Introduction
Brucellosis, a globally distributed zoonotic disease caused by Brucella species, presents significant socioeconomic and public health challenges to humanity [1,2]. Narok County has a substantial livestock population [3]. In Narok County, livestock are raised in both pastoral systems characterized by large herds of cattle and small ruminant flocks, and agro-pastoral systems in which livestock herds and flocks are usually in the range of 2–10 animals. In both production systems, natural breeding is practiced whereby one bull or ram is shared by all the females in the herd or flock. In this country, the success of dairy and beef value chains is threatened by diseases such as brucellosis. This county also has well-established wildlife conservancies including the famous Maasai Mara Game Reserve and increased mixed farming (agro-pastoral) in areas further away from the game reserves [4]. Moreover, some of the areas with high cultural and economic dependence on livestock are close to wildlife-protected areas. During the dry season, livestock and wildlife share pastures in some of these areas. The county has no brucellosis control program and vaccination against brucellosis has not been carried out. Moreover, knowledge of the epidemiology of these pathogens is limited in livestock, wildlife, and human populations in this county due to a lack of prioritization, poor surveillance systems, and diagnostic capacities [4].
Previous studies on brucellosis in Narok County have reported variable seroprevalences in both humans (28%) and livestock (12.44%–36.9%) [5,6,7]. Since these studies were undertaken in a few regions of Narok County, the reported seroprevalence rates may not be a true reflection of the situation in the whole county but do indicate the presence of brucellosis in livestock of this county. This study aimed to fill this gap by conducting a thorough assessment of the serological and molecular prevalence of brucellosis among livestock populations and its geographical distribution within Narok County, Kenya.
Materials and Methods
Study area and design
A cross-sectional study was conducted in 13 administrative wards within 5 sub-counties of Narok County (Fig. 1) from December 2019 to December 2022. 49 flocks (25 ovine and 24 caprine) and 31 bovine herds were randomly selected with guidance from the County Veterinary Staff.
Figure 1.
Narok County map (A) showing study sites (sub-counties and Wards-red stars).
Sample size calculation
The sample size was calculated according to [8] formula, considering brucellosis prevalence ranges in Kenya [9].
Sample size, n=[Z2 × p × (1-p)] /d2
where n=sample size
d=Precision of prevalence
p=Expected prevalence
Z=1.96 (confidence interval)
Therefore, the calculated minimum sample sizes were: 215 bovine, 182 ovine, and 247 caprine.
Inclusion criteria
Households with bovine herds and small ruminant flocks were included, with herds and flocks randomly selected. All adult animals were tested in small herds or flocks (≤ 10 animals); in herd/flocks with more than 10 animals, at least 20 animals were randomly tested, prioritizing females that had given birth. All animals tested had not been vaccinated. A total of 762 livestock (254 bovines, 250 ovines, and 258 caprines) were used in the study (Table 1).
Table 1.
Sample size by livestock species in Narok County.
| Sub-County | Ward | Animal species | ||
|---|---|---|---|---|
| Bovine | Ovine | Caprine | ||
| Narok West | Siana | 6 | 0 | 10 |
| Mara | 65 | 0 | 0 | |
| Narok South | Ololulunga | 22 | 30 | 20 |
| Majimoto/Naroosura | 29 | 11 | 9 | |
| Narok East | Mosiro | 38 | 10 | 5 |
| Suswa | 13 | 5 | 10 | |
| Narok North | Narok Town | 21 | 54 | 56 |
| Olorropil (Kasiriri) | 7 | 27 | 5 | |
| Nkareta | 0 | 2 | 4 | |
| Melili | 2 | 6 | 0 | |
| Transmara West | Lolgorian Central | 12 | 65 | 66 |
| Kilgoris Central | 13 | 21 | 42 | |
| Kimintet | 26 | 17 | 27 | |
| Total | 254 | 250 | 258 | |
Blood sample collection
Certified veterinary professionals and skilled laboratory technicians from the University of Nairobi and County Satellite Laboratories collected blood samples. Trained animal handlers assisted during sampling. Five to ten milliliters of blood was drawn from the jugular vein of each animal using aseptic methods and placed in 10-milliliter plain vacutainers. Before testing for Brucella antibodies, serum samples were stored in 2 ml cryovials at −20°C, at the Microbiology Laboratory at the University of Nairobi. They were then moved to a −80°C freezer for long-term storage.
