Serological investigation of bovine brucellosis in three cattle production systems in Yewa Division, south-western Nigeria.

Limited data are available on the risk factors responsible for the occurrence of brucellosis amongst different cattle production systems in Nigeria despite its significant impact on livestock production. Consequently, a cross-sectional study was conducted to determine the prevalence of bovine brucellosis in three cattle production systems in Yewa Division of Ogun State, south-western Nigeria. A total of 279 blood samples (sedentary = 88; transhumance = 64; trade = 127) were examined for antibodies to Brucella sp. using the Rose Bengal test (RBT) and competitive enzyme-linked immunosorbent assay (cELISA). Overall, 24 (8.6%) and 16 (5.7%) of the animals tested seropositive for Brucella using RBT and cELISA, respectively. The herd seroprevalences based on RBT and cELISA were 31.6% and 15.8%, respectively. The results using cELISA reveal higher seroprevalence in the trade cattle (7.9%; confidence intervals [CI] = 3.2% - 12.6%) and those in a sedentary system (5.7%; CI = 0.9% - 10.5%) than in cattle kept under a transhumant management system (1.6%; CI = 1.5% - 4.7%). Age (> 3 years; p = 0.043) and breed (Djali; p = 0.038) were statistically significant for seropositivity to brucellosis based on cELISA, but sex (female, p = 0.234), production system (trade and sedentary; p = 0.208) or herd size (> 120; p = 0.359) was not. Since breeding stock is mostly sourced from trade and sedentary cattle, it is important that routine serological screening should be conducted before introducing any animal into an existing herd.


Introduction
Brucellosis is a disease of major public health importance, causing significant economic losses (Abdou 2000) to the animal industry because of abortion and infertility, and extensive chronic morbidity in humans.Brucellosis as a zoonosis poses a serious hazard for human health worldwide (Hamidy & Amin 2002;Zinsstag et al. 2007).The available evidence, although incomplete, shows that bovine brucellosis is widespread (McDermott & Arimi 2002).It is a serious, debilitating disease in humans, causing fever, headaches and further complications if left untreated (Abdou 2000;Zinsstag et al. 2007).Whilst some countries have eliminated or substantially reduced the disease by means of extensive eradication programmes, it remains endemic in Africa in particular (Omer et al. 2000).Some of the risk factors responsible for the spread of bovine brucellosis have been studied in some developing countries (Berhe, Belihu & Asfaw 2007;Dinka & Chala 2009;OIE 2011).Recent investigations based on serological studies have shown that bovine brucellosis is endemic in Nigeria (Abdou 2000;Ate et al. 2007;Cadmus et al. 2010) and that there is an apparent increase in its occurrence in the country (Junaidu, Oboegbulem & Salihu 2011;Ocholi et al. 2004).Previous reports have indicated that the prevalence of bovine brucellosis may be influenced by herding different species together (Bale, Nuru & Addo 1982;Junaidu, Oboegbulem & Salihu 2008;Ocholi et al. 2004), use of common pastures and water sources (Bertu et al. 2012), age (Cadmus, Adesokan & Stack 2008;Junaidu et al. 2011;Ocholi et al. 1996), breed (Cadmus et al. 2008;Esuruoso 1974), sex (Atsanda & Agbede 2001;Junaidu et al. 2011), lactation status (Junaidu et al. 2011) and season (Bertu et al. 2012;Nuru & Dennis 1976).However, the role of different cattle production systems needs to be assessed given the uncontrolled free movement of animals, especially as practised by the Fulani herdsmen, and unsupervised introduction of new animals into herds.
This study was therefore conducted to determine the seroprevalence of bovine brucellosis and identify associated risk factors in different cattle production systems in Yewa Division, southwestern Nigeria.

Study area
The Yewa Division of Ogun State (7°15'N, 3°3'E) was used for this study (Figure 1).Yewa is a town on the Nigerian border with the Republic of Benin.It is home to 60% of the cattle population

