- Email: [email protected]
Evaluation of whole cell and subcellular vaccines against Brucella ovis in rams
Veterinary Immunology and Immunopathology, Elsevier Science Publishers B.V.. Amsterdam
31 ( 1993) 251-270
257
Evaluation of whole cell and subcellular vaccines against Brucella ovis in rams J.M. Blasco”,C. Gamazob,A.J. Winter’, M.P. Jimhez de BagtiC?, C. Marin”, M. BarberAnd,I. Moriy6nb, B. Alonso-Urmenetaband R. Diazb “Servicio de Investigacidn Agraria. Diputacidn General de Aragdn, 50080 Zaragoza, Spain “Departamento de Microbiologia, Universidad de Navarra, 31080 Pamplona, Spain CDepartment of Veterinary Microbiology, Immunology and Parasitology Cornell University, Ithaca, NY 14853, USA ‘Departamento de Patologia Animal, Universidad de Zaragoza. 50080 Zaragoza, Spain (Accepted 5 October 1992)
ABSTRACT Blasco, J.M., Gamazo, C., Winter, A.J., Jimenez de Bagiits, M.P., Marin, C., Barberan, M., Moriyon, I., Alonso-Urmeneta, B. and Diaz, R., 1993. Evaluation of whole cell and subcellular vaccines against Brucella ovis in rams. Vet. Immunol. Immunopathol., 37: 251-270 Five antigen preparations from Brucella ovis strain REO 198 were incorporated with the pluronic polymer L-l 2 1 and muramyl dipeptide and tested as vaccines against B. ovis infection of rams. The antigenic preparations were: ( 1) a fraction enriched in outer membrane proteins and rough lipopolysaccharide (hot saline extract, HS); (2) the proteins from HS substantially free of lipopolysaccharide; (3 ) outer membrane blebs; (4) outer membrane-peptidoglycan complexes extracted with detergent; (5) killed whole cells. The experimental vaccines were compared with two standard vaccines, rough Bruce/la abortus 45120 whole killed cells in an oil based adjuvant, and live Brucella meiitensis Rev 1. Immunizations with non-living vaccines were performed on two occasions, I8 weeks apart. The rams were challenged with a virulent strain of B. ovis 3 1 weeks after the second vaccination and slaughtered I5 weeks thereafter. Rates of infection in groups vaccinated with Rev 1 (33%), and HS (40%) were significantly lower (PcO.005 and PcO.025, respectively) than that in the non-vaccinated control group (871). Strain 45/20 was the only other vaccine that conferred a significant level of protection (50%) (PcO.05). The organ distribution of the infection and the level of colonization of infected organs did not differ significantly between infected animals in the various vaccine groups and those in the unvaccinated control group. No statistically significant relationship was detected between the magnitude of the antibody responses to the HS extract, to outer membrane proteins, or to the rough lipopolysaccharide, and freedom from infection. The results indicate that the HS extract of B. ovis may represent a useful alternative to B. melitensis Rev 1 or B. abortus 45120 as a vaccine against B. ovis. ABBREVIATIONS ABTS, 2,2’-azino-bis, 3 ethylbenzthiazoline-6-sulfonic acid; HS, hot saline; LPS, lipopolysacharide; OMP, outer membrane protein; PBS, phosphate buffered saline. Correspondence to: J.M. Blasco, Servicio de Investigation de Aragori, 50080 Zaragoza, Spain.
0 1993, Elsevier Science Publishers
Agraria,
Ap 727 Diputacion
B.V. All rights reserved 0165-2427/93/$06.00
General
258
J.M. BLASCO
ETAL.
INTRODUCTION
Brttcellu ovis is the causative agent of contagious ram epididymitis, an infectious diseasecausing abortion and a decreasein ram fertility with severe economic losses.Although several control methods have been proposed, vaccination representsthe only practical measure in countries with a high incidence of the disease (Blasco, 1990). Brzrcella meliferzsisRev 1, a smooth attenuated strain, is generally consideredthe best vaccine available. The rate of protection afforded by Rev 1 against experimental infection varies from 60 to 1OO%,depending on the criteria used to assessprotection (Van Heerden and Van Rensburg, 1962;Gradwell and Van Zyl, 1975;Garcia-Carrillo, 1981; Fensterbank et al., 1982; Blasco et al., 1987; Marin et al., 1990a). However, Rev 1 evokesa serological responseinterfering with the interpretation of serological tests used to diagnose infection by B. melitensis and B. ovis (Ris, 1967; Gradwell and Van Zyl, 1975; Erasmus and Bergh, 1985; Blasco et al., I987 ), although conjunctival vaccination reducesthese problems (Marin et al., 1990). Additional disadvantagesof Rev 1 are that its use is not allowed in countries free from B. melitensis, and that this strain is pathogenic in humans. Several bacterins have been used with less success (Biberstein et al., 1962:Buddle, 1962;Buddle et al., 1963;Claxton, 1968;Swift and Maki, 1968; Haughey and Hughes, 1969;Afzal et al., 1984). The efficacy of B. ovis bacterins is improved by the simultaneous inoculation of live B. abortzrsstrain 19 (McGowan and Harrold, 1979)) but this combined vaccine elicits long lasting serological responsesthat interfere with serological diagnosis (Ris, 1967) and can cause sequelae such as epiphysitis and epididymitis (West et al., 1970). In contrast to B. melitensis, B. ovis is a naturally rough form lacking the Opolysaccharide side chain that is the major antigen of smooth brucellae (Diaz and Bosseray, 1973; Moreno et al., 1984). However, both speciesshare soluble antigens (Diaz et al., 1967) and part of the core-lipid A lipopolysacharide (LPS) determinants (Diaz and Bosseray, 1973;Moreno et al., 1984). Moreover, B. melitensis and B. ovis strains have a quantitatively similar outer membrane protein (OMP) profile, and the OMPs show strong immunological cross reactivity (Gamazo et al., 1989a). Along with rough LPS (R-LPS), thesegroups of OMPs are major components of the hot saline extracts (HS) of B. 0vi.sused in the standard diagnostic tests, and B. ovis-infected rams developed an antibody responsenot only to the R-LPS but also to OMPs (RiezuBoj et al., 1986, 1990). Since B. melitensis and B. ovis share OMPs, but not the 0-polysaccharide as surface antigens, it is possible that OMPs play a significant role in the protection afforded by Rev 1 against B. ovis. If so, a subcellular vaccine containing the outer membrane components present in HS could be effective against B. ovis and would not interfere with the diagnosis of B. melitensis. The objective of the present work was to compare the effi-
V,4CCINES
t\GAINST
N. 01 ‘IS IN R.AklS
259
cacy of subcellular vaccines, bacterins, and live Rev 1 against an experimental challenge with virulent B. ovis in rams. MATERIALS
AND METHODS
.4ntigensjiw vaccinesand irnmrnological assays All antigens were obtained from B. ovis REO 198 cells cultivated in liquid medium (Gamazo et al., 1989a). To prepare the B. ovis whole cell vaccine (bacterin ), cells were killed by incubation in 0.5% phenol-20 mM phosphate buffered saline (PBS) (pH 7.2) for 2 days at 4°C washed three times with distilled water and lyophilized. A fraction rich in peptidoglycan-outer membrane (PG-OM ) was prepared by Sarkosyl extraction of cell envelopes (Verstreate et al., 1982). Outer membrane blebs were obtained by ultracentrifugation ( 100 OOOxg, 1h) of spent culture broths as described previously (Gamazo and Moriyon, 1987). Crude hot saline extracts (HS), the protein fraction of HS substantially freed from LPS by extraction with petroleum ether-chloroform-phenol (HS-PCP), and the purified R-LPS were prepared by methods described previously (Gamazo et al., 1989a; Riezu-Boj et al., 1990). The percentages of LPS (estimated by the content of 2-keto-3-deoxyoctonate) and protein in each of the subcellular vaccines were as follows: HS, LPS 34.6%, protein 65.4%; HS-PCP, LPS < I%, protein 93.0%; outer membrane blebs, LPS 13.6%, protein 34.9%: PG-OM, LPS 27.4%, protein 66.1%.
Preparation and administration qf vaccines Subcellular vaccines and B. ovis bacterins were administered in an adjuvant containing pluronic polymer L- 12 1 (BASF Wyandotte, Parsippany, NJ ), squalane (Sigma Chemicals, St. Louis, MO), and muramyl dipeptide (N-acetylmuramyl-L-cu-aminobutyryl-D-isoglutamine; Choay Chimelab, Paris, France). The emulsions were prepared by a modification of the method of Byars and Allison ( 1987). Briefly, 0.3 ml of L-121 were mixed with 4.2 ml of an 0.4% solution of Tween 80 by vortexing with 6 mm glass beads. Squalane (0.5 ml) was added to this mixture and vortexing was continued until no oil droplets larger than approximately 40 pm diameter could be detected by phase contrast microscopy. The emulsion was then supplemented with 10 mg of muramyl dipeptide and mixed thoroughly with an equal volume of an antigen solution in PBS. Rams received subcutaneously 1 ml of vaccine containing 30 ,~l of L- 12 1, 50 ~1 of squalane, 1 mg of muramyl dipeptide, and the corresponding antigen. The amounts of antigen administered per ram were, bacterin, 10.0 mg; PG-OM, 2.0 mg; outer membrane blebs, 2.0 mg; HS, 2.0 mg; and HS-PCP, 1 .O mg. Preliminary studies done in Finn Dorset lambs showed
260
J.M. BL.-\SCO
ET.AL.
that these doses in adjuvant induced strong antibody responses (Gamazo et al., 1989b). B. abortus strain 45/20 bacterin in a mineral oil-based adjuvant (Duphavac) was purchased from Duphar (Duphar B.V., Amsterdam, Netherlands). One ml of this vaccine (2 x 10’ ’ cells ) was administered intramuscularly to each animal. B. melitensis strain Rev 1 was kindly provided by Dr. J.M. Verger (Station de Pathologie de la Reproduction, Nouzilly, France). To prepare the vaccine, smooth Rev 1 was grown for 3 days on Tryptone Soya Agar (Oxoid London, UK) in Roux flasks, washed twice with sterile saline, suspended in a milklactose vehicle and lyophilized. After lyophilization there were 2 x 10” c.f.u. per ml. The lyophilized vaccine was diluted in saline and animals were vaccinated subcutaneously with 5 x 1OSc.f.u. in 1 ml.
