Investigation of Genetic Relatedness of Brucella Strains in Countries Along the Silk Road

In this study, MLVA (multiple-locus variable-number tandem repeat analysis) genotype data of Brucella strains from 11 countries along the Silk Road were downloaded from the MLVAbank. MLVA data of strains were applied to the constructed Minimum Spanning Tree to explore the species/biovars distribution, geographic origins, and genetic relationships of the strains analyzed. Moreover, whole-genome sequencing–single-nucleotide polymorphism (WGS-SNP) phylogenetic analysis of the genome of Brucella melitensis strains from GenBank was performed to discriminate the relatedness of strains further and investigate the transmission pattern of B. melitensis brucellosis. A total of 1,503 Brucella strains were analyzed in this study: 431 Brucella abortus strains (29.8%), 1,009 B. melitensis strains (65.7%), and 63 Brucella suis strains (4.5%). B. melitensis biovar 3 was the dominant species and was shown to be widespread in all of the examined regions, suggesting that the prevention and surveillance of the B. melitensis population are a main challenge in these countries. A wide host spectrum was observed for this Brucella population; many animal reservoirs are a potential reason for the continuous brucellosis circulation in these countries. Although the B. abortus strains from the examined regions had common geographic origins, only a few shared genotypes were observed in different countries. These data revealed that the majority of B. abortus strains were spreading within the national borders. However, the B. melitensis strains from Italy originated from a Western Mediterranean lineage; strains from the other 10 countries originated from Eastern Mediterranean lineage, and this lineage was shared by strains from three to nine different countries, suggesting that the introduction and reintroduction of the disease in the 10 countries might have occurred in the past. Furthermore, the most shared MLVA-16 genotypes were formed in the B. melitensis strains from China, Kazakhstan, and Turkey, suggesting that the introduction and trade in sheep and goats have occurred frequently in these countries. WGS-SNP analysis showed that the B. melitensis in this study originated from the Malta (Italy) region. According to their territorial affiliation between four clade strains from these countries in genotype B, the absence of a clear differentiation suggests that strains continuously expand and spread in countries along with Silk Road. Active exchange and trade of animals (sheep and goats) among these countries are reasonable explanations. B. suis strains from different nations showed unique geographic origins and epidemiological characteristics. Therefore, there is an urgent need for the control of transfer and trade of infected sheep (goats) in countries along the Silk Road, namely, the strengthening of the entry–exit quarantine of sheep and goats and improvements in the diagnosis of animal brucellosis.

In this study, MLVA (multiple-locus variable-number tandem repeat analysis) genotype data of Brucella strains from 11 countries along the Silk Road were downloaded from the MLVAbank. MLVA data of strains were applied to the constructed Minimum Spanning Tree to explore the species/biovars distribution, geographic origins, and genetic relationships of the strains analyzed. Moreover, whole-genome sequencingsingle-nucleotide polymorphism (WGS-SNP) phylogenetic analysis of the genome of Brucella melitensis strains from GenBank was performed to discriminate the relatedness of strains further and investigate the transmission pattern of B. melitensis brucellosis. A total of 1,503 Brucella strains were analyzed in this study: 431 Brucella abortus strains (29.8%), 1,009 B. melitensis strains (65.7%), and 63 Brucella suis strains (4.5%). B. melitensis biovar 3 was the dominant species and was shown to be widespread in all of the examined regions, suggesting that the prevention and surveillance of the B. melitensis population are a main challenge in these countries. A wide host spectrum was observed for this Brucella population; many animal reservoirs are a potential reason for the continuous brucellosis circulation in these countries. Although the B. abortus strains from the examined regions had common geographic origins, only a few shared genotypes were observed in different countries. These data revealed that the majority of B. abortus strains were spreading within the national borders. However, the B. melitensis strains from Italy originated from a Western Mediterranean lineage; strains from the other 10 countries originated from Eastern Mediterranean lineage, and this lineage was shared by strains from three to nine different countries, suggesting that the introduction and reintroduction of the disease in the 10 countries might have occurred in the past. Furthermore, the most shared MLVA-16 genotypes were formed in the B. melitensis strains from China, Kazakhstan, and Turkey, suggesting that the introduction and trade in sheep and goats have occurred frequently in these countries. WGS-SNP analysis showed that the B. melitensis in this study originated from the Malta (Italy) region. According to their territorial affiliation between four clade strains from these countries in genotype B, the absence of a clear differentiation suggests that strains continuously expand and spread in countries along with Silk Road. Active exchange and trade of animals (sheep and goats) among these countries are reasonable explanations. B. suis strains from different nations showed unique geographic origins and epidemiological characteristics. Therefore, there is an urgent need for the control of transfer and trade of infected sheep (goats) in countries along the Silk Road, namely, the strengthening of the entry-exit quarantine of sheep and goats and improvements in the diagnosis of animal brucellosis.

