Identification and characterization of corpuscular, soluble and secreted antigens of a Venezuelan isolate of Anaplasma marginale

Veterinary Parasitology 94 (2000) 1–15

Identification and characterization of corpuscular, soluble and secreted antigens of a Venezuelan isolate of Anaplasma marginale Martha Leal a , Alfredo Noda a , Armando Reyna-Bello b , Beatriz Casas a , Eric Précigout c , Pedro M. Aso a , André Gorenflot c , Mary Isabel Gonzatti a,∗ b

a Departamento de Biolog´ıa Celular, Universidad Simón Bol´ıvar, Caracas 1080, Venezuela Centro de Estudios Biomédicos y Veterinarios, Universidad Simón Rodr´ıguez, Caracas 1041, Venezuela c Laboratoire de Biologie Cellulaire et Moleculaire, Faculté de Pharmacie, Université de Montpellier, Montpellier 1, Venezuela

Received 13 March 2000; received in revised form 3 August 2000; accepted 10 August 2000

Abstract Anaplasma marginale is the etiological agent of anaplasmosis, a tick-transmitted disease with an important economic impact that affects cattle throughout the world. Although, North American isolates of A. marginale and their antigens have been extensively studied, relatively little information is available on the antigenic composition of South American isolates. The characterization of diverse geographical isolates of A. marginale will result in a thorough antigenic profile and may lead to the identification of additional diagnostic and immunoprophylactic tools. Short-term cultures of a Venezuelan isolate (Ta) of A. marginale were maintained for up to 13 days in vitro. During that period, the A. marginale remained viable and were propagated in the bovine erythrocyte culture system. During the initial days of culture, cell division and reinvasion were evidenced by a significant rise in parasitemia up to a 50%. A. marginale antigens were identified by metabolic labeling with (35 S) methionine, followed by fractionation and immunoprecipitation with homologous and heterologous bovine sera. This yielded a complete antigenic set for the Ta isolate of A. marginale, including soluble, secreted and corpuscular polypeptide antigens. Fifteen immunodominant polypeptides were recognized by the bovine sera in the soluble and corpuscular fractions with relative molecular weights of 200, 150, 100–110, 86, 60, 50, 47, 40, 37, 33, 31, 25, 23, 19 and 16 kDa. Seven polypeptides were present in the exoantigen fraction. The 31 and 19 kDa antigens were recognized by the ANAR76A1 and ANAF16C1 monoclonal antibodies, respectively which are specific for MSP-4 and MSP-5 from North American isolates of A. marginale. Metabolic labeling ∗ Corresponding author. Tel.: +58-29063075; fax: +58-29063064. E-mail address: [email protected] (M.I. Gonzatti).

0304-4017/00/$ – see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 0 1 7 ( 0 0 ) 0 0 3 7 1 - X

2

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

with (14 C) glucosamine prior to immunoprecipitation with bovine sera allowed the identification of glycoprotein antigens of 200, 100–150, 60, 55, 50, 45–43, 37, 33, 31, 22, 19 and 16 kDa in the soluble fraction. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Anaplasma marginale; Antigens; Glycoproteins; Metabolic labeling; Short-term cultures

1. Introduction Anaplasmosis is a tick-borne rickettsial disease that affects cattle, causing serious economic losses, especially in tropical and subtropical areas but also in parts of the US (Ristic, 1960; McCallon, 1973; Guglielmone, 1995). Theiler (1910) was the first to recognize Anaplasma marginale as the etiological agent of anaplasmosis, a disease characterized by fever, depression, anorexia and acute anemia, concomitant with the appearance of internal marginal bodies near the periphery of the red blood cells (RBCs) (Ristic, 1960). Sequence analyses of small subunit ribosomal RNA places A. marginale within the ␣-proteobacteria group, next to Cowdria ruminantum and Ehrlichia chaffeensis (Weisburg et al., 1991; La Scola and Raoult, 1997). A. marginale belongs to genogroup II of pathogenic, tick-transmitted ehrlichia (Mahan et al., 1999). The comparison of these sequences with those of mitochondrial DNA shows that this group of obligate intracellular parasites are evolutionarily related to mitochondria (Gray, 1998). The disease process of anaplasmosis is initiated with the introduction of A. marginale either by ticks (Ristic, 1960; Kocan, 1995) or by biting flies which appear to be an important vector throughout Latin America (De R´ıos et al., 1990; Alonso et al., 1992; Guglielmone et al., 1997). Following the initial attachment to the surface of the RBC, A. marginale is internalized by an erythrophagocytic process and divides by binary fission inside a vacuole producing from two to eight ‘initial bodies’, surrounded by a parasitophorous membrane. (Francis et al., 1979; Giardina et al., 1983). The infective rickettsiae are subsequently released from the erythrocyte surface without discernible cell damage and proceed to infect neighboring RBCs. Several membrane surface polypeptides of North American isolates of A. marginale have been identified as protective antigens (Palmer and McGuire, 1984; Palmer et al., 1986). Four major surface proteins MSP-1, 2, 3 and 4 were initially recognized by (125 I)-cell surface iodination of a Florida strain of A. marginale, followed by immunoprecipitation with bovine immune sera (Palmer and McGuire, 1984). An additional membrane component MSP-5, was first detected in a Zimbabwe strain and later shown to be present and broadly conserved in North American (Tebele et al., 1991) and Venezuelan isolates (Reyna-Bello et al., 1998), as well as in other Anaplasma spp. isolates (Visser et al., 1992). Davis et al. (1978) demonstrated the incorporation of (14 C) methionine and (3 H) thymidine in A. marginale short-term cultures maintained in bovine erythrocytes. Kessler and Ristic (1979) showed replication of this ehrlichia parasite in bovine erythrocytes for up to 42 days. However, a long term in vitro culture of A. marginale has only been achieved in bovine erythrocytes co-cultured with endothelial cells (Waghela et al., 1997). At the present time, relatively little information is available on the antigenic characteristics of South American isolates. The analyses of isolates from this region will complement

