WILL L. GOFF,'* WILLIAM C. DAVIS,2 GUY H. PALMER,2 TERRY F. McELWAIN 2t. WENDELL C. JOHNSON,' JOHN F. BAILEY,3 AND TRAVIS C. McGUIRE2 ... Three Babesia bovis merozoite surface proteins with relative molecular weights ...
INFECTION
AND IMMUNITY,
Sept. 1988, p. 2363-2368
Vol. 56, No. 9
0019-9567/88/092363-06$02.00/0 Copyright © 1988, American Society for Microbiology
Identification of Babesia bovis Merozoite Surface Antigens by Using Immune Bovine Sera and Monoclonal Antibodies GOFF,'* WILLIAM C. DAVIS,2 GUY H. PALMER,2 TERRY F. McELWAIN 2t WENDELL C. JOHNSON,' JOHN F. BAILEY,3 AND TRAVIS C. McGUIRE2 Animal Disease Research Unit, Agricultural Research Service, U.S. Department of Agriculture, Pullman, Washington 99164-70301; Department of Veterinary Microbiology and Pathology, College of Veterinary Medicine, Washington State University, Pullman, Washington 99164_70302; and Electron Microscopy Center, University of Idaho, WILL L.
Moscow, Idaho 838433 Received 4 April 1988/Accepted 30 May 1988
Three Babesia bovis merozoite surface proteins with relative molecular weights of 37,000, 42,000, and 60,000 identified by indirect immunofluorescence of live merozoites and by immunoprecipitation of '251-surfacelabeled merozoite proteins with immune bovine sera and monoclonal antibodies. These proteins were clearly of parasite origin, as evidenced by immunoprecipitation of metabolically labeled ([35SJmethionine) merozoites from cultures with specific antimerozoite monoclonal antibodies. In addition, two other proteins were identified with these methods. An 85-kilodalton protein was considered to be of parasite origin based on fluorescence reactivity with a monoclonal antibody. However, this protein was not detected after immunoprecipitation of metabolically labeled parasites, and thus, the exact nature of its origin is equivocal. A fifth protein of 145 kilodaltons was detected by immunoprecipitation after metabolic labeling but was not directly apparent on the surfaces of live merozoites. Since merozoite surface proteins may be important in the induction of protective immunity, those identified here are candidates for vaccine studies. were
parasitemia in peripheral blood was approximately 1.0%. Asynchronous cultures of B. bovis were maintained by previously described techniques (6, 13). Isolation of merozoites. Merozoites were harvested from cultures after the relative percentage of parasitized erythrocytes was increased by sequential reduction of the concentration of erythrocytes (6). For collection of merozoites, the contents of flasks containing >15% parasitized erythrocytes were centrifuged at 400 x g for 10 min at 4°C. The supernatant was centrifuged at 3,000 x g for 15 min at 4°C to pellet the merozoites. The merozoites were suspended in Puck saline-glucose (saline-G), and 2 ml was overlaid on 10 ml of a preformed continuous gradient of 65% Percoll-35% Puck saline-G. The gradient was centrifuged in a swinging bucket rotor at 3,000 x g for 20 min at 4°C. The merozoites were isolated from a band with an approximate density of 1.069 g/ ml between erythrocyte ghosts at the Percoll-Puck saline-G interface and the residual intact erythrocyte pellet. The merozoites were washed once in 0.15 M NaCl containing 0.01 M sodium citrate (CS), suspended in CS, and stored on ice until used (within 2 to 4 h). Quantitation and viability estimation of merozoites. An equal volume of the isolated merozoite suspension was mixed with 6-carboxy fluorescein diacetate (6-CFDA; final concentration in CS, 10 jig/ml; Calbiochem-Behring, La Jolla, Calif.) (18). The mixture was incubated at room temperature for 20 min, followed by centrifugation at 1,000 x g for 10 min, and was then suspended in phosphatebuffered saline (PBS; 0.15 M, pH 7.2) for counting on a hemacytometer. The sample was examined with phase microscopy and epifluorescence with a 40x oil objective and fluorescein filter (450 to 520 nm). Viability was assessed as the percentage of total merozoites emitting fluorescence. Electron microscopy of isolated merozoites. Merozoites harvested as described above were fixed in 1.5% glutaraldehyde. After 15 min of incubation on ice, the suspension was pelleted at 3,000 x g, postfixed in 1.0% osmium tetroxide,
