Molecular and Biochemical Parasitology, 47 ( 1991 ) 213-222
213
© 1991 Elsevier Science Publishers B.V. /0166-6851/91/$03.50 ADONIS 016668519100247Y MOLBIO 01565
Molecular characterization and immunogenicity of neutralization-sensitive Babesia bigemina merozoite surface proteins T e r r y F. M c E l w a i n ~'2, L a n c e E. P e r r y m a n ~, A n t o n y J. M u s o k e 3 and Travis C. M c G u i r e ~ IDepartment of Veterinary Microbiology and Pathology, Washington State University, Pullman, WA, U.S.A.; 2Departmentof Infectious Diseases, College of Veterinary Medicine, University of Florida, Gainesville, FL, U.S.A.; and 3International Laboratory for Research on Animal Diseases, Nairobi, Kenya (Received 29 November 1990; accepted 13 March 1991)
Monoclonal antibodies binding to the surface of live Mexico isolate Babesia bigemina merozoites have defined 4 parasite-encoded surface antigens (36, 45, 55, and 58 kDa) that are potential targets for immune-mediated neutralization of merozoites. In this study, we have characterized the post-translational modification, antigenic polymorphism, and immunogenicityof these 4 proteins. Monoclonal antibody immunoaffinity-purified36- and 55-kDa polypeptides were identical in gel electrophoresis to immunoprecipitated radiolabeled proteins, while the purified 45-kDa protein consisted of 2 closely spaced polypeptides with relative molecular weights of 45 and 43 kDa. The 36-, 45-, and 55-kDa proteins were post-translationally modified by incorporation of [3H]glucosamine and [3H]myristic acid, suggesting they are integral membrane proteins secured by a phosphatidylinositol anchor. Cross-reactivity studies with monoclonal and monospecific polyclonal antibodies revealed marked antigenic polymorphism of these 3 glycoproteins among diverse geographic isolates. In contrast, none of the polypeptides bound by anti-p58 monoclonal antibody were glycosylated or myristilated. Both monoclonal and monospecific polyclonal antibodies recognizing p58 bound to similar molecular weight proteins in 4 additional isolates ofB. bigemina from Mexico, Puerto Rico, St. Croix, and Kenya, suggesting widespread conservation of p58 immunogenic epitopes among geographic isolates. Calves immunized with immunoafffinitypurified gp45, gp55, or p58 antigens were able to neutralize the infectivity of merozoites as indicated by significant reductions in the peak parasitemia after experimental challenge. Precise definition and appropriate presentation of neutralization sensitive epitopes on gp45, gp55, or p58 may enhance the merozoite neutralizing immune response in immunized cattle. Key words: Babesia bigemina; Merozoite surface protein; Antigenicdiversity; Post-translational modification; Immunogenicity
Introduction Bovine babesiosis is an intraerythrocytic protozoan disease that imposes a severe constraint on livestock production, especially in areas where enzootic stability is not maintained [ 1]. In tropical and subtropical climates the disease is caused by 2 species, Babesia bigemina and Babesia bovis, that are Correspondenceaddress: Terry F. McElwain, Department of Veterinary Microbiology and Pathology, Washington State University, Pullman, WA 99164-7040, U.S.A.
Abbreviations: SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; mAb, monoclonal antibody; PBS, phosphate-buffered saline; IFA, indirect fluorescent antibody assay; ELISA, enzyme-linkedimmunosorbentassay
transmitted by Boophilusspp. ticks [ 1]. The impact of babesiosis is currently reduced by a combination of acaricide treatment to decrease disease transmission by ticks, chemotherapy to ameliorate clinical signs, and premunition using an in vivo derived and attenuated live vaccine which prevents serious clinical disease after challenge [2]. Each of these techniques has biological and technical drawbacks that limit its utility, especially within countries where the prevalence of disease is high [3]. Our hypothesis is that an antigenically defined vaccine can target the immune response to neutralization-sensitive epitopes, enabling practical and effective immunoprophylaxis against bovine babesiosis. Accessible targets of the protective immune response against babesial parasites include extra-
214
cellular infective stages from tick salivary glands, extracellular merozoites, and infected erythrocytes. Surface exposed regions of membrane proteins on any of these targets may contain epitopes sensitive to neutralization. Exposed merozoite membrane antigens are therefore being investigated for their ability to induce protective immunity. Five B. bigemina-encoded merozoite membrane proteins with surface exposed epitopes have been previously defined with monoclonal antibodies [4]. All 5 proteins are recognized by antibodies in sera from cattle that are immune to babesiosis, but no further evidence of an individual antigen's ability to induce a protective immune response is available. In the current study, 4 of the 5 B. bigemina merozoite proteins with surface exposed epitopes are further characterized as to size, epitope conservation among isolates, and post-translational modification. Each of them can be purified by immunoaffinity chromatography in quantities sufficient for immunizing cattle. Their potential as immunogens is evaluated by their ability to induce an immune response that neutralizes merozoites, reducing the peak level of parasitemia in experimentally challenged cattle.