Read online:
in the state and houses a major cattle market that attracts cattle merchants from other African countries (e.g.Burkina Faso, Mali, Togo and Chad) travelling into Nigeria through the Republic of Benin.Cattle production in the region is extensive and includes sedentary, transhumant and trade cattle husbandry systems, which intermingle during grazing or at the market.
Sedentary farming is characterised by herds of cattle resident in the region for a minimum of six years.The transhumant system comprises Fulani 1 herds that migrate southwards from northern Nigeria in search of water and pastures during the dry season (between November and April) and hence can intermingle with resident herds at watering points.They then migrate back to the north at the beginning of the rainy season.Trade cattle comprise mainly cattle brought to the market from neighbouring West African countries that share boundaries and socioeconomic as well as cultural ties with the study site (Figure 1).These cattle are brought directly to the market for sale and hence have no contact with cattle from other production systems until they reach the market.Therefore, the simultaneous presence of cattle from the different production systems in Yewa at the time of this study provided an opportunity to determine the prevalence and other risk factors associated with brucellosis in these animals.
1.The Fulani is a major tribe that deals in livestock production in Nigeria.

Animal sampling, sample collection and handling
Herds in the sedentary and transhumant production systems were sampled, as well as trade cattle brought to the cattle market.Livestock owners in the study area were informed about the purpose of the study and its associated benefits to their operations.However, there were limitations to the number of herds/animals that could be screened, as some owners were unsure of the consequences of having their animals tested.
Every third herd from both the sedentary and the transhumant cattle production systems was randomly selected.Of the selected herds, 5% of each population was sampled (sedentary = 88; transhumance = 64).A group representing 5% of the trade cattle (n = 127) was randomly selected at the cattle market by picking one in every 20 animals (Table 1).A blood sample of 10 mL was collected from the jugular vein from each selected animal, using a 15-mL sterile vacutainer  tube.The breed, age and sex of the animals, as well as the herd size, were recorded.History of abortion amongst the cattle in the herds was also documented.
The blood samples were allowed to clot and centrifuged at 3000 g for 5 min.Serum samples were decanted and stored at -20 °C until assay.The sera were examined for antibodies to Brucella spp.using the Rose Bengal test (RBT) (Alton et al. 1988) and the competitive enzyme-linked immunosorbent assay (cELISA) (MacMillan et al. 1990).Both tests were sourced from the Veterinary Laboratories Agency (Surrey, United Kingdom) and standardised according to the stipulations set by the World Organisation for Animal Health (OIE 2008).
For the cELISA, some wells were used as controls for the monoclonal antibody as a basis for determining the positive/ negative optical density cut-off criteria for each plate.The plates were read at 450 nm using an automated Multiscan reader (Anthos Labtech, Austria) to determine whether a sample was positive or negative.Cut-off points for positive samples were set at 60% or lower when compared with the wells containing the monoclonal antibody, as determined in a previous study (Stack et al. 1999).

Statistical analysis
Data entry and analysis were performed using the statistical software package SPSS for Windows.Data were analysed to determine the association between seropositivity to Brucella antibodies and the different cattle production systems.Frequencies and proportions were stratified according to seropositivity and reported for individual, herd and cattle production systems.Group differences were tested using chi-square statistics for categorical variables.All variables significant at the 10% significance level were included in the multivariate logistic regression model.The odds ratios were reported with their 95% confidence intervals (CI).

Ethical considerations
The protocols for this study were approved by the Ethics Committee, Faculty of Veterinary Medicine, University of Ibadan, Nigeria (Ethics/12/10/03).

Potential benefits and hazards
There was no appreciable risk for the animals as only 10 mL of blood was collected from each animal.By participating in the study, owners would be aware of their animals' brucellosis status and act accordingly to minimise the spread of the disease.

Recruitment procedures
Cattle owners' participation was voluntary.The purpose of the study was explained to the cattle owners and they were allowed to withdraw from the study without any attached penalty.

Informed consent
The cattle owners who allowed their cattle to be sampled consented verbally as most of them could not read or write.

Data protection
The data from the respective herds were kept separately.Herd-specific results were conveyed to the relevant owners.

Rose Bengal test
Of the 279 serum samples examined using RBT, 24 (8.6%) were seropositive to smooth Brucella species.The individual and herd seroprevalences obtained were 8.6% and 31.6%,respectively.Seropositivity to Brucella was lowest in cattle kept under a transhumant management system (4.7%),followed by those in sedentary systems (7.9%).The highest seropositivity was detected in the trade cattle (11.0%) (p = 0.208).
Higher age-specific prevalence was recorded in animals older than three years (12.5%), with this group being more than twice as likely to be seropositive for brucellosis than those younger than three years (p = 0.033).Female animals also displayed a higher seroprevalence (11.2%) than male animals (5.5%), although the result was not significant (p = 0.226).Female animals therefore were almost twice as likely to be seropositive to Brucella (Table 2).
Animals sampled from large herds showed a higher seropositivity rate than those from small herds (33.3% vs 25.0%; p = 0.670).Cattle originating from herds larger than 120 animals were approximately one and a half times more likely to be seropositive for Brucella than those from herds with fewer than 120 animals.A chi-square test of significance and multivariate logistic regression showed that age (p = 0.033) and breed (p = 0.038) were statistically significant with regard to seropositivity to Brucella, whilst sex (p = 0.226), production system (p = 0.208) and herd size (p = 0.670) were not significant.