Experimental design One hundred and twenty crossbred brucella-free rams, 5 months old, were divided into eight groups ( 15 lambs per group). Individual groups were vaccinated with the following antigens in L- 12 1-muramyl dipeptide adjuvant: B. ovis bacterin, PG-OM, blebs, HS, and HS-PCP. In addition one group was vaccinated with Rev 1, one with 45/20 bacterin, and a control group remained unvaccinated. Revaccinations were performed 18 weeks later, except for the group that received Rev 1 which was not revaccinated. At regular intervals after vaccination and revaccination, blood samples were taken and animals were examined for genital lesions. Challenge inoculations were performed 3 1 weeks after the second vaccination with a virulent strain of B. ovis obtained from a naturally infected ram. This strain was grown on Tryptone Soya Agar supplemented with 0.5% yeast extract and 5.0% sterile bovine serum for 72 h at 37°C in a 10% CO? atmosphere. The cells were then harvested, washed and resuspended in sterile PBS. Each ram received a total volume of 60 ~1 ( 1 x lo9 c.f.u., as shown by viable cell counts made on the day of inoculation), both conjunctivally (30 ~1) and preputially (30 ~1). This dose was chosen on the basis of previous experiments (Fensterbank et al., 1982; Blasco et al., 1987; Marin et al., 1990a). The efficacy of the challenge infection was determined by examining the shedding of B. ovis in the semen of unvaccinated controls. Two weeks after challenge, ten of the 15 control rams excreted B. ovis. All animals were slaughtered 15 weeks after the challenge inoculation and subjected to bacteriological and histological examinations. A ram was classified as protected if no B. ovis was isolated from any of the 13 organs sampled at necropsy. The serological response to HS was analyzed on all serum samples taken prior to treatment and at the following weeks after the first vaccination: 1, 2,
VACCINES
AGAINST
B. OVIS IN RAMS
261
4, 8, 12, 18 (just prior to revaccination), 19, 21, 25, 27, 31, 35, 39, 41, 42, 43,45,46,48 ( 1 week prior to challenge inoculation), 50,52,53, and 55. The presenceof antibodies to OMPs was tested on sera from four representative rams of each treatment group. The four rams were selectedby the intensity of the infection after challenge, with one animal being completely negative (absenceof B. ovis in all organs) and three animals showing different degreesof infection: low ( + ) , intermediate ( + + ) and severe( + + + ), basedon the proportion of bacteriologically positive organsand the numbers of B. ovis per organ. In the vaccinated groups, the sera analyzed were taken prior to vaccination, at Weeks 2 and 12 after the first vaccination, Weeks 1 and 30 after revaccination, and Weeks 3, 10, and 15 after challengeinoculation. In the control group, analyseswere performed on sera taken prior to infection and at Weeks 3, 10, and 15 after challengeinoculation. At the same sampling periods the serological responseto R-LPS was determined on sera from the uninfected and the severelyinfected ( + + + ) animals of eachgroup. Bacteriology Cultures were performed on the liver, spleen,testes, epididymides, vesicular glands, bulbourethral glands, ampullae, and cranial (submaxillary, parotid, retropharyngeal), iliac, precrural and prescapular lymph nodes. After removing small portions for histological studies, the whole organs (except liver, of which only one lobe wasprocessed),werehomogenizedby using either a blendor or a Stomacher (Seward Medical Division, London, UK) after adding 20 ml of saline for liver, spleen,testes and epididymides, or 5 ml for the other organs. One ml of each homogenate was seededonto each of two plates of modified Thayer-Martin’s medium (Brown et al., 1971). Colonies were counted after incubation for 10 days at 37’ C in a 10%COZ atmosphere. The number of colonies per organ was derived from the mean of the counts on the two plates. Brucella colonies were identified by standard procedures (Alton et al., 1988). Pathology Portions of epididymides, vesicular glands and ampullae were fixed for 24 h in 10%buffered formalin and embeddedin paraffin. Sections 5-6 pm thick were cut and stained with haematoxylin and eosin by standard procedures. Immunological procedures The ELISA was performed according to protocols described previously (Riezu-Boj et al., 1986; Marfn et al., 1989). Polystyrene plates were coated by overnight incubation at 37oC with a solution of HS extract ( 5 pg per well)
267
J.M. BL.k!KO
ET .AL.
in carbonate buffer (pH 9.6 ). Peroxidase-conjugated rabbit anti-sheep IgG (H + L specificity, Nordic Immunological Laboratory, Tilburg, Netherlands) was used at a dilution of 1 : 3500. ABTS (2,2’-azino-bis, 3 ethylbenzthiazoline-6-sulfonic acid) was used as the cromophore in the developing reaction. The response to R-LPS was measured by the same ELISA with purified RLPS ( 5 ,ug ml- ’ in coating buffer). The response to OMPs was assessed by immunoblot using PCP-HS as a fraction rich in OMPs (Riezu-Boj et al., 1990). The following modifications of the methods of Towbin et al. ( 1979) and Burnette ( 198 1) resulted in a satisfactory OMP transfer and immunoenzymatic detection. SDS-PAGE was carried out in 13.5% acrylamide (Laemmli, 1970) on 1.5 mm thick slabs with an antigen load of 20 /cg per well. Proteins were transferred to Immobilon membranes (Immobilon P, Millipore, Bedford, MA) using a semi-dry transblotter (Bio-Rad Laboratories, Richmond, CA) with a constant current of 200 mA for 30 min. Blots were blocked by overnight incubation at room temperature with PBS containing 3.0% skim milk and 0.05% Tween 20. Incubation with sera ( 1: 50 in blocking buffer without skim milk) was performed overnight at room temperature. Antibodies to blotted OMPs were detected by a 2 hour incubation with the peroxidase-conjugated rabbit anti-sheep IgG diluted at 1: 500 in the same buffer. Finally, peroxidase activity was detected with H,O? and 4-chloro-1-naphtol.
Chemical analyses Total protein was determined by the method of Lowry et al. ( 195 1 ), with bovine serum albumin as the standard. Analysis for 2-keto-3-deoxyoctonate corrected for 2-deoxyaldoses was performed by the method of Warren as modified by Osborn (Warren, 1959; Osborn, 1963 ).
Statistics The majority of analyses were performed by the Chi square method with correction for continuity (Snedecor and Cochran, 1989 ). The Wilcoxson Rank Sum test was used to compare numbers of culture positive organs per infected ram among treatment groups. ELISA data and immunoblot reactions were analyzed in a one way analysis of variance. For this purpose titers were transformed to integers from one to ten corresponding to titers of 100 to 5 1 200, respectively, and immunoblot reactions to the Group 3 protein were graded from 0 (no visible band) to 5 (most intense reaction).
V,vXINESAG.AINST
H. UI'ISIN
263
R.AMS
RESULTS
Clinical observations No clinical abnormalities were associated with vaccination aside from the formation of small granulomas at the sites of adjuvant inoculation. Two rams (one vaccinated with 45/20 bacterin and one vaccinated with HS-PCP) died from enterotoxemia during the course of the experiment.
Bacteriological e,uamination Significant levels of protection in comparison with the control group were obtained only by vaccination with Rev 1 (P
I
Evaluation ofvaccinssagainst Vaccim?
No. infecled rams per group (O/o1
Rev I HS 45/20 bactcrin PG-OM REO bactcrin Blebs HS-PCP None
B~rcllu
cn,is in rams based upon bacteriological
isolationsat
Statistical difference from control P(Chi square. 1 d.f.)
No. organs culture positive per infected ram (mean*SD)
5/15 (33.3%) 6/15 (40.0%) l/l5 (46.7%)
< 0.005 co.025 < 0.05
3.6t2.3 4.8-11.3 4.9+ 3.2
9/ I5 (60.0%) II/l5 (73.3%)
-co.10 z 0.3
6.Ok3.6 5.4t4.0
> 0.20 > 0.20
I l/l5 (73.3%) 13/14 (92.8%) l3/15 (86.7%)
> 0.25 > 0.15
7.s+3.1 5.2 AI3.4 6.Ok3.1
> 0.20 > 0.30
necropsy”
Statistical difference from control P( rank sum test) 0.03 0.10 0.15
“Groups of I5 rams were vaccinated twice at an IS week interval. escept for one group that received a single dose of Rev I. Animals received a challenge inoculation 3 I weeks after the second vaccination and were nccropsied I5 weeks later. “Rev I. B. rt~elircr~sis strain Rev I; HS, hot saline extract; 45/20 bacterin, from strain 45/20 of B. nbo~r~rs: PC-OM, outer membrane-peptidoglycan complex; REO bacterin. from B. oris strain REO 198: blebs. from outer membranes; HS-PCP. HS extracted with ether petroleum-chloroform-phenol: none. unvaccinated rams. All subcellular vaccines were prepared from B. ovis strain REO 198.
