BABEslA
BO VLS: DEXTRAN SULPHATE AS AN ADJUVANT PRECIPITANT OF PROTECTIVE IMMUNOG~NS
FOR AND
B. V. GOODGER,* D. J. WALTISBUWL, M. A. COMMINS and I. G. WRIGHT CSIRO, Division of Tropical Animal Production, Long Pocket Laboratories, Queensland 4068, Australia
Private Bag No. 3, P.O., Indooroopilly,
(Received 12 June 1991; accepted 20 January 1992)
AbStnWh-&ODGER B. V., WALTISBUHLD. .I., COMMINS M. A. and WRIGHT I. G. 1992. Babesia bovis: dextran sulphate as an adjuvant for and precipitant of protective immunogens. International Journal for Parasitology U: 465-469. Dextran sulphate, a chemical with some specificity for lipoproteins, was used to precipitate a fraction from a soluble extract of Babe&a bows-infected erythrocytes. The precipitate, in combination with dextran sulphate as an adjuvant, was used to vaccinate naive calves. The vaccinates and a group of control calves were challenged with virulent homologous B. bovb. The vaccinates showed delayed and decreased parasitaemias comparative to the controls. The antibody response to vaccination was primarily against the infected erythrocyte being of both IgG, and IgG2 classes. We believe this is the first report of B. bovis antibody being detected in the IgG, class. Lipase inhibition and chemical analysis suggested babesial lipid or lipoprotein was sufficiently immunogenic to produce serologically detectable antibody and presumably to elicit immunity.
INDEX KEY WORDS: Bbesia
bovis; dextran sulphate; adjuvant; lipid; cattle; vaccination.
INTRODUCTION studies have shown conclusively that immunity to Babesia bovis can be elicited by vaccinating cattle with fractions from B. bovis-infected erythrocytes (Mahoney & Wright, 1976; Goodger, Wright & Waltisbuhl, 1983; Wright, White, TraceyPatte, Donaldson, Goodger, Waltisbuhl8z Mahoney, 1983). With one exception (Goodger et al., 1983), all the studies used Freund’s complete adjuvant which, although efficacious, is not commercially acceptable and has adverse side effects such that an alternative adjuvant is urgently required. Such an adjuvant has to be commercially and ethically acceptable, produce no deleterious reaction at the vaccination site and be capable of producing a predo~nantly humoral response of long memory. Dextran sulphate as an adjuvant fulfils these criteria as well as having some chemical specificity for lipoproteins (Diamantstein, Wagner, Odenwald & Schulz, 1971; Weiland & Seidel, 1973). A recent vaccination study suggested lipoprotein of B. bavis parasite origin was involved in production of antibodies but whether the lipoprotein elicited a protective immune response was not
determined (Goodger, Commins, Wright, Waltisbuhl & Mirre, 1987). Accordingly, this manuscript reports a vaccination study in which both adjuvancy and lipoprotein immunogenicity were addressed. Dextran sulphate was used to precipitate lipoproteins from a soluble fraction of B. bovis-infected erythrocytes and the resultant precipitate then emulsified in dextran sulphate and used as inoculum.
SEVERAL
* To whom all correspondence should be addressed.
MATERIALS AND METHODS Calves. Calves, 3-6 months of age, were obtained from areas known to be free of Boophilus microplus, the tick vector for B. bovis. On arrival at the laboratory their blood was screened for B. bovis and other protozoa by thick film analysis (Mahoney & Saal, 1961) while serum was tested for B. bovis antibody by an enzyme-linked immunosor~nt assay (ELISA) (Waltisbuhl, Goodger, Wright, Commins & Mahoney, 1987). Organism. The Samford (S) strain of B. bows, maintained as a stabilate at these laboratories, was used throughout the experiment (Mahoney &Wright, 1976). Preparation of vaccination materiai. Acute B. bovis (S) infections were produced in splenectomized calves and suspensions containing 95100% intact infected erythrocytes were obtained from infected blood by preferential hypotonic lysis of non-infected erythrocytes (Mahoney, 1967). The infected erythrocytes were washed in 0.02 M-PBS, pH 7.2, lysed in distilled water and the resultant stroma-parasite
465
466
B. V.
GOODGER.
