Evaluation of EryC as Molecular Marker for Differential Identification of Brucella abortus from Vaccine Strain S 19

Brucellosis is a widespread, economically devastating and highly infectious zoonotic disease of animals prevalent throughout the world (Dawood, 2008). The disease causes serious economic losses to the livestock industry in terms of loss of calves, sterility, infertility, reduction or complete loss of milk yield after the abortion (Chahotal et al., 2003). Brucellosis continues to be of great health significance and economic importance in many countries.

Brucellosis can be eradicated from animals by quarantine of infected herds, vaccination, test and culling policy and regular surveillance. Use of attenuated live Brucella abortus strain S19 vaccine has been introduced in several states in the country for controlling the disease. One of the major drawbacks of this vaccine is the development of postvaccinal antibody response that interferes with the interpretation of serodiagnostic tests in eradication programmes (Nicoletti, 1980). To overcome this problem, attention has been focused on the development of diagnostic tools for differentiating vaccinated cattle from infected cattle (Nielsen et al., 1989).

In order to develop a tool to differentiate between B. abortus S19 from virulent Brucella strains, the studies have recently been focused on the eryC gene encoding D-erythrulose 1-phosphate dehydrogenase, which is involved in the erythritol metabolism in virulent B. abortus strains. However, there is a deletion of 702 bp sequence in eryC gene of B. abortus S19 (Sangari et al., 1994; Patil et al., 2014). This deletion of 702 bp sequence of eryC gene in B. abortus S19 strain thus provides a method to distinguish animals infected with virulent B. abortus strain from B. abortus S19 vaccinated animals by PCR amplification of this genomic region using an appropriate pair of primers (Eoh et al., 2010). 

The present study was planned with a view to develop a tool to differentiate B. abortus S19 vaccinated cattle from those infected with wild strain of B. abortus using EryC PCR assay.

The permission was taken from Institutional Biosafety Committee for conducting research as per letter ref No.BVC/Dean/VPH/IBSC/221/2016 dated 02/07/2016. Reference strains of Brucella i.e Brucella abortus 544, Brucella melitensis Rev 1 and Brucella abortus S-19 were purchased from Division of Biological Standardization, IVRI, Izatnagar, U.P. Along with that, two isolates received from J.J. Hospital Mumbai and three isolates received from BLDE University, Bijapur are included in the study. A total 1145 clinical samples derived from unvaccinated cattle and buffaloes including (200) blood samples, (710) sera, (190) vaginal swabs and (20) abomasal contents of foetus and (25) foetal tissues were collected in Tryptose broth from dairy farms in and around Mumbai and Pune. Also 146 blood samples were collected from vaccinated animals from Mumbai and Pune region. The sera samples were screened by RBPT. The vaginal swabs, abomasal content and foetal tissues were processed for isolation of Brucella by standard methods. The isolates suspected to be of Brucella were subjected biochemical tests for identification (Oie, 2009). The vaginal swabs, abomasal content and foetal tissue samples were used for isolation of Brucella organisms and further subjected to PCR. The blood samples, vaginal swabs, abomasal content and foetal tissue samples were also used for direct DNA extraction and further characterized by EryC PCR assay. The extraction of genomic DNA of B. abortus from the blood samples, bacterial cultures, Brucella reference strains and  Foetal tissue, abomasal contents was carried out by Leal-Klevezas et al., (1995), Romero et al., (1995a) and OLeary et al., (2006) and Leal-Klevezas et al., (1995) with slight modifications, respectively.

The nucleotide sequence of eryC gene of B. abortus was retrieved from NCBI GenBank and multiple sequence alignment of eryc gene of B. abortus S 19 and virulent B. abortus 544 was carried using Clustal W tool. The designing of primers for amplification of the eryC gene was undertaken using bioinformatics tools available online using the Primer Express Software Version 2.0.at ABI. The newly designed oligonucleotide primer sequences were synthesized and supplied by Urofins India Pvt. Ltd. The details of primer sequences designed is EryC-F (5'-CATGACACGCGGCA TATAAC-3') and EryC-R (5'- GACCTCCAGCTTACCCATGA -3'). The reaction mixtures were prepared using 200 ng/μl genomic DNA, 2.50 µl 10X PCR buffer, 2.0 µl 25 mM MgCl2, 0.50 µl 10 mM dNTPs, 1 µl (10 picomols) of primers each, 0.2 µl 5U Taq polymerase. The reaction mixtures prepared as above were subjected to cyclic conditions of Initial denaturation at 94^oC  for 3 min; 35 cycles of denaturation at 94^oC for 3 min, annealing at 60^oC for 45 Sec, extension at 72^oC for 1 min and final extension at 72^oC for 6 min. In order to assess the ability of EryC PCR assay to detect Brucella spp., blood samples spiked with different concentrations of Brucella abortus 544 and S19 were subjected to EryC PCR assay. EryC PCR assay was also applied on all the field isolates and directly on clinical specimens like blood samples (unvaccinated and vaccinated animals), vaginal swabs, aborted foetus tissues and abomasal contents. 

Serological detection of brucellosis
 
A total of 710 bovine sera samples were processed for detection of Brucella antibodies using RBPT. Of these, 200 samples were found positive by RBPT with an overall serological prevalence of 28.16%.