364
J.M. BLASCO
El-AL.
bulbourethral glands, ampullae, epididymides and spleen. No single organ served as a consistent indicator of infection. Although in almost all of the infected animals organisms were isolated from one or more genital organs, there were three animals in which the infection was exclusively extragenital. The organ distribution of B. orjs among infected rams from the various vaccine groups did not differ significantly from that in the unvaccinated group. The numbers of B. oris isolated from a majority of organs were low ( I25 c.f.u.), while the percentageof organs containing high numbers of bacteria ( 2 I26 c.f.u. ) rangedfrom 15to 33% in the various groups (data not shown ). There were no significant differences in the percentagesof infected organs with low ( 125 c.f.u.), intermediate (26-125 c.f.u.), or high ( > 125 c.f.u.) numbers of bacteria between the unvaccinated control group and any of the vaccine groups. Histopathology Lesionswere in accord with previous descriptions (Blasco, 1990) and were classified as severe (marked edema, interstitial fibrosis, epithelial hyperplasia, and widespread infiltration of mononuclear cells including plasma cells) or mild to moderate (vacuolation of basilar portions of the epithelium, and mild to moderate degreesof epithelial hyperplasia, mononuclear cell inliltration, and fibrosis). Severe lesions were restricted to some of the rams that were infected at the time of necropsy. The percentageof infected rams with severelesions in one or more organs was significantly greater (PcO.05) in the group vaccinated with Rev 1 than in those vaccinated with HS, 45/20 bacterin, REO bacterin, and HS-PCP (data not shown). Antibody responses Prior to treatment antibodies to HS antigens were undetectable in the majority of rams and low (titer= 100) in the remainder (data not shown). Vaccination with live Rev 1 or with two injections of the non-living products causeda rapid and sustaineddevelopment of antibodies for HS antigens (data not shown). The highest titers achieved after revaccination and the highest titers at Week 48, just prior to challengeinfection, were obtained in the groups that received the protective vaccines HS and 45/20 (Table 2). However, the lowest titers to HS were obtained in the Rev 1 vaccine group (Table 2 ). Moreover, in none of the groups was there a significant difference in antibody titers between protected and unprotected rams (Table 2). Challenge inoculation resulted in the formation of high antibody titers in all of the treatment groups, again with no significant differences in peak responsesbetween the infected and uninfected rams within a group (Table 2). In accord with prior results (Gamazo et al., 1989b), weak immunoblot re-
TABLE 1 ELISA titers against HS extract in rams that were culture positive or culture negative following challcngc inoculation” Vacci&
Rev I HS 45/20 bactcrin PC-OM REO bacterin Blebs HS-PCP None
Infection status<
+ + + + + + + + -
No. of rams
5 IO 6 9 7 7 9 6 II 4 II 4 I3 I I3 ?.
ELISA titers (mean k SD)d Highest titer after vaccination’
Titer just prior to challenge ( Week 48 )’
Highest titer after challenge”
5.sI!zo.45 j.SkO.32 l0.0I!I0.0 9.9LO.33 9.9+0.3s 10.0+0.0 7.8 + 0.44 7.7+ I.2 8.0+0.63 S.2&0.96 7.0+0.77 7.2 k 0.50 9.2+0.69 9.0 nd nd
5.4kO.55 5.4kO.53 7.st 1.4 8. I z!I0.93 9.1* I.1 9.7kO.49 6.SkO.67 6.SkO.75 7.1 I!I I .3 s.o+o.s1 5.4fO.52 6.0+0.0 7.2+ I.1 7.0 nd nd
s.s*o.s4 8.3+ 1.6 9.9rto.41 9.7kO.71 9.9+0.3s 10.0+0.0 9.2 k 0.44 9.3+o.s2 9.4 k 0.69 9.s*o.50 9.5fO.U 9.oI!Io.o 9.9+0.x 10.0 9.4kO.67 9.7kO.56
“Groups of I5 rams were vaccinated twice at an IS week interval, escrpt for one group that rcccivcd a single dose of Rev I. Animals received a challenge inoculation 3 I weeks after the second vaccination and were nccropsied I5 weeks later. “Rev I, B. r~lilorsis strain Rev I: HS. hot saline extract; 45/20 bacterin. from strain 45/20 of R. c10o~trt.s:PC-OM, outer membrane-peptidoglycan complcs: REO bacterin, from B. 0~;s strain REO 198: blcbs. from outer membranes; HS-PCP. HS estracted with ether petroleum-chloroform-phenol; none. unvaccinated rams. All subcellular vaccines were prepared from B. Otis strain REO 19s. ‘( + ). culture positive and ( - ), culture negative at the time of necropsy. “ELISA titers from I00 to 5 I 200 were converted into integers from one to ten. respectively. nd. not done. ‘In no paired comparison were thcrc significant differences in titers between culture positive and culture negative animals.
actions to Group 3 protein (graded 1.O or less) were present in some rams ( 17 of 32 ) prior to vaccination or infection (data not shown). Almost all of the animals demonstrated anti-OMP responsesafter vaccination, with the most intense responsesdirected against the group 3 protein (Fig. 1). Rev lvaccinated rams showed the weakest responsesto OMPs (Fig. 1). Control rams produced OMP antibody responsesfollowing challengeinoculation that were very similar to those observed in naturally infected rams (Riezu-Boj et al., 1990), with generally more intense responsesin animals that were more heavily infected (Fig. 1). In some comparisons between vaccinated rams that
266
J.M. BLASCO
567
567
567
3+
2+
+
C
123457
123457
123457
3+
REV1
ET AL.
123457
3+
HS
Fig. 1. Immunoblot analysis of the antibody responseto membrane proteins of B. ovis in sera of representative vaccinated and unvaccinated rams. C, control unvaccinated rams that presented high (3+ ) intermediate (2+ ) or low ( + ) degrees of infection as judged by bacteriological results at necropsy. Rev 1, B. tnelilensis Rev 1 vaccinated rams that became heavily infected (3 + ) or were uninfected; HS, HS vaccinated rams that became heavily infected (3 + ) or were uninfected. The immunoblots were performed with HS-PCP (20 pg per lane) using serum samples taken ( 1) 2 weeks after the first vaccination; (2) I2 weeks after the first vaccination; (3) 1 week after revaccination; (4) 30 weeks after revaccination ( 1 week prior to challenge inoculation); (5) 3 weeks after challenge inoculation; (6) 10 weeks after challenge inoculation; (7) 15 weeksafter challengeinoculation. Bars at left represent positions of standard molecular weight markers at 94, 67,43, 30,2 I, and 14.3 kDa.
became heavily infected or remained uninfected following challengeinoculation, there was an apparent direct association between the intensity of OMP antibody responsesjust prior to challenge and culture-negative status thereafter (Fig. 1,Lanes 4). However, statistical analysisof pooled data from seven culture negative rams and sevenrams with ‘severe’ ( + + + ) infections (one from each vaccine group) revealed no significant differences between infection status and the intensity of the OMP Group 3 immunoblot reactions at 1 week after revaccination, at 30 weeks after revaccination ( 1 week prior to challenge inoculation), or at 3 weeks after challenge inoculation. The same analyseslikewise demonstratedno significant differencesbetweenthesegroups in the magnitude of ELISA titers for HS or R-LPS antigens at any of the three time points (data not shown). DISCUSSION
Vaccination of rams with HS, a subcellular extract of B. ovis, conferred a level of protective immunity against virulent B. ovis which closely approximated that provided by the live vaccine Rev 1 (Table 1). Experiments currently being performed in the mouse model confirm this finding. Vaccination of mice with HS reduced mean numbers of B. ovis in the spleen at 2 weeks
VACCINES
AGAINST
6. OVIS IN RAMS
261
after challenge infection by over 1000 fold, and provided the same level of protection afforded by vaccination with live Rev 1 (M.P. Jimenez de Bagties, J.M. Blasco and A.J. Winter, unpublished data, 1992). HS therefore represents a potentially useful vaccine against B. ovis, and its use would offer a solution to the problems associatedwith live Rev 1. The bacteriological data confirm prior reports (Fensterbank et al., 1982; Blasco et al., 1987;Marin et al., 1990b) that an accurateevaluation of B. ovis infection requires a complete cultural examination of genital as well as extragenital organs. Prior evaluations of B. ovis vaccines that relied on cultures only of semenand/or testes (Van Heerden and Van Rensberg,1962;Claxton, 1968; Swift and Maki, 1968; Gradwell and Van Zyl, 1975; McGowan and Harrold, 1979; Garcia-Carrillo, 1981) were therefore inadequate. The low numbers of B. ovis in a majority of infected organs also point out the need to culture large samples of tissue homogenatesfor a reliable assessmentof infection status. In this connection, it is evident that vaccines that provided significant protection neverthelessfailed to modify the organ distribution of the infection or the levels of colonization of infected organsin those animals that remained culture positive. The results of this investigation are consistent with our original hypothesis that a subcellular vaccine rich in outer membrane components would be successful against B. ovis. However, not all the vaccines containing outer membrane protein antigens conferred protection. Whereas HS extracts, enriched in R-LPS and OMPs, were protective, the protein fraction of HS (HS-PCP) which contained less than 1% R-LPS was not (Table 1) . It thus appearsthat OMPs alone could not induce protective immunity against B. ovis and that, as in infections with smooth brucellae (Plommet, 1987; Winter et al., 1988, 1989), the LPS may be an important protective antigen. However, R-LPS seemedlikewise insufficient to induce protective immunity, as illustrated by the fact that the PG-OM, outer membrane blebs, and B. ovis bacterin did not provide protection (Table 1) . The subcellular fractions did not differ substantially in protein profile (Gamazo et al., 1989a), but did vary in the percentagecontents of LPS and protein, which gave ratios of 0.53 for HS, 0.41 for PG-OM, 0.39 for outer membrane blebs, and less than 0.01 for HS-PCP. Therefore, the different proportions of OMPs and R-LPS might explain the different levels of protection provided by thesevaccines. Experiments are in progressto assessthe respective contributions of R-LPS and OMPs to protective immunity against B. ovis. We found no statistical associations in this experiment between the levels of antibodies to the HS extract or its antigenic components (OMPs and RLPS) and protection from infection. These negative findings must be interpreted cautiously, and do not invalidate our hypothesis (Riezu-Boj et al., 1986) that antibodies directed against the surface exposedOMP and R-LPS antigens mediate protection against B. ovis. In the murine model of B. abortus
268
J.M. t3L.GCO
ET AL.
infection, for example, we have likewise been unable to demonstrate a simple statistical relationship between protection and serum levels of 0 polysaccharide-specific antibodies (Winter et al., 1988, 1989) even though we know from passivetransfer experiments that theseantibodies confer excellent protection (Montarez et al., 1986;Winter et al., 1989). An evaluation of cell mediated immunity induced by the vaccines was beyond the scope of this study. At the present time there is no information on the protective role of cell mediated immunity in ram epididymitis. Cell mediated immune responsesmay have been induced by vaccines incorporated with L-l 2 1 and MDP, as was demonstrated by Byars and Allison ( 1987). Even though antibody responsesinduced by Rev 1 were relatively weak (Table 2, Fig. 1), we hypothesize that this vaccine stimulated vigorous cell mediated immune responsesthat provided good protection. We are currently employing the mouse model system to examine the functions of humoral and cell mediated immune responsesin protection against B. mix ACKNOWLEDGMENTS
We are deeply indebted to Santiago Lazaro and Marisa Pardo for excellent technical assistance.This work was supported in part by the US-Spain Joint Committee (CCA-85 1O/O85), and CICYT and CONAI from Spain.