Serological Tests
Rose bengal plate test (RBPT)
This test was conducted for bovine samples as per the OIE manual [10]. A modified RBPT method [11] was used for ovine and caprine samples. Any amount of agglutination was scored positive regardless of the degree. All plates showing no evidence of agglutination were scored negative. Every plate test included a positive and negative serum sample as controls.
Indirect enzyme-linked immunosorbent assay (i-ELISA)
Brucellosis i-ELISA kits (Elabscience® Biotechnology Inc., 2018-2019) were used following the manufacturer’s instructions. Results were determined on the basis of optical density (OD) at 450 nm of both test and control sera. Positive and negative control sera were supplied together with the other reagents in the ELISA kit. An OD ≥ 0.38 indicated a positive result, while an OD < 0.38 indicated a negative result.
Competitive ELISA (c-ELISA)
This test was conducted using COMPELISA kits (APHA Scientific) for all samples. A plate was considered valid if the mean OD of the 6 negative controls at 450 nm was >0.700 and the mean OD of the 6 positive controls was <0.100. The difference between the positive and negative controls’ OD had to be ≥0.300. The cut-off was set at 60% of the mean OD of the four conjugate control wells [12]. OD values less or equal to the cut-off were taken as positive, while values above were taken as negative. Positive and negative control sera were supplied together with the COMPELISA kit.
DNA extraction and purification
Genomic DNA from 184 serum samples comprising 104 bovine, 43 ovine, and 37 caprine (86 seropositive and 98 seronegative animals) was extracted using QIAamp® DNA Blood Mini Kit (Qiagen, Germany) according to the manufacturer’s guidelines. DNA quality and quantity were checked using a NanoDrop™ Spectrophotometer.
Conventional PCR for Brucella genus
This was conducted using B4/B5 primers (B4- 5’-TGGCTCGGTTGCCAATATCAA-3’ and B5- 5’-CGCGCTTGCCTTTCAGGTCTG-3’), targeting the bcsp31 gene as previously described [13]. PCR reactions were performed in a Veriti 96 wells Thermal Cycler (Applied Biosystems) with the following conditions: initial denaturation at 95°C for 5 minutes, 40 cycles of amplification (denaturation at 95°C for 30 seconds, annealing at 60°C for 30 seconds, and elongation at 72°C for 60 seconds), and final elongation at 72°C for 10 minutes. The amplified product (223 bp) was electrophoresed in 2% agarose gel stained with ethidium bromide and imaged using a UVP GelMax 125 Imager. Escherichia coli ATCC 25922 DNA from the stock cultures of the Department of Veterinary Pathology, Microbiology and Parasitology, University of Nairobi, was used as a negative control, whereas Brucella abortus used as a positive control was donated by Dr. James Akoko of ILRI Laboratories, Kenya.
Species-specific PCR
Detection of Brucella species was done using UF1/UR1 species-specific primers (UF1: 5’-GGCTATCGGCTGGGAAAGG-3’ and UR1: 5’-CCTTCCGAAAAGTCCCCC-3’) as previously described [14]. PCR reactions were performed under the following conditions: initial denaturation at 95°C for 5 minutes, 40 cycles of amplification (denaturation at 94°C for 30 seconds, annealing at 52°C for 30 seconds, and elongation at 72°C for 45 seconds), and a final elongation at 72°C for 10 minutes.
Gel electrophoresis of amplified Brucella DNA
PCR-amplified DNA samples plus a 100-bp DNA ladder were electrophoresed in 2% agarose gel at 100 V for 60 minutes. Gels were imaged using a UVP GelMax 125 Imager and results were interpreted by comparing the presence and size of the PCR products with the control. The control used was B. abortus DNA donated by Dr. James Akoko of ILRI Laboratories, Kenya.
Statistical analysis
Data from serological and molecular tests were entered into a spreadsheet (Microsoft Excel 2010) and analyzed using chi-square tests and logistic regression with STATA® version 16 and Epitools applications [15]. Sensitivity, specificity, concordance percentage, and agreement (kappa statistic) between tests were assessed [16,17]. County, sub-county, individual animal species, herd/flock, and test prevalence rates were calculated based on serological and PCR analysis results.
Results
The results indicated varying positivity rates across different animal species and serological tests (Table 2).
Table 2.