D. J. WALTISBLJHL, M. A. COMMINSand I. G. WRIGHT
matrix sonically disrupted and ultracentrifuged. The ultracentrifugal supernatant was depleted of oxy-haemoglobin by gel filtration on Biogel A-5m, adjusted in volume to that of the starting volume of infected erythrocytes and designated as gel filtration fraction (GFF). These procedures have been previously described in detail (Goodger, Wright, Waltisbuhl & Mirre. 1985). A similar fraction was prepared from normal erythrocytes for use as a control. A precipitate enriched for lipoproteins was obtained from 20 ml of GFF by mixing it at 37°C for 15 min with 800 ~1 of 5% aqueous sodium dextran sulphate 2000 (Pharmacia) and 2 ml of 11% CaCI, (Wieland & Seidel, 1973). The resulting precipitate was sedimented by centrifugation at 2000 g for 10 min at 22”C, washed twice in the precipitating solution, suspended in 20 ml of 5% sodium dextran sulphate, designated as DSP, and stored at 4°C until required. Vaccination and challenge. Four calves were randomly selected and each injected subcutaneously with 2 ml of DSP. The injections were repeated at 4 weeks. At 6 weeks, the calves, along with four control calves, were splenectomized and, at 8 weeks, challenged with 1 x 10’ B. bovis (S)-infected erythrocytes. Blood and titrated plasma samples were obtained prior to each inoculation, prior to challenge and on each day following challenge. Calves were treated with lmizol (Coopers Animal Health, Australia) when it was deemed from clinical and pathological criteria that they would not survive infection. Analysis of vaccination material. The blood parasitaemias were determined by thick film analysis (Mahoney & Saal, 196 1). Antibody titre and cellular specificity were obtained by indirect fluorescent antibody (IFA) assays (Johnston, Pearson & Leatch, 1973) while antibody concentration was scored on a matrix mode from 0 to 10 by ELISA (Waltisbuhl et al., 1987) For immunoblotting studies, GFF was firstly solid-phase adsorbed with glutaraldehyde-treated normal
bovine plasma whereas antisera and control negative sera were adsorbed with normal erythrocyte-stroma. Such absorptions are necessary to eliminate or reduce isoantigenantibody reactions and the methodology has been described previously (Goodger, Wright & Waltisbuhl, 1985). Immunoblotting, following SDS electrophoresis of absorbed, denatured and reduced antigen, was performed as described in detail previously (Goodger, Wright, Waltisbuhl & Mirre, 1985). If required, specific primary antibody was obtained from preparative immunoblots (80 mm x 80 mm) by cutting transverse 2 mm wide strips from the blots. washing three times for 15 min with PBS and eluting with 2 ml of 0.1 Mglycine-HCl, pH 2.2 for 5 min. The pH of the eluate was immediately adjusted to neutrality with 2 M-Tris. Antiserum was enriched for IgG by gel filtration on Sephadex G200 (Pharmacia), and the IgG separated into specific IgG, and enriched IgG, fractions by passage through a Protein A column (Pharmacia) equilibrated with PBS. The IgG, and other plasma proteins passed through the column while the IgG, bound specifically and was later eluted with 0.01 M-glycine_HCl, pH 2.8 (Goudswaard, Van der Donk, Noordzij, Van Dam & Vaerman, 1978). Both fractions were brought to neutrality with 2 M-Tris and adjusted in volume to that of the original antiserum. The protein concentration of the dextran sulphate precipitate was estimated by the method of Bradford (1976) while lipid was estimated by the method of Zollner & Kirsch (1962). Both assays were recorded as mg per ml ofpacked infected erythrocytes. Aliquots of GFF were incubated at 37°C for 30 min with equal volumes of PBS or PBS to which was added 0.01% pancreatic lipase (Calbiochem, California, U.S.A.), 0.01% phenylmethylsulphonyl chloride and 0.01% benzamidine. The GFF aliquots were diluted l/160 and four-fold dilutions then tested by ELISA (Waltisbuhl et al., 1987) using bovine anti-DSP and normal plasma as primary test sera at l/1000 dilutions.
60 -
70 e 6 e f ? w
60 -
.--I
O~dmn sulpk+*
50 -
_
Conh.i
z 8
40-
\ z f
30 -
e 20CT!
4
5
6
7
6
9
Days after Challenge FIG. 1. Mean daily parasitaemias following B. bovis homologous challenge of a group of control cattle () and a group previously vaccinated with dextran sulphate precipitate from B. bovis-infected erythrocytes (-------).
10
B. bovis and dextran sulphate
467
on the last 2 days of the challenge whereas none was evident in the vaccinates. In addition all the controls required treatment with babesiacide on the last day of the experiment whereas the vaccinates did not. Analysis
-67
-25
FIG. 2. Immunobiot of B. bovis antigen foilowing SDS-ME electrophoresis and probing with bovine antisera to the dextran sulphate precipitate of B. bovis-infected erythrocytes. The bands at 200 kDa were the only ones detected with control negative bovine sera. Molecular weight standards in kDa are marked on the right.
