Brucellosis is a zoonotic disease affecting human and animal species, including livestock, wild animals, and marine mammals. In animals, it causes abortion, mastitis, reproductive disorders, and reduced milk production. In human, brucellosis can cause various symptoms, from a mild flu-like illness to severe complications such as arthritis and endocarditis, making early detection and control critical. The disease is considered as a main public health concern due to its impact on livestock productivity and human health. ^{1– 3}

In Oman, brucellosis remains the most common zoonosis. The disease is endemic in Dhofar Governorate due to its humid climate, especially during the monsoon season. A common practice of keeping animals, such as cattle, sheep, and goats, in proximity significantly increases the chances of Brucella spillover. Moreover, human-to-animal contact increases the risk of transmission, necessitating robust surveillance and control measures. ^{4– 6}

Brucella is an intracellular Gram-negative coccobacillus bacterium. There are various Brucella species. The classic known six species are B. abortus, B. melitensis, B. ovis, B. canis, B. suis, and B. neotomae. ⁷ More species were recognized later like, B. ceti, B. pinnipedialis, B. microti, B. inopinata, B. papionis, and B. vulpis. ^{8– 10} Brucella melitensis has been widely reported in humans and livestock in Oman. ^{4, 6}

The Brucella genome encodes virulence factors and metabolic pathways crucial for intracellular survival. Brucella species show similarity higher than 90% at the genome level, making it difficult to differentiate between Brucella strains through conventional genetic analysis techniques. ^{11– 13}

Molecular methods are widely used for Brucella diagnosis such as conventional PCR, real-time PCR, multiplex PCR, multiple loci variable number tandem repeat analysis (MLVA), and single nucleotide polymorphism (SNP) analysis. Conventional PCR involves DNA amplification of a single target (singleplex PCR) or multiple targets (multiplex PCR). AMOS and Bruce-ladder are multiplex PCR methods that amplify multiple Brucella target genes in a single reaction. These assays use multiple sets of primers, each specific to a different region in the genome, to differentiate between Brucella species. ^{14– 17}

MLVA is widely used to identify Brucella genetic diversity. A tandem repeat is a sequence of two or more DNA base pairs repeated and directly adjacent to each other in the genome. These repeats can vary in length and number. MLVA can be used in various genetic studies, including genetic fingerprinting, studying genetic diversity, and identifying hereditary diseases. By analyzing tandem repeats within the genome, MLVA enables high-resolution differentiation of strains, and offering valuable insights into the genetic makeup of Brucella. ^{16, 18} The faster accumulation of genetic variation in tandem repeat markers compared to SNP-based variation, allows genotyping and distinguishing between closely related Brucella strains. ¹⁹ MLVA technique is cost-effective and serves as an excellent alternative to other sequencing methods. ¹⁶

No previous studies in Oman have used MLVA to analyze the genetic diversity of Brucella melitensis in both human and milk samples, limiting understanding of transmission and control. Given the zoonotic nature of brucellosis, accurate strain differentiation is essential for surveillance, treatment, and vaccine development. This study addresses this gap by investigating the genetic makeup of Brucella strains Brucella strains circulating in different geographical locations in Oman in both human and livestock using MLVA.

Peaks corresponding to all 16 markers were detected in all isolates (human and goat) by capillary electrophoresis. However, only 12 out of 42 milk samples showed peaks for all 16 markers. The remaining samples lacked at least one locus. Therefore, MLVA-14 was selected as a typing approach since it allowed the inclusion of samples missing up to two loci, while excluding those missing three or more loci. As a result, 27/42 milk samples were amplified for all 14 markers (Bruce8, Bruce11, Bruce43, Bruce45, Bruce21, Bruce6, Bruce42, Bruce18, Bruce4, Bruce9, Bruce16, Bruce30, Bruce19). No sample showed multiple alleles at any locus. MLVA-14 genotyping of 49 B. melitensis strains showed that Bruce8, Bruce11, Bruce43, Bruce45 and Bruce21 loci were homogenous. Conversely, the most discriminatory loci were Bruce6, Bruce42, Bruce18, Bruce4, Bruce9, Bruce16, Bruce30 and Bruce19 ( Table 1). Subsequent cluster analysis allowed to recognize two main B. melitensis clusters including 31 different MLVA-14 genotypes ( Figure 2). The distances between strains within and between clusters were calculated based on the number of matching and differing VNTRs. In Dhofar governorate, B. melitensis strains extracted from camel milk (K187, K189, K185, K188) and goat milk (K197) in 2023 shared the same genotype. Moreover, B. melitensis strains extracted from goat milk (K43, K44, K47, K48, K50) had the same genotype as strains recovered from cattle milk (K55, K56, K60). In addition, strains retrieved from camel milk (K102, K103, K104, K107) shared the same genotype as strains from cattle milk (K112, K113). On the other hand, two human isolates (K219, K220) dating 2024 had the same genotype; also, the other two human isolates (K212, K216), identified in the same year, shared a unique genotype. The rest of the strains showed different genotypes thus forming subclusters ( Figure 2).

