This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and

Transcription

1 This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier s archiving and manuscript policies are encouraged to visit:

2 Available online at Veterinary Microbiology 132 (2008) Protection levels in vaccinated heifers with experimental vaccines Brucella abortus M1-luc and INTA 2 M.A. Fiorentino a, *, E. Campos b, S. Cravero b, A. Arese b, F. Paolicchi a, C. Campero a, O. Rossetti b a EEA Balcarce Instituto Nacional de Tecnología Agropecuaria, 7620 Balcarce, Argentina b Inst. de Biotecnología, CICVyA, INTA-Castelar, Buenos Aires, Argentina Received 13 December 2007; received in revised form 8 April 2008; accepted 6 May 2008 Abstract Brucella abortus M1-luc is a mutant strain derived from S19 vaccine strain in which most of bp26 sequence has been replaced by the luciferase coding gene. Strain I2 is a double mutant derived from M1-luc in which most of omp19 has been deleted without introduction of any genetic markers. In BALB/c mice, M1-luc presented equivalent performance to S19 regarding persistence, splenomegaly and protection against challenge. Interestingly, I2 was more attenuated than S19, with no reduction of protection against challenge. In order to evaluate the potential for vaccine use of these strains in the natural host, four groups of 15 heifers, 6-month old, were either non-vaccinated or vaccinated with S19, M1-luc or I2. To at reached 17-month old, heifers were synchronized with two doses of PGF2a and received natural service during 60 days with two bulls. Pregnant heifers were challenged at approximately six gestation months with virulent B. abortus S2308. Blood samples post-challenge of heifers were collected for serologic test as well as specimens of aborted fetuses and premature calves for bacterial isolation and histopathological analyses. Protection levels against abortion were 78.6% for S19, 81.8% for M1-luc and 45.5% for I2, compared to the 25% that did not abort from the non-vaccinated group. These results indicate that in bovines BP26 had no influence in protective capacity of S19, correlating with the results obtained in mice. However, contrarily to what was previously observed in mice, lack of expression of Omp19 rendered in less protection capacity of S19 in the natural host. # 2008 Elsevier B.V. All rights reserved. Keywords: Brucella abortus; Vaccine; Bovine; M1-luc; INTA 2 1. Introduction * Corresponding author at: Lab. Bacteriología, Dpto. Producción Animal, EEA INTA Balcarce, RN226 Km 73.5 CC276, 7620 Balcarce, Buenos Aires, Argentina. Tel.: ; fax: addresses: mafiorentino@balcarce.inta.gov.ar, mafiorentino@yahoo.com.ar (M.A. Fiorentino). Pregnancy losses due to infectious agents produce a severe economic impact on the profitability of the herds worldwide. In developing countries bovine brucellosis causes severe economic losses and represents a public health problem (Godfroid et al., 2005; Pappas et al., 2006). Brucellosis of cattle is /$ see front matter # 2008 Elsevier B.V. All rights reserved. doi: /j.vetmic

3 M.A. Fiorentino et al. / Veterinary Microbiology 132 (2008) caused by Brucella abortus, a facultative intracellular bacterium, which causes abortion, birth of weak and sick calves and decreased fertility (Nielsen, 2002). Brucellosis is a zoonoses causing a chronic disease in man (Pappas et al., 2006). At present, the smooth attenuated strain B. abortus S19 vaccine is used in many parts of the world conferring about 70 80% of protection to vaccinated cattle (World Health Organization, 1997). The main drawbacks of this live vaccine include abortions in pregnant animals and inability to distinguish vaccinated from infected animals (Nielsen, 2002). Additionally this vaccine also has the disadvantage of being pathogenic for humans (Alton et al., 1988). Several approaches have been followed up to overcome the difficulty in diagnosis. One approach is the use of live attenuated rough Brucella strains, particularly B. abortus strain RB51. Vaccination with RB51 does not result in O antigen-specific antibodies, serological differentiation of infected from vaccinated animals is facilitated (Schurig et al., 1991). Another approach consists in the identification of diagnostic protein antigens and the deletion of its corresponding genes in live attenuated smooth Brucella vaccine strains that are currently employed. Following this approach, construction of mutant strains derived from S19, B. abortus M1-luc and INTA2 (I2) was previously reported (Campos et al., 2002). These strains were constructed with the objective of obtaining a candidate vaccine strain that might be compatible with a