More than half a billion cattle are estimated to be at risk of acquiring the tick-borne hemoprotozoan disease babesiosis (16). Current vaccine strategies include the use of attenuated live Babesia bovis parasites and various inactivated preparations (7, 10, 11, 15, 20, 21, 27, 28). The attenuated vaccine provides the best protection against challenge with both homologous and heterologous strains, although there are a number of serious disadvantages, including a short shelf-life, variation in virulence, contamination with host erythrocyte stroma, and perpetuation of the life cycle by creation of a carrier state (1). Inactivated vaccines induce protection against challenge with homologous strains; however, only partial protection occurs against challenge with heterologous strains (20, 28). Protective immunity in babesiosis may be directed against one or more surface antigens associated with sporozoites, infected erythrocytes, and/or merozoites. B. bovis merozoite proteins were selected for study because (i) malaria merozoite surface proteins induce protective immune responses (22, 26), (ii) in vitro cultures provide a reliable source of infective B. bovis merozoites with intact surfaces, and (iii) B. bovis merozoites have a morphologically distinct surface coat outside the unit membrane which is recognized by immune bovine sera (27). In this paper we identify three, possibly four, B. bovis merozoite surface proteins. MATERIALS AND METHODS Source and cultivation of parasites. The isolate of B. bovis was obtained in Mexico in 1979 from a Boophiluis microplus tick-induced infection of cattle. Blood stabilates were prepared and stored in liquid nitrogen (14) until 1983, when an infection was induced with a stabilate in a splenectomized calf. Blood was collected for in vitro culturing when the * Corresponding author. t Present address: Department of Infectious Diseases, University of Florida, Gainesville, FL 32610.
2363
2364
GOFF ET AL.
and taken through a graded series of ethanol solutions. The final pellet was embedded in Med-Cast (Ted Pella Co., Tustin, Calif.) embedding medium and polymerized at 70°C for 10 h. Thin sections were stained with uranyl acetate and lead citrate and examined with a Zeiss EM-10A transmission electron microscope. Preparation of MAbs. Approximately 107 harvested merozoites were inoculated intramuscularly into BALB/c mice with an equal volume of Freund complete adjuvant on day 1 and with Freund incomplete adjuvant on days 7 and 14. Four days prior to fusion, a similar number of merozoites was inoculated intravenously without adjuvant. Hybridomas were prepared as previously described (3, 19). The ratio of spleen cells to myeloma cells (X63 Ag8.653) was 2.5:1. Mouse thymocytes were used as feeder cells during hybridoma growth. Initial screening for antibody-producing hybrids was done 2 weeks after fusion by an indirect fluorescent-antibody test with acetone-fixed B. bovis-infected erythrocyte smears (5). Hybrids from positive wells were cloned twice by limiting dilution. Grow-to-die supernatants were prepared in culture flasks with each hybrid. Ascites fluid was prepared in pristane-primed BALB/c mice. Ouchterlony double diffusion was used to determine the monoclonal antibody (MAb) isotype, and the MAb concentration was determined by radial immunodiffusion. All selected MAb culture supernatants and ascites fluids were adjusted to 0.4 mg of immunoglobulin per ml. Preparation of immune bovine sera. Two spleen-intact Holstein-Friesian steers, 14 months of age and indirect fluorescent-antibody test negative for B. bovis, Babesia bigemina, and Anaplasma marginale, were inoculated intravenously with approximately 6 x 108 B. bovis-infected erythrocytes from the same blood stabilate used to initiate in vitro cultures. On day 9 postinoculation each steer developed detectable parasitemia and a febrile response which persisted through day 13 postinoculation. Antibody specific for B. bovis was detected with the indirect fluorescentantibody test on day 10 postinoculation. The steers were challenge inoculated as before on days 48 and 80 postinoculation, and although the animals did not develop a fever or parasitemia, the antibody titer increased after each challenge. Sera were collected and stored at -70°C after the final challenge, when the indirect fluorescent-antibody test titer was 1:10,000. Immunofluorescence of live merozoites. Viable merozoites were collected as described above and reacted with the various antibodies by a modification of a previously described technique (18). The MAb-containing ascites fluids were diluted 1/10 (40 p±g/ml) in PBS. Immune bovine sera were diluted 1/10 in PBS. An equal volume of each antibody preparation was added to 100 ,ul of a merozoite suspension and incubated on ice for 1 h. Each sample was centrifuged at 3,000 x g for 10 min, and the merozoites were washed twice in cold PBS. The samples were then suspended in the appropriate rhodamine-conjugated second antibody (1/40 dilution in PBS) (Kirkegaard and Perry, Inc., Gaithersburg, Md.) and incubated on ice for 1 h. After being washed, the merozoites were suspended in 6-CFDA and incubated for 20 min at room temperature. The merozoites were then centrifuged at 3,000 x g for 10 min at 4°C, suspended in 50 ptl of PBS, and examined in wet mounts with appropriate filters for rhodamine (antibody binding) and fluorescein (6-CFDA viability) (546 to 590 nm and 450 to 520 nm, respectively). To determine whether trypsinization of merozoites would alter the antibody reactivity under live immunofluorescence conditions, isolated merozoites were incubated for 8 min at
INFECT. IMMUN.