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

3

the information on A. marginale antigens that is available from studies on North American isolates and may lead to the identification of additional antigenic candidates for diagnosis and immunoprophylaxis against anaplasmosis. The following studies were aimed at the identification of A. marginale antigens from a Venezuelan isolate by in vitro labeling, using a short-term bovine RBC culture system. Secreted, soluble and corpuscular antigens of A. marginale were identified and characterized by metabolic labeling followed by immunoprecipitation with sera from infected cattle 2. Materials and methods 2.1. Experimental infections The isolate of A. marginale used in this study was initially collected from a naturally infected calf in Táchira (Ta) state, in the western part of Venezuela. The Ta field isolate was cryopreserved in liquid nitrogen (Love, 1972) and used to infect splenectomized, male Holstein-Zebu or Brown-Swiss Zebu calves (4–10 months old). Calves used for the experimental infections, were clinically healthy and serologically negative for anaplasmosis by the ELISA test. For the experimental infections, 5 × 108 –109 parasites were inoculated, half intravenous and half intramuscularly. The animals were monitored every other day for clinical signs of the disease such as fever, decrease in the hematocrit and appearance of A. marginale organisms in blood smears. The percentage of parasitemia (%P) in the blood smears was determined by fluorescent staining, as previously described (Caballero, 1993). Briefly, a small aliquot of infected blood was mixed with a 100 ␮g/ml solution of ethidium bromide and acridine orange in phosphate buffer saline (PBS) pH 7.2 at a 9:1 ratio. A blood smear was prepared with this mixture and examined by fluorescent microscopy. 2.2. Short-term cultures Fresh bovine blood was collected on day 22 post-infection at a parasitemia of 17% and a hematocrit of 16. The infected RBCs were centrifuged at 1470×g for 10 min at room temperature and the buffy coat removed. The cells were washed twice with PBS and RPMI-1640 media (GIBCO Life Technologies) supplemented with 0.029% glutamine, 100 ␮g/ml of streptomycin and 100 IU/ml of penicillin (Sigma). Duplicate cultures were initiated by adding 0.5 ml of infected cells to 4.5 ml of RPMI-1640 media, supplemented with 20% fetal bovine serum (FBS). The cultures were established in T25 flasks (Costar) using a 5% CO2 incubator at 37◦ C. Every 24–48 h, an aliquot of infected RBCs from the culture was removed to determine the percentage of rickettsemia by fluorescent microscopy, as described (Caballero, 1993). On days 3 and 7, the cultured RBCs were centrifuged, the supernatant was removed and fresh, non-infected erythrocytes were added to restore the 10% hematocrit. 2.3. Antibodies and ELISA Two of the homologous, anti-Ta sera used for the immunoprecipitation assays (2432 and 2639) were collected from experimentally infected animals on days 6 or 7 post-peak of