diagnostic test and have less residual virulence than vaccine strain S19. B. abortus M1-luc was obtained replacing most of bp26 gene with the luciferase (luc) coding gene. Strain I2 was obtained by deletion of most of omp19 coding sequence in strain M1-luc, without the introduction of any other genetic marker. BP26 is a 26 kda periplasmic protein that is strongly recognized by sera from infected cattle, goats, sheep and humans (Cloeckaert et al., 1996, 2001; Linder et al., 1996; Rossetti et al., 1996; Arese et al., 1999). Lack of expression of BP26 in S19 resulted in a strain with the same residual virulence and protective capability as S19 when assayed in BALB/c mice (Boschiroli et al., 1997), therefore rendering a potential vaccine strain that would allow differentiation of vaccinated from infected animals using BP26 as diagnostic antigen. OMP19 is an outer membrane lipoprotein involved in virulence of B. abortus. Lack of expression of OMP19 in S19, resulted in a mutant strain with similar protection capacity against challenge in BALB/c mice. M1-luc and I2 strains showed to be smooth and stable both in vitro and in BALB/c mice (Campos et al., 2002). The objective of the present study was to investigate the potential value of two experimental vaccines, named M1-luc and I2, in vaccinated and challenged heifers. 2. Materials and methods 2.1. Animals and experimental design Sixty Aberdeen Angus and Hereford crossbreed heifers with 6-month old were obtained from brucellosis-free herd of the Instituto Nacional de Tecnología Agropecuaria (INTA), Buenos Aires Province, Argentina. Prior to experiment started, blood samples were taken from all animal and tested negative for evidence of exposure to infectious diseases of reproduction. All heifers were randomly allocated into four experimental groups: animals from Group S19 (n = 15) were immunized with a commercial vaccine B. abortus S19 with colonyforming units (CFU); Group M1-luc (n = 15) received the B. abortus M1-luc vaccine with CFU and Group I2 (n = 15) were vaccinated with B. abortus I2 with CFU. All animals were vaccinated with 2 ml dose applied by subcutaneous route (SC) in the neck area. Control group (n = 15) was used as nonvaccinated control and received sterile saline solution, 2 ml SC in the neck area. Experimental groups were kept together until challenge under extensive management in natural pasture with adequate forage quality and availability in order to maintain a body condition scoring of 6 (scales 1 9) during all this study. Clinical and genital tract examination, body condition and pelvic area measurements were performed to the heifers before breeding time (Mortimer et al., 1997) Strains vaccines Batches of a commercial S19 vaccine were used (Lab. San Jorge Bagó, Argentina). Strain S19

4 304 M.A. Fiorentino et al. / Veterinary Microbiology 132 (2008) lyophilized vaccine was used according to procedures recommended by Servicio Nacional de Sanidad Animal (SENASA). Prior and just after vaccination, each batch was evaluated for purity and number CFU by plate count. The Brucella strains M1-luc and I2 were grown at 37 and 30 8C, respectively (Campos et al., 2002), for 48 h on tryptose agar (Difco, Detroit, MI, USA), scraped and re-suspended in sterile saline solution (0.85% NaCl) and diluted to the appropriate concentration. Doses of bacteria were calculated to be within limits of current S19 vaccine guidelines and previous experiments in cattle (Cheville et al., 1992, 1993). Studies in bovines had been approved and monitored by the Comisión Nacional Asesora de Biotecnología Agropecuaria of Argentina (CONA- BIA) and SENASA (Res /00). All animal usage was according to protocols from the Animal Ethics Committee at INTA-Balcarce Breeding protocol All heifers were allocated in the same pasture and breed at 17 months of age. Two weeks before the service time, all females were synchronized hormonally with two doses of a synthetic PGF2a analogue (Ciclase 1, Syntex, Buenos Aires, Argentina) given 11 days apart. After 48 h of the last PGF2a injection, heifers received natural service during 60 days with two Aberdeen Angus 2 years old bull. Pregnancy was confirmed by transrectal palpation and ultrasound. Pregnancy status was also monitored approximately at 5-gestation months previous conjunctival challenge Experimental challenge At the moment of experimental challenge, all the pregnant heifers were transferred to BL3 facilities until the end of the experiment. The animals were challenge with the virulent B. abortus strain 2308 obtained from SENASA. After growth in tryptose agar (Difco, Detroit, MI, USA) for 48 h in 5% CO 2, a saline suspension of