Materials and Methods
B. bigemina isolates. Mexico, Texcoco, St. Croix, Puerto Rico, and Kenya isolates were maintained as cryopreserved stabilates as described [5]. Texcoco, St. Croix, and Puerto Rico isolates were provided by G. Buening, University of Missouri, Columbia, MO [6].
Microaerophilous stationary phase culture/ biosynthetic radiolabeling. Parasite proteins were radiolabeled with [35S]methionine in short term culture as previously described [4], or by incorporation of [35S]methionine, [3H]amino acids, [3H]glucosamine, or [3H]myristic acid into parasites grown in continuous culture. Cultures were initiated with Mexico isolate B. bigemina and maintained as described [7]. For metabolic radiolabeling, 1.0 ml of culture medium overlying the infected erythrocytes was replaced with 200400 flCi of lyophilized [35S]methionine (>800 Ci mmol -~ in 20 mM potassium acetate containing
0.1% 2-mercaptoethanol, Amersham, Arlington Heights, IL), [3H]L-amino acid mixture (in 0.1N hydrochloric acid, Dupont, Wilmington, DE), [3H]glucosamine hydrochloride ( 3 0 ~ 0 Ci mmol in aqueous buffer, Dupont), or [3H]myristic acid ( 10-60 Ci mmol -~ in ethanol, Dupont) resuspended in 1.0 ml medium. Radiolabeled parasites were harvested 16-18 h later, washed, and lysed in 1% (v/v) Nonidet P-40 (LKB, Bromma, Sweden) as previously described [4].
Radioimmunoprecipitation, immunoblotting and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Immunoprecipitation of radiolabeled antigen was performed as previously reported [4]. For immunoblotting, 1 x l0 s gradientseparated merozoites [4] were electrophoresed in 7.5-17.5% gradient SDS-PAGE and transferred to nitrocellulose [8]. Immunoblots using bovine antibodies detected with 125I-labeled protein-G or rabbit antibodies detected with ~25I-labeled protein-A were performed as described [4,9,10]. Nitrocellulose was blocked with either 0.25% (v/v) Tween20/0.25% (w/v) gelatin in veronal-buffered saline [9] or 10% (w/v) milk in 20 mM Tris-HC1/150 mM NaCI. Gradient polyacrylamide gel techniques, SDS-PAGE sample buffer, and autoradiography have been previously described [11]. ~4C-labeled proteins used for molecular weight comparisons (Amersham, Arlington Heights, IL) consisted of myosin, 200 kDa; phosphorylase b, 92.5 kDa; bovine serum albumin, 69 kDa; ovalbumin, 46 kDa; carbonic anhydrase, 30 kDa; and lysozyme, 14.3 kDa.
Purification of monoclonal antibodies and coupling to Sepharose beads. The preparation and screening of mAbs 14.1, 14.16, 14.20, and 14.52, which recognize B. bigemina merozoite surface proteins, have been described elsewhere [4]. Ascitic fluid was generated in pristane-primed BALB/c mice with twice cloned hybridoma cells. Purification ofmAbs was performed by diethylaminoethyl cellulose chromatography (DE-52; Whatman Ltd., Maidstone, U.K.) of 50% ammonium sulfate precipitated immunoglobulin (from ascitic fluid) in 0.032 M Tris, pH 7.4 [ 12]. Purity of the isolated mAbs was established in Coomassie bluestained gradient polyacrylamide gels loaded with
215 50/tg of protein from column fractions in each lane. Purified mAbs were coupled to cyanogen bromideactivated Sepharose 4B (10 mg protein to 1 ml swollen beads) according to manufacturer's recommendations (Pharmacia Fine Chemicals, Uppsala, Sweden).
lmmunoaffinity chromatography.