4

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

parasitemia (ppP). Three other anti-Ta sera (2686, 2690 and 2692) were collected from a calf that was experimentally infected with 7–9 × 108 rickettsias on days 0, 30 and 126. Following the second and third infection, the animal developed two parasitemia maxima (15 and 10%) and was treated with oxytetracycline. Sera were collected during 6 months and tested by ELISA against purified anaplasma bodies. The 2686, 2690 and 2692 sera from days 3, 5 and 7 post-second peak of parasitemia, respectively, were selected for the immunoprecipitations on the basis of their high reactivity in the ELISA test (O.D. ≥ 1). Two anti-Yaracuy (Ya) sera were used as heterologous probes in the immunoprecipitation assays. These sera were obtained from the same bovine inoculated three times with a soluble antigenic fraction of A. marginale (Ya isolate), obtained as described by Aso (1985). The animal was challenged with the Ya isolate and sera (1972 and 1976) were collected on days 8 and 16 ppP and used in the immunoprecipitations. The ANAR76A1 and ANAF16C1 monoclonal antibodies which recognize the MSP-4 and MSP-5, respectively, were a kind gift of Dr. Guy H. Palmer from Washington State University. Sera from healthy, non-infected bovines that scored negative for anaplasmosis in ELISA, were used as negative control for the immunoprecipitation assays. 2.4. (35 S) methionine labeling A. marginale were metabolically labeled to distinguish intrinsic, rickettsia proteins from RBC components. Infected RBCs were isolated from bovines infected with the Ta isolate of A. marginale and labeled in vitro for 24 h using (35 S) methionine (NEN Life Science). In two separate experiments, bovine blood was collected at 16 or 32% parasitemia on days 24 or 32 post-infection. The infected RBCs were washed twice as described for the in vitro culture, except that supplemented RPMI 1640 media without methionine (GIBCO Life Technologies) was used for the final wash. The infected RBCs were distributed in T25 flasks at 10% hematocrit in 5 ml of supplemented RPMI 1640 media. The cultures were pre-incubated in the absence of methionine for 1–2 h at 37◦ C. Labeling was initiated by the addition of (35 S) methionine (50 ␮Ci/ml) and proceeded for an additional 24 h at 37◦ C. To confirm that the labeled proteins were indeed de novo synthetic products of A. marginale, control experiments were performed in the presence of 50 ␮g/ml oxytetracycline. This antibiotic blocks prokaryotic protein synthesis selectively and has no effect on the incorporation of (35 S) methionine into mammalian cells. 2.5. A. marginale fractionation and immunoprecipitation At the end of the 24 h period, the (35 S) methionine labeled cultures were fractionated according to the scheme presented in Fig. 1. All the steps were carried out at 4◦ C and three fractions S1, S2 and P1 were obtained and immunoprecipitated. The A. marginale cultures were first centrifuged at 1650 × g for 15 min, to separate the infected RBC pellet from the supernatant fraction S1. The S1 fraction was further clarified by centrifugation at 12 000 rpm for 10 min in an Eppendorf centrifuge, concentrated 10× through Centriprep (Amicon Corp.) and filtered (0.22 ␮m membrane) prior to immunoprecipitation. The pellet, containing the infected RBCs, was washed three times with Tris buffer saline (TBS 10 mM, Tris–HCl, pH 7.4 and 150 mM NaCl) and further fractionated into a soluble

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

5

Fig. 1. Fractionation scheme of (35 S) methionine labeled culture of A. marginale. Blood was collected from a bovine experimentally infected with the Ta isolate and the isolated RBCs were labeled with (35 S) methionine for 24 h as described.

(Aso, 1985) and a corpuscular fraction (Palmer and McGuire, 1984). The infected RBCs were lysed by freeze-thawing, followed by osmotic shock at 37◦ C for 5 min, using 3 volumes of TSHi (10 mm, Tris–HCl, pH 7.4, 20 mm NaCl, 1 mm PMSF and 5 mm EDTA) with 10% glycerol. The resulting homogenate was centrifuged at 6780 × g for 20 min and the supernatant was recentrifuged at 100 000×g for 1 h. The proteins present in this supernatant

6

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

were partially purified by double 60% (NH4 )2 SO4 precipitation to remove hemoglobin. The (NH4 )2 SO4 precipitate was collected by centrifugation at 10 280×g for 15 min, resuspended in PBS, concentrated and dialyzed against PBS by centriprep to render the soluble S2 antigenic fraction. Following erythrocyte lysis, the initial pellet (P1), containing the anaplasma bodies, was washed twice with TSHi buffer and recovered by centrifugation at 1650 × g for 15 min. The P1 was solubilized for 1 h on ice with lysis buffer (50 mm Hepes, pH 7.0, 0.1% SDS; 1% NP-40, 0.5% sodium deoxycholate, 1.25 mm PMSF and 5 mm EDTA) and clarified by centrifugation at 12 000 rpm for 15 min in an Eppendorf centrifuge. The radioactivity incorporated into protein in the S1, S2 and P1 fractions was quantitated by precipitation with trichloroacetic acid (TCA), followed by scintillation counting. Aliquots of the P1, S1 and S2 fractions, containing 4–8×106 TCA-precipitable counts, were immunoprecipitated overnight at 4◦ C, using either bovine sera (1:50 dilution), anti-MSP-4 (1:25 dilution) or anti-MSP-5 monoclonal antibodies (1:70 dilution). The final volume was adjusted to 0.5 ml with lysis buffer. The immune complexes were precipitated by the addition of protein A-Sepharose (Sigma) for 1–2 h at 4◦ C, collected by centrifugation and washed consecutively with lysis buffer (2×), high-salt buffer (2×) (50 mm Hepes, pH 7.0, 0.5 M LiCl, 1% ␤ME, 1.25 mm PMSF and 5 mm EDTA lysis buffer (1×) and buffer 50 mm Hepes pH 7.0 (1×) (Gonzatti-Haces et al., 1988). Immunoprecipitated proteins were analyzed by 12% SDS-PAGE (Laemmli, 1970) followed by fluorography. The gels were dried and exposed to autoradiography at −70◦ C. 2.6. (14 C) glucosamine labeling and immunoprecipitation Infected RBCs were collected at 30% parasitemia from a bovine experimentally infected with A. marginale (Ta). The (14 C) glucosamine labeling was performed as described for (35 S) methionine, except that 70 ␮Ci of (14 C) glucosamine (NEN Life Science) and complete RPMI 1640 media were used. Normal RBCs from a healthy, seronegative bovine were used as control. A soluble fraction of A. marginale (S2), containing 1.8 × 106 TCA-precipitable cpm, was immunoprecipitated with homologous and heterologous bovine sera at a 1:30 dilution.