strain 2308 was prepared and adjusted to a concentration of CFU/mL, which was confirmed by plate counts. Animals were exposed to the challenge strain between 5 and 6 months of pregnancy by conjunctiva instillation of 100 ml (50 ml/eye) of the strain 2308 suspension, resulting in CFU/animal. The isolation of the challenge strain from different cows specimens and/or the serological titles persistent were used as criterion of brucellosis infection Blood sampling Blood samples were collected by venipuncture in 10 ml tubes at 30 days before vaccination, the day of vaccination (day 0), 7 days post-vaccination (PV), every 2 weeks up to 3 months PVand monthly until the end of the experiment. In addition, calves serum was taken prior to the colostrums ingestion, and the serum from their dams was also taken for serological tests Serologic tests Blood samples were centrifuged and serum separated in aliquots (2 ml), and stored at 20 8C until further use. Serum was assayed for detection of anti-b. abortus antibodies by buffered plate antigen (BPA) test, the standard tube agglutination (SAT) test, 2-mercaptoethanol (2-Me) test, complement fixation (CTF) test and anti-bp26 ELISA. BPA was previously described by Angus and Barton (1984). Results were expressed as positive or negative. The SAT and 2-ME tests were performed in parallel as described by Alton et al. (1988). The interpretation of the results was made following the procedures recommended by SENASA. Sera with titers 1:200 to SAT and/or 1:50 to 2-Me were considered positive. The CTF was performed following Hill s method of 50% hemolysis (Alton et al., 1988) using the second International Standard anti-b. abortus Serum (ISAbS) as reference. Sera with 30 International Units were considered positive. The presence of IgG antibodies against BP26 was determined by indirect ELISA as previously described (Arese et al., 1999). Briefly, recombinant BP26 was diluted to a 1 mg/ml concentration in carbonate buffer 0.06 M (ph 9.6) and used to coat the wells of polystyrene plates (200 ml/well; Polysorp, Nunc- Immuno). After overnight incubation at room temperature, the plates were washed four times in washing buffer (PBS at ph 7.2, 0.05% Tween 20) and blocked with 5% low fat milk in washing buffer. After 1 h at 37 8C, the blocking solution was discarded, and the

5 M.A. Fiorentino et al. / Veterinary Microbiology 132 (2008) plate was washed five times with washing buffer. Afterwards, diluted bovine serum samples (1:200) were added to the wells (200 ml/well). The plates were incubated 1 h at 37 8C, washed six times in washing buffer and goat anti-igg bovine horseradish peroxidase conjugate (ACCURATE) was added at a 1:2000 dilution. After 1 h of incubation at room temperature, the plates were washed thoroughly, and 200 ml of substrate solution (ABTS 0.04 M in citrate buffer 0.05 M ph 5 and 1 mm H 2 O V) was added to each well. After 10 min of incubation at room temperature, the A 450 was recorded with a microplate reader (Titirtek, Multiskan, ICN) Reproductive losses Abortion is defined who fetal loss occurred between 42 and 260 days of gestation and from day 260 until term, premature deliveries (Hubbert, 1972). However, in the present study all birth of a nonviable calf was defined as abortion. Fetal age was determined using crown-rump length (Hubbert, 1972; Campero et al., 2003) and breeding date. The period of neonatal mortality was defined as calf loss occurring between 1 and 7 days of life (Hubbert, 1972). Aborted fetuses and premature weak calves were collected within 12 h after abortion or death. Specimens of lung, spleen and abomasal content were collected from the aborted fetuses and premature calves for bacterial isolation. The viability of the calves since parturition until weaning was recorded. Specimens of milk, cervico-vaginal mucus (CVM) and placenta were collected from aborted cows or cows with normal calves to microbiological studies. Protection was measured as number of animals that did not abort in relation to the total of pregnant animals per group. For the purpose of analyses in the present study, premature weak calves and neonatal losses were considered as abortions Microbiologic studies Samples arrived at laboratory within 8 h after collection. Samples were transported on ice. All the tissues were cultured on Columbia Blood Agar (CBA) (Oxoid Ltd., Wad Road, Basingstoke, UK) with 5% sterile defrinated bovine blood and Skirrow (SK) agar with antibiotics (Terzolo et al., 1991). CBA and SK plates were incubated at 37 and 30 8C under 10% CO 2 and microaerobic atmospheres and examined for grow daily for 7 days. All isolates were identified by routine