Brucellosis seropositivity by the 3 serological tests in livestock of Narok County.
| Test Category | Number positive | % positive | ||||||
|---|---|---|---|---|---|---|---|---|
| Bovine (n=254) |
Ovine (n=250) |
Caprine (n=258) |
All 3 species | Bovine | Ovine | Caprine | All 3 species | |
| RBPT overall | 54 | 0 | 1 | 55 | 21.3 | 0 | 0.4 | 7.23 |
| c-ELISA overall | 72 | 6 | 1 | 79 | 28.35 | 2.4 | 0.4 | 10.37 |
| i-ELISA overall | 65 | 0 | 0 | 65 | 25.6 | 0 | 0 | 8.53 |
| RBPT plus i-ELISA | 8 | 0 | 0 | 8 | 3.15 | 0 | 0 | 1.05 |
| RBPT plus c-ELISA | 5 | 0 | 0 | 5 | 1.97 | 0 | 0 | 0.66 |
| i-ELISA plus c-ELISA | 7 | 0 | 0 | 7 | 2.76 | 0 | 0 | 0.92 |
| RBPT, i-ELISA plus c-ELISA | 29 | 0 | 0 | 29 | 11.42 | 0 | 0 | 3.81 |
| RBPT only | 13 | 0 | 1 | 14 | 5.12 | 0 | 0.4 | 1.84 |
| i-ELISA only | 22 | 0 | 0 | 22 | 6.66 | 0 | 0 | 2.9 |
| c-ELISA only | 32 | 6 | 1 | 39 | 12.6 | 2.4 | 0.4 | 5.12 |
| Overall seropositivity | 115 | 6 | 2 | 123 | 45.3 | 2.4 | 0.78 | 16.14 |
Individual test positivity rates
RBPT identified brucellosis in 21.3% of bovines (54/254), with no positive cases in ovines and a single positive case in caprines (0.4%) (Table 2). The overall positivity rate across all species was 7.23%; c-ELISA had a higher detection rate, with 28.35% positivity in bovines (72/254), 2.4% in ovines (6/250), and 0.4% in caprines (1/258). The total positivity across all species was 10.37%, and i-ELISA detected 25.6% of bovines as positive (65/254) but did not identify any positive cases in ovines or caprines. The overall positivity rate was 8.53% (Table 2).
Combined test positivity rates
RBPT and i-ELISA together identified 3.15% of bovines (8/254) as positive, with no positive cases in ovines and caprines. The combined positivity rate for all species was 1.05%; RBPT and c-ELISA detected 1.97% positivity in bovines (5/254), with no positive cases in other species. The overall positivity was 0.66%; and i-ELISA and c-ELISA together identified 2.76% of bovines (7/254) as positive, with no positive cases in ovines or caprines. The overall positivity rate across all species was 0.92%. All three tests (RBPT, i-ELISA, and c-ELISA) identified 11.42% of bovines (29/254) as positive, with no positive cases in ovines or caprines. The total positivity rate across all species was 3.81% (Table 2).
Single test positivity rates
RBPT only detected 5.12% of bovines (13 out of 254), with no positive cases in ovines and one positive case in caprines (0.4%). The overall positivity was 1.84%; i-ELISA only detected 6.66% of bovines (22 out of 254), with no positive cases in ovines or caprines. The overall positivity was 2.9%; and c-ELISA only had the highest single-test detection rate, identifying 12.6% of bovines (32 out of 254), 2.4% of ovines (6 out of 250), and 0.4% of caprines (1 out of 258). The overall positivity rate was 5.12%. The total or cumulative seropositivity across all tests showed that 45.3% of bovines (115/254), 2.4% of ovines (6/250), and 0.78% of caprines (2/258) were positive for brucellosis. The overall seropositivity across all species was 16.14% (Table 2).
Sensitivity, specificity, and accuracy of the three serological Tests
Sensitivities, specificities, and accuracy of the three serological tests were determined using either c-ELISA or i-ELISA as diagnostic reference tests. Test sensitivities were 47.54%–56.86% for RBPT, 47.54% for i-ELISA, and 56.86% for c-ELISA, while the specificities were 84.24%–93.6% for RBPT, 88.6% for i-ELISA, and 93.6% for c-ELISA. (Table 3).
Table 3.