RESULTS Vaccination
The first injection with dextran sdphate produced no site reaction. However the second injection produced moderate swelling at the site but this disappeared within 2 days. Plasma samples taken on the day prior to challenge showed vaccinated calves had a mode IFA titre of l/3200. The staining was predominantly of the infected erythrocyte being both membrane and cytoskeleton. Parasites stained in 5-10% of infected erythrocytes but the mode titre of such staining was only l/400. Uninfected erythrocytes did not stain. The mode ELISA reaction had a matrix score of 10. Following challenge the vaccinates showed delayed and decreased parasitaemias comparative to controls, the differences being statistically significant (PcO.05) from Day 7 (Fig. 1). All the control calves had visual haemoglobinaemia
Immunoblots of GFF probed with pooled bovine antisera to DSP showed a major reaction around 80 kDa. Normally one, but sometimes up to three bands were produced depending on the batch of antigen. In addition, minor bands were detected from 30-38 kDa and at 20 kDa with non-specific spurious bands being found at N 200 kDa (Fig. 2). The latter were detected weakly with normal plasma whereas the former were not seen at all. Similar bands were not detected in analogous fractions in normal erythrocytes. The blotting patterns were similar, both qualitatively and quantitatively, when developed with either IgG, or IgG, fractions of bovine antiserum. Likewise, IFA titres and staining specificities were similar with both IgG fractions. Antibody to the individual 80 kDa antigens, when eluted from preparative immunoblots, avidly stained infected erythrocytes by IFA as well as reacting not only with the 80 kDa bands on immunoblots but also weakly with the 3&38 kDa bands. ELISA reactivity of GFF was reduced by at least 60% following lipase treatment in the presence of protease inhibitors (Fig. 3). Chemical analyses indicated DSP contained 4.00 mg ml-’ of lipid and 2.5 mg ml-’ of protein. DISCUSSION The initial studies by D. F. Mahoney (unpublished) into B. bovis vaccination indicated clearly that an adjuvant was required and, of a considerable number tested, only Freund’s complete and to a lesser extent, saponin, were capable of inducing antibody formation and a protective immune response. The initial studies also clearly showed that neither normal red cell extracts nor adjuvant alone elicited non-specific immunity and accordingly relevant control groups were not included in the present study. As a precaution, however, all relevant serology was performed on antisera adsorbed with normal red cell extracts to ensure serological specificity. The present results strikingly indicate that dextran sulphate acted as a powerful adjuvant in stimulating a protective immune response against B. bovis infection, albeit homologous. Furthermore, it seems reasonable to conclude that the protection was due primarily to antibody, as dextran sulphate is a powerful stimulus for the humoral system as was evidenced by the high antibody titres obtained (Diamantstein, Ruhl, Vogt &
468
B. V.
GOODGER,
D. J. WALTISBUHL, M. A. COMMINSand I. G. WRIGHT
f-----t
0.0
1 l/160
P +
Lipora
I
1
1
1
l/640
l/2560
l/l0240
l/40960
Antigen Dilution FIG. 3. Reduction of ELBA activity following prior incubation of B. bovis antigen with lipase in the presence of protease inhibitors. P and N represent bovine antiserum to B. bovis dextran sulphate precipitate and negative serum, respectively. The solid line denotes activity prior to lipase treatment while the dotted line represents activity following lipase treatment.
Bochert, 1973). The antibody classes produced were both IgG, and IgG, which is of interest as IgG, antibody to B. bovis has not been detected by serological studies of either vaccinated cattle or cattle with natural immunity (Mahoney, Kerr, Goodger & Wright, 1979; Goff, Wagner, Craig & Long, 1982). Surprisingly, the IgG, had similar affinity and serological specificity as the IgG,. Its unexpected production may have been due to the polyanionic nature of dextran sulphate and to the charge of the antigen-dextran sulphate emulsion. This aside, the more important aspect is whether the IgGz is protective and this warrants future passive immune and/or vaccination studies, because IgG, has different biological properties to IgG,, such as induction of phagocytosis (McGuire, Musoke & Kurtti, 1979), and this could confront the parasite with entirely new immunological attack mechanisms. The major antigen, as detected by immunoblotting, had a size of N 80 kDa and was located mainly in the infected erythrocyte. From the immunoblotting studies it appears to exhibit molecular size polymorphism as well as having a weak crossrelationship with the 3&38 kDa antigens. Whether it is the protective moiety is currently being investigated but its predominantly cellular location suggests its main mechanism for protection might be the removal of infected erythrocytes from the circulation, more so than parasite destruction per se. The removal
of infected erythrocytes has been proposed as the main protective mechanism operating in B. bovis immunity (Mahoney et al., 1979). The precipitation of protective antigens by dextran sulphate and the reduction of ELISA activity following incubation of antigen with lipase strongly indicate that lipoprotein has a defined role in the antigenic repertoire of B. bovis. Whether the lipid or the protein moiety of lipoprotein/proteolipid complexes is protective requires elucidation and is currently being investigated. However, if the lipid is the protective moiety it may not augur well for its production by genetic engineering. Dextran sulphate, as an adjuvant, presents a fascinating enigma, as in combination with antigen it induces a depot effect. In contrast it does not necessarily have to be injected at the same site as the antigen (Nakashima, Matsuura, Nagase, Yokochi & Kato, 1981). Moreover, it can be injected intravenously to boost the humoral response (Bradfield, Souhami & Addison, 1974). There is great scope to modify its effect on the immune response to B. bovis antigen(s) not only in dosage rate but also in route of application. Indeed, its sulphate moiety, with its potential for activating polyclonal B cells (Diamanstein et al., 1973), should be ideal for stimulating B. bovis immunity which is thought to be primarily humoral (Mahoney et al., 1979).
