Molecular detection of Brucella spp. in ruminant herds in Greece

Brucellosis is a worldwide distributed infectious disease. Ruminants and other animal species (swine, dogs, equids, etc.), as well as wild mammals, can be affected. The disease can be transmitted to humans through the food chain or by direct contact with infected animals. Because of the relatively high economic burden due to abortions within a herd, significant efforts have been employed and hence the disease in most European countries has been eradicated. Accordingly, Greece applies both control and eradication programs concerning small ruminants (sheep and goats) and bovines depending on the geographical area. Current challenges in the standard antibody-based laboratory methods used for Brucella detection are the failure to differentiate antibodies against the wild strain from the ones against the vaccine strain Rev1 and antibodies against B. melitensis from those against B. abortus. The aim of the study was to reexamine and combine previously published protocols based on PCR analysis and to generate a rapid, not expensive, and easy to perform diagnostic tool able to confirm the doubtful results delivered from serology. For this reason, 264 samples derived from 191 ruminants of the farm and divided in 2 groups (male/female) were examined with a modified DNA extraction and PCR protocol. Molecular examination revealed the presence of Brucella spp. in 39 out of 264 samples (derived from 30 animals). In addition, Brucella spp. was detected in infected tissues such as testicles, inguinal lymph nodes, fetal liver, and fetal stomach content.


Introduction
Brucellosis, also known as Malta fever, is one of the most common bacterial zoonoses in Mediterranean countries (Wareth, et al., 2019) with intensive farming (Memish, et al., 2004, FAO/Brucellosis in human and animals., 2006). The main causative agent of human Brucellosis is Brucella melitensis (Alton, et al., 1996) (Brucellosis in human and animals., 2006). The disease is transmitted mainly from infected ruminants to humans and from the consumption of unpasteurized dairy products (Cooper, 1992 (Godfroid, et al., 2013, Berger, 2016. It is mostly an occupational zoonosis (International Labour O ce, 2010). Even if human mortality rate is low, the burden caused by the disease is substantial because of its chronic nature and the frequent relapses that occur with systemic manifestation (Colmenero, et al., 1996). Therefore, an early and accurate diagnosis is of utmost importance (Pappas, et al., 2005).
The natural reservoirs of the bacterium in the environment are the domestic ruminants of the farm, such as sheep, goats and cattle. Brucellosis in farm animals represents an equally great burden with nancial consequences in most parts of the world (Rossetti, et al., 2017). Several countries though in western and northern Europe, Canada, Japan, Australia and New Zealand are believed to be free from the pathogen (OIE, https://www.oie.int/en/disease/brucellosis/ 2021). Given the fact that there is no human vaccine available, the key-player to prevention of human disease is the control of zoonosis in animal populations either by vaccination programs or by eradication strategy.
Since 1977, the basic concept for controlling and eradicating brucellosis in Greece has always been vaccination of herds combined with blood testing and slaughtering of all animals found positive in laboratory tests such as Rose Bengal Test (RBT) and Complement Fixation Test (CFT) (seropositive animals). The State Veterinary Services from the Ministry of Rural Development & Food and from the Ministry of Interior have implemented a control and eradication policy, based on systematic vaccination of female animals over the age of 3 months old and blood-sampling of males over the age of 6 months old on the mainland (vaccination zone-VZ), as well as a test and slaughter policy on most of the islands (eradication zone-EZ). The vaccination of sheep and goats was based on the conjunctival administration of the attenuated live vaccine strain, Rev1 (OCUREV ® CZV), once in their life of female sheep and goats. According to the above program, it is strictly forbidden to move animals from the VZ to the EZ, or to move animals which have not been vaccinated within VZ (Fig. 1). Animals found positive in RBT and/or CFT are slaughtered separately from healthy ones, taking all necessary biosecurity measures to prevent the spreading of infection (Tzani, et al., 2012, Greek Ministerial Decision on sheep & goat brucellosis programme, 3339/117339 2016). Unfortunately, this approach was unsuccessful due to various reasons (Katsiolis, et al., 2018).
Detection of Brucella spp. by culture method is sometimes unsuccessful. The current golden standard in terms of laboratory techniques for the screening of brucellosis in ruminants are Rose Bengal Test (RBT) and Complement Fixation Test (CFT). Although both methods are o cially recommended by the EU, fail to discriminate the produced antibodies between the wild and the vaccinate strains resulting in generation of false positive results (Stournara, et al., 2008) that provoke wrongful slaughtering. Therefore, the generation of a novel, rapid, sensitive and speci c molecular diagnostic tool such as PCR directly on tissues of suspected animals (mainly aborted fetuses and associated maternal tissues, spleen of seropositive animals) (Bricker, 2002) should be developed in order to accurately detect the pathogen. The most common di culties comprise the preparation of samples and puri cation of PCR inhibitors (Amin, et al., 2001).
For all the above reasons, a reliable PCR test able to detect the bacterium in infected tissues of seropositive animals would minimize the false positive results and would boost the laboratory involvement. Once reconstituted one intraocular dose (1drop ≈ 35μl) of the vaccine Rev1, contains 1-2x10 9 cfu of live attenuated B. melitensis strain each dose of the vaccine Rev1 (smooth phase). Accordingly, each dose (2ml, sc) of the vaccine RB51, contained 10-34x10 9 cfu/ml of live attenuated B. abortus strain RB51 (Tittarelli, et al., 2008) (rough phase). These two types of vaccines are widely used in national brucellosis eradication programs in Greece. B. melitensis strains (16M, Rev1) were cultured aerobically while the B. abortus strains (544, RB51) were cultured under microaerophilic conditions, respectively. For cultivation Brucella agar (Oxoid, Hampshire, UK) supplemented with Brucella Selective Supplement SR0083A (Oxoid, Hampshire, UK) and Columbia sheep blood agar plates (Oxoid, Hampshire, UK) were used.