3. Results 3.1. Short-term in vitro cultures The Ta isolate of A. marginale remained viable under the culture conditions that were used and was successfully maintained for up to 13 days in vitro (Fig. 2). The cultures were initiated at 17% parasitemia from fresh infected red blood cells and three maxima were observed. Every 3–4 days, the cultures were diluted down to 10% parasitemia, by the addition of non-infected bovine erythrocytes. During the initial 3 days of culture, cell division and reinvasion were evidenced by a significant 2.9× increase in parasitemia to a 50% peak. At that point, fresh normal RBCs were added and a new wave of rickettsemia followed, up to 40%. Following the second addition of RBCs, on day 7, a significant rise

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

7

Fig. 2. In vitro culture of bovine erythrocytes infected with A. marginale. On days 3 and 7, the parasitemia in the cultures was diluted to 10% by the addition of normal RBCs.

in the parasitemia was again observed with a 23% maximum by day 11. Afterwards, the parasitemia decreased and the cultures were terminated on day 13. 3.2. Metabolic labeling with (35 S) methionine, fractionation and immunoprecipitation In order to identify potentially relevant A. marginale antigens from the Ta isolate, proteins were labeled for 24 h with (35 S) methionine and three fractions were obtained (Fig. 1). Each fraction was immunoprecipitated with anti-Ta (homologous) or anti-Ya (heterologous) bovine sera. The labeled A. marginale antigens present in the S1 and S2 fractions were identified by immunoprecipitation with homologous anti-Ta sera (Fig. 3). Two major exoantigens of 100–110 and 19 kDa, as well as faint polypeptides of 86, 60 and 16 kDa were recognized by the anti-Ta serum 2686 in the culture media supernatant (S1) (Fig. 3(A), lane 1). When the same serum was used on the S2 soluble fraction, a much stronger immunoprecipitation was observed (Fig. 3(A), lane 2). The following polypeptides were observed: 100–110, 86, 60, 50, 25, 23, 19 and 16 kDa. The S1 and S2 fractions appear to share several antigens, including the 100–110, 86, 60, 19 and 16 kDa polypeptides (Fig. 3(A), lanes 1 and 2). Under these conditions, the pre-immune serum showed a few non-specific bands (Fig. 3(A), lane 3). In a separate labeling-immunoprecipitation experiment, two other homologous, anti-Ta sera 2639 and 2690 were tested against the S2 soluble fraction (Fig. 3(B), lanes 2 and 3). New S2 antigens were identified by these sera, including the 150, 47, 40, 37, 33 and 31 kDa. The signal corresponding to the 50 kDa antigen appears as a broad band with a slight compression, caused by albumin or Ig Heavy chain present as contaminant. None of these polypeptides were recognized by the pre-immune serum (Fig. 3(B), lane 1). The monoclonal antibody ANAF16C1 unequivocally identified the 19 kDa antigen present in S2 as MSP-5 (Fig. 3(B), lane 4).

8

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

Fig. 3. A. marginale excreted/secreted and soluble antigens recognized by homologous anti-Ta bovine sera. Panel A: the S1 (lane 1) and S2 fractions (lane 2) were immunoprecipitated with anti-Ta 2686. As a control, the S2 fraction was precipitated with preimmune serum (lane 3). Panel B: the S2 fraction was precipitated with preimmune serum (lane 1), anti-Ta 2639, 2690 (lanes 2 and 3) or with the ANAF16C1 monoclonal antibody (lane 4).

In order to analyze whether similar antigens were recognized by heterologous sera, the S1 and S2 fractions were immunoprecipitated using anti-Ya sera (Fig. 4). Four exoantigens of 60, 33, 25 and 19 kDa were recognized by the anti-Ya serum in the S1 fraction (Fig. 4(A), lane 1) and not by the pre-immune serum (Fig. 4(A), lane 2). In the presence of oxytetracycline, an insignificant level of (35 S) methionine incorporation was incorporated and no antigens were immunoprecipitated (data not shown). Immunoprecipitation of the S2 soluble fraction with the anti-Ya serum showed immunodominant antigens with relative molecular weights of 31, 25 and 19 kDa as well two broad bands between 37–60 and 80–110 kDa (Fig. 4(B), lane 1). In addition, two minor bands of 150 and 200 kDa were revealed. None of these proteins were recognized by the preimmune serum (Fig. 4(B), lane 2). Two homologous anti-Ta sera were then tested against the P1 fraction which contains Anaplasma bodies (Fig. 5(A)). One of the anti-Ta sera 2432 recognized five major antigens of relative molecular weights 100–110, 86, 60, 40 and 37 kDa and minor bands of 47, 33, 23 and 19 kDa (Fig. 5(A), lane 2). The other anti-Ta serum 2690 gave a strong positive signal for all these polypeptides and it also recognized two larger proteins of 200 and 150 kDa and two low molecular weight bands of 31 and 16 kDa (Fig. 5(A), lane 3). A very similar pattern was obtained when another anti-Ta serum 2686 was used, except for a weaker reactivity towards the 23, 19 and 16 kDa antigens (Fig. 5(B), lane 1).