INTRODUCTION
Brucellosis is a zoonotic infectious disease caused by members of the genus Brucellae, circulating in more than 170 countries, especially in the Mediterranean, Asia, and Central and South America (1). The disease is primarily contracted through direct or indirect contact with infected animals and their contaminants, although the consumption of contaminated animal products can also cause infection. Low fever, sweating, fatigue, and joint pain are the most common clinical manifestations of patients (2). Chronic brucellosis is challenging to cure, which affects the quality of life of patients and causes great family and socioeconomic burdens (3). In females, abortion and stillbirth are the main manifestations, whereas orchitis and sterility are common in males (4). Therefore, brucellosis not only has a significant impact on the husbandry and health of the human population, but is also closely related to many fields such as public safety, food hygiene, and foreign trade (5).
Although human brucellosis (HB) is one of the most common bacterial zoonotic diseases, its occurrence greatly differs between geographic areas. The highest annual incidences of HB per million of the population are observed in Syria and Mongolia (1); for instance, 39,838 human cases were reported in Syria in 2007. Mongolia has a high incidence of brucellosis in humans and animals due to livestock husbandry; livestock seroprevalence ranged from 0.5 to 1.8% in 2011 (6). Moreover, a total of 684,380 HB cases were reported in mainland China during 1950-2018, and the incidence of HB peaked in 2014 (4.32/100,000) (7). Kazakhstan is a hyperendemic area, considering the very high incidence rates in the human population and farm animals (1). From 1999 to 2016, 38,557 new cases of HB were recorded, with annual counts of cases ranging from 1,443 (2014) to 3,596 (2004) (8). In Pakistan, the high age-wise prevalence was recorded as 32.25% (98/304) by enzyme-linked immunosorbent assay (ELISA) (P < 0.05) in 21-to 30-year-old females (9). In India, the seroprevalence of brucellosis among cattle and goats was estimated to be 1.1 and 11.2%, respectively (10). Brucellosis is still reported from all types of farms and domestic livestock in Russia; between 1999 and 2009, the number of infected sites increased, ranging from 54 in 2001 to 112 in 2009 (11). The incidence of HB in Turkey is 26 per 100,000 (12), and 18,264 new brucellosis cases were reported in 2004 (13); moreover, pediatric brucellosis may constitute 20% to 30% of all brucellosis cases in the world (14). Previous research reported that 22 (5.1%) tested by Rose Bengal Test (RBT) and 58 (13.5%) tested using the blocking ELISA assay were positive for Brucella in wild boar hunted in the Campania region of Italy during 2016-2017 (15). Each year a relatively small number of cases are reported in Germany, but within the last 2 years, an increase of cases was observed (16). Many comprehensive brucellosis controls have been developed in the majority of countries, including testing-slaughter, restricted infected animal transfer, and vaccination of livestock. However, brucellosis is still an endemic zoonosis in these nations (17).
The Silk Road Economic Belt has had a profound effect on economic prosperity, cultural diversity, foreign trades, and many other characteristics of nations along the road; therefore, it might also provide opportunities for the spread of zoonotic diseases (18). Indeed, zoonotic diseases such as plague and anthrax are associated with commercial activities related to the Silk Road (19,20), especially in countries along the line of the road. Indeed, these countries have high amounts of husbandry, and brucellosis circulates between humans and animals in Italy, Germany, Turkey, Iran, and Kazakhstan (21)(22)(23)(24)(25). Accordingly, there is the potential for transmission of brucellosis during processing of animals and trade of animal products. Currently, MLVA-16 genotyping of Brucella strains can be used for trace-back and trace-forward investigations, as well as the identification of the spreading route (26), and MLVA-11 is often used to trace back the geographic origin of strains (27). Moreover, whole-genome sequencing-singlenucleotide polymorphism (WGS-SNP) is likely to be a suitable tool for trace-back analysis of Brucella melitensis, as it is capable of suggesting the potential geographic origin of a given strain (16). Therefore, this study was conducted to better understand the epidemiological characteristics of brucellosis in countries along the Silk Road, provide valuable information for strengthening trade cooperation, and launch a corresponding strategy of prevention and control of brucellosis.

