MOLECULAR DETECTION AND GEOSPATIAL DISTRIBUTION OF BRUCELLA INFECTIONS IN HORSES IN KANO METROPOLIS, KANO STATE, NIGERIA

: Brucellosis is a major zoonotic disease that affects several domestic, wild and marine mammals, and man. It causes serious economic damage in livestock productivity, especially in developing countries, including Nigeria. The study aimed to determine the species of Brucella that infect horses and establish the spatial distribution of the infection in Kano, Nigeria. A total of 328 blood samples were collected from horses. Sera were collected from the samples and kept at -20 0 C until tested. The sera were successively subjected to RBPT, SAT-EDTA and multiplex-PCR. Seventy-nine serum samples (24.0%) were positive for Brucella antibodies on RBPT. Out of the RBPT positive samples, 39 (49.4%) were also positive for SAT-EDTA. There was no difference between the results of the two tests (RBPT: p =0.4420; SAT-EDTA: p =0.124). Of the 39 SAT-EDTA positive sera, 32 (82.1%) were confirmed to be Brucella abortus . Fifteen of the 16 mares tested were positive for B. abortus while 73.9% of the stallion were positive. All horses in the age groups of 1-5 and 11-15 years were 100% positive, while 72.0% of the horses in the 6-10 year age group were also positive. The study showed 81.8 % detection rate of B. abortus using AMOS-PCR and that 15 (93.7%) of the female horses tested (16) had B. abortus while 17 (73.9%) of the 23 male horses tested had B. abortus.

Different methods have been used for the diagnosis of brucellosis.However, in most cases serological tests are being employed due to the nature of the organism as its confirmation is mostly done by competent laboratories due to its zoonotic nature (Bale, 1980).Serological tests mostly employed in the diagnosis of Brucella infections include RBPT, SAT, SAT-EDTA, Combs test, 2-Mercaptoethanol, MRT, SPT, CFT, Buffered Antigen Test (BPAT), Lateral Flow Test and Fluorescence Polarization Assay (FPA), among others (Baba, 2016).Molecular techniques have also been used in the identification of Brucella species.(Fekete et al., 1990;Bricker & Halling, 1994;Huber et al., 2009;Bertu, 2014).Similarly, Ohtsuki et al. (2008) used loop-mediated isothermal amplification method to detect Brucella spp.In addition, Pan et al. (2011) developed and applied the novel visual loop-mediated isothermal amplification of omp25 sequence for the rapid detection of Brucella spp.
Serology has remained the main means for identifying infection with Brucella in Nigeria with less attention in identifying the actual bacterium causing the disease due to poor laboratory backup for clinicians in the field.Where laboratory backup is available, the risks accompanied by working directly in identifying the organism have made it difficult for routine confirmation of the organism.This has made the disease to spread among cattle herds and other animal species with potential consequences.
Another problem in the diagnosis of the disease due to Brucella infection is the poor knowledge, attitude and practices of livestock owners and even the general public, especially horse owners and their grooms of other animals apart from cattle.This makes them to come hand-in-hand with the disease agents unnoticed and even facilitate their spread and transmission.The poor knowledge of the actual Brucella species infecting horses and other domestic animals in Nigeria, especially due to the poor disease surveillance network, poor laboratory back-up and inadequate clinical services to horse owners can also allow the disease to continue to flourish among horses and others animals.In the process, the disease agents, Brucella species will continue to ravage the livestock industry in Nigeria, especially that the disease has been reported to be endemic in Nigeria (Esuruoso, 1974;Falade, 1974;Falade et al., 1975;Falade and Shonekan, 1982;Avong, 2000;Bertu et al., 2010;Ior et al., 2013).The lack of knowledge on the Brucella species infecting livestock and particularly horses in Nigeria and the risk of zoonosis, especially due to B. abortus and B. melitensis in humans will certainly be a cause for concern due to the serious zoonotic implications of these particular Brucella species.In Nigeria, horses are being widely used in leisure, marriage ceremonies, horse racing, playing polo in addition to meat and skin.These activities lead to closer contact of horses with humans and hence increase the risk of acquiring pathogens such as Brucella.The objectives of the study were to detect the Brucella species present in horses using the Polymerase Chain Reaction (PCR) technique and to determine the geographic distribution of the equine brucellosis in Kano metropolis, Kano State, Nigeria.

