Patho-epidemiological and molecular diagnosis of swine brucellosis in Punjab

Swine brucellosis is a serious problem for progressive swine farmers as it leads to economic losses in swine industry with birth of dead or weak piglets or abortion in sows and orchitis in boar (Ogundipe et al., 2001; Olsen and Tatum, 2017). The Brucella, Gram-negative, facultative intracellular, pathogenic bacteria, infects a wide variety of domestic and wild mammalian species. Brucella suis is the only spp. of Brucella genus whose infection in swine leads to systemic or generalized infection that results into reproductive failure and orchitis while other spp. of Brucella infect only up to regional lymph node (Deyoe and Manthei, 1967). Four members of the Brucella genus, B. melitensis, B. suis, B. abortus, and B. canis are zoonotic agents and effective serological surveillance is essential to be taken for its control (Erdenebaatar et al., 2004). To eradicate brucellosis from the livestock populations a rapid, definitive and accurate diagnosis of Brucella infection is very important (Surucuoglu et al., 2009).There are various tests employed for the diagnosis of brucellosis in pigs viz. culture, serological, immune-histochemical and molecular methods. Although culture of organism is considered as gold standard but due to low sensitivity, time consuming and biohazardous, it is not commonly used as routine diagnostic test (Ferris et al., 1995). Various serological tests are used with varying degree of sensitivity and specificity for screening herds and individual animals. However, cross reactions with other Gram- negative bacteria especially Y. enterocolitica serotypes O: 9 are a major problem of the serological assays (Munoz et al., 2005).Immunohistochemical techniques have been used for diagnosis of brucellosis in formalin fixed, paraffin-embedded tissues of aborted fetuses.PCR has now been widely used for diagnosis of brucellosis in pig in cases of abortion.But there is scanty information about the various conventional and advanced bio-molecular tests for diagnosis of brucellosis in swine in India. In the present study, a comprehensive diagnosis of porcine brucellosis was done using c-ELISA, histopathological, IHC and PCR techniques.

Seroprevalence studies
 
Ninety adult pigs were randomly selected from 15 swine farms (Fig 1) located in Punjab by using random number table. Total swine population of these selected farms was 800, out of which 90 (approximately 11%) belonging to various swine owners were randomly selected.The permission of sampling and other procedures was duly approved by institutional animal ethics committee (IAEC). About 5 ml of blood was collected from each of the pig aseptically from the ear vein in a test tube. The test tube was kept in slanting position at room temperature till clotting. The serum was collected from clotted blood by centrifuging at 3000 rpm for 10 min. The separated serum was collected in a screw capped plastic vial and stored at -20°C until they were tested for antibodies to Brucella (Rose Bengal Plate Test and Svanovir Competitive ELISA kit). Ninety serum samples of animals of different sex and age group i.e. 73 females (30 gilts and 43 adults) and 17 males were collected.
 

 
Pathological studies
 
Tissue samples from reproductive tract of pig (n=32), placental tissue (n=10) and aborted fetuses (n=18) were collected from Postmortem Hall of Department of Veterinary Pathology, GADVASU, field out breaks and from butcher houses in and around Ludhiana. Tissue samples from reproductive tract of pig and aborted fetuses viz. lung, heart, liver, spleen, lymph node, kidneys, brain, intestine etc. including placenta (n=10)were collected in 10% neutral buffered formalin for histopathology and  immunohistochemistry. Tissues were later embedded in paraffin wax and 4-5µ sections were cut and stained with routine haematoxylin and eosin technique (Bancroft and Gamble, 2002).
 
Immunohistochemical studies were performed as per Mahajan et al., (2013). Briefly, 4-5µ thick paraffin embedded tissue sections were cut and mounted on Superfrost Plus, positively charged microscopic slides (Fisher Scientific, Hudson, New Hampshire,USA). After heat induced antigen retrieval and endogenous peroxidase blocking, the slides were incubated with commercially available monoclonal antibody against Brucella abortus (Abcam) in a dilution of 1:50 followed by incubation with secondary antibody (Universal, Vector) and colour developed using 3, 3-diaminobenzidine (DAB) and counterstained with Gill’s haematoxylin. As negative control, sections were incubated with Phosphate Buffer Saline (PBS) instead of the primary antibody.
 