procedures (Alton et al., 1988). Additionally, specimens of fetuses and cows were investigated for viral and protozoal causes of reproductive losses as described (Campero et al., 2003) Histopathology and immunohistochemical analyses Multiple tissues from the aborted fetuses including brain, lung, heart, liver, kidney, adrenal, spleen, thymus, lymph node, abomasum, intestine, skeletal muscle and placenta (when available) were sampled and fixed in neutral-buffered 10% formalin, dehydrated and embedded in paraffin. For histopathology examination tissues were sectioned at 4 5 mm thick and stained with haematoxylin and eosin (H&E). Histopathologic changes were identified and lesions were recorded and classified (Campero et al., 2003). Formalin-fixed paraffin sections of lung and placenta were also stained with an avidin biotin peroxidasa complex immunoenzyme technique, as described (Hsu et al., 1981). The primary polyclonal antibody (B. abortus antiserum, Difco, Detroit, MI, USA) was used at 1:200 with 45 min incubation at 37 8C. Formalin-fixed sections of bovine fetal lung culture positive for B. abortus were used as positive tissue control and to titer the primary antibody. Negative controls consisted of normal rabbit serum applied to the same positive control slides. Additional negative controls consisted of omitting the secondary antibody in the IHC procedure on positive control slides Statistical analysis The relative risk (RR) was used as measure of association between exposure (non-vaccinated group) and the cumulative incidence of abortion and cow infection in the vaccinated groups. Confidence intervals of this ratio were calculated using the logarithmic approximation (Poester et al., 2006). Vaccine efficacy was estimated in the form of an attributable fraction in the exposed group [(RR 1)/ RR], were the non-vaccinated group is the exposed group or risk factor positive. It can be interpreted as

6 306 M.A. Fiorentino et al. / Veterinary Microbiology 132 (2008) the proportion of cases that would have occurred in the vaccinated groups should the vaccine have not been used (Poester et al., 2006). 3. Results 3.1. Post-vaccination All heifers in this study presented free serologic evidence of infection with Brucella, as assayed by BPA, before vaccination. The control group heifers remained negative during the whole experience until challenge. All heifers vaccinated with S19, M1-luc or I2 developed antibodies against Brucella after 1 week postvaccination evidenced by SAT and 2-ME tests. Those heifers that received vaccines developed peak antibody titer in 2 weeks, as determined by serological standard test. A similar antibody response was seen in animals from the three groups until 8 weeks PV. After that, negative animals outnumbered the positive ones (Figs. 1 and 2). Also, no antibodies against luciferase could be detected by western blot with pools of sera from the different groups at either 1, 2 or 6 months after vaccination (data not shown), indicating that Fig. 2. Serologic response in bovines vaccinated with S19, INTA 2 or M1-luc strains and non-vaccinated controls. IgG antibody response determined by 2-mercaptoethanol (2-ME) test. Results are expressed as the mean of 15 individual heifers. luciferase does not interfere with serologic reaction of the host. Heifers vaccinated with S19, M1-luc or I2 strains, did not produce antibodies to BP26 when serum samples were examined by an ELISA using specific recombinant antigen BP26. The ELISA OD readings of serum samples taken from these vaccinated heifers were negative (0.115) and were not different from OD readings from non-vaccinated control heifers serum samples (0.106). Two positive control sera obtained from naturally infected heifers gave ELISA OD readings of and 1.700, respectively. During the breeding time, all heifers maintained an adequate good body condition (score 6). After 2 months of natural service, 48/60 heifers became pregnant. Therefore, the experimental groups numbers were S19 n = 14, M1-luc n = 11, I2 n = 11 and control group n = 12 heifers Post-challenge Fig. 1. Serologic response in bovines vaccinated with S19, INTA 2 or M1-luc strains and non-vaccinated controls. Serum antibody response against Brucella as determined by serum agglutination test (SAT). Results are expressed as the mean of 15 individual heifers. At day of challenge 3/14 heifers of the S19 group and 3/11 of the M1-luc group were detected with low residual post-vaccination titles by SAT, but negative to 2-ME and FC. On the other hand, heifers of I2 and non-vaccinated control groups animals were serologically negative to all the tests at the day of challenge.