Whole blood collected from splenectomized calves experimentally infected with the Mexico isolate of B. bigemina was collected at peak parasitemia, washed 3 times with 0.01 M sodium phosphate/0.15 M NaC1, pH 7.2 (PBS) to remove leukocytes, and stored at -70°C as a blood stabilate containing packed erythrocytes 1:1 (v/v) with a cryopreservant of 10% (w/v) polyvinylpyrollidone and 2% (w/v) glucose (crude antigen) [5]. Crude antigen washed 3 times at 22 000 x g with cold PBS containing 50 mM phenylmethylsulfonylfluoride was solubilized in buffer containing 50 mM Tris/5 mM EDTA/5 mM iodoacetamide/ 1 mM phenylmethylsulfonylfluoride/ 0.1 mM N-ct-p-tosyI-L-lysyl chloromethyl ketone/ 1% (v/v) Nonidet P-40 (lysis buffer) and centrifuged for I h at 135 000 × g. The supernatant was filtered through a 0.45-micron membrane and sonicated on ice. Solubilized antigen was passed twice through the mAb-coupled Sepharose 4B columns hooked in series (750 ~1 of packed wet beads each). The columns were washed sequentially with 0.02 M Tris/0.005 M EDTA/0.1 M NaC1/ 0.015 M NAN3, pH 7.6 containing I% (v/v) NP-40 and 0.1 mM phenylmethylsulfonylfluoride; and the same buffer without detergent. They were then preeluted individually with 10 ml of 0.1 M glycineNaOH/ 1 M NaC1, 0.5% (w/v) deoxycholate, pH 10.0 [ 13]. Bound proteins were eluted from individual columns with 5 ml of 0.05 M diethylamine/ 0.5% (w/v) deoxycholate, pH 11.5 directly into siliconized test tubes containing 0.5 ml of 1 M Tris, pH 8.5; and dialyzed against PBS to remove diethylamine and detergent. Eluted, dialyzed proteins were quantitated by Lowry protein microassay according to Peterson [ 14], performed in microtiter plates after elimination of the precipitation step and 4-fold reduction of reagent volumes, or bicinchoninic acid technique (Pierce Chemical, Rockford, IL). Purity of the eluted proteins was established in silver stained [15] gradient polyacrylamide gels loaded with 5 ~tg protein per lane.
[35S]methionine biosynthetically labeled B. bigemina proteins were purified by immunoaffinity chromatography exactly as above.
Immunization of rabbits with purified sulface proteins. Individual rabbits were immunized subcutaneously with 25 ~tg of one of the 4 purified surface proteins in Freund's complete adjuvant. Three booster immunizations of 15 ~tg each in Freund's incomplete adjuvant were given at 1-2-week intervals, and serum was harvested one week later. All sera were tested by ELISA against the respective immunogen and had endpoint titers of 104. Specificity of the antibody response was confirmed by immunoprecipitation and immunoblot as above.
Immunization trial.
Twenty-five 3-month-old Holstein calves were randomly assigned to 5 groups of 5 calves each. Each group was immunized intramuscularly with 1 of the 4 purified merozoite surface proteins or with ovalbumin, 50/,tg per calf, in Freund' s complete adjuvant, followed by intramuscular immunizations with 50 ~g protein in Freund's incomplete adjuvant at 2-week intervals until 5 total immunizations had been administered. The serum antibody response was characterized by immunoprecipitation (as above), enzyme linked immunosorbent assy (ELISA) (see below), and indirect immunofluorescence assay (IFA) of live merozoites. The gradient separation of live B. bigemina merozoites, preparation of bovine ghost erythrocytes, and technique of IFA have been previously reported [4]. One week following the last immunization, all calves were challenged by intravenous inoculation of freshly collected, heparinized whole blood containing 3 x 109 blood stage B. bigemina from a splenectomized calf experimentally infected with the Mexico isolate. Parasitemia was monitored daily by calculating the percentage of 1000 erythrocytes containing parasites in a modified Wright's stained blood smear. Means of the peak parasitemia for each group of calves immunized with a different protein were compared using one way analysis of variance and Duncans's method for multiple comparison of group means [16,17].
ELISA.
Sera collected from experimental animals were titered by ELISA against their respective
216 immunogens by established techniques [18]. Microtiter plates were coated with purified surface protein at 25 ng p58, 50 ng gp55, 50 ng gp45, or 100 ng gp36 per well. Bound antibody was detected after addition of 50 gl peroxidase conjugated rabbit antibovine immunoglobulin or peroxidase conjugated protein-A (for rabbit sera). A positive ELISA reaction was defined as any A450/630 reading that was over 3 times the reading using the same dilution of serum from ovalbumin immunized calves or normal rabbit serum. Results
Post-translational modification of merozoite surface proteins. At least 10 parasite proteins were post-translationally modified by incorporation of 3H
3H
MYR. ACID
G L UC.