The seroprevalence of brucellosis in animals in India has been studied by several workers in India. Handa et al., (1998) reported prevalence of 6.6% in Delhi; Thakur and Thapliyal (2002) observed prevalence of 4.97% in Uttaranchal, Kumar and Nanu (2005) reported prevalence of 2.45% in Kerala. Similarly, Rana et al., (1985) reported a prevalence of 27.7% and Kumar et al., (1997) found prevalence of 28.57% in Delhi. As far as the studies conducted in Maharashtra are concerned, Aher (2010) studied the seroprevalence of brucellosis in bovines using RBPT and recorded prevalence of 64.81% in Pune and 16.16% prevalence in Thane. Differences in the serological test sensitivity, infection stage, duration and design of study and variations within infected flocks may be the possible explanation for these variations among different studies (Al-Talafhah et al., 2003).
 
Isolation
 
A total of 26 (11.06%) brucella isolates were recovered from 235 clinical samples. Five human isolates are also included in the study. All the isolates exhibited morphology and staining characteristics typical of Brucella spp. i.e. they were Gram negative coccobacilli, showed acid-fastness in MZN staining and appeared red coloured. The colonies of organisms isolated on BAM appeared round, glistening and smooth; those on MacConkey agar were non lactose fermenting and on blood agar the colonies were non haemolytic. The isolates recovered were further confirmed as members of Brucella spp. employing different biochemical tests. All isolates produced oxidase, catalase, urease, reduced nitrates while none produced indole.

From 235 clinical samples processed, 26 isolates were recovered including 4 from foetal tissues, 6 from abomasal content and 16 from vaginal swabs. Highest proportions of (30%) isolates were recovered from abomasal contents of aborted fetuses followed by 16% isolates from foetal tissues. The isolation rate was relatively low (8.42%) from vaginal swabs.

Several workers in India have made an attempt towards isolation of B. abortus with varying rates of isolation. Jeyaprakash et al., (1999) recorded isolation of B. abortus in 15 per cent cases from vaginal swabs employing tryptose agar while Das et al., (1990) using Brucella selective medium recorded an isolation rate of 38.18 per cent in cows and 14.28 per cent in buffaloes. Isolation rates are much higher during first two weeks of symptoms (80 - 90-% in acute form and 30-70% in chronic form) (Al Dahouk et al., 2003). Relatively low isolation rate (11.06%) recorded in the present study is in agreement with previous report of 6.4% (Celebi and Otlu, 2011; Kala et al., 2018). Slow growing and fastidious nature of the pathogen could be another explanation for a relatively lower isolation rate (Seleem et al., 2010).
 
Differentiation of Brucella abortus S 19 and wild strains by EryC PCR
 
In order to assess the ability of EryC PCR assay to detect Brucella spp., blood samples spiked with different concentrations of Brucella abortus 544 and S19 were subjected to EryC PCR assay. The results are presented in Table 1. The results showed that no amplification was generated at a concentration of 4 x 103 organisms per ml of blood while specific amplification product of 555 bp and 1257 bp was generated at the concentrations of 2 x 104 and 3 x 104 organisms per ml of blood (Plate 1). In blood samples spiked with B. abortus 544 a product of 1257 bp was generated while in those spiked with B. abortus S19 a product of 555 bp was generated. In samples spiked with both i.e. B. abortus 544 and B. abortus S19 two bands one each of 1257 bp and 555 bp was generated. The results clearly suggested that the EryC PCR assay developed during the present work is effective in differentiating between B. abortus virulent strains and vaccine strain S19 and could be used for differentiation of vaccinated and infected animals.





All the isolates and reference strains were also simultaneously subjected to EryC PCR. Brucella abortus S 19 revealed amplicon of 555 bp while B. abortus 544 and B. melitensis Rev1 showed amplicon of 1257 bp. All the 31 isolates showed amplification product of 1257 bp (Plate 2). This EryC PCR was also applied directly on clinical samples i.e blood samples (146) of vaccinated animals. Twenty one blood samples showed amplification product of 1257 bp while two blood samples showed amplicon of 555 bp. The results of EryC PCR assay are presented in Table 2. Thus B. abortus S19 specific amplification product of 555 bp was noticed in only two out of the total 146 vaccinated animals. The detection rate of B. abortus S19 from vaccinated animals was relatively low by EryC PCR which could possibly be attributed to the stage of collection of blood after vaccination. In the present investigation, the blood samples were collected from vaccinated animals randomly without following any specific timeline between collection of blood and vaccination. Among the other clinical samples tested including 200 blood samples of unvaccinated animals, 190 vaginal swabs and 45 abomasal contents and foetal tissues, none was found positive for B. abortus S19 specific amplification product of 555 bp (Plate 3).







Perusal of literature on development of PCR assay for differentiation of B. abortus S19 from wild type of Brucella strains indicates that several workers have attempted development of assay targeting EryC gene (Tuba et al., 2012, Chavarria et al., 2006, Sangari et al., 1994). The results of EryC PCR regarding its efficacy in differentiating the vaccinated and infected animals are in agreement with the findings of Patil et al., (2014).

The EryC PCR assay developed during the present study proved effective in differential identification of animals infected with virulent Brucella strains from those vaccinated with B. abortus S19.  

In countries where vaccination program is implemented to eradicate Brucellosis, it is important to have a tool that can differentiate animals infected with wild strain from those vaccinated with S19 strain. The reliable differentiation of vaccine strains from field isolates is an important element in brucellosis control programs. The conventional methods cannot meet this requirement however B. abortus S19 vaccine strain can be readily differentiated from field strains by EryC PCR. The EryC PCR assay developed during the present investigation thus could be an effective tool in differentiation of S19 vaccine strain from wild type of strains of Brucella spp.