1a in calves C749 and C751, 8-, 14-, and 18-week postimmunization sera were tested by immunoblotting with isotype-specific, anti-bovine IgG1 and IgG2 MAb antibodies. The serum antibodies binding to native MSP1a were restricted to the IgG1 isotype (Fig. 3). Binding of IgG2 antibodies to native MSP1a was not detected, even at a 1:10 dilution. In contrast, positive-control serum, obtained from cow B523 immunized with affinity-purified recombinant MSP1a in saponin adjuvant (31), contained both MSP1a-specific IgG1 and IgG2 antibodies (Fig. 3A and B, lanes 8).

FIG. 3.

IgG isotype analysis of MSP1a antibody in calves C749 and C751 at 18 weeks after the initial immunization. Each lane contains 7.5 μg of A. marginale homogenate. Homogenate was reacted with anti-bovine IgG1 MAb (A) or anti-bovine IgG2 MAb (B). (A and B) Lanes: 1, preimmunization serum (1:10) (C749); 2 and 3, postimmunization serum (C749) at 1:10 and 1:50 dilutions; 4, preimmunization serum (1:10) (C751); 5 and 6, postimmunization serum (C751) at 1:10 and 1:50 dilutions; 7, preimmunization B523 negative-control bovine serum (1:100); 8, postimmunization B523 positive-control bovine serum (1:100). (C) Homogenate was reacted with antibody controls. Lanes: 1 and 2, postimmunization serum from C749 and C751, respectively (1:10); 3, B523 postimmune serum (1:100); 4, anti-bovine IgG1 MAb; 5, anti-bovine IgG2 MAb; 6, donkey anti-mouse IgG. Molecular mass markers are indicated on the left in kilodaltons.

MSP1a-specific T-lymphocyte proliferation.

Within 2 months following the fourth immunization, short-term T-lymphocyte lines were established from both calves by stimulation of PBMC with A. marginale homogenate. Two or more lymphocyte lines from each calf, tested several times during the 4-week culture period, proliferated in a dose-dependent manner in response to both crude A. marginale homogenate and affinity-purified native MSP1 complex. Phenotype analysis of T lymphocytes cultured with A. marginale for 5 weeks revealed equal numbers of CD4⁺ α/β T lymphocytes and γ/δ T lymphocytes. Proliferative responses to an optimal (5.0-μg/ml) concentration of A. marginale homogenate, native MSP1, or control B. bovis membrane antigen are presented in Table 2.

TABLE 2.

Proliferation of T lymphocytes from calves immunized with pVCL/MSP1a

Cell lineb	[3H]thymidine incorporation (SI)a in response to:
	Medium	B. bovis	A. marginale	MSP1
C749 (26)	808	943 (1.2)	40,571 (50.2)	3,648 (4.5)
C749 (28)	1,089	2,036 (1.9)	11,532 (10.6)	5,032 (4.6)
C751 (11)	966	1,203 (1.2)	6,313 (6.5)	8,123 (8.4)
C751 (28)	2,706	5,848 (2.2)	16,574 (6.1)	17,759 (6.6)

^aData are presented for T lymphocytes cultured for 4 days with 5 μg of antigen/ml or with medium alone, radiolabeled, harvested, and counted as described in Materials and Methods. Results are presented as mean counts per minute of duplicate cultures, with the SI given in parentheses. SI values greater than 3.0 (boldfaced) were considered significant (5). 

^bDesignated by the number of the animal, with the number of days the lines were in culture at the time of the assay given in parentheses. 

T-lymphocyte lines established from calves C749 and C751 at 4 weeks after the last immunization were tested for IFN-γ production. Levels of IFN-γ secreted at 4 weeks by these lymphocyte lines stimulated for 24 h with antigen and APC were 17 to 23 U/ml. The IFN-γ levels in the supernatants of irradiated PBMC from calves C749 and C741 cultured for 24 h with 5 μg of A. marginale/ml were below the detection limit (0.3 U/ml) of the assay.