glucosamine, while 7 were myristilated (Fig. 1). Monoclonal antibodies 14.1, 14.16, 14.20, and 14.52, which bind to exposed epitopes on the surface of B. bigemina merozoites, immunoprecipitated 4 major surface proteins of 58, 55, 45, and 36 kDa, respectively, from [35S]methionine and 3Hamino acid-labeled B. bigemina antigen (Fig. 2) [4]. Five of the 10 glycosylated and 7 myristilated B. bigemina proteins could be immunoprecipitated by surface protein specific monoclonal antibodies (Fig. 2). These included 2 closely spaced polypeptides of 45 and 43 kDa immunoprecipitated by mAb 14.1, the 55-kDa polypeptide precipitated by mAb 14.20, and the 36- and 16-kDa polypeptides immunoprecipitated by mAb 14.36. The signal for the 36kDa glycoprotein labeled with [3H]myristic acid was better visualized in longer autoradiographic exposures. The surface protein defined by mAb 14.16 (58 kDa) was not labeled with [3H]glucosamine or [3H]myristic acid (Fig. 2D). As expected from previous results [4], uninfected erythrocytes in culture did not incorporate radiolabel (data not shown).
Purification of surface proteins. 200,000 •
92,500 •
69,000 •
4 6,000 •
30,000 •
Fig. 1. Gradient SDS-PAGEofB. bigeminapolypeptidesbiosynthetically incorporating [3H]myristic acid and [3H]glucosamine. Surface proteins are indicated by arrows on the right. Molecular weight standards are indicated on the left.
Immunoaffinity chromatography columns were used to purify [35S]methionine-labeled B. bigemina merozoite surface proteins. Major proteins of 58, 55, 45 and 36 kDa corresponding exactly to the proteins immunoprecipitated by the same mAbs, were isolated from the columns (data not shown), confirming the parasite origin and specificity of isolated polypeptides. For immunization of cattle, surface proteins were also purified from crude unlabeled B. bigemina antigen by immunoaffinity chromatography. Proteins eluted from the columns were separated by SDS-PAGE and detected with silver stain. The relative mobility and purity of isolated proteins are illustrated in Fig. 3. Purified preparations contained the same proteins as immunoprecipitates and [35S]methionine affinity chromatography eluates with the following exceptions: (1) 2 proteins with a relative mobility greater than 200 kDa were occasionally present with p58 (lane 2); and (2) an additional 43-kDa protein was infrequently isolated with gp55 (not shown).
lsolate cross-reactivity of merozoite surface proteins. To determine whether immunogenic
217
1
A
B
2
2
C 1
3
2
200,000
92,500
69,000
46,000
30,000
14,300
D 1
2
E 3
1
2
3
200,000~
92,500
--
69,000
-
46,000,-
30,000
,,-
B 14,300
--
" ~
•
Fig. 2. lmmunoprecipitation ofglycosylated and myristylated B. bigeminapolypeptides with surface reactive mAbs. Gradient SDSPAGE of [3H]amino acids (lane 1), [3H]glucosamine (lane 2), or [3H]myristic acid (lane 3) labeled proteins with isotype control mAb (Panel A), mAb 14.1 reactive with gp45 (B), mAb 14.20 reactive with gp55 (C), mAb 14.16 reactive with p58 (D), and mAb 14.36 reactive with gp36 (E). Molecular weight standards are indicated on the left.
epitopes of Mexico isolate merozoite surface proteins are conserved in other isolates, monospecific antisera from calves and rabbits immunized with
purified antigens (see below) were reacted in immunoblots with gradient-separated merozoites from 3 additional Latin American isolates and one
218 TABLE I Immunization of cattle with purified B. bigemina merozoite surface proteins
lmmunogen Ovalbumin
gp45
p58
gp55
gp36
Animal B258 B257 B263 B262 B235
ELISA titer -
Surface IFA titer -
Peak % parasitemia 2.5 1.7 2.4 0.8 1.4
Group peak % parasitemia~'
B264 B260 B265 B256 B261
5x10 a 104 104 104 5x 104
102 102 102 102 102
0.6 0.6 0.9 0.7 0.6
0.7±0.1 b
B279 B277 B274 B273 B272
104 104 104 5x 104 5x 104
102 102 102 102 102
0.3 0.9 1.2 0.9 0.5
0.8±0.4 b
B271 B275 B276 B278 B238
104 l() 3 104 104 104
l ()l 1() I 102 10~' 102
0.6 0.6 0.5 0.6 1.9
0.8±0.6"
B266 B267 B268
104 104 l0 ~
1()l l0 t 101
1.g 1.7 2.0
B269 B270
104 104
101 10~
0.4 1.2
1.8 ± 0.7
1.4 _+ 0.6
"Mean _+ standard deviation.