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

9

Fig. 4. A. marginale excreted/secreted and soluble antigens recognized by the heterologous anti-Ya sera. Panel A: the S1 fraction was precipitated with anti-Ya 1976 or preimmune bovine sera (lanes 1 and 2). Panel B: the S2 fraction was precipitated with anti-Ya 1972 or preimmune sera (lanes 1 and 2).

Using the same P1 fraction, immunoprecipitation with a heterologous anti-Ya serum 1972 was then examined (Fig. 5(B), lane 2). Most of the antigens, that were previously identified in P1 using the homologous anti-Ta sera, were recognized. Interestingly, the 33 kDa polypeptide appeared as a doublet. The recognition appeared to be specific, since the preimmune serum yielded faint bands (Fig. 5(B), lane 3). The AnaR76A1 monoclonal antibody recognized a single polypeptide of 31 kDa in the P1 fraction (Fig. 5(B), lane 4). 3.3. (14 C) glucosamine labeling and immunoprecipitation The in vitro labeling of RBCs infected with A. marginale using (14 C) glucosamine, followed by immunoprecipitation with bovine sera, allowed us to identify antigenic glycoproteins. The infected erythrocyte lysate containing the soluble anaplasma fraction (S2), was tested against the anti-Ya and anti-Ta sera (Fig. 6). Several antigens with relative molecular weights of 200, 100–150, 60, 50, 47–45, 40, 37, 33, 27, 25 and 19 kDa, were recognized by the anti-Ya serum 1972 (Fig. 6, lane 1). The anti-Ta serum recognized bands of 220, 100–110, 60, 45, 37, 25–23, 19 and 16 kDa (Fig. 6, lane 2). No polypeptides were detected by the pre-immune serum (Fig. 6, lane 3).

10

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

Fig. 5. A. marginale corpuscular antigens recognized by homologous/heterologous bovine sera and with the ANAR76A1 monoclonal antibody. Panel A: the P1 fraction was immunoprecipitated with preimmune, anti-Ta 2432 or with 2690 bovine sera (lanes 1–3). Panel B: the P1 fraction was immunoprecipitated with anti-Ta 2686, anti-Ya 1972 or preimmune bovine sera (lanes 1–3). P1 fraction precipitated with the anti-MSP-4 monoclonal antibody (lane 4).

4. Discussion A culture of the Ta isolate of A. marginale was initiated from fresh infected red blood cells and was propagated in bovine erythrocytes for up to 13 days. An efficient multiplication/invasion cycle was attained during the initial 24 h, however, upon dilution with normal RBCs, lower peaks of rickettsemia were observed. Similar results were reported by Kessler and Ristic (1979) using bovine and ovine erythrocytes. An alternative in vitro culture system for A. marginale has been successfully applied by Waghela et al. (1997), by co-culturing erythrocytes with an endothelial cell monolayer. The continuous in vitro culture of A. marginale may require the addition of essential growth factors or the alteration of some critical physical parameter. Examples of other intraerythrocitic pathogens that can be propagated continuously in vitro, include the bacteria that cause Carrion’s disease, Bartonella bacilliformis (Walker and Winkler, 1981) and the ethiological agents of malaria, human, bovine and canine babesiosis, P. falciparum, Babesia divergens, B. bovis and B. canis, respectively (Trager and Jensen, 1976; Levy and Ristic, 1980; Schetters et al., 1992; Grande et al., 1997). An overview of the antigens recognized in the culture supernatant (S1), the soluble fraction (S2) and the corpuscular fraction (P1) of a Venezuelan isolate (Ta) of A. marginale

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

11

Fig. 6. (14 C) glucosamine-labeled A. marginale antigens recognized by homologous and heterologous sera. The S2 fraction was immunoprecipitated either with anti-Ya 1976, anti-Ta 2692 or with control bovine sera (lanes 1–3).

is presented in the accompanying Table 1. At least 15 polypeptides with apparent molecular weights of 200, 150, 100–110, 86, 60, 50, 47, 40, 37, 33, 31, 25, 23, 19 and 16 kDa were identified in the P1 fraction by the three anti-Ta homologous sera, 12 of these bands were also recognized by the heterologous anti-Ya serum. All the immunoprecipitated polypeptides present in P1 were also detected by the anti-Ta sera in the S2 fraction, except the 200 kDa antigen. Five to seven exoantigens (100–110, 86, 60, 33, 25, 19 and 16 kDa) were identified in the S1 fraction by homologous or heterologous sera, all of them were present in P1. The 60 and 19 kDa polypeptides were present in the three fractions and recognized by all the sera that were used. The results presented here confirm the parasitic origin of the exoantigens and the soluble antigens of A. marginale and corroborate the importance and relevance of these fractions. This opens new possibilities for the development of alternative strategies to neutralize this parasite. Amerault and Roby (1964) first demonstrated the presence of A. marginale exoantigen in serum from experimentally infected animals, as well as in RBCs. Aso (1985) subsequently compared soluble antigens and supernatants from short-term cultures and found them to be identical using an immunodiffusion technique. This suggests that the release of antigens that is observed in the short-term in vitro culture might also occur in vivo. It is possible that the exoantigens present in the culture supernatant in the short-term culture of A. marginale are released, as vesicles or blebs from the surface of the rickettsia. This is consistent with electron microscopic evidence (Giardina el al., 1983). The secreted