Minimum spanning trees (MST) were generated for the 49 Omani strains based on host species and geographical location ( Figure 3). MST presents the genetic profiles, with circles representing individual or grouped strains and edges reflecting genetic distances. Circle colors indicate the host; green (human), red (goat), purple (camel), and yellow (cattle), while circle sizes correspond to the number of isolates sharing an identical genetic profile. Human isolates (green) were related, suggesting a shared genetic background. Brucella strains extracted from goat milk (red) showed distinct clustering, with some overlap with strains extracted from camel milk (purple) and cattle milk (yellow). For geographic annotations, most strains originated from Dhofar Governorate, two human isolates from AD Dakheliya Governorate and a goat isolate from Al Jabal Al Akhdhar which located in AD Dakheliya Governorate. The phylogeographic patterns of the studied Brucella strains were compared to MLVA profiles available in the international database ( Figure 4 & Figure 5). It should be noted that for MST worldwide analysis, Bruce19 locus was not included since many foreign isolates were not typed for this marker and to avoid numbering uncertainty due to the recent discovery of rare alleles at the Bruce19 locus that changed the nomenclature. ²⁵ The Omani strains belong to the East Mediterranean lineage. In the MST analysis with worldwide isolates, most Omani strains tend to separate forming a country branch, which includes two strains from the United Arab Emirates (in brown). However, another group of Oman strains clusters with Spain, Portugal, India, Turkey, and China strains. These two Omani clusters are connected to a junction node that includes isolates from Portugal, China, France and Kazakhstan. Finally, it can be noticed one “outlier” Oman strain correlated with a big cluster mostly formed by isolates from China, but also including strains from Portugal, Kazakhstan, Turkey, Mongolia, Marocco, France and Spain.

This study provides new insights into the genetic diversity of Brucella strains circulating in humans and livestock in Oman. To confirm the presence of Brucella in milk samples, a milk ring and rapid tests were used. Then, real-time PCR was also carried out, followed by conventional PCR based on three Brucella-specific primer pairs. Moreover, species-specific primers (Bruce-ladder) were employed, as described by Ref. 15.

This study, to the best of our knowledge, is the first to use DNA extracted directly from milk samples for MLVA analysis. However, DNA extracted directly from milk may result in low yield and quality, and may contain PCR inhibitors that affect amplification efficiency. This likely contributed to missing loci in some samples and justified the adoption of MLVA-14 instead of MLVA-16. In general, it should also be taken into consideration that milk samples might contain more than one Brucella isolates, and multiple alleles can be amplified at certain loci thus hampering inference of MLVA haplotypes. Such occurrence seems to be unlikely given the fact that, in this study, all milk samples investigated were infected by a single strain; however, the possibility to detect co-infections with multiple strains would obviously increase when working with bulk milk samples in endemic areas. Nonetheless, we decided to work with milk samples to avoid the risk associated with culturing Brucella. Moreover, the lack of a biosafety level 3 cabinet (BSL-3) at the microbiology laboratory in the animal and veterinary sciences department in SQU was an obstacle to dealing with the zoonotic nature of Brucella. It is worth noting that all human isolates used in this study were cultured at Sultan Qaboos University Hospital (SQUH) and Sultan Qaboos Hospital in Salalah (SQH). All these isolates were successfully typed at each locus for the 16 markers. Taken together, these observations confirm that isolates remain the most reliable source for MLVA analysis. Interestingly, even a study by Ref. 26 using isolates had reported missing loci for some markers, such as Bruce07 and Bruce19, underscoring potential challenges in MLVA analysis.