7 M.A. Fiorentino et al. / Veterinary Microbiology 132 (2008) The appearance of antibodies anti-bp26 presented great individuals variations. There was no correlation between the infected animal and the titer against BP26 (data not shown). The cumulative incidence of cows infections in the vaccinated group was 50% (7/14) for S19, 45% (5/11) for M1-luc and 100% (11/11) for I2 whereas in the control group the accumulative incidence of infections was 92% (11/12). The mean between challenge and abort or parturition (in days) was 98.4, 100.5, 76.5 and 74.3 from M1-luc, S19, I2 and control, respectively. Comparison of mean days from challenge to abort or parturition did reveal statistical differences between M1-luc and S19 vaccinated group vs. non-vaccinated control group ( p < ). However, the comparison between group vaccinated by I2 or non-vaccinated control group did not reveal any statistical difference. Protection rate against reproductive losses was 78.6, 81.8, and 45% for S19, M1-luc and I2 vaccines, respectively. In control group, there were 25% of liveborn calves (Table 1). From 2/3 of heifers in the control group that gave birth to suitable calves, it was isolated B. abortus from cervico-vaginal mucus and calostrum. One animal from this group presented temporary serological titles, B. abortus 2308 was not isolated from its fluids and gave birth to healthy calf. From 2/3 and 1/6 of aborted heifers of S19 and I2 group, respectively, B. abortus was isolated from cervico-vaginal o colostrum but B. abortus 2308 was not isolated from its fetuses (Table 1). All the isolations from aborted fetuses and cows, B. abortus 2308 was isolated but none of the vaccine strains were. Bacteria isolated were by biochemical tests (catalase, oxidase, citrate, nitrate reduction, urease), acriflavin agglutination and PCR with primers for bp26 and omp19 and western blot against BP26 and OMP19 (data not shown). A multifocal distribution of interstitial pneumonia and severe bronchopneumonia were the changes more often found in the lung of the examined fetuses. Specific Brucella immuno-staining was seen in sections of the lungs and placental tissues (Table 1). Specific Brucella immuno-staining was seen in the lungs of a fetus with negative isolation, which was the calf of a heifer from group S19 that eliminated B. abortus in milk. The results of the placenta immunohistochemical (IHQ) for the I2 and control groups coincide with the numbers of isolations. Nevertheless, in the groups S19 and M1-luc there were found discrepancies between the positive isolations and the results of the IHQ. Presenting positive immuno-staining, there were two placentas from group S19 heifers and three placentas from group M1-luc, which resulted negative to B. abortus isolation and had not serological response to brucellosis along the study. The placentas were collected next to expulsed, for that reason the elevate level of placenta s contamination was hard the B. abortus isolation from this tissues. The attributable fraction in the exposed and the relative risk of controls vs. vaccinated animals is shown in Table 2. The relative risk reveled that nonvaccinated heifers had and times higher risk of aborting than M1-luc and S19 vaccinatedanimals, respectively. If the attributable fractions are interpreted as preventable fractions, the results indicate that the vaccination with M1-luc prevented Table 1 Frequency of B. abortus isolation and positive immunohistochemistry, abortion and weaning rate of vaccinated and non-vaccinated heifers after challenge with B. abortus 2308 Group Isolation IHQ Abortion Calves weaned (%) *** Cows Fetus * Placenta ** Fetal lung * Total Colostrum * CVM * Placenta ** M1-luc 5/11 2/11 3/11 1/6 2/2 4/6 2/2 2 9 (81.1) a S19 7/14 3/14 6/14 1/5 1/3 3/4 2/ (78.6) a INTA 2 11/11 9/11 11/11 7/7 5/6 4/4 5/6 6 5 (15.5) ab Non-vaccinated 11/12 9/12 11/12 9/10 9/9 6/7 8/9 9 3 (25) b * Number of B. abortus isolation/total samples. ** Number of positive/total recovered. *** Different letters between row = significant difference ( p < 0.05).

8 308 M.A. Fiorentino et al. / Veterinary Microbiology 132 (2008) Table 2 Attributable fraction in the exposed (non-vaccinated) group and relative risk of abortion and infections in heifers in non-vaccinated vs. M1-luc and vs. S19 vaccinated groups 75.8% of abortions and 50.4% of cow infection. Whereas, S19 vaccination prevented 71.4% of abortions and 45.5% of cow infection. I2 vaccination showed no effect on relative risk of disease. 4. Discussion Attributable fraction in the nonvaccinated group Relative risk Confidence interval 95% M1-luc Abortion Cow infection S19 Abortion Cow infection INTA2 Abortion Cow infection In our study, we evaluated the levels of protection of two new vaccines (M1-luc and I2) in bovine brucellosis. One of the main aims of animal brucellosis research is the development of new vaccines which, because they are effective and keep an adequate persistence in the host to induce solid and durable immunity, may easily enable the differentiation between infected and vaccinated animals. In Argentina, one of the pillars of the National Plan for the Control and Eradication of the Bovine Brucellosis carried out by SENASA, is the vaccination of all the heifers between 3 and 8 months old with the strain S19. This is a smooth strain, and thus induces antibodies interfering in the conventional serological test. To overcome this problem, the live attenuated B. abortus RB51 vaccine, a rough strain, was developed and is used in cattle in some countries in replacement of B. abortus S19 (Schurig et al., 1991). However, the use of B. abortus RB51 is not permitted in Argentina (SENASA, resolution 1048/2002). As a consequence, we initiated studies with the mutants M1-luc and I2 in search of an alternative to the problem described above. The protection in brucellosis is measured by a decrease either in abortions or birth of weak calves, and a significant decrease in colonization of vaccinated cattle when compared with non-vaccinated controls after challenge (Elzer et al., 1998). In this study, the protection rate against reproductive losses conferred by M1-luc (81.8%) was similar that conferred by S19 (78.6%), whereas in I2 group it was lower (45%). In the non-vaccinated control group only 25% of the heifers had a normal parturition and viable calves. In the same way, protection against infection in heifers