12

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

Table 1 Overview of the A. marginale antigens present in the S1, S2 and P1 fractions that are recognized by homologous (Ta) and heterologous (Ya) bovine sera kDa

S1 fraction 2686-Ta

200 150 100–110 86 60 50 47 40 37 33 31 25 23 19 16

+ +/− +/−

S2 fraction 1976-Ya

+

2686-Ta

+ + + +/−

+ + + +/−

+

+ + ++ ++

P1 fraction 2690-Ta

1972-Ya

2686-Ta

+/− +/− ++ ++ ++ ++

++ + + + +

+ ++ ++ + + +/− +/− +/− +/− + +

+ +

++

++

++ ++

++ ++ + + +/− +/− +/− +/−

2690-Ta + + ++ + + + + ++ ++ + + + + +

1972-Ya + ++ ++ + + + + + +/− +/− +/− +/−

proteins may play a role as virulence factors as has been shown in other pathogenic bacteria (Cao et al., 1998; Ibrahim et al., 1998; Pugsley, 1993; Vanet and Labigne, 1998). However, we cannot completely rule out that, along with the actively secreted proteins, the culture supernatant includes some somatic molecules released into the medium during replication or by autolysis. Here, we describe several immunodominant antigens of a Venezuelan isolate of A. marginale which were identified in the culture supernatant, soluble and corpuscular fractions. Some of these antigens appear to correspond to the 105, 86, 61, 31 and 15 kDa membrane proteins, previously described in a Florida isolate of A. marginale (Palmer and McGuire, 1984), suggesting that MSPs may be present in the parasite in more than one form, as integral membrane proteins, but also as soluble/released components. The results obtained in the immunoprecipitation assays with the ANAF16C1 monoclonal antibody suggest that the MSP-5 antigen, previously identified by Tebele et al. (1991) and cloned from North American strains of A. marginale (Visser et al., 1992) is a major immunodominant component of the S2 soluble fraction. The MSP-5 was first used as a diagnostic tool in a competitive inhibition ELISA in the United States (Knowles et al., 1996; Torioni de Echaide et al., 1998), other studies were performed in Brazil (Vidotto et al., 1998) and Australia (Molloy et al., 1999). A recombinant MSP-5, cloned from a Venezuelan isolate, has also been successfully used in an indirect ELISA (Reyna-Bello et al., 1998). The immunoprecipitation with the AnaR76A1 identified a 31 kDa polypeptide, that shares at least an epitope with the MSP-4 antigen, that has been identified and extensively studied in North American isolates of A. marginale (Palmer and McGuire, 1984; Oberle and Barbet, 1993). This protein is the product of a single gene and is highly conserved among different geographical isolates of A. marginale (Oberle et al., 1993). The MSP-4 is homologous to

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

13

the gene that encodes the immunodominant 32 kDa MAP1 protein of Cowdria ruminantium (van Vliet et al., 1994). The MSP-4 has also been found to be homologous to the multiple copy gene families p28 and omp1 from Ehrlichia chaffensis and Ehrlichia canis, respectively (Ohashi et al., 1998a; Ohashi et al., 1998b). The in vitro labeling with (14 C) glucosamine allowed the first identification of several antigenic glycoproteins in A. marginale. Polypeptides of 200, 100–150, 60, 55, 50, 45, 43, 37, 33, 31, 22, 19 and 16 kDa were labeled with (14 C) glucosamine in RBC infected with A. marginale and immunoprecipitated by homologous and heterologous sera. Some of these glycoproteins, namely the 125, 55, 33 and 31 kDa polypeptides were observed upon purification of A. marginale initial bodies by ConA-Sepharose chromatography (data not shown). In spite of the traditional view that no glycoproteins were present in prokaryotes, several recent reports of prokaryotic glycoproteins have been presented (Moens and Vanderleyden, 1997; Dobos et al., 1995). To our knowledge, this constitutes the first evidence for the presence of antigenic glycoproteins in A. marginale.

Acknowledgements We thank Dr. Guy H. Palmer from Washington State University, for his generous gift of the monoclonal antibodies. This work was supported by the BID-CONICIT project BTS-53 and the PCP program ‘Anticuerpos Monoclonales’, between the French Government and CONICIT.