Competitive enzyme-linked immunosorbent assay
Results from the cELISA showed the seroprevalence of individual animals and herds to be 5.7% and 15.8%, respectively.The seropositivity results for the different production systems follow a similar trend to those for RBT.
Seroprevalences according to age and sex were higher in older (8.8%; p = 0.043) and female (7.3%; p = 0.234) animals, following a similar trend to that of results from RBT.A similar pattern was also obtained with respect to herd size (p = 0.359), with cattle originating from herds larger than 120 animals being approximately three times more likely to be seropositive for Brucella than those from herds with fewer than 120 animals.A chi-square test of significance and multivariate logistic regression show that age (p = 0.043) and breed (p = 0.038) were statistically significant for seropositivity, whilst sex (p = 0.234), production system (p = 0.537) and herd size (p = 0.359) were not significant (Table 3).Abortion or sterility was not significantly associated with seropositivity to Brucella in individual animals (χ 2 = 1.5, p > 0.05) (Table 4).

Discussion
In this study, 5.7% and 8.6% seroprevalence to brucellosis were obtained as measured by cELISA and RBT, respectively.This implies that the disease is endemic amongst the cattle population in Yewa Division of Ogun State and corroborates the results of previous studies in other parts of Nigeria (Abdou 2000;Ate et al. 2007;Ishola & Ogundipe 2000).
Although there were differences in the results obtained with the two tests, they were not statistically significant (p > 0.05) (Table 2); RBT is, however, known to be a less specific test than cELISA (Nuru & Dennis 1976) owing to its poor ability to discriminate between antibodies from a cross-reacting organism (Biancafiori et al. 2000;Samartino et al. 1999).The observation of fewer cELISA positives than RBT positives has also been noted in another study that demonstrated higher specificity of cELISA than RBT in cattle and humans (Mainar-Jaime et al. 2005).It has been recommended that when using two serological tests, animals should test positive to both in any serial testing strategy (Mainar-Jaime et al. 2005).Therefore, the 5.7% seroprevalence obtained with the cELISA is considered valid as all the samples testing positive with cELISA also tested positive with the RBT.
Overall, our findings reveal higher seroprevalence rates in trade and sedentary cattle than in the transhumance production system, a possible epidemiological factor to be considered in the spread of the disease.The higher seroprevalence recorded amongst cattle in the trade and sedentary systems is contrary to reports that showed highest incidences in pastoral production systems in sub-Saharan Africa (Bekele et al. 2000;Dinka & Chala 2009;McDermott & Arimi 2002;Schelling et al. 2003).This may be attributed to keeping large herds of cattle with a high turnover of animals infected with Brucella (Musa 1995), as replacement stock was often from similar husbandry systems.Keeping flocks of other species (such as goats) in close contact with cattle herds, especially under sedentary conditions, could also have contributed to this finding.
Furthermore, the higher seroprevalence of Brucella infection in the sedentary system may be due to the introduction of newly purchased animals from similar husbandry systems  These results did not include the trade cattle.
into existing herds without prior screening.This is often the case, as farmers' decisions in such transactions are based on the cost of the animal, with lower prices being favoured.Again, the sedentary herds used in this study had been established in the same region for more than the past six years.Given the years of existence and close contact with other potentially infected animals, it was therefore possible that Brucella circulated within the herds.
The difference in breed-specific prevalence is contrary to the findings of Cadmus et al. (2010), who showed that the breed of cattle was not significantly associated with the disease.Meanwhile, age-specific prevalence was higher in animals older than three years (8.8%) than in younger animals (2.8%), which is consistent with several reports (Bekele et al. 2000;Berhe et al. 2007;Kubuafor, Awumbila & Akanmori 2000).
Sexually mature and pregnant cattle have been found to be more susceptible to infection by Brucella than sexually immature animals (Walker 1999).Younger animals tend to be more resistant to infection and frequently clear infections, although re-infection could occur at a later time (Radostits, Blood & Gay 1995).The higher prevalence of brucellosis in older cattle could be attributed to consistent exposure of the cattle to the infectious agent.
Similar to the findings in this study, other studies also recorded a higher seroprevalence in female animals than in male animals (Bekele et al. 2000;Berhe et al. 2007;Dinka & Chala 2009;Kebede, Ejeta & Ameni 2008;Kubuafor et al. 2000;Tolosa, Regassa & Belihu 2008).According to Kebede et al. (2008), male animals are generally kept in the breeding herd for a shorter time than female animals, thus making the chances of exposure lower for male animals.Berhe et al. (2007) also stated that the serological response of male animals to Brucella infection is limited and that the testes of serologically positive male animals were usually observed to be culture negative.
The difference in the herd sizes did not significantly affect the number of seropositive animals in this study.However, this is contrary to reports by some authors, who asserted that large herd size is one of the major risk factors for bovine brucellosis (Berhe et al. 2007;McDermott & Arimi 2002).It has also been reported that a large herd size increases the exposure potential when a large number of animals are in contact with each other at common feeding and watering points, with higher risk following cases of abortion (Dinka & Chala 2009).Nonetheless, the overriding factor for infection in this study may be common exposure of these animals, irrespective of the herd size at watering and feeding points, particularly during the calving period.
The findings of this study had some limitations.Firstly, the number of animals sampled was small; more samples would probably have provided better epidemiological information about each production system.Consequently, limited data were obtained regarding the roles of age, sex and breed as risk factors in each group.Secondly, there were no records of the pregnancy status of the cows screened.
This could have introduced some bias in the observed ratio of infections amongst female and male animals, as it has been noted that pregnant cattle at more than five months of gestation are more susceptible to Brucella infection.This is due to the preferential localisation of Brucella in the uterus, where allantoic fluid factors such as erythritol stimulate the growth of Brucella (Godfroid et al. 2004).Thirdly, we obtained different infection rates from the RBT and cELISA, which might be due to the unknown vaccination status of some of the animals, especially those within the trade cattle production system.Vaccination against Brucella spp.using Brucella S19, which is the vaccine available in Nigeria, is not routinely carried out.However, many of the trade cattle were from neighbouring African countries, whose practice regarding routine vaccination is not known.It is therefore possible that some of the animals were already vaccinated, resulting in the different rates found with the RBT and cELISA.Furthermore, cultures were not performed; this would have helped to confirm the true status of the animals with Brucella infection.