B. bovis and de7ctran sulphate REFERENCES BRADFIELD J. W., SOUHAMI R. L. & ADDISONI. E. 1974. The mechanism of the adjuvant effect of dextran sulphate. Immunology 26: 383-392. BRADFORD M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Analytical Biochemistry 12: 248-254. DIAMANTSTEIN T., WAGNERB., ODENWALD M. V. &SCHULZG. 1971. Stimulation of humoral antibody formation by polyanions. IV. The effects of dextran sulfate on the kinetics of secondary immune response in mice. European Journal of Immunology 1: 426429. DIAMANTSTEIN T., RUHL H., V~GT W. & B~CHER~G. 1973. Stimulation of B-cells by dextran sulphate in vitro. Immunology 25: 743-741. GOFF W. L., WAGNERG. G., CRAIGT. M. & LONGR. E. 1982. The bovine immune response to tick-derived Babesia bovis infection: serological studies of isolated immunoglobulins. Veterinary Parasitology 11: 109-120. GRUDGER B. V., WRIGHTI. G. & WALTISBUHL D. J. 1983. The lysate from bovine erythrocytes infected with Babesia bovis. Analysis of antigens and a report on their immunogenicity when polymerized with glutaraldehyde. Zeitschrift fir Parasitenkunde 69: 473482. GWDGER B. V., WRIGHTI. G. & WALTISBUHL D. J. 1985. Babesia bovis: the effects of acute inflammation and isoantibody production in the detection of babesial antigens. Experientia 41: 1577-1579. GRUDGER B. V., WIUGHTI. G., WALTISBUHL D. J. & MIRREG. B. 1985. Babesia bovis: successful vaccination against homologous challenge in splenectomized calves using a fraction of haemagglutinating antigen. International Journal for Parasitology 15: 175-l 79. GRUDGER B. V., COMMONS M. A., WRIGHTI. G., WALTISBUHL D. J. & MIRRE G. B. 1987. Successful homologous vaccination against Babesia bovis using a heparin-binding fraction of infected erythrocytes. International Journal for Parasitology 17: 935-940. GOUDSWAARD J., DONK J. A. VAN DER, NWRDZIJ A., DAM R. H. VAN & VAERMANJ. P. 1978. Protein A reactivity of various mammalian immunoglobulins. Scandinavian Journal of Immunology 8: 21-28. JOHNSTON L. A. Y., PEARSONR. D. & LEAXH G. 1973.
469
Evaluation of an indirect fluorescent antibody test for detecting Babesia argentina infection in cattle. Australian Veterinary Journal 49: 373-377. MAHONEY D. F. & SAALJ. R. 1961. Bovine babesiosis: thick blood films for the detection of parasitaemia, Australian Veterinary Journal 37: 4447. MAHONEYD. F. 1967. Bovine babesiosis: preparation and assessment of complement fixing antigens. Experimental Parasitology 20: 232-241. MAHONEYD. F. & WRIGHTI. G. 1976. Babesia argentina: immunisation of cattle with a killed antigen against infection with a heterologous strain. Veterinary Parasitology 2: 273-282. MAHONEYD. F., KERRJ. D., GOOM~ER B. V. &WRIGHT I. G. 1979. The immune response of cattle to Babesia bovis (syn. B. argentina). Studies on the nature and specificity of protection. International Journal for Parasitology 9: 297-306. MCGUIRET. C., MU~~KEA. J. & KURT~IT. 1979. Functional properties of bovine IgG, and IgG,: interaction with complement, macrophages, neutrophils and skin. Immunology 38: 249-256. NAKASHIMAI., MATSUURAA., NAGASEF., YOK~CHIT. & KATON. 198 1. Adjuvant actions of polyclonal lymphocyte activators. IV. Augmentation of antigen retention occurring early and transiently at the site ofinjection and in the draining lymph node. Cellular Immunology 57: 477485. WAL~~SBUHL D. J., GOODGER B. V., WRIGHTI. G., COMMINS M. A. & MAHONEY D. F. 1987. An enzyme linked immunosorbent assay to diagnose Babesia bovis infection in cattle. Parasitology Research 73: 126-l 3 1. WIELANDH. & SEIDELD. 1973. Improved techniques for assessment of serum lipoprotein patterns. II. Rapid method for diagnosis of type III hyperlipoproteinemia without ultracentrifugation. Clinical Chemistry 19: 11391141. WRIGHTI.G., WHITE M., TRACEY-PATTE P. D., DONALDSONR. A., GOODGERB. V., WALT~SLIUHL D. J. & MAHONEYD. F. 1983. Babesia bovis: isolation of a protective antigen by using monoclonal antibodies. Infection and Immunity 41: 244-250. ZOLLNERN. & KIRSCHK. 1962. Microdetermination of lipids by the sulfophosphovanillin reaction. Zeitschrift fiir die Gesamte Experimenteile Medizin 135: 545-561.