Samples collection
In this study, 264 samples were collected from 191 domestic ruminants from farms/herds located in different regions of Greek mainland, during the period 2016-2018. All samples were immediately forwarded to the Department of Microbiology and Infectious Diseases (Aristotle University of Thessaloniki, Greece) for DNA extraction. The extract was then stored at -20 o C for further analysis.
The samples were divided into two different groups. The rst group included 121 samples from tissues (49 testicles, 56 spleens and 16 inguinal lymph nodes) that were collected from 83 non-vaccinated, slaughtered male animals which were found positive in RBT and/or CFT (47 rams, 29 billy goats and 7 bulls).

DNA extraction
All tissue samples were processed under Biosafety Level three (BSL3) with high personal protections (WHO, 2004). Tissue samples were processed aseptically by removing extraneous material and chopped into small pieces and macerated using a tissue grinder with 200μl of sterile Phosphate Buffered Saline (PBS-Sigma Aldrich, Germany).
Genomic DNA from reference bacterial strains was extracted using the High Pure PCR template preparation kit (Roche, Basel, Switzerland) according to the instructions of manufacturer but also with some adaptions as previously described (Karponi, et  For the PCR, a thermal cycler with heated lid was used (Techne, Touchgene, Gradient). After initial denaturation at 95 o C for 7 min, the PCR pro le was as follows: 1 min of template denaturation at 95 o C, 1 min of primer annealing at 59 o C and 1 min of primer extension at 72 o C, for a total of 40 cycles, with a nal extension at 72 o C for 10 min. After PCR ampli cation, 12 μl of PCR product and 4 μl of bromophenol blue (loading buffer) were loaded into wells in 1.5% regular agarose gel (Ultra Pure Agarose, Invitrogen) in Tris base, Acetic acid and Ethylene Diethyl Tetracetic Acid (EDTA) (TAE) buffer in a cuvette ooded with TAE 1x slightly covering the gel. 1 kb DNA ladder (ready to use, GeneRuler, ThermoScienti c) was used as molecular marker. Sterile ultrapure water was used as negative control and reference strain 16M, Rev1, reference strain 544 or RB51 were used as positive controls. Bands were visualized by staining with GelRed, Biotium. The electrophoresis equipment was set to run at 110 V for 40 min. Finally, the gel was visualized under UV light and the bands were observed and recorded. Initially, all samples were tested with primers 428f/428r in order to distinguish those which were positive in Brucella spp. as screening method. Then, the positive samples of sheep and goats were tested with the pair 752f/752r in order to discriminate Rev1 strain. The positive samples of cattle were tested with 843f/844r in order to distinguish B. melitensis from B. abortus, and once negative they were further tested with 998f/997r in order to distinguish B. abortus from RB51 (positive in case of B.abortus) ( Table 1).

PCR primers
Footnote: [a] ATCC: American Type Culture Collection

Results
In total 264 samples were examined derived from 191 animals. Thirty-nine (39) out of 264 derived from 30 animals were found PCR positive (Tables 2, 3,

& 5).
In the rst group of the 83 seropositive male animals, the results were as follows: 13 out of 121 samples derived from 12 seropositive animals (2 rams, 5 billy goats and 5 bulls) were found positive. In detail, in 1 ram B. melitensis and in another ram Rev1 strain were detected respectively, out of the total of 47 seropositive rams (4,26%). In 2 billy goats was found B. melitensis and in other 3 ones both B. melitensis and Rev1 strain were detected respectively, out of the 29 seropositive billy goats (17,24%) examined, Finally, 5 bulls were found positive to B. abortus out of the 7 seropositive ones (71,43%) (Tables  2 and 3).

