Abstract: Efficacy of vaccination with Brucella abortus S19 vaccine as control measure against bovine brucellosis has been controversial; therefore, it is necessary to know the efficacy of this vaccine under different field conditions. In this study, a clinical assay was performed to establish the efficacy of this vaccine on double purpose cattle. Two groups of one hundred animals each were formed. Infected cattle were not eliminated or segregated. One herd was identified as infected, with four animal reactors to Card Test (CT) and confirmed by Rivanol Test (RT) with a serum reaction rate of 1.2%. Confirmation of infected herd was carried out by isolation and identification of Brucella abortus colonies and PCR on milk samples from RT reactor animals. In 18 months, the number of infected animals increased to eight females, seven within the non-vaccinated group and one in the vaccinated group for a serum reaction rate in the non-vaccinated group of 5.8% and the vaccinated one of 0.8%. Thus, in this period the accumulated serum reaction rate for both groups was increased from 1.2 to 3%. Vaccination efficacy of strain S19 was 86% and the risk of getting the disease in these animals was very low (RR = 0.112; I.C.95% 0.014-0.887). It is concluded that strain S19 vaccine is efficacious in the control of brucellosis in herds with a 3% prevalence of the disease; yet, before its use, interference of diagnosis problems that are produced should be assessed to properly evaluate economics and vaccination efficiency.
INTRODUCTION
In Mexico, bovine brucellosis control has been carried out since 1951 using the S19 strain of Brucella abortus, nevertheless, when in 1997 the RB51 strain appeared, its use tapered down significantly even though today there is still a large amount of cows that have been vaccinated with RB51 vaccine (Bustamante et al., 2000).
A disadvantage of the S19 strain is that it induces permanence of agglutinins in serum and milk, propitiating interference with diagnosis since there is no discrimination of cattle infected with field strains and that vaccinated with S19 vaccine (Saldarriaga and Rugeles, 2002; Castro et al., 2005). To solve this problem, diagnostic tests should be used that are more specific, such as Radial Immunodiffusion (RID) that has proven in strain S19 vaccinated adult cattle that it is more sensitive and specific than the complement fixation test, since it has the capacity to identify individuals that do not have the disease by the determination of the Native Hapten (NH) that is an important antigenic polysaccharide in nature, component of Brucella abortus (Bustamante et al., 2000; Gonzalez et al., 2006).
Notwithstanding this diagnostic disadvantage, Aparicio et al. (2003) and Samartino (2005) stated that contribution and benefits in the use of S19 strain in Campaign programs to control ruminant brucellosis outweigh such disadvantage. Also, they reiterate that in view of the controversy that preventive failure reported in immunized herds has caused it is necessary that its use be reevaluated to establish their efficacy in the control of the disease when exposed to different field conditions and challenges. Therefore, the objective of this study was to evaluate Brucella abortus S19 vaccine in cattle naturally infected with brucellosis under tropical conditions.
MATERIALS AND METHODS
Study Site
Research was carried out in the El Desengaño community, Municipality
of Las Choapas, Veracruz, Mexico between August 2006 and February 2008 (18 months);
in the first stage, a transverse epidemiological study was carried out to identify
herds naturally infected with brucellosis (Toma et al.,
1999).
Inclusion Criteria
Units that in tropical climate had a double purpose, grazing production
system, which had not vaccinated animals against brucellosis, were selected.
Samples were taken from animals six months or older. Infected herds were those
that had reactors to the tampon-antigen test or card test with antigen at 8%
concentration (CT) and at least one positive case to Rivanol Test (RT).
Serological Diagnosis
Five milliliter of blood were obtained from the coccygeal vein by vacutainer
without anticoagulant. Samples were transported in refrigeration to the Microbiology
Laboratory of the Veterinary Medicine and Animal Husbandry Faculty of the University
of Veracruz. Sera were then placed in vials identified with sample number and
preserved at -20°C until they were processed by the Animal Microbiology-CENID
Laboratory of INIFAP in Palo Alto, D.F., by CT and RT according to NOM-041-ZOO-1995
National Campaign against Brucellosis in Animals. Positive animals were those
that were reactors to CT, confirmed by RT, when there was titer of 1:25 or above
in non-vaccinated cattle or those that had a titer of 1:50 or above in vaccinated
cattle (SAGDR, 1996).
