Latex agglutination using the periplasmic proteins antigen of Brucella melitensis is a successful,rapid,and specific serodiagnostic test for ovine brucellosis

Animals and sampling

This study was conducted between February 2013 and January 2015 in nine sheep flocks located in the central and western regions of the Kingdom of Saudi Arabia. The western region was historically designated as areas of concern for ruminant brucellosis and five sheep flocks from Turrabah and Taif governorates were selected. The naturally infected animals from these flocks were used as positive reference populations. Four flocks from the center region, located in Riyadh governorate, were historically Brucella-free and used as negative reference populations. Consequently, this population provided information about test specificity. All treatments of the animals and the collection of samples were in compliance with the guidelines for the welfare of animals and those of the concerned ethical authorities.

The flocks were subjected to careful clinical and laboratory investigations for application of the inclusion and exclusion criteria. The inclusion criteria included: (1) flocks that have never been vaccinated and were considered as free or not free of infection, so any positive reaction in the serological tests was considered to be due to infection with a field strain of Brucella; (2) animals with a history of abortion or stillbirth; (3) a group of clinically healthy animals to be used as a control in bacteriological examination. The exclusion criteria included: (1) flocks vaccinated with any type of Brucella vaccines; and (2) animals with surgical interference. All animals of the selected herds were sampled. A total of 1777 blood samples were collected from sheep of both sexes (females, n = 1420; males, n = 357) at different ages, that represented the applied criteria in the western (n = 1130) and central (n = 647) regions. However, in some instances, access to animals and flocks was difficult or restricted resulting in small sample sizes in some areas. The sample sizes according to governorates were as follows: Turrabah (n = 630), Taif (n = 500), and Riyadh (n = 647). Approximately 3 mL of blood was collected from each animal by jugular venipuncture with disposable needles and venoject tubes. The tubes were labeled and the blood was allowed to clot at ambient temperature before transport to the laboratory on ice. The sera were separated by centrifuging at 5000×g for 7 min. Serum samples were heated at 56°C for inactivation of non-specific antibodies. Each sample was identified by a code number and stored in aliquots at −20°C until used for Bayesian and test agreement analysis in all the five tests (LAT, RBT, BPAT, SAT, and i-ELISA). Sera from both Brucella-infected and Brucella-free individuals act as positive and negative reference samples. A pretested questionnaire designed to collect animal and flock level data during blood sampling was administered. Whole blood (3 mL), milk samples (20 mL), and vaginal swabs (after abortion) were collected from all animals (n = 150) in flocks with clinical infection in the western region and submitted to bacteriological study. All infected tissues, cultures, and potentially contaminated materials handled at containment level 3.

Bacteriological examination

Infection confirmed individually in all cases by culture on Brucella selective media (Oxoid) and the pathogen isolated from whole blood, milk samples, and vaginal swabs. ¹⁰ The species and biovars of Brucella were identified by cultural and serological criteria as previously described. ¹

Diagnostic serological tests

We evaluated the SBPPs antigen in LAT by comparison with four other validated well established serological tests (RBT, BPAT, SAT, and i-ELISA). All tests were carried out in accordance with the Office International des Epizooties (OIE) Manual of Standards for Diagnostic Tests. ¹ Tests were performed by personnel blind to the results of the other tests.

LAT

The SBPPs were prepared as previously described ¹¹ with some modifications. Five milliliters of B. melitensis serotype 3 suspension (10¹⁰/mL) were added to 800 mL of sterile Trypticase soy broth (BBL Microbiology Systems, Cockeysville, MD, USA) and incubated at 37°C in a Co₂ incubator for 48 h. Brucella melitensis was harvested and washed once with normal saline. Hot saline extracts were obtained by suspending B. melitensis in normal saline and autoclaving at 121°C for 20 min. The autoclaved suspension was centrifuged at 12,000×g for 20 min at 4°C. Then the ammonium sulphate was added to supernatant and the solution was centrifuged at 10,000×g for 20 min. The precipitate was dissolved in 0.01 M phosphate buffered saline (PBS) (pH 7.2). Dialysis occurred against PBS (pH 7.2) in a dialysis bag overnight at 4°C. This preparation was designated SBPPs. The protein concentration was determined by a commercial kit using bovine serum albumin as standard (Pierce, Rockford, IL, USA). The SDS-PAGE was done for determining the different pattern of protein bands in SBPPs. The SBPPs were stored at −80°C until use.