was higher in the groups M1-luc and S19. However, the I2 vaccination did not prevent the infection in cows. In brucellosis the protection against abortion is translated into lower direct losses, while protection against infection indicates less risk of dissemination of B. abortus to other animals and men. The number of abortions observed in the nonvaccinated control group coincides with the reports made by other authors, who applied B. abortus 2308 as the strain of the challenge (Schurig et al., 1991; Cheville et al., 1993). B. abortus infected fetuses and premature calves developed histological changes similar to those reported in experimental and natural infections in cattle (López et al., 1984; Sözmen et al., 2004). The discrepant results between IHQ and the isolation in the placenta samples could be a consequence of a low efficacy of the latter, due to high levels of contamination frequently present in these tissues. Nevertheless, the coincidence between the isolation and the presence of serological titres suggests that we could be dealing with healthy animals and that the results of the IHQ could be false positives. In agreement with previous results in the mouse model and with other Brucella species in mouse and ruminant models, the lack of expression of BP26 in S19 resulted in a strain with the same residual virulence and protective capability as the original strain (Cloeckaert et al., 2004; Boschiroli et al., 1997; Guilloteau et al., 2006; Jacques et al., 2007). In our study a similar protection capacity was presented by M1-luc and S19, indicating that BP26 is not necessary for the development of a protective immunity. In a previous study, the lack of expression of OMP19 in S19, resulted in a mutant strain with significantly less capacity to colonize and replicate in the spleen of BALB/c mice, although maintaining a

9 M.A. Fiorentino et al. / Veterinary Microbiology 132 (2008) similar protection capacity against challenge (Campos et al., 2002). However, in our study, the lack of expression of OMP19 rendered in a mutant strain with less protective capacity, did not correlate with previous results obtained in BALB/c mice (Campos et al., 2002). Some authors have reported the importance of this protein as a stimulant antigen of the immune cellular response (Vemulapalli et al., 2000). Since this type of response is significant in the protection against pathogenic brucellae, strains of B. abortus that do not express OMP19 could have a less protective capacity. The discrepancy between the levels of protection observed in mice (Campos et al., 2002) and in our work when we used the I2 strain in cattle could be the result of a large contribution of the OMP19 in the generation of a cellular immune response in one species as compared to the other one. Because of the high costs and long time span of the experiments in natural hosts, the mouse model has been used as a preliminary step in the analyses of vaccines against brucellosis, thus allowing a prediction of safety and immunogenicity. In our experience, the mouse model was not predictive of the results using I2 in cattle. Ideally, Brucella mutants should not induce antibody responses against a target protein, which may be used in new diagnostic tests. This protein should be immunodominant during infection, allowing the detection of infected animals. Some Brucella deletion mutants, devoid of proteins with potential use in diagnosis of brucellosis, have been studied. Deletion of the P39 coding gene from B. abortus S19 or B. melitensis Rev.1 does not have an effect on residual virulence or protective capability of both vaccine strains, as evaluated in a mouse model (Tibor et al., 1996) although no experiments in natural hosts have been carried out yet. B. abortus S19 mutants deleted for Cu Zn superoxide dismutase or BSCP31 a 31 kda periplasmic protein have been constructed and have provided a protective immunity against B. abortus in cattle, which is similar to that induced by the parental strain (Cheville et al., 1992, 1993). However, the proteins chosen in this work were not immunodominant in infected cattle and therefore were unsuitable to be used as diagnostic reagents. BP26 has been shown to be strongly recognized by sera from Brucella-infected cattle, but is non or poorly recognized by sera from S19 vaccinated animals (Rossetti et al., 1996; Arese et al., 1999; Cloeckaert et al., 2001). As expected, none of the heifers vaccinated with S19, M1-luc or I2 strains, produced antibodies against BP26 when serum samples were examined by an ielisa using specific recombinant antigen BP26. However, in this experience all the groups (vaccinated and control) failed to produce antibodies against BP26 or were poorly produced when they were challenge-exposed with B. abortus Previous works have demonstrated that a combination of immunodominant Brucella proteins is able to detect most of the infected animals (Cloeckaert et al., 1992; Hemmen et al., 1995; Limet et al., 1993; Tabatabai and Hennager, 1994). The results of the conventional serological tests applied in our work indicate that both mutants M1-luc and I2 strains are capable of inducing anti-lps antibodies in vaccinated heifers in the same way as S19 vaccine, correlating with the fact that both are smooth strains. As evaluated in the present study, strain S19 single deletion for the bp26 gene confer protective immunity against infection and abortion by B. abortus in cattle. Although protein BP26 is a protein recognized by the immune system of naturally infected bovines, under the conditions of the present study, our data suggest that the BP26 antigen will not allow serologic separation of vaccinated and infected cattle. In the present work the results obtained with the I2 vaccine in the murine model cannot be extrapolated to natural host. The protection afforded by the mutant strain I2 was smaller than strain S19 or M1-luc, stressing the necessity to evaluate vaccines in the natural host. Acknowledgments We thank to A. Mendez for media preparation, R.C. Malena, M.A. Poso, D. Cano and M.R. Leunda for your efficient assistance and J.L. Pereyra, J. Llamas and S. Danino for animal care (INTA Balcarce). M. A. F. is Doctoral fellow of CONICET. This work is part of the Doctoral Thesis of the first author. Financial support was provided by project BID 802 OC/AR-PID 616 from Agencia Nacional de Promoción Científica y Técnica.