Strain Isolation and Identification
A total of 1,503 Brucella strains were collected from samples of patients and animals from 11 different countries. According to standard bacteriology approaches, all of the strains were isolated and identified (28). Conventional identification methods were used to assign the strains to biotypes. All of the strains were gram-negative, agglutinated with polyvalent brucellosis serum, had oxidase and catalase activity, synthesized urease, and were capable of growing in atmospheric conditions. Subsequently, BCSP-31 polymerase chain reaction (PCR) (29), AMOS-PCR (30), and/or Brucella ladder PCR (31) were applied to verify the results from the biotyping assays.

DNA Preparation and Genotyping
A fresh single Brucella strain clone was picked up from a solid agar medium surface and then heat-inactivated at 80 • C for 20 min. Subsequently, DNA was isolated using a phenol/chloroform extraction method and/or the QIAamp DNA Mini Kit (Qiagen, United States) according to the manufacturer's instructions. The MLVA-16 assay was performed as previously described (32,33). Briefly, 16 loci were divided into three panels: Panel 1 (also called MLVA-8), Panel 2A, and Panel 2B. The combination of panels MLVA-8 and 2A was called MLVA-11, whereas the combination of all three panels (16 loci) was designated MLVA-16. The MLVA-11 panel allows for tracing the geographic origin of the strains analyzed, whereas the Panel 2B loci are highly discriminatory, and their combination with MLVA-11 was used for molecular epidemiology investigation. PCR was used to determine the number of repeats from each sample, and its products were purified and directly sequenced using an ABI Prism Big Dye Terminator. Size analysis of VNTR repeats was performed using GeneMapper 4.1 (Applied Biosystems).  1, 188 in biovar 3, 79 in biovar 6, four in biovar 7, one in rough strain, and another 98 strains have not been identified ( Figure 1A). Among the B. melitensis strains, 99 strains in biovar 1, 50 strains in biovar 2, 532 strains in biovar 3, five strains in rough, and nine strains were atypical, and 313 strains of unknown biovars were observed ( Figure 1B). Among the 63 B. suis strains, 18 were in biovar 1, 44 in biovar 2, and one in biovar 3 ( Figure 1C). All three Brucella species were observed in China, Germany, and Italy (

Host Lineages Profiles of Brucella Strains
In this study, a total of 13 kinds of hosts were observed in all of the countries analyzed, including humans, cattle, sheep, camels, buffalo, goats, bharals, dogs, deer, swine, boars, hares, and wild swine, of which 583 strains were found in humans, 448 in cattle, and 363 in sheep ( Table 2). The hosts of B. abortus strains included cattle, sheep, camels, buffalo, and humans, of which 92.0% of strains were collected from cattle (Figure 2A). The hosts of B. melitensis strains included goats, sheep, cattle, deer, bharals, camels, humans, and dogs. Remarkably, approximately 56.8% (573/1,009) of strains were from humans, and this host was distributed widely in all of the countries; moreover, 35.3% (353/1,009) of the strains were obtained from sheep ( Figure 2B, Table 2). Deer, humans, sheep, swine, boars, hares, and wild swine were common hosts from isolated B. suis strains, and 55.6% (35/63) of the strains from Germany and Italy were found in wild swine ( Figure 2C). However, humans, cattle, and sheep were the main hosts of the B. suis strains in China ( Table 2). The most extensive hosts of Brucella in this study were observed in China, followed by Italy, Kazakhstan, and India ( Table 2). genotypes. In contrast, no shared genotype was found in other countries ( Figure 3C).

Geographic Origins and Genetic Relatedness of B. melitensis Strains
There were 62 MLVA-11 genotypes in 1,009 strains of B. melitensis, of which 116, 125, 87, 96, 111, and 91 were the most common genotypes (Figure 4A).  Figure 4B). MST analysis based on MLVA-16 data of B. melitensis strains indicated that 1,009 strains of B. melitensis were clustered into two groups (A and B). The strains in Italy have formed an independent group A, whereas the other 10 countries' strains were clustered into group B (Figure 4C) Figure 4C).