The procedure for the sensitization of latex microspheres was a slight modification of a published method. ¹² A latex polystyrene microsphere particle with a diameter of 0.81 µm (LB8, Sigma) was used in the covalent binding reaction. Prior to use, the beads were suspended in deionized water and vigorously vortexed to ensure even distribution, to break up any large particles, and to obtain a 10% suspension. An equal volume of latex particles and SBPPs antigen (0.6 mg protein/mL) was incubated over night at 37°C. The mixture was centrifuged at 6000×g for 10 min. The supernatant was removed and the pellet was washed and brought to an original volume in glycine sodium chloride buffer (final pH 8.2) containing 1:500 bovine serum albumin to keep the latex suspended (7.505 g of glycine, 5.85 g of sodium chloride, and 0.1% sodium azide in 1 L of deionized water, adjust the pH to 8.2 with 1 N NaOH). The latex particles were colored by methylene blue (0.005%), and the latex suspension was kept at 4°C.

Agglutination reactions were performed using the following procedure. A drop of the sensitized latex solution and sample was placed on a glass slide. The reaction was mixed by a side-to-side movement. The agglutination reaction appears in less than 2 min at room temperature in positive cases. ⁹

i-ELISA

LPS from B. melitensis serotype 3 was extracted ¹ by heating 5 g dry weight (or 50 g wet weight) of cells suspended in 170 mL distilled water to 66°C followed by the addition of 190 mL of 90% (v/v) phenol at 66°C. The mixture was stirred continuously at 66°C for 15 min, cooled, and centrifuged at 10000 g for 15 min at 4°C. The brownish phenol in the bottom layer was removed with a long cannula and large cell debris was removed by filtration. The sLPS was precipitated by the addition of 500 mL cold methanol containing 5 mL methanol saturated with sodium acetate. After a 2-h incubation at 4°C, the precipitate was removed by centrifugation at 10,000 g for 10 min. The precipitate was stirred with 80 mL of distilled water for 18 h and centrifuged at 10,000 g for 10 min. The supernatant solution was kept at 4°C. The precipitate was resuspended in 80 mL distilled water and stirred for an additional 2 h at 4°C. The supernatant solution was recovered by centrifugation as above and pooled with the previously recovered supernatant. Next, 8 g of trichloroacetic acid was added to the 160 mL of crude sLPS. After stirring for 10 min, the precipitate was removed by centrifugation and the translucent supernatant solution was dialyzed against distilled water (two changes of at least 4000 mL each) and then freeze-dried. The freeze-dried sLPS was weighed and reconstituted to 1 mg/mL in 0.05 M carbonate buffer, pH 9.6, and sonicated in an ice bath using approximately 6 watts three times for 1 min each. The sLPS was then freeze-dried in 1 mL amounts and stored at room temperature. The freeze-dried sLPS was reconstituted to 1 mL with distilled water and was further diluted 1/1000 (or to a dilution predetermined by titration against the OIE ELISA Standard Sera) in 0.05 M carbonate buffer, pH 9.6.

The test procedure was performed as previously described. ¹ The results were recorded on an ELISA reader (Reader 270, BioMerieux, France) at 490 nm. The color development is directly proportional to the bound antigen-antibody complex. Samples with a value double the optical density of the mean negative control were considered positive.

BPAT

Antigen for the BPAT is prepared from B. melitensis according to the procedure previously described. ¹ Serum samples (80 µL of each) was placed on a glass plate marked in 4 × 4 cm squares. Then, 30 µL of antigen was placed near each serum spot and the serum and antigen was thoroughly mixed to produce a circular zone approximately 3 cm in diameter. The plate was rotated three times in a tilting motion to ensure even dispersion of the reagents, and then incubated for 4 min in a humid chamber at ambient temperature. The plate was removed and rotated as above, and then returned for a second 4-min incubation. Agglutination was read immediately after the 8-min period is completed. Any visible reaction was considered to be positive. A control serum that gives a minimum positive reaction was tested before each day’s tests to verify the sensitivity of test conditions.