bSignificantly different from ovalbumin controls, P
213
© 1991 Elsevier Science Publishers B.V. /0166-6851/91/$03.50 ADONIS 016668519100247Y MOLBIO 01565
Molecular characterization and immunogenicity of neutralization-sensitive Babesia bigemina merozoite surface proteins T e r r y F. M c E l w a i n ~'2, L a n c e E. P e r r y m a n ~, A n t o n y J. M u s o k e 3 and Travis C. M c G u i r e ~ IDepartment of Veterinary Microbiology and Pathology, Washington State University, Pullman, WA, U.S.A.; 2Departmentof Infectious Diseases, College of Veterinary Medicine, University of Florida, Gainesville, FL, U.S.A.; and 3International Laboratory for Research on Animal Diseases, Nairobi, Kenya (Received 29 November 1990; accepted 13 March 1991)
Monoclonal antibodies binding to the surface of live Mexico isolate Babesia bigemina merozoites have defined 4 parasite-encoded surface antigens (36, 45, 55, and 58 kDa) that are potential targets for immune-mediated neutralization of merozoites. In this study, we have characterized the post-translational modification, antigenic polymorphism, and immunogenicityof these 4 proteins. Monoclonal antibody immunoaffinity-purified36- and 55-kDa polypeptides were identical in gel electrophoresis to immunoprecipitated radiolabeled proteins, while the purified 45-kDa protein consisted of 2 closely spaced polypeptides with relative molecular weights of 45 and 43 kDa. The 36-, 45-, and 55-kDa proteins were post-translationally modified by incorporation of [3H]glucosamine and [3H]myristic acid, suggesting they are integral membrane proteins secured by a phosphatidylinositol anchor. Cross-reactivity studies with monoclonal and monospecific polyclonal antibodies revealed marked antigenic polymorphism of these 3 glycoproteins among diverse geographic isolates. In contrast, none of the polypeptides bound by anti-p58 monoclonal antibody were glycosylated or myristilated. Both monoclonal and monospecific polyclonal antibodies recognizing p58 bound to similar molecular weight proteins in 4 additional isolates ofB. bigemina from Mexico, Puerto Rico, St. Croix, and Kenya, suggesting widespread conservation of p58 immunogenic epitopes among geographic isolates. Calves immunized with immunoafffinitypurified gp45, gp55, or p58 antigens were able to neutralize the infectivity of merozoites as indicated by significant reductions in the peak parasitemia after experimental challenge. Precise definition and appropriate presentation of neutralization sensitive epitopes on gp45, gp55, or p58 may enhance the merozoite neutralizing immune response in immunized cattle. Key words: Babesia bigemina; Merozoite surface protein; Antigenicdiversity; Post-translational modification; Immunogenicity
Introduction Bovine babesiosis is an intraerythrocytic protozoan disease that imposes a severe constraint on livestock production, especially in areas where enzootic stability is not maintained [ 1]. In tropical and subtropical climates the disease is caused by 2 species, Babesia bigemina and Babesia bovis, that are Correspondenceaddress: Terry F. McElwain, Department of Veterinary Microbiology and Pathology, Washington State University, Pullman, WA 99164-7040, U.S.A.
Abbreviations: SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; mAb, monoclonal antibody; PBS, phosphate-buffered saline; IFA, indirect fluorescent antibody assay; ELISA, enzyme-linkedimmunosorbentassay
transmitted by Boophilusspp. ticks [ 1]. The impact of babesiosis is currently reduced by a combination of acaricide treatment to decrease disease transmission by ticks, chemotherapy to ameliorate clinical signs, and premunition using an in vivo derived and attenuated live vaccine which prevents serious clinical disease after challenge [2]. Each of these techniques has biological and technical drawbacks that limit its utility, especially within countries where the prevalence of disease is high [3]. Our hypothesis is that an antigenically defined vaccine can target the immune response to neutralization-sensitive epitopes, enabling practical and effective immunoprophylaxis against bovine babesiosis. Accessible targets of the protective immune response against babesial parasites include extra-
214
cellular infective stages from tick salivary glands, extracellular merozoites, and infected erythrocytes. Surface exposed regions of membrane proteins on any of these targets may contain epitopes sensitive to neutralization. Exposed merozoite membrane antigens are therefore being investigated for their ability to induce protective immunity. Five B. bigemina-encoded merozoite membrane proteins with surface exposed epitopes have been previously defined with monoclonal antibodies [4]. All 5 proteins are recognized by antibodies in sera from cattle that are immune to babesiosis, but no further evidence of an individual antigen's ability to induce a protective immune response is available. In the current study, 4 of the 5 B. bigemina merozoite proteins with surface exposed epitopes are further characterized as to size, epitope conservation among isolates, and post-translational modification. Each of them can be purified by immunoaffinity chromatography in quantities sufficient for immunizing cattle. Their potential as immunogens is evaluated by their ability to induce an immune response that neutralizes merozoites, reducing the peak level of parasitemia in experimentally challenged cattle.