Geographical Origin and Genetic Relatedness of B. suis Strains
There were 14 MLVA-11 genotypes in 63 strains of B. suis, and MLVA-11 genotypes 33, 57, and 44 were predominant, accounting for 28.6% (18/63), 22.2% (14/63), and 7.0% (9/63), respectively; the remaining 11 genotypes contained one to five strains from the respective country ( Figure 6A). Genotypes 33, 31, 57, and 44 were unique to Chinese, India, German, and Italian strains, respectively ( Figure 6B). MST analysis based on MLVA-16 data revealed that 63 strains of B. suis from Germany, Italy, China, and India formed three independent branches (A-C). The majority of strains from Germany clustered in branch A, and all Italy strains and a few German strains clustered in branch B. Strains from China and India (n = 1) were grouped in branch C. No shared MLVA-16 genotypes were found in strains from different countries (Figure 6C).   characteristic of pathogenic bacteria population and Brucella biovars was similar to the previous study; B. melitensis is the most important zoonotic agent, followed by B. abortus and B. suis (35,36). This correlates with the fact that the worldwide control of bovine brucellosis (due to B. abortus) has been achieved to a greater extent than the control of sheep and goat brucellosis (due to B. melitensis), these latter species being the most important domestic animals in many developing countries (17). B. melitensis strains were distributed widely in all of the countries examined; 56.8% (573/1,009) strains were obtained from humans, and 35.3% (353/1,009) were from sheep. These data indicated that B. melitensis strains are the predominant pathogen responsible for human and animal brucellosis in these countries. These findings suggest that identical or similar epidemiological characteristics of brucellosis might be found in countries along the Silk Road (37). Globally, despite the remarkable results achieved by the majority of industrialized countries, where bovine brucellosis has been eradicated or controlled, small ruminant brucellosis remains a problem in some of these countries (38). Our current study confirmed that B. melitensis is a latent "travel bacterium" that continuously spreads and expands from Northern China to Southern China (39). From 1970 to 2015, B. melitensis has been the pathogen isolated most frequently in human cases, accounting for more than 99% of Brucella spp. isolated from humans in Italy (40). Similarly, the species that causes the most frequent brucellosis and produces the most severe disease picture is B. melitensis in Turkey (13). Although 28.7% of total strains were of B. abortus, most of them were obtained from cattle, and its pathogenicity to humans was significantly lower than B. melitensis strains (17). Moreover, B. suis biovar 2 strains in this study were from a wild animal; B. suis biovar 1 and biovar 3 from humans and B. suis strains were observed in limited countries; fewer strains were isolated from humans, and the B. suis biovar 2 is not an essential pathogen for humans in contrast to B. suis biovars 1, 3, and 4 (17). Therefore, to solve the brucellosis problem in countries along the Silk Road, we should focus more on the B. melitensis strains than B. abortus and B. suis populations, and it is urgent to strengthen the detection and management of the infected sheep and goats in these countries.
In the present study, B. abortus strains and B. melitensis strains had a wide host spectrum, including many livestock and wild animals. Some shared genotypes were observed in different hosts among each species, indicating that these wild animals might become a source for reintroducing Brucella strains in domestic ruminants and humans (41). B. abortus and B. melitensis are the most regularly transmitted species between wild and domestic ungulates. They are most frequently associated with the conflicting needs of wildlife and agriculture and the risk of human disease (42). This confirmed the widespread existence and natural epidemic origin of brucellosis in countries along the Silk Road (43,44), which poses severe challenges to the prevention and surveillance of brucellosis in these countries.
Based on the MLVA-11 analysis of B. abortus strains, genotypes 72, 201, 82, and 70 were more common; each of them was shared by strains from two to five countries, especially 72 being the predominant genotype and shared by strains from Kazakhstan, China, Italy, Mongolia, and Turkey. These findings demonstrate that the B. abortus strains from these countries have high homogeneity and common geographic origins (45,46). MST analysis of B. abortus strains revealed that there were bovine brucellosis outbreaks in Kazakhstan, Italy, and China, but the number of cases of cross-infection among countries was limited (47). This was not only closely related to the reduction in virulence of B. abortus strains, but also restricted by factors such as larger size in cattle, higher transportation costs, and trade policies.
Among the 1,009 B. melitensis population, genotypes 116, 125, and 111 originated from the Eastern Mediterranean lineage, and genotypes 87, 96, and 91 originated from Western Mediterranean lineage. These findings demonstrate that the B. melitensis strains from countries along the Silk Road have two different geographical origins. The strains from Italy originated from a Western Mediterranean lineage, and strains from the other 10 countries originated from an Eastern Mediterranean lineage. Moreover, genotypes 116, 125, and 111 were shared by strains from nine, nine, and three countries, respectively, indicating that the strains from these countries had the same geographical origin. This suggests that the introduction and reintroduction of the disease among the 10 countries might have occurred in the past (48,49). Based on MST analysis revealed that B. melitensis strains clustered into two groups, with strains in Italy forming a relatively independent group A, whereas strains in the remaining 10 countries clustered into group B. In addition, there were at least 10 shared MLVA-16 genotypes in strains from group B. The most shared genotypes were formed in strains from China, Kazakhstan, and Turkey. It was previously suggested that B. melitensis strains in China, Turkey, and Kazakhstan were epidemiologically correlated, and the prevalence of B. melitensis brucellosis in these countries resulted from a common infectious source (16,50,51). The breeding and trade of small ruminants were the main economic source and industrial pillar of the vast agricultural and pastoral areas in Turkey, Kazakhstan, and China, and animal introduction and trade were extremely frequent in these countries. The results were consistent with those of previous studies that showed the majority of B. melitensis strains in Kazakhstan were genetically related to strains transmitted in China and were closely related to the long-term trading partnership between the two countries (52). Therefore, WGS-SNP analysis indicated that B. melitensis were originated from Malta (Italy) regions, strains from group B forming the four subclades, but do not have a clear differentiation according to their territorial affiliation, indicating the frequent penetration of the B. melitensis strains from one country to another (53). Active exchange and trade of animals (sheep and goats) among these countries could promote pathogen dissemination, and there was continuous expansion and spread in countries along with Silk Road. Recently, many infectious diseases have emerged, and most of their pathogens originated from zoonosis; some of the diseases spread worldwide by the international flow of goods, people, and animals (54). We suggest that implementing a larger scale of vaccine coverage of small ruminants would significantly reduce infection in livestock and humans. Moreover, restricting movement in positive sheep and goats among different countries and the regular disinfection of the farm environments could effectively reverse the trend of increasing brucellosis in countries along the Silk Road.
B. suis strains in Germany, China, Italy, and India represent a unique predominated genotype, respectively, suggesting that each of these strains had an exclusive geographic origin. MST analysis revealed that MLVA-16 shared genotypes were not found in different countries' strains, demonstrating that B. suis strains were only prevalent in their respective countries. This was possibly related to environmental adaptability and economic and social factors influencing hosts of B. suis strains. B. suis biovar 2 is largely restricted to continental Europe and is maintained in wild animal (Suidae and hares) populations (40). We suggest that enhanced surveillance regulation for wild animals in Europe is essential for the prevention of strain spill over into the livestock. Moreover, further genome comparison analysis of more B. suis strains from all over the world will increase our understanding of B. suis brucellosis epidemiology.
Our research has some limitations. Most of the data were derived from highly concentrated time and location, rather than from continuous time points and wide coverage areas, so they may only partly reflect the truth situation of brucellosis in the regions examined. Second, data on animals' epidemiology and movements in these countries were not available. A focus on risk analysis or statistical analysis of hosts from the spatiotemporal or species-location perspective is essential.