RBT

This test is a simple spot agglutination test using antigen stained with rose Bengal and buffered to a low pH, usually 3.65 ± 0.05. Serum samples (25–30 µL of each) were placed on a WHO hemagglutination plate. Equal volume of antigen was placed near serum spots. The serum and antigen were mixed thoroughly to produce an oval zone approximately 2 cm in diameter. The mixture was agitated gently for 4 min at ambient temperature on a rocker. The agglutination was read after the 4-min period. Any visible reaction was considered to be positive. A control serum that gives a minimum positive reaction was tested before each day’s tests to verify the sensitivity of test conditions.

SAT

The antigen represents a bacterial suspension in phenol saline (NaCl 0.85% [w/v] and phenol at 0.5% [v/v]). The ethylene diamine tetraacetic acid was added to the antigen suspension to 5 mM final test dilution to reduce the level of FP results. Subsequently, the pH of 7.2 was readjusted in the antigen suspension. The sensitivity of the antigen was standardized in relation to the OIE. Serum agglutination test was performed in microtiter plates. ¹ The mixture of antigen and serum dilutions was incubated for 6 h at 37°C. At least three dilutions were prepared for each serum in order to refute prozone negative responders. The degree of Brucella agglutination in a serum was expressed in IU per mL. Samples showing more than 30 IU/mL were considered positive.

Diagnostic test evaluation and validation

The performance and validation of the various tests was analyzed using classical means of calculation with respect to the test agreement and bacteriological culture.^10,13 With respect to the test agreement, ¹³ sera from both Brucella-infected and Brucella-free individuals act as positive and negative reference samples. The test sample was considered a true positive if a positive result was given by at least three tests. Similarly, a sample was considered a true negative if a negative result was given by at least three tests. The number of true-positive (TP) and false-negative (FN) cases in the diseased group were counted. The number of false-positive (FP) and true-negative (TN) cases in the non-diseased group were counted. From these counted number of cases, we calculated sensitivity (Se), specificity (Sp), positive likelihood ratio (PLR), negative likelihood ratio (NLR), positive predictive value (PPV), negative predictive value (NPV), and disease prevalence (DP) using the following equations: Se=TP/(TP+FN); Sp=TN/(TN+FP); PPV=TP/(TP+FP); NPV=TN/(TN+FN); PLR=Se/(100-Sp); NLR=(100-Se)/Sp; and DP=(TP+FN)/(TP+FP+FN+TN). Sensitivity means probability that a test result will be positive when the disease is present. It is also called the TP rate. Specificity means probability that a test result will be negative when the disease is not present. It is also called the TN rate. PLR is the ratio between the probability of a positive test result given the presence of the disease and the probability of a positive test result given the absence of the disease. NLR is the ratio between the probability of a negative test result given the presence of the disease and the probability of a negative test result given the absence of the disease. PPV means the probability that the disease is present when the test is positive. NPV means the probability that the disease is not present when the test is negative. Sensitivity, specificity, PPV, and NPV as well as DP are expressed as percentages for ease of interpretation.

With respect to bacteriological culture, ¹⁰ the infected animals that gave positive results in bacteriological culture were compared to their serological results in the selected five tests to calculate the sensitivity of each test. Therefore, sensitivity with reference to bacteriological culture was calculated by dividing the number of animals showing positive results in the serological tests with the number of the same animals showing positive results in the bacteriological test. The non-infected animals who gave negative results in the bacteriological culture were compared to their serological results in the five selected tests to calculate specificity of each test. Therefore, specificity with reference to bacteriological culture was calculated by dividing the number of animals showing negative results in the serological tests with the number of the same animals showing negative results in the bacteriological tests.

Statistical analysis

SPSS 17 statistical software (SPSS v. 17.0, SPSS Inc., Chicago, IL, USA.) was used to analyze the collected data. Comparisons among serological tests were performed using the χ2 (Chi-square) test. A difference was considered to be significant when P ⩽0.05.

Clinical picture

Some clinical signs appeared in naturally infected sheep (n = 150) in the Turrabah and Taif governorates including reduced appetite, abortion, retained placenta, or stillbirth.