References Alonso, M., Camus, E., Rodr´ıguez Diego, J., Bertaudière, L., Tatareau, J.C., Liabeuf, J.M., 1992. Situation actuelle des hemoparasitoses bovines en Martinique (Antilles francçaises). Rev. Elev. Méd. vét. Pays trop. 45, 9–14. Amerault, T.E., Roby, T.O., 1964. An exo-antigen of A. marginale in serum and erythrocytes of cattle with acute anaplasmosis. Am. J. Vet. Res. 25, 1642–1647. Aso, P.M., 1985. Isolation and immunochemical characterization of soluble A. marginale antigen. Ph.D. Thesis. University of Illinois, Urbana-Champaign. Caballero, H., 1993. A rapid, sensitive, simple and economical method for detection of A. marginale by fluorescent microscopy with acridine orange and ethidium bromide. In: Proceedings of the Ninth International Veterinary Hemoparasite Disease Conference. Merida, Mexico, p. 1. Cao, P., McClain, M.S., Forsyth, M.H., Cover, T.L., 1998. Extracellular release of antigenic proteins by Helicobacter pylori. Infect. Immun. 66, 2984–2986. Davis, W.C., Talmadge, J.E., Parish, S.M., Johnson, M.I., Vibber, S.D., 1978. Synthesis of DNA and protein by A. marginale in bovine erythrocytes during short-term culture. Infect. Immun. 22, 597–602. De R´ıos, L.G., Aguirre, D.H., Gaido, A.B., 1990. Infection naturelle par A. marginale chez deux troupeaux de bovins avec différents niveaux d’infestation par la tique Boophilus microplus. Rev. Elev. Med. vét. Pays trop. 43, 447–452. Dobos, K.M., Swiderek, K., Khoo, K., Brennan, P.J., Belisle, J.T., 1995. Evidence for glycosylation sites on the 45 kDa Glycoprotein of Mycobacterium tuberculosis. Infect. Immun. 63, 2846–2853. Francis, D.H., Kinden, D.A., Buening, G.M., 1979. Characterization of the inclusion limiting membrane of A. marginale by immunoferritin labeling. Am. J. Vet. Res. 40, 777–782. Giardina, S., Bretaña, A., Márquez, Q.N., 1983. Ultrastructural aspects of intraerythrocytic development of a Venezuelan strain of A. marginale. Tropenmed. Parasitol. 34, 7–10.

14

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

Gonzatti-Haces, M.I., Seth, A., Park, M., Copeland, T., Oroszlan, S., Vande Woude, G.F., 1988. Characterization of the TPR-MET oncogene p65 and the MET proto-oncogene cell surface p140 tyrosine kinases. Proc. Natl. Acad. Sci. USA 85, 21–25. Grande, N., Précigout, E., Ancelin, M.L., Moubri, K., Carcy, B., Lemesre, J.L., Vial, H., Gorenflot, A., 1997. Continuous in vitro culture of Babesia divergens in a serum-free medium. Parasitology 115, 81–89. Gray, M.W., 1998. Rickettsia, typhus and the mitochondrial connection. Nature 396, 109–110. Guglielmone, A.A., 1995. Epidemiology of babesiosis and anaplasmosis in South and Central America. Vet. Parasitol. 57, 109–119. Guglielmone, A.A., Abdala, A.A., Anziani, O., Mangold, A.J., Volpogni, M.M., Vanzini, V.R., 1997. Different seasonal occurrence of anaplasmosis outbreaks in beef and dairy cattle in an area of Argentina free of Boophilus microplus ticks. Vet. Quart. 19, 32–33. Ibrahim, A.S., Filler, S.G., Sanglard, D., Edwards, J.E., Hube, B., 1998. Secreted aspartyl proteinases and interactions of Candida albicans with human endothelial cells. Infect. Immun. 66, 3003–3005. Kessler, R.H., Ristic, M., 1979. In vitro cultivation of A. marginale: invasion of and development in noninfected erythrocytes. Am. J. Vet. Res. 40, 1777–1780. Knowles, D.P., Torioni de Echaide, S., Palmer, G.H., McGuire, T.C., Stiller, D., McElwain, T.C., 1996. Antibodies against an A. marginale MSP-5 epitope common to tick and erythrocyte stages identifies persistently infected cattle. J. Clin. Microbiol. 34, 2225–2230. Kocan, K.M., 1995. Targeting ticks for control of selected hemoparasitic diseases of cattle. Vet. Parasitol. 57, 121–151. La Scola, B., Raoult, D., 1997. Laboratory diagnosis of rickettsioses: current approaches to diagnosis of old and new rickettsial diseases. J. Clin. Microbiol. 35, 2715–2727. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 227, 680–685. Levy, M.G., Ristic, M., 1980. Babesia bovis: continuous cultivation in a microaerophilus stationary phase culture. Science 207, 1218–1220. Love, J.N., 1972. Cryogenic preservation of A. naplasma marginale with dimethyl sulfoxide. Am. J. Vet. Res. 33, 2557–2559. Mahan, S.M., Allsopp, B., Kocan, K.M., Palmer, G.H., Jonegan, F., 1999. Vaccine strategies for Cowdria ruminantum infections and their application to other ehrlichial infections. Parasitol. Today 15, 290–294. McCallon, B.R., 1973. Prevalence and economic aspects of Anaplasmosis. In: Jones, E.W. (Ed.), Proceedings of the sixth National Anaplasmosis Conference. Las Vegas, NV, USA, pp. 1–3 Moens, S., Vanderleyden, J., 1997. Glycoproteins in prokaryotes. Arch. Microbiol. 168, 169–175. Molloy, J.B., Bowles, P.M., Knowles, D.P., McElwain, T.F., Bock, R.E., Kingston, T.G., Blight, G.W., Dalgliesh, R.J., 1999. Comparison of a competitive inhibition ELISA and the card agglutination test for detection of antibodies to A. marginale and A. centrale in cattle. Aust. Vet. J. 77, 245–249. Oberle, S.M., Barbet, A.F., 1993. Derivation of the complete MSP-4 gene sequence of A. marginale without cloning. Gene 136, 291–294. Oberle, S.M., Palmer, G.H., Barbet, A.F., 1993. Expression and immune recognition of the conserved MSP-4 outer membrane protein of A. marginale. Infect. Immun. 61, 5245–5251. Ohashi, N., Zhi, N., Zhang, Y., Rikihisa, Y., 1998a. Immunodominant major outer membrane proteins of Ehrlichia chaffeensis are encoded by a polymorphic multigene family. Infect. Immun. 66, 132–139. Ohashi, N., Unver, A., Zhi, N., Rikihisa, Y., 1998b. Cloning and characterization of multigenes encoding the immunodominant 30 kDa major outer membrane proteins of Ehrlichia canis and application of the recombinant protein for serodiagnosis. J. Clin. Microbiol. 36, 2671–2680. Palmer, G.H., McGuire, T.C., 1984. Immune serum against A. marginale initial bodies neutralizes infectivity for cattle. J. Immunol. 133, 1010–1015. Palmer, G.H., Barbet, A.F., Kuttler, K.L., McGuire, T.C., 1986. Detection of an A. marginale common surface protein present in all stages of infection. J. Clin. Microbiol. 23, 1078–1083. Pugsley, A.P., 1993. The complete general secretory pathway in gram-negative bacteria. Microbiol. Rev. 57, 50–108. Reyna-Bello, A., Cloeckart, A., Vizca´ıno, N., Gonzatti, M.I., Aso, P.M., Dubray, G., Zygmunt, M., 1998. Evaluation of an enzyme-linked immunosorbent assay using recombinant major surface protein 5 (MSP-5) for serologic diagnosis of bovine anaplasmosis in Venezuela. Clin. Diagn. Lab. Immunol. 5, 259–262.