REFERENCES Afzal. M.. Tengerdy. R.P., Ellis, R.P., Kimberling, C.V. and Morris, C.J., 1984. Protection of rams against epididymitis by a B~trcellu oris vitamin E adjuvant vaccine. Vet. Immunol. Immunopathol.. 7: 293-304. Alton, G.G.. Jones, L.M.. Angus. R.D. and Verger. J.M.. 1988. Techniques for the brucellosis laboratory. Institute National de la Recherche Agronomiquc, Paris, I90 pp. Biberstcin, E.L., McGowan, B.. Robinson, E.A. and Harrold, D.R.. 1963. Epididymitis in rams. Studies on immunity. Cornell Vet., 52: 2 14-2 17. Blasco, J.M., 1990. Bruce/la oris. In: K. Nielsen and J.R. Duncan (Editors), Animal Bruccllosis. CRC, Boca Raton, FL, pp. 35 l-374. Blasco. J.M., Marin, C.M.. Barberan, M., Moriyon, I. and Diaz. R.. 1987. Immunization with Brmdlu /rre/i[crtsis Rev I against Brucella oris infection of rams. Vet. Microb.. 14: 38 l-392. Brown, G.M., Ranger, C.R. and Kelly, D.J.. I97 I. Selective medium for the isolation ofBrrd/a oris. Cornell Vet., 6 I: 265-280. Buddlc. M.B., 1962. Production of immunity in rams against Brudln uris infection. NZ. Vet. J.. IO: I l-15. Buddle, M.B., Calverley, F.K. and Boyes, B.W., 1963. Brlrcd/u oris vaccination of rams. NZ. Vet. J.. I I: 90-93. Burnette. W.N.. I98 I. ‘Western blotting’: electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radiolabeled protein A. Anal. Biochem.. I 12: 195-203.
VACCINES
AGAINST
B. OVIS IN RAMS
269
Byars, N.E. and Allison, A.C., 1987. Adjuvant formulation for use in vaccinesto elicit both cellmediated and humoral immunity. Vaccine, 5: 223-228. Claxton, P.D., 1968. Brucella ovis vaccination of rams. Comparison of two commercial vaccines and two methods of vaccination. Aust. J. Vet., 44: 48-54. Diaz, R. and Bosseray,N., 1973. Identification d’un composCantigenique specifique de la phase rougueuse (R) des Brucella. Ann. Rech. V&t., 4: 283-292. Diaz, R., Jones, L.M. and Wilson, J.B., 1967. Antigenic relationship of Brucellu ovis and Brucella melitensis. J. Bact., 93: 1263-1268. Erasmus, J.D. and Bergh, E.C., 1985. Ovine brucellosis: repeated vaccination with Rev 1 vaccine and the prevalence of the diseasein the Winburg district. J. S. Afr. Vet. Assoc., 56: 205209. Fensterbank, R., Pardon, P. and Marly, J., 1982. Efficacy of Brucella Melitensis Rev 1 vaccine against Brucellu ovis infection in rams. Ann. Rech. Vet., 13: 185-190. Gamazo, C. and Moriybn, I., 1987. Release of outer membrane fragments by exponentially growing Brucella melitensis. Infect. Immun., 55: 609-6 15. Gamazo, C., Winter, A.J., Moriybn, I., Riezu-Boj, J.I., Blasco,J.M. and Diaz, R., 1989a. Comparative analyses of proteins extracted by hot saline or released spontaneously into outer membrane blebs from field strains of Brucellu ovis and Brucella melitensis. Infect. Immun., 57: 1419-1426. Gamazo, C., Winter, A.J., Riezu, J.I., Blasco, J.M. and Jimtnez de BagtiCs,M.P., 1989b. Respuesta inmune humoral inducida por las preparaciones antigbnicas subcelularesde Brucella ovis en corderos. Vol. 2, Abs. 1 p. 89, Abstr. XII Congreso National de la Sociedad espaiiola de Microbiologia. Garcia-Carrillo, C., 1981. Protection of rams against Brucella ovis infection by Brucella melitensis Rev 1. Zblt. Vet. .Med. B, 28: 425-429. Gradwell, D.V. and van Zyl, F.E., 1975. Effectivity of Rev 1 vaccine in rams against Brucella ovis infection. J. S. Afr. Vet. Assoc.,46: 349-354. Haughey, K.G. and Hughes, K.L., 1969. The apparent failure of a vaccine to protect rams against naturally occurring Brucella ovis infection. Aust. Vet. J., 45: 1O-l 6. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227: 680-685. Lowry, O.H., Rosebrough, N., Farr, A.L. and Randall, R.J., 195 1. Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193: 265-275. Marin, C., Jimtnez de BagtiCs,M.P., Blasco,J.M., Gamazo, C., Moriybn, I. and Diaz, R., 1989. Comparison of three serological tests for Brucellu ovis infection of rams using different antigenic extracts. Vet. Rec., 125: 504-508. Marin, C.M., BarberBn, M., JimCnez de Bag%, M.P. and Blasco, J.M., 1990a. Comparison of subcutaneous and conjunctival routes of Rev I vaccination for the prophylaxis of Brucella ovis infection of rams. Res. Vet. Sci., 48: 209-2 15. Marin, C., JimCnez de Bag&, M.P., Barber&n, M. and Blasco, J.M., 1990b. Efficacy of long acting oxytetracycline combined or not with streptomycin for treating Brucellu ovis infection of rams. Am. J. Vet. Res., 50: 560-563. McGowan, B. and Harrold, D.R., 1979. Epididymitis in rams: studies on vaccine efficacy. Cornell Vet., 69: 73-76. Montaraz, J.A, Winter, A.J., Hunter, D.M., Sowa, B.A., Wu, A.M. and Adams, L.G., 1986. Protection against Brucelh abortus in mice with 0-polysaccharide-specific monoclonal antibodies. Infect. Immun., 5 1: 961-963. Moreno, E., Jones, L.M. and Berman, D.T., 1984. Immunochemical characterization of rough Brucella lipopolysaccharides. Infect. Immun., 43: 778-782. Osborn, M.J., 1963. Studies on the gram-negative bacterial cell wall. I. Evidence for the role of
270
J.M. BLASCO
ET AL.
2-keto-3 deoxyoctonate in the lipopolysaccharide of Salmonella typhimurium. Proc. Natl. Acad. Sci. USA, 50: 499-506. Plommet, M., 1987. Brucellosisand immunity: humoral and cellular components in mice. Ann. Inst. Pasteur Micro., 138: 105-l 10. Riezu-Boj, J.I., Moriyon, I., Blasco, J.M., Marin, C. and Diaz, R., 1986. Comparison of lipopolysaccharideand outer membrane protein-lipopolysaccharide extracts in an enzyme-linked immunosorbent assayfor the diagnosis of Brucella ovis infection. J. Clin. Micro., 23: 938942. Riezu-Boj, J.I., Moriyon, I., Blasco, J.M., Gamazo, C. and Diaz, R., 1990. Antibody response to Brucella ovis outer membrane proteins in ovine brucellosis. Infect. Immun., 58: 489-494. Ris, D.R., 1967. The persistenceof antibodies against Brucella ovis and Brucella abortus in rams following vaccination: a field study. NZ. Vet. J., 15: 94-100. Snedecor, G.W. and Cochran, W.G., 1989. Statistical methods. 8th edn. Iowa State University Press,Ames, IA, 503 pp. Swift, B.L. and Maki, L.R., 1968. Immunological studies on three ram epididymitis bacterins. Cornell Vet., 58: 659-665. Towbin, H., Staehelin, T. and Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets:procedure and some applications. Proc. Natl. Acad. Sci. USA, 16: 4350-4354. Van Heerden, K.M. and van Rensburg, S.W., 1962. The immunization of rams against ovine brucellosis. J.S. Afr. Vet. Assoc., 33: 143-148. Verstreate, D.R., Creasy, M.T., Caveney, N.T., Baldwin, C.L., Blab, M.W. and Winter, A.J., 1982. Outer membrane proteins of Brucella abortus: isolation and characterization. Infect. Immun., 35: 979-989. Warren, L., 1959. The thiobarbituric acid assayof sialic acids. J. Biol. Chem., 234: 1971-1975. West, D.M., Johnstone, A.C., Bruere, A.N. and Chapman, H.M., 1970. Epiphysitis in rams following vaccination against Brucella ovis infection. NZ. Vet. J., 26: 133-l 34. Winter, A.J., Rowe, G.E., Duncan, J.R., Eis, M.J., Widom, J., Ganem, B. and Morein, B., 1988. Effectivenessof natural and synthetic complexes of porin and 0 polysaccharide as vaccines against Brucella abortus in mice. Infect. Immun., 56: 2808-28 17. Winter, A.J., Duncan, J.R., Santisteban, C.G., Douglas, J.T. and Adams, L.G., 1989. Capacity of passivelyadministered antibody to prevent establishment of Brucella abortus infection in mice. Infect. Immun., 57: 3438-3444.