BO VLS: DEXTRAN SULPHATE AS AN ADJUVANT PRECIPITANT OF PROTECTIVE IMMUNOG~NS
FOR AND
B. V. GOODGER,* D. J. WALTISBUWL, M. A. COMMINS and I. G. WRIGHT CSIRO, Division of Tropical Animal Production, Long Pocket Laboratories, Queensland 4068, Australia
Private Bag No. 3, P.O., Indooroopilly,
(Received 12 June 1991; accepted 20 January 1992)
AbStnWh-&ODGER B. V., WALTISBUHLD. .I., COMMINS M. A. and WRIGHT I. G. 1992. Babesia bovis: dextran sulphate as an adjuvant for and precipitant of protective immunogens. International Journal for Parasitology U: 465-469. Dextran sulphate, a chemical with some specificity for lipoproteins, was used to precipitate a fraction from a soluble extract of Babe&a bows-infected erythrocytes. The precipitate, in combination with dextran sulphate as an adjuvant, was used to vaccinate naive calves. The vaccinates and a group of control calves were challenged with virulent homologous B. bovb. The vaccinates showed delayed and decreased parasitaemias comparative to the controls. The antibody response to vaccination was primarily against the infected erythrocyte being of both IgG, and IgG2 classes. We believe this is the first report of B. bovis antibody being detected in the IgG, class. Lipase inhibition and chemical analysis suggested babesial lipid or lipoprotein was sufficiently immunogenic to produce serologically detectable antibody and presumably to elicit immunity.
INDEX KEY WORDS: Bbesia
bovis; dextran sulphate; adjuvant; lipid; cattle; vaccination.
INTRODUCTION studies have shown conclusively that immunity to Babesia bovis can be elicited by vaccinating cattle with fractions from B. bovis-infected erythrocytes (Mahoney & Wright, 1976; Goodger, Wright & Waltisbuhl, 1983; Wright, White, TraceyPatte, Donaldson, Goodger, Waltisbuhl8z Mahoney, 1983). With one exception (Goodger et al., 1983), all the studies used Freund’s complete adjuvant which, although efficacious, is not commercially acceptable and has adverse side effects such that an alternative adjuvant is urgently required. Such an adjuvant has to be commercially and ethically acceptable, produce no deleterious reaction at the vaccination site and be capable of producing a predo~nantly humoral response of long memory. Dextran sulphate as an adjuvant fulfils these criteria as well as having some chemical specificity for lipoproteins (Diamantstein, Wagner, Odenwald & Schulz, 1971; Weiland & Seidel, 1973). A recent vaccination study suggested lipoprotein of B. bavis parasite origin was involved in production of antibodies but whether the lipoprotein elicited a protective immune response was not
determined (Goodger, Commins, Wright, Waltisbuhl & Mirre, 1987). Accordingly, this manuscript reports a vaccination study in which both adjuvancy and lipoprotein immunogenicity were addressed. Dextran sulphate was used to precipitate lipoproteins from a soluble fraction of B. bovis-infected erythrocytes and the resultant precipitate then emulsified in dextran sulphate and used as inoculum.
SEVERAL
* To whom all correspondence should be addressed.
MATERIALS AND METHODS Calves. Calves, 3-6 months of age, were obtained from areas known to be free of Boophilus microplus, the tick vector for B. bovis. On arrival at the laboratory their blood was screened for B. bovis and other protozoa by thick film analysis (Mahoney & Saal, 1961) while serum was tested for B. bovis antibody by an enzyme-linked immunosor~nt assay (ELISA) (Waltisbuhl, Goodger, Wright, Commins & Mahoney, 1987). Organism. The Samford (S) strain of B. bows, maintained as a stabilate at these laboratories, was used throughout the experiment (Mahoney &Wright, 1976). Preparation of vaccination materiai. Acute B. bovis (S) infections were produced in splenectomized calves and suspensions containing 95100% intact infected erythrocytes were obtained from infected blood by preferential hypotonic lysis of non-infected erythrocytes (Mahoney, 1967). The infected erythrocytes were washed in 0.02 M-PBS, pH 7.2, lysed in distilled water and the resultant stroma-parasite
465
466
B. V.
GOODGER.