M. Leal et al. / Veterinary Parasitology 94 (2000) 1–15

15

Ristic, M., 1960. Anaplasmosis. Adv. Vet. Sci. 6, 111–192. Schetters, T.P., Kleuskens, J., Scholtes, N., Bos, H.J., 1992. Vaccination of dogs against Babesia canis infection using parasite antigens from in vitro culture. Parasite Immunol. 14, 295–305. Tebele, N., McGuire, T.C., Palmer, G.H., 1991. Induction of protective immunity by using A. marginale initial body membranes. Infect. Immun. 59, 3199–3204. Theiler, A., 1910. A. marginale (genus nov. et species nov.). Un nouveau protozoaire du bétail. Bull. Soc. Pathol. Exot. 3, 135–137. Torioni de Echaide, S., Knowles, D.P., McGuire, T.C., Palmer, G.H., Suarez, C.E., McElwain, T.F., 1998. Detection of cattle naturally infected with A. marginale in a region of endemicity by nested PCR and a competitive enzyme-linked immunosorbent assay using recombinant major surface protein 5. J. Clin. Microbiol. 36, 777– 782. Trager, W., Jensen, J.B., 1976. Human malaria parasites in continuous culture. Science. 193, 673–676. Vanet, A., Labigne, A., 1998. Evidence for specific secretion rather than autolysis in the release of some Helicobacyter pylori proteins. Infect. Immun. 66, 1023–1027. van Vliet, A.H., Jongejan, F., van Kleef, M., van der Zeijst, B.A., 1994. Molecular cloning, sequence analysis and expression of the gene encoding the immunodominant 32 kDa protein of Cowdria ruminantium. Infect Immun. 62, 1451–1456. Vidotto, M.C., Vidotto, O., Andrade, G.M., Palmer, G., McElwain, T., Knowles, D.P., 1998. Seroprevalence of A. marginale in cattle in Parana State, Brazil, by MSP-5 competitive ELISA. Ann. N. Y. Acad. Sci. 29, 424–426. Visser, E.S., McGuire, T.C., Palmer, G.H., Davis, W.C., Shkap, V., Pipano, E., Knowles, D.P., 1992. The A. marginale MSP-5 gene encodes a 19 kDa protein conserved in all recognized Anaplasma species. Infect. Immun. 60, 5139–5144. Waghela, S.D., Cruz, D., Droleskey, R.E., DeLoach, J.R., Wagner, G.G., 1997. In vitro cultivation of A. marginale in bovine erythrocytes co-cultured with endothelial cells. Vet. Parasitol. 73, 43–52. Walker, T.S., Winkler, H.H., 1981. Bartonella bacilliformis: colonial types and erythrocyte adherence. Infect. Immun. 31, 480–486. Weisburg, W.G., Barns, S.M., Pelletier, D.A., Lane, D.J., 1991. 16S ribosomal DNA amplification for phylogenetic study. J. Bacteriol. 173, 697–703.