Conclusion
This study highlighted the endemicity of brucellosis in the different cattle production systems screened in Yewa, south-western Nigeria.It also highlighted the fact that trade and sedentary cattle production systems are risk factors associated with exposure of animals to Brucella.As our findings show lower seroprevalence amongst transhumant cattle, it is suggested that additional vaccination efforts should be considered for this group of cattle to protect them from other infected groups.In addition, further work should be carried out using a larger sample size in order to achieve the best control measures in countries with endemic bovine brucellosis.
Finally, given the close contact of cattle herders with their animals further epidemiological studies to investigate the link between bovine and human brucellosis in the present study area are recommended.This will go a long way towards formulating strategies that will help in the control of brucellosis, given its public health implications in Nigeria and its neighbours.

TABLE 1 :
Population size and number of animals sampled from each cattle production system in Yewa Division, Nigeria.

TABLE 2 :
Risk factors associated with seroprevalence of bovine brucellosis as measured by the Rose Bengal test and competitive enzyme-linked immunosorbent assay in three cattle production systems in Yewa Division, Nigeria.
RBT, Rose Bengal test; n, sample size; cElisa, competitive enzyme-linked immunosorbent assay.*, There was no significant difference in the use of the RBT and cELISA in the diagnosis of brucellosis.

TABLE 3 :
Risk factors associated with seroprevalence of bovine brucellosis as measured by competitive enzyme-linked immunosorbent assay in three cattle production systems in Yewa Division, Nigeria.This did not include the trade cattle as they are not organized into herds.
*, Risk factors statistically significant for brucellosis.

TABLE 4 :
Relationship between animals with history of abortion/sterility and brucellosis as measured by competitive enzyme-linked immunosorbent assay amongst the transhumant and sedentary cattle populations screened in Yewa Division, Nigeria.