31 ( 1993) 251-270
257
Evaluation of whole cell and subcellular vaccines against Brucella ovis in rams J.M. Blasco”,C. Gamazob,A.J. Winter’, M.P. Jimhez de BagtiC?, C. Marin”, M. BarberAnd,I. Moriy6nb, B. Alonso-Urmenetaband R. Diazb “Servicio de Investigacidn Agraria. Diputacidn General de Aragdn, 50080 Zaragoza, Spain “Departamento de Microbiologia, Universidad de Navarra, 31080 Pamplona, Spain CDepartment of Veterinary Microbiology, Immunology and Parasitology Cornell University, Ithaca, NY 14853, USA ‘Departamento de Patologia Animal, Universidad de Zaragoza. 50080 Zaragoza, Spain (Accepted 5 October 1992)
ABSTRACT Blasco, J.M., Gamazo, C., Winter, A.J., Jimenez de Bagiits, M.P., Marin, C., Barberan, M., Moriyon, I., Alonso-Urmeneta, B. and Diaz, R., 1993. Evaluation of whole cell and subcellular vaccines against Brucella ovis in rams. Vet. Immunol. Immunopathol., 37: 251-270 Five antigen preparations from Brucella ovis strain REO 198 were incorporated with the pluronic polymer L-l 2 1 and muramyl dipeptide and tested as vaccines against B. ovis infection of rams. The antigenic preparations were: ( 1) a fraction enriched in outer membrane proteins and rough lipopolysaccharide (hot saline extract, HS); (2) the proteins from HS substantially free of lipopolysaccharide; (3 ) outer membrane blebs; (4) outer membrane-peptidoglycan complexes extracted with detergent; (5) killed whole cells. The experimental vaccines were compared with two standard vaccines, rough Bruce/la abortus 45120 whole killed cells in an oil based adjuvant, and live Brucella meiitensis Rev 1. Immunizations with non-living vaccines were performed on two occasions, I8 weeks apart. The rams were challenged with a virulent strain of B. ovis 3 1 weeks after the second vaccination and slaughtered I5 weeks thereafter. Rates of infection in groups vaccinated with Rev 1 (33%), and HS (40%) were significantly lower (PcO.005 and PcO.025, respectively) than that in the non-vaccinated control group (871). Strain 45/20 was the only other vaccine that conferred a significant level of protection (50%) (PcO.05). The organ distribution of the infection and the level of colonization of infected organs did not differ significantly between infected animals in the various vaccine groups and those in the unvaccinated control group. No statistically significant relationship was detected between the magnitude of the antibody responses to the HS extract, to outer membrane proteins, or to the rough lipopolysaccharide, and freedom from infection. The results indicate that the HS extract of B. ovis may represent a useful alternative to B. melitensis Rev 1 or B. abortus 45120 as a vaccine against B. ovis. ABBREVIATIONS ABTS, 2,2’-azino-bis, 3 ethylbenzthiazoline-6-sulfonic acid; HS, hot saline; LPS, lipopolysacharide; OMP, outer membrane protein; PBS, phosphate buffered saline. Correspondence to: J.M. Blasco, Servicio de Investigation de Aragori, 50080 Zaragoza, Spain.
0 1993, Elsevier Science Publishers
Agraria,
Ap 727 Diputacion
B.V. All rights reserved 0165-2427/93/$06.00
General
258
J.M. BLASCO
ETAL.
INTRODUCTION
Brttcellu ovis is the causative agent of contagious ram epididymitis, an infectious diseasecausing abortion and a decreasein ram fertility with severe economic losses.Although several control methods have been proposed, vaccination representsthe only practical measure in countries with a high incidence of the disease (Blasco, 1990). Brzrcella meliferzsisRev 1, a smooth attenuated strain, is generally consideredthe best vaccine available. The rate of protection afforded by Rev 1 against experimental infection varies from 60 to 1OO%,depending on the criteria used to assessprotection (Van Heerden and Van Rensburg, 1962;Gradwell and Van Zyl, 1975;Garcia-Carrillo, 1981; Fensterbank et al., 1982; Blasco et al., 1987; Marin et al., 1990a). However, Rev 1 evokesa serological responseinterfering with the interpretation of serological tests used to diagnose infection by B. melitensis and B. ovis (Ris, 1967; Gradwell and Van Zyl, 1975; Erasmus and Bergh, 1985; Blasco et al., I987 ), although conjunctival vaccination reducesthese problems (Marin et al., 1990). Additional disadvantagesof Rev 1 are that its use is not allowed in countries free from B. melitensis, and that this strain is pathogenic in humans. Several bacterins have been used with less success (Biberstein et al., 1962:Buddle, 1962;Buddle et al., 1963;Claxton, 1968;Swift and Maki, 1968; Haughey and Hughes, 1969;Afzal et al., 1984). The efficacy of B. ovis bacterins is improved by the simultaneous inoculation of live B. abortzrsstrain 19 (McGowan and Harrold, 1979)) but this combined vaccine elicits long lasting serological responsesthat interfere with serological diagnosis (Ris, 1967) and can cause sequelae such as epiphysitis and epididymitis (West et al., 1970). In contrast to B. melitensis, B. ovis is a naturally rough form lacking the Opolysaccharide side chain that is the major antigen of smooth brucellae (Diaz and Bosseray, 1973; Moreno et al., 1984). However, both speciesshare soluble antigens (Diaz et al., 1967) and part of the core-lipid A lipopolysacharide (LPS) determinants (Diaz and Bosseray, 1973;Moreno et al., 1984). Moreover, B. melitensis and B. ovis strains have a quantitatively similar outer membrane protein (OMP) profile, and the OMPs show strong immunological cross reactivity (Gamazo et al., 1989a). Along with rough LPS (R-LPS), thesegroups of OMPs are major components of the hot saline extracts (HS) of B. 0vi.sused in the standard diagnostic tests, and B. ovis-infected rams developed an antibody responsenot only to the R-LPS but also to OMPs (RiezuBoj et al., 1986, 1990). Since B. melitensis and B. ovis share OMPs, but not the 0-polysaccharide as surface antigens, it is possible that OMPs play a significant role in the protection afforded by Rev 1 against B. ovis. If so, a subcellular vaccine containing the outer membrane components present in HS could be effective against B. ovis and would not interfere with the diagnosis of B. melitensis. The objective of the present work was to compare the effi-
V,4CCINES
t\GAINST
N. 01 ‘IS IN R.AklS
259
cacy of subcellular vaccines, bacterins, and live Rev 1 against an experimental challenge with virulent B. ovis in rams. MATERIALS
AND METHODS
.4ntigensjiw vaccinesand irnmrnological assays All antigens were obtained from B. ovis REO 198 cells cultivated in liquid medium (Gamazo et al., 1989a). To prepare the B. ovis whole cell vaccine (bacterin ), cells were killed by incubation in 0.5% phenol-20 mM phosphate buffered saline (PBS) (pH 7.2) for 2 days at 4°C washed three times with distilled water and lyophilized. A fraction rich in peptidoglycan-outer membrane (PG-OM ) was prepared by Sarkosyl extraction of cell envelopes (Verstreate et al., 1982). Outer membrane blebs were obtained by ultracentrifugation ( 100 OOOxg, 1h) of spent culture broths as described previously (Gamazo and Moriyon, 1987). Crude hot saline extracts (HS), the protein fraction of HS substantially freed from LPS by extraction with petroleum ether-chloroform-phenol (HS-PCP), and the purified R-LPS were prepared by methods described previously (Gamazo et al., 1989a; Riezu-Boj et al., 1990). The percentages of LPS (estimated by the content of 2-keto-3-deoxyoctonate) and protein in each of the subcellular vaccines were as follows: HS, LPS 34.6%, protein 65.4%; HS-PCP, LPS < I%, protein 93.0%; outer membrane blebs, LPS 13.6%, protein 34.9%: PG-OM, LPS 27.4%, protein 66.1%.
Preparation and administration qf vaccines Subcellular vaccines and B. ovis bacterins were administered in an adjuvant containing pluronic polymer L- 12 1 (BASF Wyandotte, Parsippany, NJ ), squalane (Sigma Chemicals, St. Louis, MO), and muramyl dipeptide (N-acetylmuramyl-L-cu-aminobutyryl-D-isoglutamine; Choay Chimelab, Paris, France). The emulsions were prepared by a modification of the method of Byars and Allison ( 1987). Briefly, 0.3 ml of L-121 were mixed with 4.2 ml of an 0.4% solution of Tween 80 by vortexing with 6 mm glass beads. Squalane (0.5 ml) was added to this mixture and vortexing was continued until no oil droplets larger than approximately 40 pm diameter could be detected by phase contrast microscopy. The emulsion was then supplemented with 10 mg of muramyl dipeptide and mixed thoroughly with an equal volume of an antigen solution in PBS. Rams received subcutaneously 1 ml of vaccine containing 30 ,~l of L- 12 1, 50 ~1 of squalane, 1 mg of muramyl dipeptide, and the corresponding antigen. The amounts of antigen administered per ram were, bacterin, 10.0 mg; PG-OM, 2.0 mg; outer membrane blebs, 2.0 mg; HS, 2.0 mg; and HS-PCP, 1 .O mg. Preliminary studies done in Finn Dorset lambs showed
260
J.M. BL.-\SCO
ET.AL.
that these doses in adjuvant induced strong antibody responses (Gamazo et al., 1989b). B. abortus strain 45/20 bacterin in a mineral oil-based adjuvant (Duphavac) was purchased from Duphar (Duphar B.V., Amsterdam, Netherlands). One ml of this vaccine (2 x 10’ ’ cells ) was administered intramuscularly to each animal. B. melitensis strain Rev 1 was kindly provided by Dr. J.M. Verger (Station de Pathologie de la Reproduction, Nouzilly, France). To prepare the vaccine, smooth Rev 1 was grown for 3 days on Tryptone Soya Agar (Oxoid London, UK) in Roux flasks, washed twice with sterile saline, suspended in a milklactose vehicle and lyophilized. After lyophilization there were 2 x 10” c.f.u. per ml. The lyophilized vaccine was diluted in saline and animals were vaccinated subcutaneously with 5 x 1OSc.f.u. in 1 ml.
