2.1. Preparation of B. melitensis B115 extracts
B. melitensis attenuated strain B115 was provided by the Veterinary Laboratories Agency (VLA) of Weybridge (U.K.) and cultured to prepare the bacterial extract according to previously described protocols [21; 22].
Briefly, the bacteria were cultured in 1 liter of Brucella broth (Becton Dickinson, France) at 37°C in aerobic conditions under stirring for 3 days. When the culture reached an optical density (OD) of 2.080, the broth was centrifuged at 5,000 g (ALC PK131R centrifuge, Milan, Italy) for 20 min. The pellet was washed with saline solution, inactivated at 100°C for 10 min, sonicated, centrifuged and the supernatant was dialysed against distilled water before measurement of the protein content as previously described by Corrente et al. 2015.
2.2. Herds and serum samples
One hundred eighty herds from the South of Italy (Apulia and Basilicata regions) were recruited over an extended period of time, i.e. from 2014 to 2018. All animals were screened by the IZS during official brucellosis survey. A total of 648 sera were collected and subjected to the official diagnostic assays (RBPT and CFT) before testing them with the B. melitensis B115-based ELISA. The sera were subdivided into 4 different groups.
Group A: serum samples, both RBPT and CFT negative (n=259), that were collected from 11 officially brucellosis-free herds;
Group B: serum samples from SR animals (n=150) tested positive in RBPT and negative in CFT; they were collected from 102 different officially brucellosis-free herds. These animals were slaughtered according to the local Regulations and the IZS tested for the absence of Brucella spp infection by using both bacteriological testing (PT/DIA/004) and real-time PCRs [9; 27]. In addition, fecal swabs were screened for Yersinia enterocolitica O:9 by official tests performed at the IZS (UNI EN ISO 10273:2017).
Group C: serum samples (n=134), that tested positive in both serological conventional tests and were collected from 50 officially brucellosis-free herds; although epidemiological data indicated that they could be FP, these animals were slaughtered to fulfill the local Regulations. As for Group B, the absence of Brucella spp. infection was tested post-mortem and the screening for Y. enterocolitica O:9 was performed by IZS as previously described.
Group D: serum samples (n=105) were both RBPT and CFT positive and were collected from 17 Brucella-infected herds. The animals were slaughtered and infection with Brucella spp. was confirmed by bacteriological and PCR analyses performed post-mortem by the IZS as described elsewhere [9; 27].
2.3. Serological tests
Serological conventional tests were performed by the IZS according to international standard procedures  while the ELISA was carried out in the Laboratory of Bacteriology at the University of Bari according to a previously described protocol with some modifications . Briefly, polysorp microtiter plates (Nunc, Milan, Italy) were coated with 100 μl of bacterial extract (25 μg of proteins/ ml) in carbonated buffer and incubated overnight at 4°C under gentle shaking. The plates were then washed four times with PBS containing 0.05% Tween 20 (PBS-T) and wells were blocked for 150 min at 37°C with 0.2% gelatin in carbonate buffer. After repeated washes, 100 μl of serum, diluted 1:100 in PBS-T, were added and the plates were incubated at 37°C for 120 min. After washings, a rabbit anti-bovine antibody labeled with peroxidase (Sigma Aldrich, Milan, Italy) was diluted 1:3000 in PBS-T and added to the plates which were then incubated for 60 min at 37°C. After final washings, an ABTS [2.2’-Azino-di-(3-ethylbenzothiazoline sulfonate)] solution (Sigma Aldrich, Milan, Italy) was added to each well and the plate was incubated at room temperature, without light for 20 min. The O.D. was measured at 405 nm using an automated ELISA reader. The negative samples (group A, n= 259) were used to determine the cut-off value of the ELISA test (i.e., the arithmetic mean of the O.D. of all negative samples plus 3 standard deviations).
The OIE international standard serum, supplied by the OIE Reference Laboratory for brucellosis at the VLA (Weybridge, UK), was used as positive control serum.
2.4. Data analysis
The Microsoft Excel® 2010 program was employed to evaluate: i) the arithmetic mean and the standard deviation of O.D. values within a single group of animals; ii) the comparison of O.D. values between different animal groups (by chi square test with Yates correction); iii) the linear regression analysis between ICFTU and O.D. values in Brucella-infected animals (Group D); the percentile (99%) of O.D. values obtained in uninfected animals (Group A).
A p value below 0.05 was considered significant.