Discussion
Brucellosis is a severe zoonosis, that affects many animal species, with most important the domestic ruminants, such as sheep, goats and cattle. Both The results of our study using the above PCR protocol indicates that detection of Brucella spp. DNA directly from testicles and spleen of seropositive male animals was not a convenient procedure confronting with the results derived from serology tests (RBT/CFT) in the rst group of 83 seropositive animals. In fact, only 14,45% of the seropositive animals were found positive in PCR analysis. Testes and spleen as biological samples-target for the detection of brucellosis via PCR is mostly rare. According to the literature, the most frequently tested tissues from male animals in order to detect Brucella spp. are serum, semen, nasal secretions, feces, lymph nodes and whole blood (Bricker, 2002, Ilhan, et al., 2008, Yu, et al., 2010.
Respectively, in the second group of 108 suspected animals, 16.67% of the female animals and fetuses were found positive in PCR analysis. The relatively few Brucella PCR positive samples from female ruminants with clinical symptoms from the reproductive system in the present study might also suggest the existence of other causes of abortion than brucellosis and it warrants further studies .
PCR-based tests are proving to be faster and more sensitive than conventional methods. Several PCR assays have been described for the detection of Brucella DNA using pure cultures or animal or human clinical samples. However, the sensitivity and speci city of the PCR for Brucella spp. varies between laboratories, standardization of sample preparation, PCR inhibitors on tissues, target genes, selected primers and clinical status of the animal (acute/chronic) (Amin, et al., 2001, Tekle, et al., 2019. The challenge lies in simplifying the procedure for detecting Brucella spp. at the species level in specialized laboratories (Navarro, et al., 2004, Ghorbani, et al., 2012.
Τhe modi ed PCR protocol used in our study was time consuming in terms of samples preparation resulting to a minimum of three days for the results to be delivered while the cost was much higher than the one required for serology.
An ideal diagnostic method should be speci c, sensitive, rapid, not expensive and easy to perform (Geresu, et al., 2016). According to the presented study, it seems that for the time being this PCR protocol described does not ful l all these requirements. Finally, the purchase of reagents is fairly expensive compared with RBT/CFT, but cost may decrease as more PCR diagnostic assays are developed.
In both groups, the prevalence of the Brucella strains detected in samples derived from goats ( 12,5% and 50%) and bovines ( 71,43% and 25,56%) was much higher than the prevalence of the same detected bacterium in sheep ( 3,03% and 8,22%). Actually, goats are approximately four times more likely to be infected than sheep (Sintayehu, et al., 2015). Generally, sheep are less susceptible to Brucella infection than goats, and this could be partly explained because sheep excrete the bacteria for shorter periods. This may reduce the potential for spread of the disease within and between sheep ocks (Radostits, et al., 2007).
In the rst group, preferred tissue for detecting the bacterium in rams and billy goats is testicle (8,7% and 19,23%) instead of spleen (0% and 4,76%), even if it is commonly known that spleen is the most heavily colonized organ, during brucellosis (Silva, et al., 2011). The stage of infection may very well in uence the number and location of Brucella organisms in white blood cells and lymphoid tissue glands (Gupta, et al., 2006). Concerning bulls, it seems that inguinal lymph nodes give encouraging results (71%), while no other kind of tissues were provided in this study.
The fact that vaccine strain Rev1 was detected in non-vaccinated rams and billy goats, indicates that possibly errors during the vaccine administration in females may occurred resulting in accidental vaccination of the male animals which were found seropositive during the surveillance program. Finally, due to the above vaccine administration errors male animals were slaughtered while the farm was quarantined for a period of at least 2 to 6 weeks. The development of a rapid and cost-effective laboratory method that will discriminate the vaccine antibodies from those caused by wild-type strains is of paramount importance. Alternatively, the development of a new vaccine, like the RB51 strain (Dorneles, et al., 2015), which is implemented in cattle and the produced antibodies are not detectable by RBT/CFT may solve this inconvenience equally.
Among female small ruminants and cattle from the second group which aborted, we suggest as preferred tissues to examine and reach to a positive PCR result tissues from fetal liver or/and stomach content from fetuses than other fetal tissues (fetal membranes, fetal spleens, placenta and posterior mammary lymph nodes).

Conclusions
Therefore, the use of the suggested PCR protocol in our study can play an important role in terms of con rmation before the suspected female animals Con icts of Interest: The authors declare no con ict of interest.