D. J. WALTISBLJHL, M. A. COMMINSand I. G. WRIGHT
matrix sonically disrupted and ultracentrifuged. The ultracentrifugal supernatant was depleted of oxy-haemoglobin by gel filtration on Biogel A-5m, adjusted in volume to that of the starting volume of infected erythrocytes and designated as gel filtration fraction (GFF). These procedures have been previously described in detail (Goodger, Wright, Waltisbuhl & Mirre. 1985). A similar fraction was prepared from normal erythrocytes for use as a control. A precipitate enriched for lipoproteins was obtained from 20 ml of GFF by mixing it at 37°C for 15 min with 800 ~1 of 5% aqueous sodium dextran sulphate 2000 (Pharmacia) and 2 ml of 11% CaCI, (Wieland & Seidel, 1973). The resulting precipitate was sedimented by centrifugation at 2000 g for 10 min at 22”C, washed twice in the precipitating solution, suspended in 20 ml of 5% sodium dextran sulphate, designated as DSP, and stored at 4°C until required. Vaccination and challenge. Four calves were randomly selected and each injected subcutaneously with 2 ml of DSP. The injections were repeated at 4 weeks. At 6 weeks, the calves, along with four control calves, were splenectomized and, at 8 weeks, challenged with 1 x 10’ B. bovis (S)-infected erythrocytes. Blood and titrated plasma samples were obtained prior to each inoculation, prior to challenge and on each day following challenge. Calves were treated with lmizol (Coopers Animal Health, Australia) when it was deemed from clinical and pathological criteria that they would not survive infection. Analysis of vaccination material. The blood parasitaemias were determined by thick film analysis (Mahoney & Saal, 196 1). Antibody titre and cellular specificity were obtained by indirect fluorescent antibody (IFA) assays (Johnston, Pearson & Leatch, 1973) while antibody concentration was scored on a matrix mode from 0 to 10 by ELISA (Waltisbuhl et al., 1987) For immunoblotting studies, GFF was firstly solid-phase adsorbed with glutaraldehyde-treated normal
bovine plasma whereas antisera and control negative sera were adsorbed with normal erythrocyte-stroma. Such absorptions are necessary to eliminate or reduce isoantigenantibody reactions and the methodology has been described previously (Goodger, Wright & Waltisbuhl, 1985). Immunoblotting, following SDS electrophoresis of absorbed, denatured and reduced antigen, was performed as described in detail previously (Goodger, Wright, Waltisbuhl & Mirre, 1985). If required, specific primary antibody was obtained from preparative immunoblots (80 mm x 80 mm) by cutting transverse 2 mm wide strips from the blots. washing three times for 15 min with PBS and eluting with 2 ml of 0.1 Mglycine-HCl, pH 2.2 for 5 min. The pH of the eluate was immediately adjusted to neutrality with 2 M-Tris. Antiserum was enriched for IgG by gel filtration on Sephadex G200 (Pharmacia), and the IgG separated into specific IgG, and enriched IgG, fractions by passage through a Protein A column (Pharmacia) equilibrated with PBS. The IgG, and other plasma proteins passed through the column while the IgG, bound specifically and was later eluted with 0.01 M-glycine_HCl, pH 2.8 (Goudswaard, Van der Donk, Noordzij, Van Dam & Vaerman, 1978). Both fractions were brought to neutrality with 2 M-Tris and adjusted in volume to that of the original antiserum. The protein concentration of the dextran sulphate precipitate was estimated by the method of Bradford (1976) while lipid was estimated by the method of Zollner & Kirsch (1962). Both assays were recorded as mg per ml ofpacked infected erythrocytes. Aliquots of GFF were incubated at 37°C for 30 min with equal volumes of PBS or PBS to which was added 0.01% pancreatic lipase (Calbiochem, California, U.S.A.), 0.01% phenylmethylsulphonyl chloride and 0.01% benzamidine. The GFF aliquots were diluted l/160 and four-fold dilutions then tested by ELISA (Waltisbuhl et al., 1987) using bovine anti-DSP and normal plasma as primary test sera at l/1000 dilutions.
60 -
70 e 6 e f ? w
60 -
.--I
O~dmn sulpk+*
50 -
_
Conh.i
z 8
40-
\ z f
30 -
e 20CT!
4
5
6
7
6
9
Days after Challenge FIG. 1. Mean daily parasitaemias following B. bovis homologous challenge of a group of control cattle () and a group previously vaccinated with dextran sulphate precipitate from B. bovis-infected erythrocytes (-------).
10
B. bovis and dextran sulphate
467
on the last 2 days of the challenge whereas none was evident in the vaccinates. In addition all the controls required treatment with babesiacide on the last day of the experiment whereas the vaccinates did not. Analysis
-67
-25
FIG. 2. Immunobiot of B. bovis antigen foilowing SDS-ME electrophoresis and probing with bovine antisera to the dextran sulphate precipitate of B. bovis-infected erythrocytes. The bands at 200 kDa were the only ones detected with control negative bovine sera. Molecular weight standards in kDa are marked on the right.