Clinical Assay
Based on the inclusion criteria of the clinical assay, a herd infected with
brucellosis was selected. To estimate the sample size from the vaccinated and
non-vaccinated groups the Win Episcope 2.0 program was used under the modality
of finding the difference between proportions by estimating the expected proportion
of 6% of animals positive to brucellosis in the vaccinated population and 20%
of animals positive to brucellosis in the non-vaccinated population, with a
confidence level of 95% and an 80% power. Thus the estimated sample size was
88 animals per group, but the results were improved with 10% more animals for
safety, foreseeing mortality, leaving 100 females per group (Thrusfield
et al., 2001). Vaccinated and non-vaccinated groups were identified
by ear tags with pair and impair numbers respectively; from the time of vaccination,
both groups were evaluated quarterly by CT and RT serology for eighteen months
as screening and confirmation tests, respectively (SAGDR,
1996). After twelve months of having applied the vaccine, all sera of the
vaccinated animals group that were positive to RT were analyzed by RID in the
Animal Microbiology-CENID Laboratory of INIFAP to confirm seropositivity by
field strains infection (Gonzalez et al., 2006).
Vaccination
All females that came out negative to CT and RT in the selection study were
vaccinated subcutaneously one only time in the middle third on the left side
of the neck. Strain S19 of Brucella abortus was applied in classic doses
(5x1010 CFU) in three to six months old females and in reduced doses
(3x108 to 3x109 CFU) in females older than six months
old, including gestating ones (SAGDR, 1996). Vaccination
of animals was carried out in the month of August 2006; at the time the experimental
groups were being formed, 41 gestating females were integrated into the vaccinated
group and 59 gestating females in the non-vaccinated group. Males were not vaccinated
and seropositive animals to RT were not segregated nor eliminated from the herd.
Statistical Analysis
Determination of seroprevalence rates, Relative Risk (RR) and Confidence
Intervals (CI) at 95% were estimated following Thrusfield (2005).
Statistical significance of observed frequencies in vaccinated and non-vaccinated
groups was estimated by chi-square and differences were considered significant
if p<0.05 (Daniel, 1999).
Vaccination Efficacy
Vaccination efficacy was estimated by the following formula:
Where:
| VE | = | Vaccination efficacy |
| CDR | = | Diseased animal rate within the control group |
| VDR | = | Diseased animal rate within the vaccinated group (Orenstein et al., 1985) |
Bacterial Isolation
Bacterial isolation was considered an inclusion criteria necessary to confirm
infection of the herd by Brucella abortus and thus, be able to evaluate
efficacy of strain S19 in the presence of field strains; therefore, in each
monitoring milk samples were collected in sterile Falcon type tubes and bacterial
isolation was carried out following procedure by Alton et
al. (1988) of all animals reactors to RT. Samples were maintained in
refrigeration from the time of collection until processed. From the milk fat,
duplicate primary seeding was carried out in Farrell selective media and incubated
in aerobiosis and micro-aerobiosis environments; media were incubated at least
during one month at 37°C and checked for colony development every other
day. Isolations suggestive of Brucella sp., by colony morphology were
seeded again in Trypticase Soy Agar (TSA) until pure culture was obtained and
identified by Biochemistry tests. Also, samples were sent refrigerated to the
Microbiology Department of the National School of Biological Sciences of the
National Polytechnic Institute, for confirmation diagnosis by simple Polymerase
Chain Reaction (PCR) (Matar et al., 1996; Hamdy
and Amin, 2002).
RESULTS
Serological Diagnosis
The transverse selection study allowed the identification of a brucellosis
infected herd by a serum reaction rate to RT of 1.2% (4/340) with agglutination
reactions between 1:25 and 1:400 that were considered positive in non-vaccinated
animals as established in NOM-041-ZOO-1995, National Campaign against Brucellosis
in Animals (SAGDR, 1996).
Clinical Assay
In the first six months of quarterly monitoring, three seropositive animals
to RT were identified in the non-vaccinated group; 75% of them (2/3) were detected
in the second quarter. Thus the number of animals with brucellosis in the non-vaccinated
group increased from four to seven; of these six were adult females (87.5%)
and the other a female one and half years old (12.5%).
In the vaccinated group a quarterly serum reaction rate to RT fluctuated between 5 and 10%, situation that implied the presence of adult reactors during the 18 months following vaccination. Nevertheless, one year after the vaccine was applied, in this group six sera were identified with positive titers to RT; to confirm if the antibodies present corresponded to a natural infection the sera were processed by RID test. Of the six sera that were analyzed, only one resulted positive so the corresponding female was diagnosed as naturally infected and the rest were identified as false positive reactors (Table 1).
During the 18 month long monitoring a total of four new infection cases were detected in the population: one in the vaccinated group and three in the non-vaccinated group, therefore the number of animals with brucellosis in the herd increased from four to eight. The presence of seropositive animals in the experimental groups propitiated an accumulated serum reaction in the vaccinated group (VDR) of 0.8% and in the non vaccinated group (CDR), the rate increased from the initial 3.3 to 5.8%. Disease dissemination in the herd favored the increase of initial serum reaction rate of 1.2% to an accumulated serum reaction rate of 3% during the study period (Fig. 1).