Experimental design One hundred and twenty crossbred brucella-free rams, 5 months old, were divided into eight groups ( 15 lambs per group). Individual groups were vaccinated with the following antigens in L- 12 1-muramyl dipeptide adjuvant: B. ovis bacterin, PG-OM, blebs, HS, and HS-PCP. In addition one group was vaccinated with Rev 1, one with 45/20 bacterin, and a control group remained unvaccinated. Revaccinations were performed 18 weeks later, except for the group that received Rev 1 which was not revaccinated. At regular intervals after vaccination and revaccination, blood samples were taken and animals were examined for genital lesions. Challenge inoculations were performed 3 1 weeks after the second vaccination with a virulent strain of B. ovis obtained from a naturally infected ram. This strain was grown on Tryptone Soya Agar supplemented with 0.5% yeast extract and 5.0% sterile bovine serum for 72 h at 37°C in a 10% CO? atmosphere. The cells were then harvested, washed and resuspended in sterile PBS. Each ram received a total volume of 60 ~1 ( 1 x lo9 c.f.u., as shown by viable cell counts made on the day of inoculation), both conjunctivally (30 ~1) and preputially (30 ~1). This dose was chosen on the basis of previous experiments (Fensterbank et al., 1982; Blasco et al., 1987; Marin et al., 1990a). The efficacy of the challenge infection was determined by examining the shedding of B. ovis in the semen of unvaccinated controls. Two weeks after challenge, ten of the 15 control rams excreted B. ovis. All animals were slaughtered 15 weeks after the challenge inoculation and subjected to bacteriological and histological examinations. A ram was classified as protected if no B. ovis was isolated from any of the 13 organs sampled at necropsy. The serological response to HS was analyzed on all serum samples taken prior to treatment and at the following weeks after the first vaccination: 1, 2,
VACCINES
AGAINST
B. OVIS IN RAMS
261
4, 8, 12, 18 (just prior to revaccination), 19, 21, 25, 27, 31, 35, 39, 41, 42, 43,45,46,48 ( 1 week prior to challenge inoculation), 50,52,53, and 55. The presenceof antibodies to OMPs was tested on sera from four representative rams of each treatment group. The four rams were selectedby the intensity of the infection after challenge, with one animal being completely negative (absenceof B. ovis in all organs) and three animals showing different degreesof infection: low ( + ) , intermediate ( + + ) and severe( + + + ), basedon the proportion of bacteriologically positive organsand the numbers of B. ovis per organ. In the vaccinated groups, the sera analyzed were taken prior to vaccination, at Weeks 2 and 12 after the first vaccination, Weeks 1 and 30 after revaccination, and Weeks 3, 10, and 15 after challengeinoculation. In the control group, analyseswere performed on sera taken prior to infection and at Weeks 3, 10, and 15 after challengeinoculation. At the same sampling periods the serological responseto R-LPS was determined on sera from the uninfected and the severelyinfected ( + + + ) animals of eachgroup. Bacteriology Cultures were performed on the liver, spleen,testes, epididymides, vesicular glands, bulbourethral glands, ampullae, and cranial (submaxillary, parotid, retropharyngeal), iliac, precrural and prescapular lymph nodes. After removing small portions for histological studies, the whole organs (except liver, of which only one lobe wasprocessed),werehomogenizedby using either a blendor or a Stomacher (Seward Medical Division, London, UK) after adding 20 ml of saline for liver, spleen,testes and epididymides, or 5 ml for the other organs. One ml of each homogenate was seededonto each of two plates of modified Thayer-Martin’s medium (Brown et al., 1971). Colonies were counted after incubation for 10 days at 37’ C in a 10%COZ atmosphere. The number of colonies per organ was derived from the mean of the counts on the two plates. Brucella colonies were identified by standard procedures (Alton et al., 1988). Pathology Portions of epididymides, vesicular glands and ampullae were fixed for 24 h in 10%buffered formalin and embeddedin paraffin. Sections 5-6 pm thick were cut and stained with haematoxylin and eosin by standard procedures. Immunological procedures The ELISA was performed according to protocols described previously (Riezu-Boj et al., 1986; Marfn et al., 1989). Polystyrene plates were coated by overnight incubation at 37oC with a solution of HS extract ( 5 pg per well)
267
J.M. BL.k!KO
ET .AL.
in carbonate buffer (pH 9.6 ). Peroxidase-conjugated rabbit anti-sheep IgG (H + L specificity, Nordic Immunological Laboratory, Tilburg, Netherlands) was used at a dilution of 1 : 3500. ABTS (2,2’-azino-bis, 3 ethylbenzthiazoline-6-sulfonic acid) was used as the cromophore in the developing reaction. The response to R-LPS was measured by the same ELISA with purified RLPS ( 5 ,ug ml- ’ in coating buffer). The response to OMPs was assessed by immunoblot using PCP-HS as a fraction rich in OMPs (Riezu-Boj et al., 1990). The following modifications of the methods of Towbin et al. ( 1979) and Burnette ( 198 1) resulted in a satisfactory OMP transfer and immunoenzymatic detection. SDS-PAGE was carried out in 13.5% acrylamide (Laemmli, 1970) on 1.5 mm thick slabs with an antigen load of 20 /cg per well. Proteins were transferred to Immobilon membranes (Immobilon P, Millipore, Bedford, MA) using a semi-dry transblotter (Bio-Rad Laboratories, Richmond, CA) with a constant current of 200 mA for 30 min. Blots were blocked by overnight incubation at room temperature with PBS containing 3.0% skim milk and 0.05% Tween 20. Incubation with sera ( 1: 50 in blocking buffer without skim milk) was performed overnight at room temperature. Antibodies to blotted OMPs were detected by a 2 hour incubation with the peroxidase-conjugated rabbit anti-sheep IgG diluted at 1: 500 in the same buffer. Finally, peroxidase activity was detected with H,O? and 4-chloro-1-naphtol.
Chemical analyses Total protein was determined by the method of Lowry et al. ( 195 1 ), with bovine serum albumin as the standard. Analysis for 2-keto-3-deoxyoctonate corrected for 2-deoxyaldoses was performed by the method of Warren as modified by Osborn (Warren, 1959; Osborn, 1963 ).
Statistics The majority of analyses were performed by the Chi square method with correction for continuity (Snedecor and Cochran, 1989 ). The Wilcoxson Rank Sum test was used to compare numbers of culture positive organs per infected ram among treatment groups. ELISA data and immunoblot reactions were analyzed in a one way analysis of variance. For this purpose titers were transformed to integers from one to ten corresponding to titers of 100 to 5 1 200, respectively, and immunoblot reactions to the Group 3 protein were graded from 0 (no visible band) to 5 (most intense reaction).
V,vXINESAG.AINST
H. UI'ISIN
263
R.AMS
RESULTS
Clinical observations No clinical abnormalities were associated with vaccination aside from the formation of small granulomas at the sites of adjuvant inoculation. Two rams (one vaccinated with 45/20 bacterin and one vaccinated with HS-PCP) died from enterotoxemia during the course of the experiment.
Bacteriological e,uamination Significant levels of protection in comparison with the control group were obtained only by vaccination with Rev 1 (P
I
Evaluation ofvaccinssagainst Vaccim?
No. infecled rams per group (O/o1
Rev I HS 45/20 bactcrin PG-OM REO bactcrin Blebs HS-PCP None
B~rcllu
cn,is in rams based upon bacteriological
isolationsat
Statistical difference from control P(Chi square. 1 d.f.)
No. organs culture positive per infected ram (mean*SD)
5/15 (33.3%) 6/15 (40.0%) l/l5 (46.7%)
< 0.005 co.025 < 0.05
3.6t2.3 4.8-11.3 4.9+ 3.2
9/ I5 (60.0%) II/l5 (73.3%)
-co.10 z 0.3
6.Ok3.6 5.4t4.0
> 0.20 > 0.20
I l/l5 (73.3%) 13/14 (92.8%) l3/15 (86.7%)
> 0.25 > 0.15
7.s+3.1 5.2 AI3.4 6.Ok3.1
> 0.20 > 0.30
necropsy”
Statistical difference from control P( rank sum test) 0.03 0.10 0.15
“Groups of I5 rams were vaccinated twice at an IS week interval. escept for one group that received a single dose of Rev I. Animals received a challenge inoculation 3 I weeks after the second vaccination and were nccropsied I5 weeks later. “Rev I. B. rt~elircr~sis strain Rev I; HS, hot saline extract; 45/20 bacterin, from strain 45/20 of B. nbo~r~rs: PC-OM, outer membrane-peptidoglycan complex; REO bacterin. from B. oris strain REO 198: blebs. from outer membranes; HS-PCP. HS extracted with ether petroleum-chloroform-phenol: none. unvaccinated rams. All subcellular vaccines were prepared from B. ovis strain REO 198.