Molecular studies
 
For molecular diagnosis 34 samples including stomach content (n=6),placental tissue (n=10) andfetal tissue samples (n=18) viz. lung, heart, liver, kidney, spleen and  brain of aborted fetuseswere processed for genus specific PCR for brucellosis. DNA was extracted using DNeasy Tissue kit (QIAGEN) as per manufacturer’s protocol. In amplification of DNA, primers JPF GCGCTCAGGCTGCCG
ACGCAA and JPR ACCAGCCATTGCGG TCGGTA that  target Omp2 gene ,specific for Brucella genus designed previously (Leal-Klevezas et al., 1995) was standardized by varying the concentration of the reaction mix i.e. 25 µl PCR Master Mix (2X), 1.25µl of both primers each, 8µl of template DNA and 14.5 µl of nuclease free water. Following a 4 min inactivation at 94°C reactions preparations were cycled in a thermocycler for 35 cycles consisting of 1 min at 94°C, 1 min at 60°C and 1 min at 72°C.  Final extension was carried out at 72°C for 3 min.The amplified product was analyzed by gel electrophoresis in a sub-marine horizontal electrophoresis unit (Power Pac Universal- Bio Rad).
 
Statistical analysis
 
The data was analyzed using Win Episcope 2.0 software, (Diagnostic test agreement), SPSS (Statistical Package for Social Sciences) for Window version 11.0.1 ©SPSS Inc. USA computer software programs. The mean histopathological score that represents the overall damage to histo-architecture was calculated and further compared with the immunohistochemical score of the respective cases using SPSS software.

Seroprevalence studies
 
Serum samples from selected animals were analyzed by RBPT and cELISA which revealed an overall apparent prevalence of brucellosis to be 16.7 and 12.5 percent, respectively. Serology is standard method for the epidemiological surveillance of brucellosis (Kushwaha et al., 2016).The seroprevalence of brucellosis (12.5%) by competitive ELISA observed in the present study was similar to/ close to that described by previous studies (Koppel et al., 2007; Leuenberger et al., 2007) with a seropositivity of 11.3% and 9.5%, respectively. Sale and purchase of pigs is quite common in this region and introduction of single infected pig in shed can spread infection in the herd. Moreover, pig farmers are not very educated and facilities of screening the pigs against brucellosis prior to purchase are not available at doorsteps which might be responsible for higher prevalence of brucellosis. However, low seropositivity (3.03%) and high seropositivity (41.04%) was reported by Jindal et al., (2017) and Shome et al., (2018) respectively, using ELISA.
 
In the present study, sows were grouped into two categories, gilt and adult females. Per cent sero-prevalence of Brucella antibody positive females in two categories was 6.66 and 11.63 per cent respectively. The seroprevalence of the disease was non- significantly higher (Chi square=2.498; P=0.114) in males (23.5 %) than females (9.58%). These results were in agreement to that described earlier (Kumar and Rao, 1980) with a seropositivity of 61.9 % and 34.2% in males and females respectively. These results were contrary to the results of Thoppil, (2000) with a seropositivity of 8.2 % and 1.6% in females and males respectively. Age is an important determinant factor to describe the distribution and risk of the disease. Non-significantly (Chi-Square= 0.502,P=0.479) higher incidence of brucellosis was found in adult sows as compared to gilts. High prevalence of brucellosis among older animals might be related to maturity with the advancing age (Kazi et al., 2005). In addition, adult females are comparatively under greater physiological stress during pregnancy which makes them more susceptible to infection (Walker, 1999). Moreover, multiple factors viz. age, sex, breed, location, herd size and living condition influences the seroprevalence of brucellosis (Ghani et al., 1998; Uddin et al., 2007).
 
In the present study, comparison was made between RBPT and c-ELISA used for serological diagnosis of brucellosis. Substantial degree of agreement (Kappa= 0.676, 95% CI= 0.372-0.826) was found between two tests (Table1) Albeit, the two tests showed degree of agreement, however the variation in prevalence by the two tests could be due to false positive. The RBPT is often used as a rapid screening test for diagnosis of brucellosis in swine. c-ELISA is a prescribed test for international trade but none of conventional serological tests including RBPT has been shown to be entirely reliable for routine diagnosis in individual pigs (OIE, 2012; Shome et al., 2016). RBPT is sensitive test but not specific while c-ELISA is both specific and sensitive test (Perrett et al., 2010; Praud et al., 2012) and can eliminate cross-reaction due to the Y. enterocolitica serotype O: 9 or other cross-reacting antibodies, such as IgM. Moreover, porcine serum samples undergo hemolysis very quickly and c-ELISA is suitable for diagnosis in such un-testable samples (Perrett et al., 2010).
 