37°C in CS containing 0.125% (final concentration) trypsin (GIBCO Laboratories, Grand Island, N.Y.). The reaction was terminated by placing the sample on ice, adding an equal volume of 1% (vol/vol) trypsin inhibitor (type 11-0; Sigma Chemical Co., St. Louis, Mo.), and washing the merozoites once in CS containing 1% trypsin inhibitor and once in CS alone. Live immunofluorescence was then carried out as before. Metabolic labeling of B. bovis. Fresh subcultures of B. bovis were allowed to grow for 24 h. The medium was then removed and replaced with methionine-free medium (6) containing 20 p.Ci of [35S]methionine (specific activity, 1,109 Ci/mmol; New England Nuclear Corp., Boston, Mass.) per ml. The labeled cultures were incubated for an additional 24 h. The cells were pelleted at 1,000 x g, washed in PBS, and dialyzed against PBS with 0.02% (wt/vol) sodium azide. Surface labeling of merozoites with 1251. In vitro-cultured merozoites were isolated as described above, washed in PBS, and surface labeled with a lactoperoxidase radioiodination technique (24). Immunoprecipitation. Metabolically labeled infected erythrocytes and surface-labeled isolated merozoites were solubilized with reagents and methods previously described (18). Antisera and MAbs were used to precipitate proteins by previously described methods (18). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography. The immunoprecipitates were electrophoresed on 7.5 to 17.5% polyacrylamide continuous gradient gels under reducing conditions, and the gels were dried and subjected to autoradiography as previously described (12). Molecular mass standards included '4C-labeled myosin (200 kilodaltons [kDa]), phosphorylase b (92.5 kDa), bovine serum albumin (69 kDa), ovalbumin (46 kDa), carbonic anhydrase (30 kDa), and lysozyme (14.3 kDa) (Amersham Corp., Arlington Heights, Ill.). RESULTS Binding of MAbs to fixed merozoites. The MAbs were first screened on fixed infected erythrocyte preparations, and four MAbs were selected for further evaluation because of their distinctive patterns of fluorescence. These patterns included staining of the merozoite cytoplasm and membrane (BABB 35), merozoite membrane (BABB 90), and merozoite cytoplasm (BABB 93) and a single, punctate reaction appearing polar in location on merozoites (BABB 75) (Table 1). The four MAbs all retained their original specificities after the hybridoma cells were cloned twice and used to produce ascites fluids. Isolation of live merozoites. Spontaneously released merozoites for surface binding and labeling experiments were isolated on Percoll gradients. A large proportion retained their surface coat (Fig. 1), and >80% were viable, as determined by 6-CFDA staining. On three occasions, 108 of these isolated merozoites were inoculated intravenously into susceptible, splenectomized calves. In each case, infection was achieved with a prepatent period similar to that in calves that received an equivalent number of infected erythrocytes from a blood stabilate. Also, in vitro cultures have been routinely reestablished after introduction of these isolated merozoites. Binding of MAbs to live merozoites. Figure 2 shows the reactions obtained with each of the four MAbs under live, dual-fluorescence conditions. BABB 35 reacted with 100% of the viable merozoites (Fig. 2A and B), BABB 75 reacted with approximately 5% of the viable merozoites (Fig. 2C and
VOL. 56, 1988
BABESIA BOVIS MEROZOlTE SURFACE ANTIGENS
2365
TABLE 1. Characterization of four MAbs specific for B. bovis merozoites
BABB BABB BABB BABB ' b
Immunofluorescence
Immunoglobulin isotype
MAb
35 75 90 93
Fixed"
Cytoplasmic and membrane Punctate, polar Membrane Cytoplasmic
G2a G2b
Gl G2a
Liveb Positive Positive Positive Negative
Mr(S) (103) of recognized protein(s) 37 and 42 60 85 145
Acetone-fixed infected erythrocyte smears. Viable merozoites as determined by 6-CFDA conversion to free fluorescein (see the text).