Materials and Methods
B. bigemina isolates. Mexico, Texcoco, St. Croix, Puerto Rico, and Kenya isolates were maintained as cryopreserved stabilates as described [5]. Texcoco, St. Croix, and Puerto Rico isolates were provided by G. Buening, University of Missouri, Columbia, MO [6].
Microaerophilous stationary phase culture/ biosynthetic radiolabeling. Parasite proteins were radiolabeled with [35S]methionine in short term culture as previously described [4], or by incorporation of [35S]methionine, [3H]amino acids, [3H]glucosamine, or [3H]myristic acid into parasites grown in continuous culture. Cultures were initiated with Mexico isolate B. bigemina and maintained as described [7]. For metabolic radiolabeling, 1.0 ml of culture medium overlying the infected erythrocytes was replaced with 200400 flCi of lyophilized [35S]methionine (>800 Ci mmol -~ in 20 mM potassium acetate containing
0.1% 2-mercaptoethanol, Amersham, Arlington Heights, IL), [3H]L-amino acid mixture (in 0.1N hydrochloric acid, Dupont, Wilmington, DE), [3H]glucosamine hydrochloride ( 3 0 ~ 0 Ci mmol in aqueous buffer, Dupont), or [3H]myristic acid ( 10-60 Ci mmol -~ in ethanol, Dupont) resuspended in 1.0 ml medium. Radiolabeled parasites were harvested 16-18 h later, washed, and lysed in 1% (v/v) Nonidet P-40 (LKB, Bromma, Sweden) as previously described [4].
Radioimmunoprecipitation, immunoblotting and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Immunoprecipitation of radiolabeled antigen was performed as previously reported [4]. For immunoblotting, 1 x l0 s gradientseparated merozoites [4] were electrophoresed in 7.5-17.5% gradient SDS-PAGE and transferred to nitrocellulose [8]. Immunoblots using bovine antibodies detected with 125I-labeled protein-G or rabbit antibodies detected with ~25I-labeled protein-A were performed as described [4,9,10]. Nitrocellulose was blocked with either 0.25% (v/v) Tween20/0.25% (w/v) gelatin in veronal-buffered saline [9] or 10% (w/v) milk in 20 mM Tris-HC1/150 mM NaCI. Gradient polyacrylamide gel techniques, SDS-PAGE sample buffer, and autoradiography have been previously described [11]. ~4C-labeled proteins used for molecular weight comparisons (Amersham, Arlington Heights, IL) consisted of myosin, 200 kDa; phosphorylase b, 92.5 kDa; bovine serum albumin, 69 kDa; ovalbumin, 46 kDa; carbonic anhydrase, 30 kDa; and lysozyme, 14.3 kDa.
Purification of monoclonal antibodies and coupling to Sepharose beads. The preparation and screening of mAbs 14.1, 14.16, 14.20, and 14.52, which recognize B. bigemina merozoite surface proteins, have been described elsewhere [4]. Ascitic fluid was generated in pristane-primed BALB/c mice with twice cloned hybridoma cells. Purification ofmAbs was performed by diethylaminoethyl cellulose chromatography (DE-52; Whatman Ltd., Maidstone, U.K.) of 50% ammonium sulfate precipitated immunoglobulin (from ascitic fluid) in 0.032 M Tris, pH 7.4 [ 12]. Purity of the isolated mAbs was established in Coomassie bluestained gradient polyacrylamide gels loaded with
215 50/tg of protein from column fractions in each lane. Purified mAbs were coupled to cyanogen bromideactivated Sepharose 4B (10 mg protein to 1 ml swollen beads) according to manufacturer's recommendations (Pharmacia Fine Chemicals, Uppsala, Sweden).
lmmunoaffinity chromatography.