When estimating observed frequencies in vaccinated and non-vaccinated groups a year and a half after vaccination, significant differences were observed between them (p<0.05). Results that were obtained indicated that S19 strain Brucella abortus vaccine did not impede infection in the total susceptible vaccinated population; nevertheless the risk of infection of vaccinated animals and getting sick was very low, due to the protection effect of the vaccine (RR = 0.112; IC95% 0.14-0.887).
| Table 1: | Animals seropositive to RT and RID identified during the monitoring of the vaccinated group a year after vaccination |
| Fig. 1: | Serum reaction rates of the herd and the vaccinated and non-vaccinated groups during quarterly monitoring follow-up |
| Table 2: | Isolation of Brucella abortus in milk obtained from animals reactors to rivanol test during the study |
| Num.: Ear tag number identification, I: Brucella abortus isolated | |
It must be underlined that none of the 41 gestating females of the vaccinated group aborted as a consequence of vaccine application.
Vaccine Efficacy
Since, only one female of the vaccinated group got infected during the six
research quarters, strain S19 of Brucella abortus vaccine had a protective
efficacy of 86%.
Bacterial Isolation
Brucella abortus colonies developed in 43% of the cases (9/21) of
the 21 milk samples cultured in micro-aerobiosis; isolations that were obtained
came from 100% (6/6) of sampled reactor animals. It is important to underline
that isolation of the bacteria in milk of the vaccinated female reactor to RID,
was also obtained during the monitoring that was carried out three months after
having been diagnosed with brucellosis (Table 2).
Isolations were confirmed as Brucella abortus through PCR carried out in collected milk using primers that amplify OMP�s 31 kDa membrane protein.
DISCUSSION
Serological Diagnosis
RT test has a relative sensitivity between 86 to 97% and therefore it is
not recommended for eradication program final stages (Dajer-Abimerhi
et al., 1998); nevertheless, due to its high specificity (100%) it
identifies IgG antibodies derived from a strong antigenic stimulus and therefore
its presence implies an active infection, or chronic infection making it useful
as a confirming test in control program or early stages of eradication campaigns
(Dajer-Abimerhi et al., 1995; Diaz
et al., 2001). The four positive cases to RT that were identified
during the transverse study were females older than three years old, situation
that coincides with what Nicoletti (2005) mentions, referring
to the fact that the infection affects bovines of all ages but persists more
frequently in sexually developed animals.
Clinical Assay
After the second sampling, subsequent monitoring did not identify new infections
probably due to that which was mentioned by Casas (2003),
who stated that vaccination of adult females induces protection against brucellosis
and the animals obtain immunity in three to four weeks after the application
of the biological product, accelerating the control process of the disease.
NOM-041-ZOO-1995 National Campaign against Brucellosis in Animals, establishes that reduced doses strain S19 vaccinated animals, may be tested with official diagnostic tests only ten months after the application of the vaccine, with the purpose of detecting serum positive titers in the animals by RT test and then go on to confirmation if the positive sera correspond to infected animals or serum converting animals by more specific laboratory tests (SAGDR, 1996) following the provisions of this standard, during the fourth quarterly sampling six adult females of the vaccinated group were found to be reactors and of these only one resulted positive to RID and therefore was diagnosed as infected, three months later when the isolation of the bacteria in a sample of milk was achieved, the diagnosis of infection was ratified.
Stevens et al. (1994) and Ramirez et al. (2002) mention that the persistence of antibody titers against the O chain of the lipopolysaccharide (LPS) of the bacteria is increased when adult animals are vaccinated with strain S19; this explains the presence of animals with serum reaction to RT in the vaccinated group during the 18 months after the application of the vaccine. On the other hand, Bustamante et al. (2000) reported that the RT test may identify the persistence of these antibodies up to 20 months after vaccination, therefore it may not be considered as a trustworthy technique to separate strain S19 vaccinated animals from infected animals; indicating also that in view of this immunological process, reactor cattle should be recognized as serum converter or false positive until it is confirmed as an infected animal by additional diagnostic tests that have greater specificity such as RID proposed by Santiago et al. (1997).
Also, there is evidence provided by Cheville et al. (1992), Stevens et al. (1995) and Molina et al. (2004) in animals vaccinated against brucellosis, that when they are exposed to field strains of Brucella sp., they serum convert even though they do not get sick; nevertheless, this new challenge could be considered as secondary exposure and therefore, antibody titers tend to be higher and more persistent (Tizard, 2004). This implies that tests such as CT and RT are not efficacious enough to discriminate between infected and vaccinated animals and therefore other diagnostic tools are needed such as molecular tests, competitive ELISA or RID, to get to know the health status of the animals, situation that happens frequently in naturally infected herds (Bustamante et al., 2000; Santiago et al., 1997).