MLVA-14 analysis identified 49 B. melitensis strains grouped into two main clusters comprising 31 distinct genotypes. Overall, MLVA-14 analysis revealed two main clusters: one relatively conserved and consistent with previously described lineages, and a second more heterogeneous cluster reflecting higher genetic variability among strains. Based on the dendrogram ( Figure 2) and ( Table 1), homogeneity was observed in Bruce8, Bruce11, Bruce43, Bruce45, and Bruce21 loci with a monomorphic profile in previous studies. ^{7, 27} On the other hand, Bruce 21 presented variable allelic types in other studies. ^{7, 25, 26, 28} In this study, Bruce6, Bruce42, Bruce12, and Bruce19 loci were moderately variable with two to three allelic types, also moderately variable in other studies. ^{24, 25} The highly discriminatory markers in this study were found in Bruce4, Bruce9, Bruce16, Bruce18, and Bruce30. Particularly, Bruce4 and Bruce9 exhibited diverse allelic types (7 types), contributing significantly to the differentiation between Brucella strains. These findings highlight both the persistence of established lineages and the presence of ongoing genetic diversification within the Omani B. melitensis population.

In a study by Tiller et al. (2019), Bruce4, Bruce9, Bruce16, and Bruce18 were also polymorphic, whereas Bruce 30 tended to be more conserved. ²⁷ Kulakov et al. (2011) reported that Bruce4 and Bruce16 were highly polymorphic, whereas Bruce9 and Bruce18 were monomorphic. ²⁹ In another study, Bruce4, Bruce30, and Bruce16 were discriminatory markers. ²⁵

Overall, the clustering patterns and genotypic differences or similarities observed in this study align with findings from previous research, emphasizing the usefulness of MLVA in studying Brucella genetic diversity and epidemiology. A similar molecular study conducted in Bangladesh reported the first genetic characterization of Brucella abortus biovar 3 using MLVA, reinforcing the value of this method in endemic regions for epidemiological insight and disease control planning. ³⁰

These results strongly support the role of livestock as a key reservoir for zoonotic transmission of B. melitensis in Oman. The genetic similarity observed among B. melitensis strains from camel milk (K187, K189, K185, K188) and goat milk (K197) samples in Dhofar Governorate suggests potential shared reservoirs or transmission pathways within livestock populations in the region. Similarly, the identical genotypes identified in goat milk strains (K43, K44, K47, K48, K50) and cattle milk strains (K55, K56, K60) highlight the interconnected nature of Brucella transmission across different animal hosts, possibly through shared environments, grazing lands, or interspecies interactions. Furthermore, the shared genotypes between camel milk strains (K102, K103, K104, K107) and cattle milk strains (K112, K113) reinforce the role of camels and cattle as significant reservoirs for B. melitensis in Dhofar Governorate. The transmission of B. melitensis between different animal species was reported previously in Oman. ^{4, 5, 31– 33} Moreover, human isolates K219 and K220 shared the same genotype, while K212 and K216 formed another group with identical genotypes. These findings are consistent with previous studies which indicate that human infection is often linked to livestock reservoirs due to zoonotic transmission, frequently through the consumption of unpasteurized dairy products or direct contact with infected animals. ^{32, 34– 37} In addition, the majority of human cases in this study reported a history of consuming unpasteurized raw milk across all age groups according to SQUH and SQH. Notably, the highest number of cases was observed among children under the age of ten years. This may be related to increased exposure and behavioral factors, such as consumption of unpasteurized milk, rather than purely biological susceptibility. Additonally, children are more likely to consume unpasteurized milk, especially in regions where it is considered a traditional dietary essential. Limited awareness of the risks associated with raw milk consumption, combined with potential exposure to contaminated environments, such as farms or infected animals, further increases their risk. Advancement of the healthcare system in Oman has led to improved diagnosis and reporting of illnesses across all age groups.

The genetic heterogeneity of other strains, divided into subclusters, reflects the diverse Brucella population in Dhofar Governorate, highlighting the need for localized control measures and management procedures. ^{4, 32, 33} Our MLVA analysis provides valuable insights also into the epidemiology of Brucella in Oman, including host specificity, geographic clustering, and potential transmission dynamics.