RESULTS Vaccination
The first injection with dextran sdphate produced no site reaction. However the second injection produced moderate swelling at the site but this disappeared within 2 days. Plasma samples taken on the day prior to challenge showed vaccinated calves had a mode IFA titre of l/3200. The staining was predominantly of the infected erythrocyte being both membrane and cytoskeleton. Parasites stained in 5-10% of infected erythrocytes but the mode titre of such staining was only l/400. Uninfected erythrocytes did not stain. The mode ELISA reaction had a matrix score of 10. Following challenge the vaccinates showed delayed and decreased parasitaemias comparative to controls, the differences being statistically significant (PcO.05) from Day 7 (Fig. 1). All the control calves had visual haemoglobinaemia
Immunoblots of GFF probed with pooled bovine antisera to DSP showed a major reaction around 80 kDa. Normally one, but sometimes up to three bands were produced depending on the batch of antigen. In addition, minor bands were detected from 30-38 kDa and at 20 kDa with non-specific spurious bands being found at N 200 kDa (Fig. 2). The latter were detected weakly with normal plasma whereas the former were not seen at all. Similar bands were not detected in analogous fractions in normal erythrocytes. The blotting patterns were similar, both qualitatively and quantitatively, when developed with either IgG, or IgG, fractions of bovine antiserum. Likewise, IFA titres and staining specificities were similar with both IgG fractions. Antibody to the individual 80 kDa antigens, when eluted from preparative immunoblots, avidly stained infected erythrocytes by IFA as well as reacting not only with the 80 kDa bands on immunoblots but also weakly with the 3&38 kDa bands. ELISA reactivity of GFF was reduced by at least 60% following lipase treatment in the presence of protease inhibitors (Fig. 3). Chemical analyses indicated DSP contained 4.00 mg ml-’ of lipid and 2.5 mg ml-’ of protein. DISCUSSION The initial studies by D. F. Mahoney (unpublished) into B. bovis vaccination indicated clearly that an adjuvant was required and, of a considerable number tested, only Freund’s complete and to a lesser extent, saponin, were capable of inducing antibody formation and a protective immune response. The initial studies also clearly showed that neither normal red cell extracts nor adjuvant alone elicited non-specific immunity and accordingly relevant control groups were not included in the present study. As a precaution, however, all relevant serology was performed on antisera adsorbed with normal red cell extracts to ensure serological specificity. The present results strikingly indicate that dextran sulphate acted as a powerful adjuvant in stimulating a protective immune response against B. bovis infection, albeit homologous. Furthermore, it seems reasonable to conclude that the protection was due primarily to antibody, as dextran sulphate is a powerful stimulus for the humoral system as was evidenced by the high antibody titres obtained (Diamantstein, Ruhl, Vogt &
468
B. V.
GOODGER,
D. J. WALTISBUHL, M. A. COMMINSand I. G. WRIGHT
f-----t
0.0
1 l/160
P +
Lipora
I
1
1
1
l/640
l/2560
l/l0240
l/40960
Antigen Dilution FIG. 3. Reduction of ELBA activity following prior incubation of B. bovis antigen with lipase in the presence of protease inhibitors. P and N represent bovine antiserum to B. bovis dextran sulphate precipitate and negative serum, respectively. The solid line denotes activity prior to lipase treatment while the dotted line represents activity following lipase treatment.
Bochert, 1973). The antibody classes produced were both IgG, and IgG, which is of interest as IgG, antibody to B. bovis has not been detected by serological studies of either vaccinated cattle or cattle with natural immunity (Mahoney, Kerr, Goodger & Wright, 1979; Goff, Wagner, Craig & Long, 1982). Surprisingly, the IgG, had similar affinity and serological specificity as the IgG,. Its unexpected production may have been due to the polyanionic nature of dextran sulphate and to the charge of the antigen-dextran sulphate emulsion. This aside, the more important aspect is whether the IgGz is protective and this warrants future passive immune and/or vaccination studies, because IgG, has different biological properties to IgG,, such as induction of phagocytosis (McGuire, Musoke & Kurtti, 1979), and this could confront the parasite with entirely new immunological attack mechanisms. The major antigen, as detected by immunoblotting, had a size of N 80 kDa and was located mainly in the infected erythrocyte. From the immunoblotting studies it appears to exhibit molecular size polymorphism as well as having a weak crossrelationship with the 3&38 kDa antigens. Whether it is the protective moiety is currently being investigated but its predominantly cellular location suggests its main mechanism for protection might be the removal of infected erythrocytes from the circulation, more so than parasite destruction per se. The removal
of infected erythrocytes has been proposed as the main protective mechanism operating in B. bovis immunity (Mahoney et al., 1979). The precipitation of protective antigens by dextran sulphate and the reduction of ELISA activity following incubation of antigen with lipase strongly indicate that lipoprotein has a defined role in the antigenic repertoire of B. bovis. Whether the lipid or the protein moiety of lipoprotein/proteolipid complexes is protective requires elucidation and is currently being investigated. However, if the lipid is the protective moiety it may not augur well for its production by genetic engineering. Dextran sulphate, as an adjuvant, presents a fascinating enigma, as in combination with antigen it induces a depot effect. In contrast it does not necessarily have to be injected at the same site as the antigen (Nakashima, Matsuura, Nagase, Yokochi & Kato, 1981). Moreover, it can be injected intravenously to boost the humoral response (Bradfield, Souhami & Addison, 1974). There is great scope to modify its effect on the immune response to B. bovis antigen(s) not only in dosage rate but also in route of application. Indeed, its sulphate moiety, with its potential for activating polyclonal B cells (Diamanstein et al., 1973), should be ideal for stimulating B. bovis immunity which is thought to be primarily humoral (Mahoney et al., 1979).