DISCUSSION

This study demonstrates, for the first time, the induction in cattle of both antibody and cellular immune responses to a rickettsial gene delivered as a DNA vaccine. The construct, pVCL/MSP1a, expressed MSP1a in transfected COS 7 cells and induced immune responses in both BALB/c mice and cattle. Characterization of the immune response in calves given four inoculations of pVCL/MSP1a at weeks 0, 3, 7, and 13 showed that antibody titers (1:100) to MSP1a were first detectable at 6 and 8 weeks and peaked at 1:1,600 and 1:800 by 16 weeks after the initial immunization. Although a positive-control bovine serum contained both IgG1 and IgG2 antibodies with specificity for MSP1a, only IgG1 antibodies were detected in sera obtained following DNA vaccination. Short-term T-lymphocyte lines established from both calves after the final immunization responded by proliferation in a dose-dependent manner in response to A. marginale homogenate and affinity-purified MSP1.

DNA vaccines mimic intracellular viral infection and induce MHC class I and class II restricted T-lymphocyte responses, in addition to humoral responses (19). The protective immune mechanisms against bovine anaplasmosis are largely undetermined. Since A. marginale is an intraerythrocytic parasite, involvement of MHC class I restricted cytotoxic T-lymphocyte responses in protection is unlikely because mature erythrocytes do not express MHC molecules. However, as shown for other intracellular parasites, including Plasmodium spp., Leishamania major, and Toxoplasma gondii, T-helper lymphocytes, especially those producing type I cytokines, may be important in the elimination of A. marginale infection by activation of macrophages and augmentation of NO synthesis (27). In a related organism, mice immunized with a DNA vaccine containing major antigenic protein 1 (surface protein) of Cowdria ruminantium had a significantly higher survival rate than controls (38). Splenocytes from these vaccinated mice secreted higher levels of IFN-γ and interleukin 2 (IL-2) than controls (38).

To enhance the uptake of pVCL/MSP1a into muscle cells, the immunization sites of mice and cattle were pretreated with cardiotoxin 5 days prior to each immunization. Cardiotoxin obtained from the venom of Naja mossambica mossambica has been shown to induce skeletal muscle degeneration that is followed by muscle cell regeneration (16). The regenerating myoblasts efficiently take up and express plasmid DNA (17). Controversy exists as to the necessity of pretreatment with cardiotoxin or other necrotizing agents such as bupivacaine for the uptake and expression of DNA plasmids in muscles. It is thought that connective tissue, which encompasses mature myofibers, may act as a barrier to the uptake of DNA into muscle cells (17). The mechanism of antigen processing and presentation by muscle cells following DNA uptake and expression is also unclear. Since muscle cells do not express MHC class II antigens or costimulatory molecules such as B7 and CD40, the pathway of antigen presentation is unknown. It has been suggested that muscle cells release antigen locally, which is then processed by professional APC. Plasmid DNA may also be taken up and protein expressed directly by APC, which are attracted to the immunization site by inflammatory mediators (19, 51). Recently it was shown that immunization of plasmid DNA within transfected dendritic cells enhanced the immune response to plasmid-encoded proteins (29).

The number of inoculations of a DNA-encoding plasmid required for successful immunization is apparently variable. In our experiments, antibody was first detected after the second immunization in one calf and after the third in the other. In contrast, cattle immunized three times with plasmids encoding the Tams1-1 and Tams1-2 genes of Theileria annulata did not develop detectable antibody titers (18). However, cattle immunized five times with a plasmid encoding glycoprotein IV of bovine herpesvirus 1 did make a specific antibody, with titers similar to those of calves C749 and C751 immunized with pVCL/MSP1a (15).

Short-term T-lymphocyte cell lines established from C749 and C751 after the fourth immunization proliferated in a dose-dependent manner in response to A. marginale homogenate and native MSP1. Although there was variation in the proliferation of T lymphocytes from C749 and C751 to crude antigen and native MSP1, the proliferative responses were significantly greater than those induced by control antigens. T lymphocytes from 4 of 10 cattle similarly immunized with DNA encoding the herpesvirus glycoprotein D (gD) had significant proliferative responses to purified gD (2).