The phylogeographic analysis of 49 Brucella melitensis strains from Oman was compared to international MLVA-15 profiles as previously described. ^{7, 24, 28} This provides significant insights into the genetic relationships and potential transmission patterns of these strains worldwide. In MST analysis integrating worldwide isolates, most Omani strains formed a distinct country-specific branch, with close association with two strains from the United Arab Emirates. Specifically, this branch included strains recovered from human (K234, K239, K244), cattle and camel milk (K53, K70, K102, K103, K107, K104, K112, K113). This finding suggests a localized genetic lineage, likely shaped by regional factors such as shared livestock trade routes or common environmental reservoirs in the Arabian Peninsula. ^{4, 38} Additionally, another cluster of Omani strains recovered from goat and cattle milk (K143, K144, K147, K148, k150, K155, K156, K160) grouped with isolates from Spain, Portugal, India, Turkey, and China. This broader clustering indicates potential historical or trade-driven links, likely reflecting the global movement of livestock and animal products that has influenced the genetic diversity of B. melitensis strains. ^{31, 32} Interestingly, one “outlier” Omani goat isolate (K179) was closely related to a cluster including mainly Chinese isolates, but also isolates from countries worldwide (i.e., Portugal, Kazakhstan, Mongolia, Turkey, Spain, Morocco, France) highlighting an exception to the broader clustering trends. This unique pattern may represent a rare introduction of a strain through trade, livestock importation, and, or human travel. These patterns may also reflect the complexity of international trade networks; for example, Oman imports dairy products, eggs, honey, and other edible products from Kazakhstan.

These results underscore the complexity of Brucella phylogeography, reflecting both regional and global transmission dynamics. The distinct clustering of most Omani strains emphasizes the potential role of localized environmental and epidemiological factors in shaping Brucella diversity. However, the observed connections with strains from other countries emphasize the importance of monitoring international trade and the movement of livestock to prevent the spreading of brucellosis. ³⁹

This study has some limitations. First, the use of milk-derived DNA may have affected typing resolution due to lower DNA quality and the presence of inhibitors. Second, sampling was geographically limited, with most samples originating from Dhofar Governorate, which may not fully represent the national epidemiological situation. Finally, MLVA, while informative, provides lower resolution compared to whole genome sequencing, which should be considered in future studies. Despite these limitations, the study provides robust epidemiological insights into Brucella circulation in Oman.

This study represents the first comprehensive genotyping and comparative analysis of B. melitensis strains from humans and livestock in Oman using MLVA-14. The results demonstrate a clear genetic relationship between strains of human and livestock origin, supporting the zoonotic nature of transmission and highlighting the interconnectedness of these populations. The distribution of genotypes across different animal species further supports established transmission routes among livestock in Dhofar and AD Dakhiliya Governorates. Isolates provided more reliable genotyping results compared to milk-derived DNA, emphasizing the importance of appropriate sample selection and handling. These findings underline the need for a One Health approach involving both veterinary and public health sectors in brucellosis control efforts. This requires coordinated policy actions, including livestock vaccination, strict enforcement of pasteurization, and increased public awareness to reduce transmission. Future research should expand sampling to additional governorates in Oman and integrate whole genome sequencing with MLVA for higher-resolution analyses. In parallel, the development of a national Brucella surveillance strategy incorporating a centralized genotype repository would significantly enhance disease monitoring and control. Finally, strengthening laboratory capacity through upgraded infrastructure, standardized diagnostics, and staff training will be essential to overcome current limitations and ensure sustainable disease surveillance in Oman.

The experiment was conducted in Dhofar and AD Dakhiliyah governorates, Oman, for milk samples. The samples were collected from grazing animals in the open, owned by private individuals with their consent and guidance, who managed animal husbandry as per rural tradition. There was minimal interaction with the animals for the collection of milk for further investigation. Minimal animal manipulation (handling & restraint) was observed for milking by experienced personnel, ensuring an aseptic collection procedure. The sample size was much smaller than the normal milking volume for each animal.

The animals were needed only for milk samples. The milking was done in a clean quiet area to ensure a peaceful environment for the animals. There were minimal handling and restraint under the consent, help, and guidance of the owners who are very familiar with their stock. Milking was carried out aseptically by experienced technicians. The clinician was around to ensure the safety of the animals during the procedure. This asserts that animal welfare was not compromised in any way more than minimal normal handling for aseptic milking under clinical supervision in the presence of animal owners.