Serological Test Specificity and Sensitivity, PPV, NPV, and Accuracy for livestock brucellosis (Narok County).
| Test | Gold standard Test | Sensitivity | Specificity | PPV | NPV | Accuracy |
|---|---|---|---|---|---|---|
| RBPT | c-ELISA | 47.54 (36.6–60.73) |
93.26 (88.86–96.37) |
85.39 (76.46–91.32) |
68.22 (62.76–73.22) |
72.15 (66.62–77.94) |
| RBPT | i-ELISA | 56.86 (42.25–70.65) |
93.6 (89.3–96.55) |
88.03 (80.49–92.91) |
72.38 65.61–78.21) |
76.96 (71.28–81.99) |
| i-ELISA | c-ELISA | 47.54 (34.60–60.73) |
88.60 (83.25–92.72) |
77.55 (68.27–84.72) |
67.10 (61.50–72.25) |
70.0 (63.96–75.57) |
| c-ELISA | i-ELISA | 56.86 (42.25–70.65) |
84.24 (78.48–88.06) |
74.92 (66.74–81.64) |
70.22 (63.11–76.47) |
71.84 (65.87–77.28) |
Herd level seropositivity of brucellosis
Herd-level seropositivirty for bovine herds was 90.3%, while it was 16.0% (n=4/25) for ovine, and 8.3% (n=2/24) for caprine flocks (Table 4). Bovine herds with a history of abortion had significantly higher seropositivity (53.85%; n=91/169) compared to those without (28.24%; n=24/85) (p=0.0001) (Table 5). All seropositive caprine flocks had a history of abortion, while seropositivity in ovine flocks was not associated with the history of abortion (Table 5).
Table 4.
Herd/Flock seroprevalence of livestock brucellosis in Narok County.
| Attribute | Bovine | Ovine | Caprine |
|---|---|---|---|
| No of Herds/Flocks | 31 | 25 | 24 |
| No of Herds/Flocks positive | 28 | 4 | 1 |
| No of Herds/flocks negative | 3 | 21 | 23 |
| % of Herds/Flocks positive | 90.3 | 28.3 | 4.17 |
Table 5.
Brucellosis seropositivity of livestock versus abortion history (Narok County).
| Attribute | Bovine | Ovine | Caprine |
|---|---|---|---|
| Number of Herds/Flocks with a history of abortion | 16 | 8 | 9 |
| Number of Herds/Flocks without a history of abortion | 15 | 17 | 15 |
| Number of Herds/Flocks positive in those with a history of abortion | 15 | 1 | 2 |
| Number of Herds/Flocks positive in those without a history of abortion | 13 | 3 | 0 |
| Number of animals in flocks with abortion history | 169 | 65 | 62 |
| Number of animals in flocks without abortion history | 85 | 185 | 196 |
| % positivity in Herds/Flocks with a history of abortion | 93.8 | 12.5 | 22.2 |
| % positivity in Herds/Flocks without a history of abortion | 86.7 | 17.65 | 0 |
| % of animals +ve in Herds/Flocks with Abortion History | 53.85 | 1.54 | 3.2 |
| % of animals +ve in Herds/Flocks without Abortion History | 28.24 | 0.54 | 0 |
Sub-county seropositivity
All 5 sub-counties of Narok had herds with seropositive bovines, and seroprevalence was highest in Narok South (58.3%; n=60), Narok East (56.9%; n=51), and Narok West (49.3%; n=71), and moderate in Narok North (23.8%; n=21) and Transmara West (21.6%; n=51) (Table 6). Small ruminant brucellosis was low (0–7.32%), with seropositive animals in all sub-counties except Narok West (Table 6).
Table 6.
Seroprevalence of livestock brucellosis in Sub-county wards of Narok County.