Although a discreet Th1/Th2 dichotomy has been described for murine CD4⁺ T lymphocytes (37), in cattle the dichotomy is not evident. Recent analysis of the cytokine profiles of B. bovis-, Fasciola hepatica-, and A. marginale-specific T-lymphocyte clones demonstrate the predominance of Th0 cells that coexpress IL-4 and IFN-γ (7, 10). Experiments with T-independent, polyclonal B-lymphocyte activation demonstrate that in cattle IL-4 enhances IgG1 and IgE production, whereas IFN-γ enhances IgG2 production (21, 22). More recent studies demonstrate stimulation of IgG2 by bovine B lymphocytes cocultured with autologous, antigen-specific CD4⁺ T-cell clones and antigen that was associated with T-cell production of IFN-γ (8). Most studies in mice show that intramuscular immunization with DNA vaccines elicits IFN-γ production and either an IgG2a-biased or a mixed IgG1 and IgG2a response (23, 46, 47). However, in certain cases IgG1 responses are induced by DNA vaccination, especially when intradermal injections are made with a gene gun (36, 48, 59). Because intramuscular inoculations were used in our studies, we were surprised by the finding that immunization of calves with pVCL/MSP1a stimulated a biased IgG1 response to MSP1a. These results may be explained by several variables, including route of delivery, the character of the antigen, the genetic background of the animals, and possibly the presence or absence of immunostimulatory sequences (ISS) within the plasmid-coding sequences (24, 47, 48).

Opsonization by peripheral blood monocytes or neutrophils in cattle is mediated by the IgG2 immunoglobin subclass (32). However, both IgG1 and IgG2 facilitated phagocytosis by peripheral blood monocytes that were cultured for 7 days (32). Also, hyperimmune serum from cattle immunized with MSP1 in complete Freund’s adjuvant resulted in opsonization and phagocytosis of A. marginale (12). Serum from animal B523 immunized with rMSP1a in saponin adjuvant contained both IgG1 and IgG2, whereas calves immunized with pVCL/MSP1a had a clearly biased IgG1 response. The ability of IgG1, elicited by immunization with pVCL/MSP1a, to mediate opsonization of A. marginale in vivo is not known.

DNA vaccines offer a method to direct the phenotype of the host’s immune response. It was recently shown that certain unmethylated ISS in bacterial DNA (CpG motifs) act to enhance a Th1 response in mice (24). This observation was further supported by the finding that immunization of mice with soluble protein and CpG-containing oligonucleotides as an adjuvant directs a Th1 response (13). It was recently demonstrated that both DNA and a CpG-containing oligonucleotide from B. bovis were mitogenic for bovine B lymphocytes (11). CpG motifs are defined as unmethylated CpG residues flanked 5′ by two purines (GA or AA) and 3′ by two primidines (TC or TT). Although there were seven CpG motifs (five GACGTC, one GACGTT, and one AACGTT) in the plasmid construct pVCL/MSP1a, the in vivo role played by these CpG motifs in this present study is unknown.

In conclusion, it was demonstrated that the plasmid vaccine encoding MSP1a of A. marginale induced antigen-specific seroconversion in mice and, more importantly, in cattle. The reactivity of immune serum with native and recombinant MSP1a suggested that the plasmid-expressed MSP1a shared B-cell epitopes with native and recombinant antigens. Furthermore, the DNA vaccine induced helper T-lymphocyte responses in vivo that were recalled ex vivo by stimulation with native antigen to produce IFN-γ. Although the antibody response to MSP1a of the two cattle reported here was restricted to IgG1, additional immunizations are necessary to determine the efficiency of pVCL/MSP1a to direct biased IgG1 responses. These studies form the foundation for designing plasmid-based vaccines for the control of anaplasmosis.