10 310 M.A. Fiorentino et al. / Veterinary Microbiology 132 (2008) References Alton, G.G., Jones, L.M., Angus, R.D., Verger, J.M., Techniques for The Brucellosis Laboratory. INRA, Paris, France. Angus, R., Barton, C., The production and evaluation of a buffered plate antigen for use in the presumptive test for brucellosis. Dev. Biol. Stand. 56, Arese, A.I., Cravero, S.L., Boschiroli, M.L., Campos, E., Samartino, L., Rossetti, O.L., Use of a recombinant protein from Brucella abortus for the diagnosis of brucellosis in different animal species. Rev. Argent. Microbiol. 31, Boschiroli, M.L., Cravero, S.L., Arese, A.I., Campos, E., Rossetti, O.L., Protection against infection in mice vaccinated with a Brucella abortus mutant. Infect. Immun. 65, Campero, C.M., Moore, D.P., Odeon, A.C., Cipolla, A.L., Odriozola, E., Aetiology of bovine abortion in Argentina. Vet. Res. Commun. 27, Campos, E., Cravero, S.L., Delgui, L., Mora, I., Kahn, N., Arese, A.I., Rossetti, O.L., Brucella abortus INTA2, a novel strain 19 Dbp26::luc Dbmp18 double mutant lacking drug resistance markers. Vet. Microbiol. 87, Cloeckaert, A., Kerkhofs, P., Limet, J.N., Antibody response to Brucella outer membrane proteins in bovine brucellosis: immunoblot analysis and competitive enzyme-linked immunosorbent assay using monoclonal antibodies. J. Clin. Microbiol. 30, Cloeckaert, A., Debbarh, H.S., Vizcaino, N., Saman, E., Dubray, G., Zygmunt, M.S., Cloning, nucleotide sequence, and expression of the Brucella melitensis bp26 gene coding for a protein immunogenic in infected sheep. FEMS Microbiol. Lett. 140, Cloeckaert, A., Baucheron, S., Vizcaino, N., Zygmunt, M.S., Use of recombinant BP26 in serological diagnosis of Brucella melitensis infection in sheep. Clin. Diagn. Lab. Immunol. 8, Cloeckaert, A., Jacques, I., Grilló, M.J., Marín, C.M., Grayon, M., Blasco, J.M., Verger, J.M., Development and evaluation as vaccines in mice of Brucella melitensis Rev. 1 single and double deletion mutants of the bp26 and omp31 genes coding for antigens of diagnostic significance in ovine brucellosis. Vaccine 22, Cheville, N.F., Jensen, A.E., Halling, S.M., Tatum, F.M., Mortiff, D.C., Hennager, S.G., Frerichs, W.M., Schurig, G.G., Bacterial survival, lymph node changes, and immunologic responses of cattle vaccinated with standard and mutant strains of Brucella abortus. Am. J. Vet. Res. 53, Cheville, N.F., Stevens, M.G., Jensen, A.E., Tatum, F.M., Halling, S.M., Immune responses and protection against infection and abortion in cattle experimentally vaccinated with mutant strains of Brucella abortus. Am. J. Vet. Res. 54, Elzer, P.H., Enright, F.M., Colby, L., Hagius, S.D., Walker, J.V., Fatemi, M.B., et al., Protection against infection and abortion induced by virulent challenge exposure after oral vaccination of cattle with Brucella abortus strain RB51. Am. J. Vet. Res. 59, Godfroid, J., Cloeckaert, A., Liautard, J.P., Kohler, S., Fretin, D., Walravens, K., Garin-Bastuji, B., Letenson, J.J., From the discovery of the Malta fever s agent to the discovery of a marine mammal reservoir, brucellosis has continuously been a re-emerging zoonosis. Vet. Res. 36, Guilloteau, L.A., Laroucau, K., Olivier, M., Grillo, M.J., Marin, C.M., Verger, J.M., Blasco, J.M., Residual virulence and immunogenicity of CGV26 and CGV2631 Brucella melitensis Rev.1 deletion mutant strains in sheep after subcutaneous or conjunctival vaccination. Vaccine 24, Hemmen, F., Weynants, V., Scarcez, T., Letesson, J., Saman, E., Cloning and sequence analysis of a newly identified Brucella abortus gene and serological evaluation of the 17- kilodalton antigen that it encodes. Clin. Diagn. Lab. Immunol. 