The genus Bartonella contains a diverse group of emerging, zoonotic, gram-negative, facultative intracellular Alphaproteobacteria that infect a wide range of wildlife and domestic animals. Rodents appear to be the most common wildlife host of Bartonella, having been associated with 20/33 described species to date, many of which have also been linked to disease in people.1,2 Transmission between animal hosts appears to be primarily through ectoparasite vectors, including fleas, mites, and ticks.2 However, it is clear from the diversity of recently described Bartonella spp. associated with a widening range of mammalian hosts and potential arthropod vectors, that the ecology of these bacteria is complex and not well understood.3,4
In people, infection with Bartonella often results in undifferentiated febrile illnesses that may be similar in clinical presentation to those caused by other pathogens (e.g., Borrelia spp.).2,5 This suggests that the global burden of Bartonella-associated diseases, although significant, may be underestimated. Some groups of people (e.g., outdoor workers, immunocompromised people, and the homeless) may be particularly vulnerable to infection, indicating that human behavior and local ecology may be significant contributors to zoonotic disease risk.2 In this study, we screened indigenous and invasive rodents found in urban, developing, and rural locations around the city of Kuching, Sarawak, for Bartonella spp., to begin to explore the role of local ecology in the presence and prevalence of Bartonella in Malaysian Borneo.
As part of a larger study on the effect of urbanization on rodent-borne diseases, we collected 316 rodents from several sites in urban, developing, and rural areas in and around the city of Kuching, Sarawak, between September 2015 and April 2016 (Supplemental Figure 1). The predominate land use type at each site was characterized by estimating the proportion of green or gray space within 10 m and 100 m radii from the trapping site and by the proportion of forest cover. Mean forest cover was estimated using QGIS v 2.14.0 (2018) and previously published forest cover and loss datasets at the Landsat pixel scale, ranked and grouped into tertiles, which were categorized as minimal, moderate, or maximal forest cover (https://earthenginepartners.appspot.com/science-2013-global-forest). Rodents were live-trapped using locally made wire-mesh traps and euthanized by over-anesthetization in isoflurane, followed by bilateral thoracotomy. Tentative species assignment, sex, breeding status, and body mass (as a proxy for age) were recorded, and tissues and ectoparasites (i.e., mites, lice, fleas, and ticks) were collected and frozen directly on dry ice. The species identity of each animal was confirmed using primers BatL5310 and R6036R, which amplify ∼750 bp of the cytochrome oxidase I gene.6 Based on the resultant sequences, rodents grouped with eight species from four genera, with most individuals falling within the Rattus rattus super-group (N = 187) or classified as Sundamys muelleri (N = 100) (Supplemental Table 1). Although three species of the R. rattus super-group were delineated by this method (i.e., Rattus sp. R3, Rattus tanezumi, and Rattus tiomanicus), we considered them collectively for this analysis, as distinct mitochondrial lineages of this super-group are known to hybridize when sympatric.7,8 All animals and samples were collected with permission from the Commonwealth Scientific and Industrial Research Organization Australian Animal Health Laboratory Animal Ethics Committee (#1750) and the Sarawak Forests Department (#NCCD.907.4.4(JLD.12)-131).
DNA was extracted from ∼30 mg of rodent spleen homogenate using the AllPrep DNA/RNA mini Kit (Qiagen Inc., Valencia, CA) and subjected to a nested polymerase chain reaction (PCR) targeting the Bartonella citrate synthase A (gltA) gene using a nested PCR reaction. Positive samples generated either a 767-bp product in round 1 (primers CS443f and CS1210r) or a 694-bp product in round 2 (primers CS443f and BhCS.1137n) of the PCR and were confirmed by Sanger sequencing.9,10 The resultant sequences (GenBank accession nos. MG807665–MG807845) were trimmed for quality and length and were manually aligned with those of a representative sample of Bartonella spp. in Geneious version 10.2.2.11 A maximum likelihood (ML) phylogenetic tree was constructed using the Generalized Time Reversible plus gamma model of nucleotide substitution in PhyML v3.1, with 1,000 bootstrap replications.12 Sequences were then trimmed to include only the 327-nt region of gltA (positions 801–1127) commonly used for taxonomic classification and compared with publicly available sequences from verified Bartonella species.13 For downstream analyses, sequences were considered to belong to a known species if they shared ≥ 96% sequence similarity and clustered with the respective species-specific clade in the ML tree (Figure 1). Statistical analyses were performed to identify significant relationships between infection status and biological and environmental variables and considered all Bartonella together, as well as each of the three predominant groups separately. Standard tests were used including the Fisher’s exact test (variables with two outcomes), Chi-square test of independence (variables with more than two outcomes), and Z test (measurement variables).