MATERIALS AND METHODS
The study was conducted in the Kano metropolis, Kano State, Nigeria.The State is located between latitude 13 0 N and 11 0 N and between longitude 8 0 W and 10 0 E. Administratively, Kano State has 44 Local Government Areas (LGAs), with Kano metropolis having eight out of these LGAs (Fig 1).The State has a total land mass of 20,760 Km 2 with and a human population of 9,383,682 (NPC, 2006).It also has an elevation range of about 650 m above the sea level but extend up to 1000 m above mean sea level at Rishi hills (Aliyu et al., 2021).The State is bounded by Jigawa State to the North and East.It shares borders with Katsina and Kaduna states from the West and Southwest, respectively.It also shares borders with Bauchi State from the extreme Southeast (Figure 1).There are an estimated 2,500 horses in Kano State out of the 1.4 million horses reported to be in Nigeria (FAO, 2019).
The horses from four of the eight LGAs in Kano metropolis were sampled.The LGAs were selected based on the purposive sampling method as most of the horses in Kano State are located within the Kano metropolis.The stables were selected based on convenience sampling method and agreement by the stable owners to participate in the study.Furthermore, the horses sampled were based on simple random sampling within the stables.The method of Michael (2005) was used to arrive at the sample size of 328 horses based on the prevalence of 20.1% for Brucella antibodies in horses in Kaduna state, Nigeria as reported by Baba (2016).A total of 10 ml of blood was collected aseptically using 18 G needles and 20 ml syringes from each horse via the jugular vein after each horse was restrained by an assistant (as contained in the ethical clearance obtained from the ethical clearance committee, University of Ilorin).The serum samples that were positive for Brucella species by SAT-EDTA were used for molecular detection of the Brucella species using AMOSPCR (The oligonucleotide primers used included: B. abortus-specific (GAG-GAA-CGG-AAT-TTT-TCC-AAT-CCC), B. melitensis-specific (AAA-TCG-CGT-CCT-TGC-TGG-TCT-GA), B. ovisspecific (CGG-GTT-CTG-GCA-CCA-TCG-TCG), B. suisspecific (GCG-CGG-TTT-TCT-GAA-GGT-TCA-GG) and IS711-specific (TGC-CGA-TCA-CTT-AAG-GGC-CTT-CAT) (Bricker and Halling, 1994).Initial denaturation was at 95 0 C for 15 min.This was followed by 35 cycles of denaturation at 94 0 C for 30 s. Then annealing at 57 0 C for 1 minand 30 s extension at 72 0 C for 2 min, followed by final extension for 10 min.The amplified products were visualized in ethidium bromide-stained agarose gel (1%) electrophoresis under UV illumination.

Data Analyses
Data generated were presented in tables.The data were also analyzed using SPSS version 17.0 (2010).Fisher's exact test was used to test for agreement between various tests such as; M-RBPT, SAT-EDTA, 2-ME, and PCR.P-value <0.05 was considered significant.

RESULTS
A total of 79 (24.1%) out of the 328 serum samples were positive for Brucella antibodies using mRBPT.Furthermore, of the 79 positive samples, 39 (49.4%) were further positive for Brucella antibodies using SAT-EDTA (Table 1).found that the highest identification of B. abortus was in polo horses as 95.45% followed by ceremonial horses with 64.71% and none in racing horses (Table 6).There was no statistically significant difference in the identification of B. abortus by use (Fisher's Exact Test= 1.824; p= 0.360).
The distribution of B. abortus using geospatial technique, within the sampling locations and the LGAs as a whole is presented in Figures 2 and 3 respectively which implies that there was geographical/spatial variation within the LGAs and between the LGAs where the horses were sampled.The prevalence decrease outward from the point of sampling.
The distribution is most concentrated in Nasarawa LGA with 14.21% to 17.00% within the immediate vicinity of the sampling location and reduced as one moved away from the location.The least distribution was in Tarauni LGA with 3 to 5.8%.