2, Hsu, S.M., Raine, L., Fanger, H., The use of avidin biotin peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J. Histochem. Cytochem. 29, Hubbert, W.T., Chairman. Committee on bovine reproductive nomenclature. Recommendations for standardizing bovine reproductive terms. Cornell Vet. 62, Jacques, I., Verger, J.-M., Laroucau, K., Grayon, M., Vizcaino, N., Peix, A., Cortade, F., Carreras, F., Guilloteau, L.A., Immunological responses and protective efficacy against Brucella melitensis induced by bp26 and omp31 B. melitensis Rev. 1 deletion mutants in sheep. Vaccine 25, Limet, J.N., Cloeckaert, A., Dezard, G., Van Broeck, J., Dubray, G., Antibody response to the 89 kda outer membrane protein of Brucella in bovine brucellosis. J. Med. Microbiol. 39, Linder, L.E., Hadfield, T.L., Tall, B.D., Snellings, N.J., Rubin, F.A., Van De Verg, L.L., Hoover, D., Warren, R.L., Cloning of a Brucella melitensis group 3 antigen gene encoding Omp28, a protein recognized by the humoral immune response during human brucellosis. Infect. Immun. 64, López, A., Hitos, F., Perez, A., Navarro-Fierro, R.R., Lung lesions in bovine fetuses aborted by Brucella abortus. Can. J. Comp. Med. 48, Mortimer, R.G., Farin, P.W., Stevens, R.D., Bovine Theriogenology Chapter 33. Reproductive examination of the nonpregnant cow. In: Youngquist, R.S. (Ed.), Current Therapy in Large Animal Theriogenology. WB Saunders Company. Press, Philadelphia, PA, USA, pp Nielsen, K., Diagnosis of brucellosis by serology. Vet. Microbiol. 90, Pappas, G., Papadimitriou, P., Akritidis, N., Christou, L., Tsianos, V.E., The new global map of human brucellosis. Lancet Infect. Dis. 6, Poester, F.P., Gonçalves, V.S.P., Paixão, T.A., Santos, S.C., Olsen, R.L., Schurig, G.G., Lage, A.P., Efficacy of strain RB51 vaccine in heifers against experimental brucellosis. Vaccine 24, Rossetti, O.L., Boschiroli, M.L., Arese, A.I., Cravero, S.L., Cloning of Brucella abortus gene and characterization of expressed 26-kilodalton periplasmic protein: potential use for diagnosis. J. Clin. Microbiol. 34, Schurig, G.G., Roop, R.M., Bagchi, T., Boyle, S., Sriranganathan, N., Biological properties of RB51: a stable rough strain of Brucella abortus. Vet. Microbiol. 28,

11 M.A. Fiorentino et al. / Veterinary Microbiology 132 (2008) Sözmen, M., Erginsoy, S.D., Genç, O., Beytut, E., Özcan, K., Immunohistochemical and microbiological detection of Brucella abortus in aborted bovine fetuses. Acta Vet. Brno 73, Tabatabai, L.B., Hennager, S.G., Cattle serologically positive for Brucella abortus have antibodies to B. abortus Cu Zn superoxide dismutase. Clin. Diagn. Lab. Immunol. 1, Terzolo, H.R., Paolicchi, F.A., Moreira, A.R., Homse, A., Skirrow agar for simultaneous isolation of Brucella and Campylobacter species. Vet. Rec. 129, Tibor, A., Saman, E., de Wergifosse, P., Cloeckert, A., Limet, J.N., Letteson, J.J., Molecular characterization, occurrence, and immunogenicity in infected sheep and cattle of two minor outer membrane proteins of Brucella abortus. Infect. Immun. 64, Vemulapalli, R., Cravero, S., Calvert, C.L., Toth, T.E., Sriranganathan, N., Boyle, S.M., Rossetti, O.L., Schurig, G.G., Characterization of specific immune responses of mice inoculated with recombinant vaccinia virus expressing an 18-kilodalton outer membrane protein of Brucella abortus. Clin. Diagn. Lab. Immunol. 7, World Health Organization (WHO), Report of the WHO Working Group meeting on Brucellosis control and research: The development of new/improved Brucellosis vaccines.