A total of 181 animals (57.3%) were positive for Bartonella spp., including 47.1% of Rattus sp. (N = 88) and 87.0% of S. muelleri (N = 87) (Supplemental Table 1). Sundamys muelleri were more likely to be infected with Bartonella than Rattus sp., and this difference was statistically significant (Table 1, P < 0.01). Bartonella spp. prevalence also varied by location, with rodents from urban and developing regions more likely to be infected than rodents from rural areas (Table 1, P < 0.01), and by body mass (Table 1, P < 0.01), but not by gender. Although heavier (older) rodents were more likely to be infected than lighter (younger) rodents, this difference was significant only when individuals from the genus Rattus were considered separately. Bartonella spp. prevalence was also found to vary by site type, with predominantly green sites more likely to harbor infected individuals than gray sites (Table 1, P < 0.01). The presence of some types of ectoparasite also correlated with Bartonella spp. infection, as individuals with mites or lice were more likely to be positive for Bartonella than those without (Table 1, P < 0.05).
Statistical tests of the associations between Bartonella spp. infection and environmental or ecological variables
|Rodent species (Rattus/Sundamys)||Fisher’s exact|
|Overall||1||P = 7.37 × 10−12||P < 0.01|
|Undescribed clade only||1||P = 3.55 × 10−26||P < 0.01|
|Mites (presence/absence)||Fisher’s exact|
|Overall||1||P = 1.06 × 10−3||P < 0.01|
|Rattus spp. only||1||P = 0.011||P < 0.05|
|Lice (presence/absence)||Fisher’s exact|
|Overall||1||P = 3.76 × 10−3||P < 0.01|
|Rattus spp. only||1||P = 0.016||P < 0.05|
|Bartonella rattimassiliensis only||P = 0.025||P < 0.05|
|Ticks (presence/absence)||Fisher’s exact|
|Undescribed clade only||1||P = 0.001||P < 0.01|
|Overall||2||χ2 = 16.992||P < 0.01|
|Undescribed clade only||2||χ2 = 12.466||P < 0.01|
|Site type (gray/green)||Fisher’s exact|
|Overall||1||P = 5.35 × 10−8||P < 0.01|
|Undescribed clade only||1||P = 3.30 × 10−13||P < 0.01|
|Forest cover (min/mod/max)||χ2|
|Undescribed clade only||2||χ2 = 19.923||P < 0.01|
|Overall||–||Z = 78.730||P < 0.01|
|Rattus spp. only||–||Z = 43.075||P < 0.01|
Sequence and phylogenetic analyses revealed the presence of six distinct groups of Bartonella in rodents, five of which were closely related to verified Bartonella species (Figure 1). Six sequences were identified by both methods as belonging to Bartonella queenslandensis, six to Bartonella tribocorum, one to Bartonella elizabethae, 31 to Bartonella rattimassiliensis, and 72 to Bartonella phoceensis (Supplemental Table 2). A further 48 sequences were < 96.0% similar to previously described species, yet clustered tightly together in the ML tree and were > 96.0% similar to each other (henceforth referred to as the “undescribed clade”). The final 18 sequences fell outside of the major clades in the ML tree and/or were unassigned by % sequence similarity (Figure 1). Differences in the distribution of the three predominant Bartonella clades were observed, as members of the undescribed clade were only detected in S. muelleri trapped in green sites and were more prevalent in sites with more forest cover, as well as in urban and developing locations (Figure 2). Positive associations were also found between the prevalence of B. rattimassiliensis and the presence of lice (P < 0.05) and the unassigned Bartonella and the presence of ticks (P < 0.01), whereas B. phoceensis was not positively associated with the presence of any ectoparasites (P > 0.05) (Table 1).
In this study, we found that Bartonella spp. infection was pervasive in both indigenous and invasive rodents and across rural, developing, and urban locations. Bartonella spp. prevalence was highest among rodents from urban sites and lowest among those from rural areas (Figure 2, P < 0.01). Notably, 85% and 100% of S. muelleri trapped at urban and developing sites, respectively, were Bartonella spp. positive. In these locations, S. muelleri rodents were restricted to remnant green patches, where they were observed to exist at a higher density than in rural environments. Therefore, high-density populations of rodents restricted to relatively small habitat patches may experience increased rates of intraspecies contact, facilitating Bartonella transmission.
It is unclear how the increased prevalence of Bartonella in rodent populations in urban and urbanizing environments might influence the risk of zoonotic transmission for people in cities; however, human infection with B. elizabethae, B. tribocorum, and B. rattimassiliensis have been documented, and all three of these species have been associated with urban rodents.14–16 Human risk of infection is likely heterogeneous in cities and governed in part by the intensity of contact between rodents, their ectoparasite vectors, and people.17 In particular, the frequent use of remnant green patches by people in and around Kuching for foraging and recreation may, therefore, increase risk of zoonotic infection, particularly from tick or flea-borne Bartonella.