DISCUSSION
From the study, the horses encountered during the study included Arewa, Argentine, Sudan and Talon breeds of horses.This has confirmed the presence of exotic breeds of horses in Nigeria.Such report has also been made by Fadimu (2014) and Baba (2016).Many workers like Kaltungo (2013) and Buhari (2014) have reported higher seroprevalence rates with SAT-EDTA than with RBPT with statistically significant differences.
The finding of seroprevalence of Brucella among the sampled horses indicates that these horses might have been mixing with other domestic animals within the area as Kaltungo (2018aKaltungo ( , 2018b) ) and Buhari et al. (2020) have reported cattle, small ruminants and camels meeting at grazing and watering points and that such mixing might have facilitated transmission and spread of infections in these animal species.The fact that these horses had seroprevalence of Brucella means that there is high risk of infection by the grooms and even the horse owners who may ride them once in a while.Another implication is that, at least for the polo and racing horses, they are usually involved in tournaments and their involvement in tournaments may mean spread of the infection to other distant areas.In another study in Kano State, Fadimu (2014) reported seroprevalence of 5.3% and 4.3% by using RBPT and SAT respectively.This study has further confirmed the existence of Brucella infection in horses in the State.
The finding of higher seroprevalence in this study might be that different horses might have been used by the two researchers.Other reports of Brucella infection in Nigeria include those of Esuruoso (1974); Bale & Kwanashie (1984); Ehizibolo et al. (2011); Bertu (2014); Baba (2016), andYakubu (2016).A report of Brucella seroprevalence in rodents in Nigeria has similarly earlier on been made (Avong, 2000).
The identification of B. abortus organisms through the use of AMOS-PCR in the blood of horses in this study means that the organisms are circulating in these horses.Thus, there is active Brucella infection in these horses.The implication of this is that the grooms and even their owners stand the risk of contracting Brucella infection anytime they initiate blood oozing out of the body of these horses.This is possible through injury of the mouths of the horses with bridle, injury of the body of the horse through the use of coarse grooming tools and such other materials that can initiate tear of the skin.Bertu (2014) and Buhari et al. (2020), both in Nigeria, also reported demonstrating B. abortus in the blood of cattle, sheep and goats during their studies.In addition, Buhari et al. (2020) Jalil (2008) has reported human brucellosis acquired through horses and that the risk is real.The exchange of grooming tools by grooms as reported by Baba, (2016) could also lead to the spread of the disease, should any of the horses groomed be infected with Brucella species and has bruises or tears on the skin.
The documentation of geospatial distribution of B. abortus infection in horses in the study area as produced in this study seems to be the first of such documentation to the best of our knowledge in the area.Similar documentations of the distribution of Brucella infection in animals in Nigeria were those by Bertu, 2014;Kaltungo, (2018a) and Buhari et al. (2020) who documented the distribution of Brucella infections in cattle and small ruminants in Kaduna, Katsina, Plateau and Sokoto States, Nigeria.

AKNOWLEDGEMENT
We would like to thank all the horse owners and grooms for allowing their animals to be part of the study.We also wish to thank the Director, Kano State Veterinary Services, for facilitating the study as well as for permitting the AVOs and ZVOs to mobilize the grooms in ensuring that the study was conducted smoothly.We are grateful to the technical staff of the Bacterial Zoonoses Laboratory, Department of Veterinary Public Health and Preventive Medicine, Ahmadu Bello University, Zaria, especially Mal.Sulaiman Yahuza for the support they gave during the course of the laboratory work.We also extend our deep appreciation to Mal Abdulmalik Shuaibu of the Central Laboratory, Usmanu Danfodio University, for the support during the molecular aspect of the study.We similarly appreciate the support and kind advice by Mal M Ismail and his colleagues in the Department of Geography and Environmental Management, Ahmadu Bello University, Few studies have reported Brucella species infecting livestock in Nigeria.These includeBertu (2014) who demonstrated Brucella abortus infecting cattle under the pastoralist herds in Kaduna and Plateau states, and Buhari et al. (2020) identified B. abortus and B. suis in the blood of sheep and goats in institutional farms.and Kaltungo (2018a, 2018b) showed B. melitensis infection in camels in a slaughter slab in Zaria.

Figure 1 :
Figure 1: Map of Kano State and the Local Government Areas Sampled in Kano metropolis (A, B, C and D).Source: Department of Geography, Ahmadu Bello University, Zaria, Nigeria.