General serological results

A total of 1777 serum samples were tested for the presence of B. melitensis antibodies using RBT, BPAT, SAT, i-ELISA, and LAT. An example of the LAT assay is shown in Figure 1. The number of animals that were positive or negative in the various serological tests and had or did not have the disease is shown in Table 1, which is based on the two-by-two contingency table. The number of total positive sera were significantly lower than the total negative sera for all five serological tests. The false positive and negative results were significantly (P ⩽0.05) lower in LAT than RBT, BPAT, SAT, and i-ELISA. Additionally, the FN result was significantly (P ⩽0.05) lower in i-ELISA than RBT. No significant differences in the FP results were detected between i-ELISA, SAT, BPAT, and RBT. The cross-classified test results were used to calculate the indices of agreement between the RBT, BPAT, SAT, i-ELISA, and LAT. There was agreement among the five tests in 1.724 cases (1.588 negative and 136 positive), with some disagreement in 53 cases.

OPEN IN VIEWER

Figure 1.

Latex agglutination test using the SBPPs antigen. A clear agglutination reaction appears in less than 2 min at room temperature in the case of a positive control serum (a), with no agglutination in the case of a negative control serum (b), and latex SBPPs alone (c).

OPEN IN VIEWER

Table 1.

Evaluation of RBT, BPAT, SAT, i-ELISA, and LAT for diagnosis of ovine brucellosis.

Variable	RBT			BPAT			SAT			i-ELISA			LAT
	D+	D–	Total	D+	D–	Total	D+	D–	Total	D+	D–	Total	D+	D–	Total
T+	144	19*	163	146	18*	164	146	20*	166	147	19*	166	149	2 †	151
T-	6*	1608	1614	4*	1609	1613	4*	1607	1611	3*	1608	1611	1 †	1625	1626
Total	150	1627	1777	150	1627	1777	150	1627	1777	150	1627	1777	150	1627	1777
Se %	96 †			97.33 †			97.33 †			98* †			99.33*
Sp %	98.83 †			98.89 †			98.77 †			98.83 †			99.88*
PPV %	88.34 †			89.02 †			87.95 †			88.55 †			98.68*
NPV %	99.63			99.75			99.75			99.81			99.94
PLR %	82.05 †			87.69 †			79.13 †			83.76 †			827.75*
NLR %	0.04			0.03			0.03			0.02			0.01
DP %	8.44			8.44			8.44			8.44			8.44
Culture Se %	81.84* †			77.77* †			70.37 †			92.59*			88.88*
Culture Sp %	86.67			86.67			80			90			93.34

Numbers carry different superscript in row 3 represent the false-positive results and those in row 4 represent the false-negative results. Numbers carry different superscript within the same row were significantly differed (P ⩽0.05).

BPAT, buffered plate agglutination test; D+, Animal has the disease; D–, Animal does not have the disease; DP, disease prevalence; i-ELISA, indirect enzyme-linked immunosorbent assay; LAT, latex agglutination test; NLR, negative likelihood; NPV, negative predictive value; PLR, positive likelihood; P(T+ | D+), True positive rate; P(T–| D–), True negative rate; PPV, positive predictive value; RBT, Rose Bengal test; SAT, serum agglutination test; Se, sensitivity; Sp, specificity; T+, Test positive; T–, Test negative.

Diagnostic test performance and validation results

Test agreement results. The estimated values of Se, Sp, PPV, NPV, PLR, NLR, and DP that based on the test agreement analysis are shown in Table 1. For diagnostic Se and Sp, LAT was significantly (P ⩽0.05) higher than i-ELISA, SAT, BPAT, and RBT. Additionally, the diagnostic specificity did not differ for i-ELISA, SAT, BPAT, and RBT. The PLR was significantly (P ⩽0.05) higher in LAT than other tests. The NLR was not significantly different between all the tests. The PPV was significantly (P ⩽0.05) higher in LAT than BPAT, i-ELISA, RBT, and SAT. No significant difference in the NPV existed for all five serological tests.

Bacteriological culture results

Natural infections were confirmed by the isolation and identification of B. melitensis serotype 3 from 27 of 150 sheep (18%). The 27 bacterial isolates were recovered from seropositive whole blood (n = 7), milk (n = 7), and vaginal swabs (13). The 27 isolates were identified as B. melitensis serotype 3. The 30 seronegative sheep were negative by cultue and serology. The sensitivity and specificity of the serological tests with respect to culture are illustrated in Table 1. The LAT and i-ELISA tests showed a significant difference in sensitivity with SAT. However, no significant difference in specificity was detected.