 
Pathological findings
 
In the present study, stillborn, mummified, dead and aborted fetuses at different stages of gestation along with live fetus were seen. Abortion due to brucellosis was reported in 5 fetuses. Gross lesions in aborted fetuses were red tinged fluid in the thoracic and abdominal cavities, enlarged haemorrhagic lymph nodes, necrotic patches on lungs, congested spleen, haemorrhagic kidney and epicardium, congestion in brain meninges.Examination of placenta revealed hemorrhages, edema and hyperemia. Histopathological examination revealed focal to diffuse mixed infiltration of cell sin the fetal organs like lung especially in bronchi and interstitial septa (Fig 2), kidney and in other organs. Large areas of hemorrhages were seen in spleenand lymph nodes with inflammatory cells.Lesions in the placenta were characterized by diffused mixed cell infiltration (Fig 3). The observed gross and histopathological changes in fetal organs may be due to bacteremia caused by invading bacteria that gain entry in host body through epithelial cells, colonize in regional lymph nodes then through infected macrophages or lymph these organism infect and persist in many other organs viz. lymph nodes, placenta and genital organs (Poester et al., 2013). These histopathological changes observed in fetal organs were similar to the B. suis infection in European hares that considered as a reservoir of the organism (Bendtsen et al., 1956). In the present study, histopathological changes suspected to be of brucellosis were observed in 8 reproductive tracts and characterized by cellular infiltration of the endometrial stroma, endometrial glands, bacterial colonies and hyperplasia of periglandular connective tissue. While nodules and abscess that are a characteristic lesion of chronic infection were not seen in any of the collected samples. Previous reports of B. suis infection in boars suggested that boars that develop genital infection from brucellosis seldom recover and showed behavioral changes that can be due to testicular involvement (Megid et al., 2010). 
 
@fiugre2
 

 
Immunohistochemical (IHC) staining revealed positive immunostaining for Brucella in bronchial epithelial cells of lung (Fig 4), endothelial cells of the spleen (Fig 5), hepatocytes in liver, cells in lymph node, endothelial cells in kidney of fetus and placenta (Fig 6) and appeared as brown, finely granular intracytoplasmic staining. In addition, positive immunostaining to Brucella was also seen in lamina propria cells of uterus (Fig 7) of 4 sows. Organ-wise IHC score of positive cases for brucellosis in aborted fetuses revealed the high immunoreactivity of lung tissue of positive fetuses for Brucella antigen followed by spleen and lymph nodes. The pearson’s correlation coefficient between the mean histopathological scoring and the immunohisto chemical scoring in lungs was found to be significantly positive(r=0.857337) (Table 2 Graph 1). Overall Histopathological scoring and the immunohistochemical scoring of uterus positive for brucellosis was compared (Table 3) and similarly, positive pearson’s correlation (r= 0. 871904) (Graph 2) was found between histopathlogical lesions and immunohistochemical expression in uterus for brucellosis.




 

 

 

 

 

 

 
Molecular diagnosis
 
In the present study, reference strains as well as 8 of the 34 samples (2 from stomach contents, 2 from placental cotyledons and 4 from pooled tissue(s) produced 193 bp amplicon which correspond to Brucella genus (Fig 8). The remaining 26 samples failed to produce the targeted amplification. Several PCR protocols have been developed for identification of infectious agents and an important alternative rapid technique that overcome problems and disadvantage of currently used conventional methods for diagnosis of brucellosis in swine (Kanani et al., 2008; Grégoireet_al2012; Pilo et al., 2015). In the present study, JPF/JPR primer pair amplified a 193 bp region of the sequence encoding an outer membrane protein (omp2).
 
@figure8
 
Comparison of diagnostic tests
 
The comparisons between diagnostic tests are presented in Table 4. Histopathological lesions suspected for brucellosis were observed in twelve cases and IHC added to the findings from examination of HE stained sections, providing specific labeling for Brucella in seven cases. Out of 6 samples of stomach content, 10 samples of placental cotyledon and 18 pooled fetal tissue samples IHC detected seven and PCR detected eight positive cases.Comparison of diagnostic tests revealed a substantial degree of agreement between histopathology and IHC (kappa value 0.641, CI- 0.251-0.6421). In the present study, PCR and IHC provide a reliable test for the diagnosis of brucellosis in abortion cases on aborted fetal tissue and placental cotyledons as revealed by kappa value of 0.904 (CI=0.438-0.904) which reflect almost perfect degree of agreement between PCR and IHC. However in autolyzed fetuses, fetus stomach content found positive for brucellosis by PCR indicating higher sensitivity of PCR over IHC.
 

Thus, PCR and IHC provide a reliable test for the diagnosis of brucellosis in pigs which supplement the gross and histopathological alteration of the disease. However, serology is important for screening and detection of Brucella positive animal in a farm.