In a prolonged exposure of the bacteria to the immune system as it happens in a field infection, antibodies against NH are produced; nevertheless, these are not synthesized in a temporary exposure to the bacteria as is the case of vaccination with S19 strain. RID test identifies these antibodies and differentiates infected from vaccinated animals becoming then an important diagnostic tool that permits certainty in the decision to eliminate a serum reactor vaccinated animal from a herd (Bustamante et al., 2000).
Brucellosis is a disease characterized by the production of abortion after the second gestation trimester, with frequent posterior consequences such as placenta retention, metritis and endometritis (Osorio, 2004); nevertheless, in this study, no abortions occurred in any of the serum positive females of the non-vaccinated group that were beyond the second third of gestation. This observation could be explained based on statements by Rodriguez et al. (2005), who estimated that in a herd, 65% of infected females abort and of these 80% do so only one time during their productive life; the rest carry their gestation full term, as it happened in animals of this group. This leads us to believe that infection in reactor animals could have been chronic and therefore abortion could have already occurred before the observation period was begun. Females with advanced gestation in the vaccinated group, also did not present abortions. Nicoletti (1976) mentioned that the application of S19 strain vaccine in females at the final stages of gestation does not cause abortion in more than 1% of the animals; this statement coincides with observations made during this field study.
Vaccine Efficacy
Experiences with reduced doses have shown that S19 vaccine induces an effective
protection in bovines. Nicoletti (1976), Casas
(2003) and Samartino (2005) indicated that cattle
immunization with this strain favors a 65 to 75% protection of vaccinated animals;
in the rest of the population, a relative defense is stimulated since even though
the bovine is vaccinated, it can become infected as it happened during this
study with one of the females of the vaccinated group. Notwithstanding this
fact, 86% protective efficacy of strain S19 applied as single doses, obtained
during this field research, is higher than what was achieved by the researchers
of reference.
Bacterial Isolation
As it is observed in Table 2, isolations that were obtained
corresponded to six RT reactor animals that were sampled. This result is higher
in frequency than what was published by Rodriguez et
al. (2005), who mentioned that approximately half of the infected cows
after aborting or calving eliminate Brucella abortus in milk during several
weeks or months In this study, in two of the six reactor females that were monitored
there was a positive isolation at least in two occasions; which in turn corroborates
the presence and elimination of the bacteria in milk during several months,
situation that favors transmission of the disease in a herd but above it all
it also implies a high risk for public health.
In Table 2, it can also be seen that from all post-vaccination sampling, at least one positive isolation was obtained except during the second monitoring. Consecutive, alternate or only isolations from milk samples coming from animal reactors to RT may be due to the interrupted presence of the bacteria in milk. In relation to this, Osorio (2004) mentioned that the bacteriological procedure is not always successful due to the intermittence of Brucella sp., elimination in milk; likewise, Renteria et al. (2005) emphasized the fact that a large amount of them are needed in the sample or that it is collected at the time of excretion. Also, it is important that the highest elimination of bacteria in milk has been recorded after calving and the lowest during lactation peak (Rodriguez et al., 2005).
According to what has been mentioned by Dajer-Abimerhi et al. (1998) and Diaz et al. (2001), bacteria isolation and identification together with the presence of new cases during the investigation, gave way to a definite diagnosis of an infected herd.
Existence of reactor animals within the population constitutes a risk factor that favors disease transmission and permanence in the herd (Moreno et al., 2002; Renteria et al., 2003). Presence of Brucella abortus as a circulating field strain coming from reactor cattle that were not eliminated, propitiated that vaccinated and non-vaccinated animals had at all times a natural, constant challenge permitting the evaluation of the protective efficacy of strain S19.
CONCLUSION
Results of this investigation allow the conclusion that strain S19 of Brucella abortus is a biological product that may be efficacious in the control of bovine brucellosis in double purpose cattle that have a disease prevalence of 3%; yet, before its application, diagnostic interference problems that characterize it should be taken into consideration in order to evaluate from the economic point of view, vaccination efficiency.
ACKNOWLEDGMENTS
This study is part of the requirements that the first author must cover to obtain the degree of Doctor in Agriculture and Livestock Sciences granted by the Autonomous University of Yucatan who carry out an economic support with a CONACYT scholarship holder. Research received support and financing from the project Comparative study of strain RB51 and strain S19 efficacy in the prevention of brucellosis in herds with different sanitary conditions of the National Forestry, Agriculture and Livestock Research Institute (INIFAP) called for by SAGARPA-CONACYT 2004 Sector fund 23.