Figure 2 :
Figure 2: Spatial Interpolation map imagery showing Brucella abortus density distribution using RBPT in the LGAs of Kano metropolis, Kano State, Nigeria.
The blood samples were collected in 20 ml blood sampling bottles and kept in a slanting position in a Coleman box with ice.Details of each horse: location, sex, age, breed, and use, were recorded.The blood samples were taken to the Bale (1980)sed to develop the geospatial map of Brucella infections in the study LGAs and Kano State as a whole.Serum samples were screened for Brucella infection using Modified Rose Bengal Plate Test (mRBPT) as described byBale (1980)and modified by Berta (2014).Serum samples that were positive for mRBPT were further screened with a Serum agglutination test with ethylene diaminotetraacetic acid (SAT-EDTA) as further evidence of actual infection.

Table 1 :
There was no statistically significant difference in the seroprevalence by test used (RBPT: Pearson Chi-Square= 2.821 a , p= 0.420; SAT-EDTA: Pearson Chi-Square= 5.763 a , p= 0.124).Of the 39 serum samples tested for characterization of Brucella species, 32 (82.1%) were confirmed to be Brucella abortus positive (Table2; Plate I).The infection of B. abortus between male and female horses is shown in Table3.Of the 39 serum sampled tested 16 were from female horses while the remaining 23 were from male horses.Of which 15 females (93.8%) were confirmed to have B. abortus.Similarly, of the 23 male horses, 17 (73.9%)werepositivefor B. abortus while the remaining 6 (26.1%) were negative for .B abortus (Table3).There was no statistically significant difference in the identification of B. abortus by sex (Fisher's Exact Test= 4.623; p= 0.150).Seroprevalence of Brucella infection using RBPT and SAT-EDTA in horses sampled in four Local Government Areas (LGAs) of Kano Metropolis, Kano State, Nigeria.
The identification of Brucella abortus by age is presented in Table4.All the 9 (100%) horses in the age group 1 to 5 years old were positive for B. abortus while of the 25 horses in the 6 to 10 years old bracket 18 (72%) were positive for B. abortus.Similarly, all the 4 (100%) horses in the 11 to 15 years old bracket and were positive for B. abortus (Table4).There was no statistically significant difference in the identification of B. abortus by age (Fisher's Exact Test= 2.824; p= 0.260).RBPT: Pearson Chi-Square= 2.821 a , p= 0.420 SAT-EDTA: Pearson Chi-Square= 5.763 a , p= 0.124

Table 2 :
Detection of Brucella species infecting horses by sex using AMOSPCR from four selected Local Government Areas (LGAs) of Kano state.

Table 3 :
Detection of Brucella species infecting horses by sex using AMOSPCR from four selected Local Government Areas of Kano state.

Table 4 :
Detection of Brucella species using the AMOS PCR from serum of sampled horses by age in four LGAs of Kano Metropolis, Kano State, Nigeria.

Table 5 :
Detection of Brucella species using the AMOS PCR from serum of sampled horses by breed in four LGAs of Kano Metropolis, Kano State, Nigeria.

Table 6 :
Detection of Brucella species using AMOS PCR from serum of sampled horses by use in Kano Metropolis, Kano State, Nigeria.
also reported detecting B. suis in sheep and goats blood samples.The existence of Brucella species in domestic animals means serious public health hazards and could be as a result of poor disease surveillance, and control strategies in place.It could also be due to the poor laboratory back up along with the relatively few Veterinary personnel that could cover the vast country to provide veterinary health care services.The poor public health services along with inadequate laboratory back up at rural areas where most of the animals are found might also result in the disease, brucellosis, being in humans without notice.Many cases in human and even veterinary clinics are routinely diagnosed based on clinical signs and symptoms.The lack of adequate public and to a certain extent private medical and veterinary care services could similarly result in many cases of human brucellosis without due attention being made for proper diagnosis.The cases routinely diagnosed in many human hospitals and clinics could actually be brucellosis as no real attention is routinely given for proper laboratory diagnosis of cases.None the less the chances of the spread of the infection could be said to be similar since both sexes are being used.The detection of B. abortus in horses by age indicated that all age groups were significantly involved with the organisms.Thus, the risk by the owners and grooms, no matter which age of horse is used is still there.The implication here is that youths, who drive pleasure in riding horses at weekends run the risk of becoming infected, especially that young horses are more involved in these weekend ride.The same inference can be deduced for identification of B. abortus by breed and use of horses since the horses are continuously used and there are chances for the horses to mix during durbar, weekend ride, polo or racing tournaments.Earlier,