D), BABB 90 reacted with approximately 90% of the viable merozoites (Fig. 2E and F), and BABB 93 failed to bind live merozoites (Fig. 2G and H). Although trypsinization of isolated merozoites did not alter their viability, the treatment did prevent BABB 35, BABB 75, and BABB 90 from reacting with live merozoites, providing further evidence for a surface location of the protein epitopes recognized by these three MAbs. Immunoprecipitation of surface-radioiodinated proteins. Radioiodinated merozoite preparations were immunoprecipitated with the four MAbs described above to confirm the outer surface or cytoplasmic location of the reactive epitopes. BABB 35 precipitated a major protein of 42 kDa and a minor protein of 37 kDa (Fig. 3, lane 1). BABB 75 and BABB 90 precipitated single proteins of 60 and 85 kDa, respectively (Fig. 3, lanes 2 and 4). In contrast, BABB 93 did not precipitate any detectable protein (Fig. 3, lane 3).
;SC
IM
PM
0
9 a
FIG. 1. Electron micrograph of a spontaneously released, Percoll-isolated B. bovis merozoite. N, Nucleus, R, rhoptry; IM, inner membrane; PM, plasma membrane; SC, surface coat.
Immunoprecipitation of metabolically labeled merozoite proteins. Three of the four MAbs each specifically precipitated [35SJmethionine-labeled proteins (Fig. 4), demonstrating the parasite origin of four reactive proteins (uninfected bovine erythrocytes failed to incorporate [35S]methionine; data not shown). These included the 42-kDa major and 37-kDa minor proteins (Fig. 4, lane 4, BABB 35), the 60-kDa protein (Fig. 4, lane 5, BABB 75), and the 145-kDa protein (Fig. 4, lane 7, BABB 93). No [35S]methionine-labeled protein was precipitated by BABB 90 (Fig. 4, lane 6). To determine which parasite proteins were recognized by the bovine immune system, we used twofold serial dilutions of immune bovine sera to immunoprecipitate metabolically labeled preparations. Proteins with relative molecular masses ranging from approximately 16 to >200 kDa were recognized by the immune bovine sera (Fig. 5). Among the proteins recognized were those identical in molecular mass to those precipitated by BABB 35 (42 and 37 kDa), BABB 75 (60 kDa), and BABB 93 (145 kDa). In addition, proteins of 145, 42, 120, and 75 kDa appeared to be immunodominant, as they were precipitated by immune bovine sera at the greatest dilution tested (Fig. 5, lanes 7 and 13).