Whole blood collected from splenectomized calves experimentally infected with the Mexico isolate of B. bigemina was collected at peak parasitemia, washed 3 times with 0.01 M sodium phosphate/0.15 M NaC1, pH 7.2 (PBS) to remove leukocytes, and stored at -70°C as a blood stabilate containing packed erythrocytes 1:1 (v/v) with a cryopreservant of 10% (w/v) polyvinylpyrollidone and 2% (w/v) glucose (crude antigen) [5]. Crude antigen washed 3 times at 22 000 x g with cold PBS containing 50 mM phenylmethylsulfonylfluoride was solubilized in buffer containing 50 mM Tris/5 mM EDTA/5 mM iodoacetamide/ 1 mM phenylmethylsulfonylfluoride/ 0.1 mM N-ct-p-tosyI-L-lysyl chloromethyl ketone/ 1% (v/v) Nonidet P-40 (lysis buffer) and centrifuged for I h at 135 000 × g. The supernatant was filtered through a 0.45-micron membrane and sonicated on ice. Solubilized antigen was passed twice through the mAb-coupled Sepharose 4B columns hooked in series (750 ~1 of packed wet beads each). The columns were washed sequentially with 0.02 M Tris/0.005 M EDTA/0.1 M NaC1/ 0.015 M NAN3, pH 7.6 containing I% (v/v) NP-40 and 0.1 mM phenylmethylsulfonylfluoride; and the same buffer without detergent. They were then preeluted individually with 10 ml of 0.1 M glycineNaOH/ 1 M NaC1, 0.5% (w/v) deoxycholate, pH 10.0 [ 13]. Bound proteins were eluted from individual columns with 5 ml of 0.05 M diethylamine/ 0.5% (w/v) deoxycholate, pH 11.5 directly into siliconized test tubes containing 0.5 ml of 1 M Tris, pH 8.5; and dialyzed against PBS to remove diethylamine and detergent. Eluted, dialyzed proteins were quantitated by Lowry protein microassay according to Peterson [ 14], performed in microtiter plates after elimination of the precipitation step and 4-fold reduction of reagent volumes, or bicinchoninic acid technique (Pierce Chemical, Rockford, IL). Purity of the eluted proteins was established in silver stained [15] gradient polyacrylamide gels loaded with 5 ~tg protein per lane.
[35S]methionine biosynthetically labeled B. bigemina proteins were purified by immunoaffinity chromatography exactly as above.
Immunization of rabbits with purified sulface proteins. Individual rabbits were immunized subcutaneously with 25 ~tg of one of the 4 purified surface proteins in Freund's complete adjuvant. Three booster immunizations of 15 ~tg each in Freund's incomplete adjuvant were given at 1-2-week intervals, and serum was harvested one week later. All sera were tested by ELISA against the respective immunogen and had endpoint titers of 104. Specificity of the antibody response was confirmed by immunoprecipitation and immunoblot as above.
Immunization trial.
Twenty-five 3-month-old Holstein calves were randomly assigned to 5 groups of 5 calves each. Each group was immunized intramuscularly with 1 of the 4 purified merozoite surface proteins or with ovalbumin, 50/,tg per calf, in Freund' s complete adjuvant, followed by intramuscular immunizations with 50 ~g protein in Freund's incomplete adjuvant at 2-week intervals until 5 total immunizations had been administered. The serum antibody response was characterized by immunoprecipitation (as above), enzyme linked immunosorbent assy (ELISA) (see below), and indirect immunofluorescence assay (IFA) of live merozoites. The gradient separation of live B. bigemina merozoites, preparation of bovine ghost erythrocytes, and technique of IFA have been previously reported [4]. One week following the last immunization, all calves were challenged by intravenous inoculation of freshly collected, heparinized whole blood containing 3 x 109 blood stage B. bigemina from a splenectomized calf experimentally infected with the Mexico isolate. Parasitemia was monitored daily by calculating the percentage of 1000 erythrocytes containing parasites in a modified Wright's stained blood smear. Means of the peak parasitemia for each group of calves immunized with a different protein were compared using one way analysis of variance and Duncans's method for multiple comparison of group means [16,17].
ELISA.
Sera collected from experimental animals were titered by ELISA against their respective
216 immunogens by established techniques [18]. Microtiter plates were coated with purified surface protein at 25 ng p58, 50 ng gp55, 50 ng gp45, or 100 ng gp36 per well. Bound antibody was detected after addition of 50 gl peroxidase conjugated rabbit antibovine immunoglobulin or peroxidase conjugated protein-A (for rabbit sera). A positive ELISA reaction was defined as any A450/630 reading that was over 3 times the reading using the same dilution of serum from ovalbumin immunized calves or normal rabbit serum. Results
Post-translational modification of merozoite surface proteins. At least 10 parasite proteins were post-translationally modified by incorporation of 3H
3H
MYR. ACID
G L UC.