364
J.M. BLASCO
El-AL.
bulbourethral glands, ampullae, epididymides and spleen. No single organ served as a consistent indicator of infection. Although in almost all of the infected animals organisms were isolated from one or more genital organs, there were three animals in which the infection was exclusively extragenital. The organ distribution of B. orjs among infected rams from the various vaccine groups did not differ significantly from that in the unvaccinated group. The numbers of B. oris isolated from a majority of organs were low ( I25 c.f.u.), while the percentageof organs containing high numbers of bacteria ( 2 I26 c.f.u. ) rangedfrom 15to 33% in the various groups (data not shown ). There were no significant differences in the percentagesof infected organs with low ( 125 c.f.u.), intermediate (26-125 c.f.u.), or high ( > 125 c.f.u.) numbers of bacteria between the unvaccinated control group and any of the vaccine groups. Histopathology Lesionswere in accord with previous descriptions (Blasco, 1990) and were classified as severe (marked edema, interstitial fibrosis, epithelial hyperplasia, and widespread infiltration of mononuclear cells including plasma cells) or mild to moderate (vacuolation of basilar portions of the epithelium, and mild to moderate degreesof epithelial hyperplasia, mononuclear cell inliltration, and fibrosis). Severe lesions were restricted to some of the rams that were infected at the time of necropsy. The percentageof infected rams with severelesions in one or more organs was significantly greater (PcO.05) in the group vaccinated with Rev 1 than in those vaccinated with HS, 45/20 bacterin, REO bacterin, and HS-PCP (data not shown). Antibody responses Prior to treatment antibodies to HS antigens were undetectable in the majority of rams and low (titer= 100) in the remainder (data not shown). Vaccination with live Rev 1 or with two injections of the non-living products causeda rapid and sustaineddevelopment of antibodies for HS antigens (data not shown). The highest titers achieved after revaccination and the highest titers at Week 48, just prior to challengeinfection, were obtained in the groups that received the protective vaccines HS and 45/20 (Table 2). However, the lowest titers to HS were obtained in the Rev 1 vaccine group (Table 2 ). Moreover, in none of the groups was there a significant difference in antibody titers between protected and unprotected rams (Table 2). Challenge inoculation resulted in the formation of high antibody titers in all of the treatment groups, again with no significant differences in peak responsesbetween the infected and uninfected rams within a group (Table 2). In accord with prior results (Gamazo et al., 1989b), weak immunoblot re-
TABLE 1 ELISA titers against HS extract in rams that were culture positive or culture negative following challcngc inoculation” Vacci&
Rev I HS 45/20 bactcrin PC-OM REO bacterin Blebs HS-PCP None
Infection status<
+ + + + + + + + -
No. of rams
5 IO 6 9 7 7 9 6 II 4 II 4 I3 I I3 ?.
ELISA titers (mean k SD)d Highest titer after vaccination’
Titer just prior to challenge ( Week 48 )’
Highest titer after challenge”
5.sI!zo.45 j.SkO.32 l0.0I!I0.0 9.9LO.33 9.9+0.3s 10.0+0.0 7.8 + 0.44 7.7+ I.2 8.0+0.63 S.2&0.96 7.0+0.77 7.2 k 0.50 9.2+0.69 9.0 nd nd
5.4kO.55 5.4kO.53 7.st 1.4 8. I z!I0.93 9.1* I.1 9.7kO.49 6.SkO.67 6.SkO.75 7.1 I!I I .3 s.o+o.s1 5.4fO.52 6.0+0.0 7.2+ I.1 7.0 nd nd
s.s*o.s4 8.3+ 1.6 9.9rto.41 9.7kO.71 9.9+0.3s 10.0+0.0 9.2 k 0.44 9.3+o.s2 9.4 k 0.69 9.s*o.50 9.5fO.U 9.oI!Io.o 9.9+0.x 10.0 9.4kO.67 9.7kO.56
“Groups of I5 rams were vaccinated twice at an IS week interval, escrpt for one group that rcccivcd a single dose of Rev I. Animals received a challenge inoculation 3 I weeks after the second vaccination and were nccropsied I5 weeks later. “Rev I, B. r~lilorsis strain Rev I: HS. hot saline extract; 45/20 bacterin. from strain 45/20 of R. c10o~trt.s:PC-OM, outer membrane-peptidoglycan complcs: REO bacterin, from B. 0~;s strain REO 198: blcbs. from outer membranes; HS-PCP. HS estracted with ether petroleum-chloroform-phenol; none. unvaccinated rams. All subcellular vaccines were prepared from B. Otis strain REO 19s. ‘( + ). culture positive and ( - ), culture negative at the time of necropsy. “ELISA titers from I00 to 5 I 200 were converted into integers from one to ten. respectively. nd. not done. ‘In no paired comparison were thcrc significant differences in titers between culture positive and culture negative animals.
actions to Group 3 protein (graded 1.O or less) were present in some rams ( 17 of 32 ) prior to vaccination or infection (data not shown). Almost all of the animals demonstrated anti-OMP responsesafter vaccination, with the most intense responsesdirected against the group 3 protein (Fig. 1). Rev lvaccinated rams showed the weakest responsesto OMPs (Fig. 1). Control rams produced OMP antibody responsesfollowing challengeinoculation that were very similar to those observed in naturally infected rams (Riezu-Boj et al., 1990), with generally more intense responsesin animals that were more heavily infected (Fig. 1). In some comparisons between vaccinated rams that
266
J.M. BLASCO
567
567
567
3+
2+
+
C
123457
123457
123457
3+
REV1
ET AL.
123457
3+
HS
Fig. 1. Immunoblot analysis of the antibody responseto membrane proteins of B. ovis in sera of representative vaccinated and unvaccinated rams. C, control unvaccinated rams that presented high (3+ ) intermediate (2+ ) or low ( + ) degrees of infection as judged by bacteriological results at necropsy. Rev 1, B. tnelilensis Rev 1 vaccinated rams that became heavily infected (3 + ) or were uninfected; HS, HS vaccinated rams that became heavily infected (3 + ) or were uninfected. The immunoblots were performed with HS-PCP (20 pg per lane) using serum samples taken ( 1) 2 weeks after the first vaccination; (2) I2 weeks after the first vaccination; (3) 1 week after revaccination; (4) 30 weeks after revaccination ( 1 week prior to challenge inoculation); (5) 3 weeks after challenge inoculation; (6) 10 weeks after challenge inoculation; (7) 15 weeksafter challengeinoculation. Bars at left represent positions of standard molecular weight markers at 94, 67,43, 30,2 I, and 14.3 kDa.
became heavily infected or remained uninfected following challengeinoculation, there was an apparent direct association between the intensity of OMP antibody responsesjust prior to challenge and culture-negative status thereafter (Fig. 1,Lanes 4). However, statistical analysisof pooled data from seven culture negative rams and sevenrams with ‘severe’ ( + + + ) infections (one from each vaccine group) revealed no significant differences between infection status and the intensity of the OMP Group 3 immunoblot reactions at 1 week after revaccination, at 30 weeks after revaccination ( 1 week prior to challenge inoculation), or at 3 weeks after challenge inoculation. The same analyseslikewise demonstratedno significant differencesbetweenthesegroups in the magnitude of ELISA titers for HS or R-LPS antigens at any of the three time points (data not shown). DISCUSSION
Vaccination of rams with HS, a subcellular extract of B. ovis, conferred a level of protective immunity against virulent B. ovis which closely approximated that provided by the live vaccine Rev 1 (Table 1). Experiments currently being performed in the mouse model confirm this finding. Vaccination of mice with HS reduced mean numbers of B. ovis in the spleen at 2 weeks
VACCINES
AGAINST
6. OVIS IN RAMS
261
after challenge infection by over 1000 fold, and provided the same level of protection afforded by vaccination with live Rev 1 (M.P. Jimenez de Bagties, J.M. Blasco and A.J. Winter, unpublished data, 1992). HS therefore represents a potentially useful vaccine against B. ovis, and its use would offer a solution to the problems associatedwith live Rev 1. The bacteriological data confirm prior reports (Fensterbank et al., 1982; Blasco et al., 1987;Marin et al., 1990b) that an accurateevaluation of B. ovis infection requires a complete cultural examination of genital as well as extragenital organs. Prior evaluations of B. ovis vaccines that relied on cultures only of semenand/or testes (Van Heerden and Van Rensberg,1962;Claxton, 1968; Swift and Maki, 1968; Gradwell and Van Zyl, 1975; McGowan and Harrold, 1979; Garcia-Carrillo, 1981) were therefore inadequate. The low numbers of B. ovis in a majority of infected organs also point out the need to culture large samples of tissue homogenatesfor a reliable assessmentof infection status. In this connection, it is evident that vaccines that provided significant protection neverthelessfailed to modify the organ distribution of the infection or the levels of colonization of infected organsin those animals that remained culture positive. The results of this investigation are consistent with our original hypothesis that a subcellular vaccine rich in outer membrane components would be successful against B. ovis. However, not all the vaccines containing outer membrane protein antigens conferred protection. Whereas HS extracts, enriched in R-LPS and OMPs, were protective, the protein fraction of HS (HS-PCP) which contained less than 1% R-LPS was not (Table 1) . It thus appearsthat OMPs alone could not induce protective immunity against B. ovis and that, as in infections with smooth brucellae (Plommet, 1987; Winter et al., 1988, 1989), the LPS may be an important protective antigen. However, R-LPS seemedlikewise insufficient to induce protective immunity, as illustrated by the fact that the PG-OM, outer membrane blebs, and B. ovis bacterin did not provide protection (Table 1) . The subcellular fractions did not differ substantially in protein profile (Gamazo et al., 1989a), but did vary in the percentagecontents of LPS and protein, which gave ratios of 0.53 for HS, 0.41 for PG-OM, 0.39 for outer membrane blebs, and less than 0.01 for HS-PCP. Therefore, the different proportions of OMPs and R-LPS might explain the different levels of protection provided by thesevaccines. Experiments are in progressto assessthe respective contributions of R-LPS and OMPs to protective immunity against B. ovis. We found no statistical associations in this experiment between the levels of antibodies to the HS extract or its antigenic components (OMPs and RLPS) and protection from infection. These negative findings must be interpreted cautiously, and do not invalidate our hypothesis (Riezu-Boj et al., 1986) that antibodies directed against the surface exposedOMP and R-LPS antigens mediate protection against B. ovis. In the murine model of B. abortus
268
J.M. t3L.GCO
ET AL.
infection, for example, we have likewise been unable to demonstrate a simple statistical relationship between protection and serum levels of 0 polysaccharide-specific antibodies (Winter et al., 1988, 1989) even though we know from passivetransfer experiments that theseantibodies confer excellent protection (Montarez et al., 1986;Winter et al., 1989). An evaluation of cell mediated immunity induced by the vaccines was beyond the scope of this study. At the present time there is no information on the protective role of cell mediated immunity in ram epididymitis. Cell mediated immune responsesmay have been induced by vaccines incorporated with L-l 2 1 and MDP, as was demonstrated by Byars and Allison ( 1987). Even though antibody responsesinduced by Rev 1 were relatively weak (Table 2, Fig. 1), we hypothesize that this vaccine stimulated vigorous cell mediated immune responsesthat provided good protection. We are currently employing the mouse model system to examine the functions of humoral and cell mediated immune responsesin protection against B. mix ACKNOWLEDGMENTS
We are deeply indebted to Santiago Lazaro and Marisa Pardo for excellent technical assistance.This work was supported in part by the US-Spain Joint Committee (CCA-85 1O/O85), and CICYT and CONAI from Spain.