| Sub-county | Ward | Number of Samples | Number positive | % positive | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Bovine | Ovine | Caprine | Bovine | Ovine | Caprine | Bovine | Ovine | Caprine | All the 3 Animal species | ||
| Narok West | Siana | 6 | 0 | 10 | 3 | 0 | 0 | 50 | 0 | 0 | 18.8 |
| Mara | 65 | 30 | 20 | 32 | 3 | 0 | 49.2 | 10 | 0 | 47.83 | |
| Narok East | Mosiro | 38 | 11 | 9 | 24 | 0 | 1 | 63.2 | 0 | 11.1 | 43.1 |
| Suswa | 13 | 10 | 5 | 5 | 0 | 1 | 38.5 | 0 | 20 | 21.43 | |
| Narok South | Ololulunga | 22 | 5 | 10 | 18 | 0 | 0 | 58.1 | 0 | 0 | 39.13 |
| Majimoto/ Naroosura |
29 | 54 | 56 | 17 | 1 | 0 | 58.6 | 1.9 | 0 | 12.95 | |
| Narok North | Narok Town | 21 | 27 | 5 | 2 | 1 | 0 | 9.5 | 3.7 | 0 | 6.82 |
| Olorropil | 7 | 2 | 4 | 3 | 0 | 0 | 42.9 | 0 | 0 | 23.08 | |
| Melili | 2 | 6 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.0 | |
| Nkareta | 0 | 2 | 4 | 0 | 0 | 0 | 0 | 0 | 0 | 0.0 | |
| Transmara West | Lolgorian Central | 12 | 65 | 66 | 2 | 0 | 0 | 16.7 | 0 | 0 | 1.4 |
| Kilgoris Central | 13 | 21 | 42 | 1 | 0 | 0 | 7.7 | 0 | 0 | 1.32 | |
| Kimintet | 26 | 17 | 27 | 8 | 1 | 0 | 30.8 | 5.9 | 0 | 12.9 | |
| Total | 254 | 250 | 258 | 115 | 6 | 2 | 33.53 | 1.35 | 1.62 | 16.14 | |
Molecular prevalence of livestock brucellosis
The overall molecular prevalence of livestock brucellosis using the genus-specific bcsp31 primer (B4/B5) was 1.63% (n=3/184). Molecular prevalences at the animal species level were 1.92% (n=2/104) for bovines, 2.7% (n=1/37) for caprines, and 0.0% (n=0/43) for ovines (Table 7 and Fig. 2). The genus-specific bcsp31 (B4/B5) primer amplified a 223 bp DNA fragment (Fig. 2). The PCR-positive caprine was seronegative in all three serological tests and came from a flock with no history of abortion (Table 8). The two PCR-positive bovines were seropositive in all the 3 serological tests and belonged to herds with abortion history (Table 8. Brucella abortus was detected in these two bovines (Fig. 3).
Table 7.
Overall Brucella PCR results for livestock in Narok County.
| No of Samples analysed by PCR | No of samples seropositive | No of the samples seronegative | No of the samples were positive in Brucella PCR | % of samples positive in Brucella PCR | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Bovine | Ovine | Caprine | Bovine | Ovine | Caprine | Bovine | Ovine | Caprine | Bovine | Ovine | Caprine | Bovine | Ovine | Caprine |
| 104 | 43 | 37 | 78 | 6 | 2 | 26 | 37 | 35 | 2 | 0 | 1 | 1.92 | 0 | 2.7 |
Table 8.
Positivity profiles of livestock serum samples in the 3 serological tests and conventional Brucella PCR.
| Animal species | Herd/Flock/ Animal ID |
Ward | Sub county | Brucella PCR | ||
|---|---|---|---|---|---|---|
| Genus- specific (B4/B5)) |
B. abortus (UF1/UR1) |
B. melitensis (UF1/UR1) |
||||
| Bovine | KjLL B1 | Keekonyokie | Kajiado West | + | − | − |
| Bovine | WB6 | Mara | Narok West | + | + | − |
| Bovine | YB6 | Mara | Narok West | + | + | − |
| Ovine | KjUS4 | Matapato North | Kajiado Central | + | − | − |
| Caprine | TmIG1 | Lolgorian Central | Transmara West | + | − | − |
Figure 2.
Brucella DNA amplicon visualization in 2% agarose gel electrophoresis, following Genus-specific PCR using B4/B5 primers. Lane M: 100-bp DNA molecular marker (Bioline-Ladder); lane1-15 are DNA samples; Lane 4=one of the positive sample (YB6).
Figure 3.
Brucella abortus DNA visualization in 2% agarose gel electrophoresis following amplification with Species-specific UFI/UF2-PCR primers. Lane M: 100-bp DNA molecular marker; lane1: TmI-G1; lane 2: KJU-S4; lane 3: Brucella arbutus (positive control); lane 4: KJLLB; lane 5: WB6; lane 6: YB6; Lane 7 and 8: Negative Control.