glucosamine, while 7 were myristilated (Fig. 1). Monoclonal antibodies 14.1, 14.16, 14.20, and 14.52, which bind to exposed epitopes on the surface of B. bigemina merozoites, immunoprecipitated 4 major surface proteins of 58, 55, 45, and 36 kDa, respectively, from [35S]methionine and 3Hamino acid-labeled B. bigemina antigen (Fig. 2) [4]. Five of the 10 glycosylated and 7 myristilated B. bigemina proteins could be immunoprecipitated by surface protein specific monoclonal antibodies (Fig. 2). These included 2 closely spaced polypeptides of 45 and 43 kDa immunoprecipitated by mAb 14.1, the 55-kDa polypeptide precipitated by mAb 14.20, and the 36- and 16-kDa polypeptides immunoprecipitated by mAb 14.36. The signal for the 36kDa glycoprotein labeled with [3H]myristic acid was better visualized in longer autoradiographic exposures. The surface protein defined by mAb 14.16 (58 kDa) was not labeled with [3H]glucosamine or [3H]myristic acid (Fig. 2D). As expected from previous results [4], uninfected erythrocytes in culture did not incorporate radiolabel (data not shown).
Purification of surface proteins. 200,000 •
92,500 •
69,000 •
4 6,000 •
30,000 •
Fig. 1. Gradient SDS-PAGEofB. bigeminapolypeptidesbiosynthetically incorporating [3H]myristic acid and [3H]glucosamine. Surface proteins are indicated by arrows on the right. Molecular weight standards are indicated on the left.
Immunoaffinity chromatography columns were used to purify [35S]methionine-labeled B. bigemina merozoite surface proteins. Major proteins of 58, 55, 45 and 36 kDa corresponding exactly to the proteins immunoprecipitated by the same mAbs, were isolated from the columns (data not shown), confirming the parasite origin and specificity of isolated polypeptides. For immunization of cattle, surface proteins were also purified from crude unlabeled B. bigemina antigen by immunoaffinity chromatography. Proteins eluted from the columns were separated by SDS-PAGE and detected with silver stain. The relative mobility and purity of isolated proteins are illustrated in Fig. 3. Purified preparations contained the same proteins as immunoprecipitates and [35S]methionine affinity chromatography eluates with the following exceptions: (1) 2 proteins with a relative mobility greater than 200 kDa were occasionally present with p58 (lane 2); and (2) an additional 43-kDa protein was infrequently isolated with gp55 (not shown).
lsolate cross-reactivity of merozoite surface proteins. To determine whether immunogenic
217
1
A
B
2
2
C 1
3
2
200,000
92,500
69,000
46,000
30,000
14,300
D 1
2
E 3
1
2
3
200,000~
92,500
--
69,000
-
46,000,-
30,000
,,-
B 14,300
--
" ~
•
Fig. 2. lmmunoprecipitation ofglycosylated and myristylated B. bigeminapolypeptides with surface reactive mAbs. Gradient SDSPAGE of [3H]amino acids (lane 1), [3H]glucosamine (lane 2), or [3H]myristic acid (lane 3) labeled proteins with isotype control mAb (Panel A), mAb 14.1 reactive with gp45 (B), mAb 14.20 reactive with gp55 (C), mAb 14.16 reactive with p58 (D), and mAb 14.36 reactive with gp36 (E). Molecular weight standards are indicated on the left.
epitopes of Mexico isolate merozoite surface proteins are conserved in other isolates, monospecific antisera from calves and rabbits immunized with
purified antigens (see below) were reacted in immunoblots with gradient-separated merozoites from 3 additional Latin American isolates and one
218 TABLE I Immunization of cattle with purified B. bigemina merozoite surface proteins
lmmunogen Ovalbumin
gp45
p58
gp55
gp36
Animal B258 B257 B263 B262 B235
ELISA titer -
Surface IFA titer -
Peak % parasitemia 2.5 1.7 2.4 0.8 1.4
Group peak % parasitemia~'
B264 B260 B265 B256 B261
5x10 a 104 104 104 5x 104
102 102 102 102 102
0.6 0.6 0.9 0.7 0.6
0.7±0.1 b
B279 B277 B274 B273 B272
104 104 104 5x 104 5x 104
102 102 102 102 102
0.3 0.9 1.2 0.9 0.5
0.8±0.4 b
B271 B275 B276 B278 B238
104 l() 3 104 104 104
l ()l 1() I 102 10~' 102
0.6 0.6 0.5 0.6 1.9
0.8±0.6"
B266 B267 B268
104 104 l0 ~
1()l l0 t 101
1.g 1.7 2.0
B269 B270
104 104
101 10~
0.4 1.2
1.8 ± 0.7
1.4 _+ 0.6
"Mean _+ standard deviation.
bSignificantly different from ovalbumin controls, P