This study was supported by a Discovery Early Career Researcher Award from the Australian Research Council (DE150101259). The authors thank Amy Hahs (The University of Melbourne), Andrew Alek Tuen (Universiti Malaysia Sarawak), Andrew Joris Noyen (Padawan Municipal Council), Basheer Ahmed (Kuching North Municipal Council), Danielle Levesque (Universiti Malaysia Sarawak), Dilop Jina (Padawan Municipal Council), Jean-Bernard Duchemin (CSIRO), Lee Trinidad (CSIRO), Nurshilawati Abdul Latip (Universiti Malaysia Sarawak), Patrick Lai Ganum (Kuching South Municipal Council), Rachel Amos-Ritchie (CSIRO), and Samuel Wong (Universiti Malaysia Sarawak).
Buffet JP, Kosoy M, Vayssier-Taussat M, 2013. Natural history of Bartonella-infecting rodents in light of new knowledge on genomics, diversity and evolution. Future Microbiol 8: 1117–1128.
Regier Y, O’Rourke F, Kempf VAJ, 2016. Bartonella spp.—a chance to establish one health concepts in veterinary and human medicine. Parasit Vectors 9: 261.
Lei BR, Olival KJ, 2014. Contrastig patterns in mammal-bacteria coevolution: Bartonella and Leptospira in bats and rodents. PLoS Negl Trop Dis 8: e2738.
Kosoy M, McKee C, Albayrak L, Fofanov Y, 2017. Genotyping of Bartonella bacteria and their animal hosts: current status and perspectives. Parasitology 145: 543–562.
Maggi RG, Mozayeni BR, Pultorak EL, Hegarty BC, Bradley JM, Correa M, Breitschwerdt EB, 2012. Bartonella spp. bacteremia and rheumatic symptoms in patients from lyme disease-endemic region. Emerg Infect Dis 18: 783–791.
Robins JH, Hingston M, Matisoo-Smith E, Ross HA, 2007. Identifying Rattus species using mitochondrial DNA. Mol Ecol Notes 7: 717–729.
Pages M, Chaval Y, Herbreteau V, Waengsothorn S, Cosson J-F, Hugot J-P, Morand S, Michaux J, 2010. Revisiting the taxonomy of the Rattini tribe: a phylogeny-based delimitation of species boundaries. BMC Evol Biol 10: 184.
Pagès M, Bazin E, Galan M, Chaval Y, Claude J, Herbreteau V, Michaux J, Piry S, Morand S, Cosson JF, 2013. Cytonuclear discordance among southeast Asian black rats (Rattus rattus complex). Mol Ecol 22: 1019–1034.
Norman AF, Regnery R, Jameson P, Greene C, Krause DC, 1995. Differentiation of Bartonella-like isolates at the species level by PCR-restriction fragment length polymorphism in the citrate synthase gene. J Clin Microbiol 33: 1797–1803.
Billeter SA, Gundi VA, Rood MP, Kosoy MY, 2011. Molecular detection and identification of Bartonella species in Xenopsylla cheopis fleas (Siphonaptera: Pulicidae) collected from Rattus norvegicus rats in Los Angeles, California. Appl Environ Microbiol 77: 7850–7852.
Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, 2012. Geneious basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28: 1647–1649.
Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordiik W, Gascuel O, 2010. New algorithms and methods to estimate maximum‐likelihood phylogenies: assessing the performance of PhyML 3.0. Syst Biol 59: 307–321.
La Scola B, Zeaiter Z, Khamis A, Raoult D, 2003. Gene-sequence-based criteria for species definition in bacteriology: the Bartonella paradigm. Trends Microbiol 11: 318–321.
Kosoy M et al. 2010. Identification of Bartonella infections in febrile human patients from Thailand and their potential animal reservoirs. Am J Trop Med Hyg 82: 1140–1145.
Tay ST, Mokhtar AS, Zain SNM, Low KC, 2014. Short report: isolation and molecular identification of Bartonellae from wild rats (Rattus species) in Malaysia. Am J Trop Med Hyg 90: 1039–1042.
Jiyipong T, Morand S, Jittapalapong S, Rolain J-M, 2015. Bartonella spp. infections in rodents of Cambodia, Lao PDR, and Thailand: identifying risky habitats. Vector Borne Zoonotic Dis 15:48–55.
Peterson AC et al. 2017. Rodent-borne Bartonella infection varies according to host species within and among cities. EcoHealth 14: 771–782.