REFERENCES Afzal. M.. Tengerdy. R.P., Ellis, R.P., Kimberling, C.V. and Morris, C.J., 1984. Protection of rams against epididymitis by a B~trcellu oris vitamin E adjuvant vaccine. Vet. Immunol. Immunopathol.. 7: 293-304. Alton, G.G.. Jones, L.M.. Angus. R.D. and Verger. J.M.. 1988. Techniques for the brucellosis laboratory. Institute National de la Recherche Agronomiquc, Paris, I90 pp. Biberstcin, E.L., McGowan, B.. Robinson, E.A. and Harrold, D.R.. 1963. Epididymitis in rams. Studies on immunity. Cornell Vet., 52: 2 14-2 17. Blasco, J.M., 1990. Bruce/la oris. In: K. Nielsen and J.R. Duncan (Editors), Animal Bruccllosis. CRC, Boca Raton, FL, pp. 35 l-374. Blasco. J.M., Marin, C.M.. Barberan, M., Moriyon, I. and Diaz. R.. 1987. Immunization with Brmdlu /rre/i[crtsis Rev I against Brucella oris infection of rams. Vet. Microb.. 14: 38 l-392. Brown, G.M., Ranger, C.R. and Kelly, D.J.. I97 I. Selective medium for the isolation ofBrrd/a oris. Cornell Vet., 6 I: 265-280. Buddlc. M.B., 1962. Production of immunity in rams against Brudln uris infection. NZ. Vet. J.. IO: I l-15. Buddle, M.B., Calverley, F.K. and Boyes, B.W., 1963. Brlrcd/u oris vaccination of rams. NZ. Vet. J.. I I: 90-93. Burnette. W.N.. I98 I. ‘Western blotting’: electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radiolabeled protein A. Anal. Biochem.. I 12: 195-203.
VACCINES
AGAINST
B. OVIS IN RAMS
269
Byars, N.E. and Allison, A.C., 1987. Adjuvant formulation for use in vaccinesto elicit both cellmediated and humoral immunity. Vaccine, 5: 223-228. Claxton, P.D., 1968. Brucella ovis vaccination of rams. Comparison of two commercial vaccines and two methods of vaccination. Aust. J. Vet., 44: 48-54. Diaz, R. and Bosseray,N., 1973. Identification d’un composCantigenique specifique de la phase rougueuse (R) des Brucella. Ann. Rech. V&t., 4: 283-292. Diaz, R., Jones, L.M. and Wilson, J.B., 1967. Antigenic relationship of Brucellu ovis and Brucella melitensis. J. Bact., 93: 1263-1268. Erasmus, J.D. and Bergh, E.C., 1985. Ovine brucellosis: repeated vaccination with Rev 1 vaccine and the prevalence of the diseasein the Winburg district. J. S. Afr. Vet. Assoc., 56: 205209. Fensterbank, R., Pardon, P. and Marly, J., 1982. Efficacy of Brucella Melitensis Rev 1 vaccine against Brucellu ovis infection in rams. Ann. Rech. Vet., 13: 185-190. Gamazo, C. and Moriybn, I., 1987. Release of outer membrane fragments by exponentially growing Brucella melitensis. Infect. Immun., 55: 609-6 15. Gamazo, C., Winter, A.J., Moriybn, I., Riezu-Boj, J.I., Blasco,J.M. and Diaz, R., 1989a. Comparative analyses of proteins extracted by hot saline or released spontaneously into outer membrane blebs from field strains of Brucellu ovis and Brucella melitensis. Infect. Immun., 57: 1419-1426. Gamazo, C., Winter, A.J., Riezu, J.I., Blasco, J.M. and Jimtnez de BagtiCs,M.P., 1989b. Respuesta inmune humoral inducida por las preparaciones antigbnicas subcelularesde Brucella ovis en corderos. Vol. 2, Abs. 1 p. 89, Abstr. XII Congreso National de la Sociedad espaiiola de Microbiologia. Garcia-Carrillo, C., 1981. Protection of rams against Brucella ovis infection by Brucella melitensis Rev 1. Zblt. Vet. .Med. B, 28: 425-429. Gradwell, D.V. and van Zyl, F.E., 1975. Effectivity of Rev 1 vaccine in rams against Brucella ovis infection. J. S. Afr. Vet. Assoc.,46: 349-354. Haughey, K.G. and Hughes, K.L., 1969. The apparent failure of a vaccine to protect rams against naturally occurring Brucella ovis infection. Aust. Vet. J., 45: 1O-l 6. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227: 680-685. Lowry, O.H., Rosebrough, N., Farr, A.L. and Randall, R.J., 195 1. Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193: 265-275. Marin, C., Jimtnez de BagtiCs,M.P., Blasco,J.M., Gamazo, C., Moriybn, I. and Diaz, R., 1989. Comparison of three serological tests for Brucellu ovis infection of rams using different antigenic extracts. Vet. Rec., 125: 504-508. Marin, C.M., BarberBn, M., JimCnez de Bag%, M.P. and Blasco, J.M., 1990a. Comparison of subcutaneous and conjunctival routes of Rev I vaccination for the prophylaxis of Brucella ovis infection of rams. Res. Vet. Sci., 48: 209-2 15. Marin, C., JimCnez de Bag&, M.P., Barber&n, M. and Blasco, J.M., 1990b. Efficacy of long acting oxytetracycline combined or not with streptomycin for treating Brucellu ovis infection of rams. Am. J. Vet. Res., 50: 560-563. McGowan, B. and Harrold, D.R., 1979. Epididymitis in rams: studies on vaccine efficacy. Cornell Vet., 69: 73-76. Montaraz, J.A, Winter, A.J., Hunter, D.M., Sowa, B.A., Wu, A.M. and Adams, L.G., 1986. Protection against Brucelh abortus in mice with 0-polysaccharide-specific monoclonal antibodies. Infect. Immun., 5 1: 961-963. Moreno, E., Jones, L.M. and Berman, D.T., 1984. Immunochemical characterization of rough Brucella lipopolysaccharides. Infect. Immun., 43: 778-782. Osborn, M.J., 1963. Studies on the gram-negative bacterial cell wall. I. Evidence for the role of
270
J.M. BLASCO
ET AL.
2-keto-3 deoxyoctonate in the lipopolysaccharide of Salmonella typhimurium. Proc. Natl. Acad. Sci. USA, 50: 499-506. Plommet, M., 1987. Brucellosisand immunity: humoral and cellular components in mice. Ann. Inst. Pasteur Micro., 138: 105-l 10. Riezu-Boj, J.I., Moriyon, I., Blasco, J.M., Marin, C. and Diaz, R., 1986. Comparison of lipopolysaccharideand outer membrane protein-lipopolysaccharide extracts in an enzyme-linked immunosorbent assayfor the diagnosis of Brucella ovis infection. J. Clin. Micro., 23: 938942. Riezu-Boj, J.I., Moriyon, I., Blasco, J.M., Gamazo, C. and Diaz, R., 1990. Antibody response to Brucella ovis outer membrane proteins in ovine brucellosis. Infect. Immun., 58: 489-494. Ris, D.R., 1967. The persistenceof antibodies against Brucella ovis and Brucella abortus in rams following vaccination: a field study. NZ. Vet. J., 15: 94-100. Snedecor, G.W. and Cochran, W.G., 1989. Statistical methods. 8th edn. Iowa State University Press,Ames, IA, 503 pp. Swift, B.L. and Maki, L.R., 1968. Immunological studies on three ram epididymitis bacterins. Cornell Vet., 58: 659-665. Towbin, H., Staehelin, T. and Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets:procedure and some applications. Proc. Natl. Acad. Sci. USA, 16: 4350-4354. Van Heerden, K.M. and van Rensburg, S.W., 1962. The immunization of rams against ovine brucellosis. J.S. Afr. Vet. Assoc., 33: 143-148. Verstreate, D.R., Creasy, M.T., Caveney, N.T., Baldwin, C.L., Blab, M.W. and Winter, A.J., 1982. Outer membrane proteins of Brucella abortus: isolation and characterization. Infect. Immun., 35: 979-989. Warren, L., 1959. The thiobarbituric acid assayof sialic acids. J. Biol. Chem., 234: 1971-1975. West, D.M., Johnstone, A.C., Bruere, A.N. and Chapman, H.M., 1970. Epiphysitis in rams following vaccination against Brucella ovis infection. NZ. Vet. J., 26: 133-l 34. Winter, A.J., Rowe, G.E., Duncan, J.R., Eis, M.J., Widom, J., Ganem, B. and Morein, B., 1988. Effectivenessof natural and synthetic complexes of porin and 0 polysaccharide as vaccines against Brucella abortus in mice. Infect. Immun., 56: 2808-28 17. Winter, A.J., Duncan, J.R., Santisteban, C.G., Douglas, J.T. and Adams, L.G., 1989. Capacity of passivelyadministered antibody to prevent establishment of Brucella abortus infection in mice. Infect. Immun., 57: 3438-3444.