CONCLUSION
In this study, the species distribution, host lineage profiles, geographical origin, and genetic relatedness and the epidemiological correlation of Brucella species in countries along the Silk Road were discussed. B. abortus was found to share the same geographic origin, but the transmission area was limited to their respective countries. However, B. melitensis strains were the dominant population in these countries and had two main geographic origins, and strains from the same geographic origin showed extensive gene sharing. B. suis strains from respective countries showed unique geographic origins and epidemiological characteristics. The surveillance and control of B. melitensis strains were identified as major problems among Brucella species in countries along the Silk Road. The control of its main host (sheep and goat) should be strengthened. Effective vaccination limits Brucella infection, restricts shedding, hampers transmission from animal to animal, and diminishes zoonosis risk (55). The corresponding regulations should be strictly implemented when introducing animals to prevent the spread of this species. Moreover, improving scientific research to achieve early diagnosis and implement joint prevention and control is recommended.

DATA AVAILABILITY STATEMENT
All of the data generated or analyzed during this study are included in this published article, and the supplementary information files will be freely available to any scientist wishing to use them for non-commercial purposes upon request via e-mail with the corresponding author.

AUTHOR CONTRIBUTIONS
ZhiL performed strain download, MLVA typing and cluster analysis. CW drafted the manuscript. KW, HW, MW, ZZ and DL conducted data processing. ZheL and ZhiL participated in design of the study and critically reviewed the manuscript and ZheL managed the project. QW revised the manuscript critically. All authors have read and approved the final manuscript.

ACKNOWLEDGMENTS
The authors thank all the staff of MLVAbank for the data collected and standardized during Brucella genotyping. The authors also thank Accdon (www.accdon.com) for linguistic assistance during the preparation of this manuscript.