B. bovis and de7ctran sulphate REFERENCES BRADFIELD J. W., SOUHAMI R. L. & ADDISONI. E. 1974. The mechanism of the adjuvant effect of dextran sulphate. Immunology 26: 383-392. BRADFORD M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Analytical Biochemistry 12: 248-254. DIAMANTSTEIN T., WAGNERB., ODENWALD M. V. &SCHULZG. 1971. Stimulation of humoral antibody formation by polyanions. IV. The effects of dextran sulfate on the kinetics of secondary immune response in mice. European Journal of Immunology 1: 426429. DIAMANTSTEIN T., RUHL H., V~GT W. & B~CHER~G. 1973. Stimulation of B-cells by dextran sulphate in vitro. Immunology 25: 743-741. GOFF W. L., WAGNERG. G., CRAIGT. M. & LONGR. E. 1982. The bovine immune response to tick-derived Babesia bovis infection: serological studies of isolated immunoglobulins. Veterinary Parasitology 11: 109-120. GRUDGER B. V., WRIGHTI. G. & WALTISBUHL D. J. 1983. The lysate from bovine erythrocytes infected with Babesia bovis. Analysis of antigens and a report on their immunogenicity when polymerized with glutaraldehyde. Zeitschrift fir Parasitenkunde 69: 473482. GWDGER B. V., WRIGHTI. G. & WALTISBUHL D. J. 1985. Babesia bovis: the effects of acute inflammation and isoantibody production in the detection of babesial antigens. Experientia 41: 1577-1579. GRUDGER B. V., WIUGHTI. G., WALTISBUHL D. J. & MIRREG. B. 1985. Babesia bovis: successful vaccination against homologous challenge in splenectomized calves using a fraction of haemagglutinating antigen. International Journal for Parasitology 15: 175-l 79. GRUDGER B. V., COMMONS M. A., WRIGHTI. G., WALTISBUHL D. J. & MIRRE G. B. 1987. Successful homologous vaccination against Babesia bovis using a heparin-binding fraction of infected erythrocytes. International Journal for Parasitology 17: 935-940. GOUDSWAARD J., DONK J. A. VAN DER, NWRDZIJ A., DAM R. H. VAN & VAERMANJ. P. 1978. Protein A reactivity of various mammalian immunoglobulins. Scandinavian Journal of Immunology 8: 21-28. JOHNSTON L. A. Y., PEARSONR. D. & LEAXH G. 1973.
469
Evaluation of an indirect fluorescent antibody test for detecting Babesia argentina infection in cattle. Australian Veterinary Journal 49: 373-377. MAHONEY D. F. & SAALJ. R. 1961. Bovine babesiosis: thick blood films for the detection of parasitaemia, Australian Veterinary Journal 37: 4447. MAHONEYD. F. 1967. Bovine babesiosis: preparation and assessment of complement fixing antigens. Experimental Parasitology 20: 232-241. MAHONEYD. F. & WRIGHTI. G. 1976. Babesia argentina: immunisation of cattle with a killed antigen against infection with a heterologous strain. Veterinary Parasitology 2: 273-282. MAHONEYD. F., KERRJ. D., GOOM~ER B. V. &WRIGHT I. G. 1979. The immune response of cattle to Babesia bovis (syn. B. argentina). Studies on the nature and specificity of protection. International Journal for Parasitology 9: 297-306. MCGUIRET. C., MU~~KEA. J. & KURT~IT. 1979. Functional properties of bovine IgG, and IgG,: interaction with complement, macrophages, neutrophils and skin. Immunology 38: 249-256. NAKASHIMAI., MATSUURAA., NAGASEF., YOK~CHIT. & KATON. 198 1. Adjuvant actions of polyclonal lymphocyte activators. IV. Augmentation of antigen retention occurring early and transiently at the site ofinjection and in the draining lymph node. Cellular Immunology 57: 477485. WAL~~SBUHL D. J., GOODGER B. V., WRIGHTI. G., COMMINS M. A. & MAHONEY D. F. 1987. An enzyme linked immunosorbent assay to diagnose Babesia bovis infection in cattle. Parasitology Research 73: 126-l 3 1. WIELANDH. & SEIDELD. 1973. Improved techniques for assessment of serum lipoprotein patterns. II. Rapid method for diagnosis of type III hyperlipoproteinemia without ultracentrifugation. Clinical Chemistry 19: 11391141. WRIGHTI.G., WHITE M., TRACEY-PATTE P. D., DONALDSONR. A., GOODGERB. V., WALT~SLIUHL D. J. & MAHONEYD. F. 1983. Babesia bovis: isolation of a protective antigen by using monoclonal antibodies. Infection and Immunity 41: 244-250. ZOLLNERN. & KIRSCHK. 1962. Microdetermination of lipids by the sulfophosphovanillin reaction. Zeitschrift fiir die Gesamte Experimenteile Medizin 135: 545-561.