DISCUSSION These studies defined three, possibly four, surface-exposed proteins of B. bovis merozoites (37, 42, 60, and possibly 85 kDa). The evidence that the proteins are surface exposed is as follows: (i) MAb binding of live merozoites, (ii) labeling by surface iodination, and (iii) sensitivity to mild
trypsinization. Two proteins (37 and 42 kDa) were immunoprecipitated by a single MAb after metabolic labeling and after surface iodination of isolated merozoites. It is possible that they are immunoprecipitated as a complex or that they are processed products of a higher-molecular-weight precursor and share a common epitope, as has been described with Plasmodium falciparum (4, 8) and B. bigemina (18) proteins. Immunofluorescence studies demonstrated that the 42- and 37-kDa proteins are present on all viable merozoites and are distributed over the entire surface of the organism. The 60-kDa protein appeared to be punctate, polar, and restricted to only about 5% of the total viable merozoite population, as assessed by live immunofluorescence. In contrast, the specific MAb reaction on fixed parasites, while punctate in appearance, occurred with virtually all merozoites. The detection of this protein on a small proportion of isolated viable merozoites may indicate that the protein is restricted to a subpopulation of merozoites free from erythrocytes. Alternatively, this protein may be rapidly processed and secreted from the majority of the merozoites, since the merozoites were free for over 2 h during the live labeling. This latter alternative has been suggested for P. falcipar-um rhoptry proteins (2, 9), which yield similar fluorescence
2366
GOFF ET AL.
INFECT. IMMUN.
200,000
-
92,500 -
-
69,000
85 KDa
_no 60 KDa .4*-42 KDa
46,0000
--^^ 37 KDa
30,000 _ 14,300
>
12 3 4
5
FIG. 3. Immunoprecipitation of 125I-surface-labeled B. bovis merozoites. Isolated merozoites were radioiodinated by the lactoperoxidase-catalyzed method. Immune complexes were precipitated after reactions with BABB-35 (lane 1), BABB-75 (lane 2), BABB-93 (lane 3), BABB-90 (lane 4), and an irrelevant MAb specific for Trypanosoma brucei (lane 5). Molecular mass standards are indicated at the left.
reactions and which induce protection against plasmodial infections (23, 25). Whether the 60-kDa protein is associated with B. bovis rhoptry proteins is currently under investigation. The 85-kDa protein met the three criteria for surface location. However, this protein was not identified by immunoprecipitation after metabolic labeling. Evidence suggests that this protein is of parasite origin because (i) live merozoites, but not erythrocyte ghosts, were labeled by live fluorescence with BABB 90; (ii) BABB 90 recognized only the parasites and not uninfected or infected erythrocyte components by fixed blood smear immunofluorescence; and (iii) the 85-kDa protein was labeled by surface iodination of merozoites. This protein may lack sufficient methionine and thus would not be labeled adequately for detection by our methods. Metabolic labeling of cultures with amino acids other than methionine may clarify this point. Until then, the origin of this protein remains equivocal. A fifth protein (145 kDa) was of parasite origin, but its location on the membrane surface was not directly apparent.
.-145 KDa :.
-460 KDa FIG. 2. Viability and MAb binding of isolated B. bovis merozoites after reaction with primary MAb ascites fluids and rhodamine-labeled second antibody, followed by incubation in 6CFDA. Panels A, C, E, and G represent a field from each MAb- and 6-CFDA-treated preparation with the wavelength optimized for fluorescein, and panels B, D, F, and H represent the corresponding fields with the wavelength optimized for rhodamine. (A and B) MAb BABB 35 (42 and 37 kDa). (C and D) MAb BABB 75 (60 kDa). (E and F) MAb BABB 90 (85 kDa). (G and H) MAb BABB 93 (145 kDa). The reactions were photographed at a magnification of 400 x. The large fluorescing particles are agglutinated merozoites, and the small particles are individual merozoites.
0 u.
-K42 KDa
m_ _I 2 3 4 5 6 7 FIG. 4. Immunoprecipitation of B. bovis-specific proteins with epitopes recognized by three individual MAbs. Proteins from solubilized B. bovis-infected erythrocytes were immunoprecipitated, after in vitro [35S]methionine metabolic labeling, with MAbs BABB35 (lane 4), BABB-75 (lane 5), BABB-90 (lane 6), and BABB-93 (lane 7). Proteins of 37, 42, 60, and 145 kDa were also precipitated by an immune bovine serum (lane 3). No parasite proteins were precipitated by the autologous preinoculation bovine serum (lane 2). Molecular weight standards are shown in lane 1 (see the text for individual molecular masses).
. ~ ~ ~ ^;~ ,jg.~ .
VOL. 56, 1988
* j[; .>E.:[4. .
W > gt
1
2
3 4 5
8
7 8 9
j
li| 0|
BABESIA BOVIS MEROZOITE SURFACE ANTIGENS
.145 :...T KDa
