Bartonella

All procedures were carried out under UK Home Office licence regulations. We studied field voles (M. agrestis) in Kielder Forest, Northumberland, United Kingdom, using live-trapping to access individual animals from natural populations. Our study was designed to permit the analysis of individual variation in condition and survival, infection status, and the expression of immune genes (additional details in Methods S1). In order to ensure representativeness, we repeated our field design at two spatially separate sites in 2008–2009 and a further two separate sites in 2009–2010 (Figure S1). The study was divided into longitudinal and cross-sectional components. Each site contained a live-trapping grid (∼0.375 ha) of 150 (10×15) regularly spaced traps (3–5 m intervals) placed in optimal habitat (Figure S1). Animals from this grid were marked with passive radio frequency transponders (AVID) and monitored over time, as sequences of capture and recaptures, forming the longitudinal component of the study. At each capture, biometric, infection, and immune expression measurements were taken (Figure S2). On each site there were also satellite transects (with traps spaced at ≥5 m intervals) from which 10 animals per month per site were sampled destructively, forming the basis for the cross-sectional component of the study. Animals from this part of the study were returned to the laboratory, where it was possible to collect a more comprehensive and detailed set of biometric, infection, and immune expression measurements (Figure S2). The transects aimed to sample a very large area of the habitat, providing data representative of the entire population at each site, while at the same time avoiding significant demographic readjustments.
Each site was monitored by monthly trapping sessions between February (in the 2008–2009 season) or April (in the 2009–2010 season) and November, during which the capture–recapture study was carried out on the grid and destructive samples were retained from the transects (Figure S3). At each site, in November at the end of the field season and again in the following March, larger numbers of animals were destructively sampled both from the transects and from the grid habitats, including some animals previously marked with AVID transponders and processed for small tail-tip blood samples as part of the capture–recapture study. These samples also contributed to the overall cross-sectional component of the study (Figure S3).
Measurements of gene expression (IFN-γ, Tbet, IL-2, Gata3, IRF5, IL-10, TGF-β1, FoxP3) in the cross-sectional element of the study focussed on cultured splenocytes and are described in detail in Jackson et al (2011) [20] (link). Through the combination of stimulatory conditions applied and the genes investigated, these measurements were intended to reflect both innate immune responses (including TLR-mediated responses) and adaptive responses (including Th1, Th2, and T-regulatory responses). In the longitudinal study we measured in vivo expression of a subset (IFN-γ, Gata3, IL-10) of the genes investigated in the cross-sectional study, in peripheral blood samples.
Using direct counts or semiquantitative abundance indices, we quantified 25 species of macroparasite in our study animals, some of which were aggregated into ecologically and phylogenetically coherent groups to facilitate analysis (additional details in Methods S1). We also recorded overt symptoms of vole TB caused by M. microti and carried out PCR diagnostics for Bartonella spp. and Babesia microti using previously established methods (see Methods S1, Table S18).
Data were analyzed using LMMs or GLMMs to relate individual variables to potential explanatory variables while allowing for nonindependence due to sampling and immunological assaying structure. SEM was used to assess patterns of interdependency among multiple variables simultaneously. Some variables used in these analyses were principal component scores (reduced from larger sets of partially redundant measurements using PCA). In the case of macroparasites, we used the first principal component (PC^{M main}) from a PCA of common taxa likely to be in strong contact with the immune system (i.e., feeding on, or dwelling in or on blood or internal tissues; >20% prevalence overall) as the main analytical variable (see Methods S1 for more details). This component was dominated by high loadings of the same sign for the main macroparasite groups (fleas, ticks, and adult cestodes). Analyses of vole return rates and survivorship were carried out using GLMMs and Cormack-Jolly-Seber (CJS) methods with individual covariates.
In order to control the type 1 error rate, the analyses proceeded according to an a priori sequential strategy, starting with a sparing number of initial main hypotheses and moving onto subsequent main hypotheses conditional upon significant results in previous rounds of testing. This strategy also made use of reduced immunological and parasitological data and multiplicity adjustments within each round of main hypothesis tests. More expansive post hoc analysis of individual variables and subsets of the data, for the purposes of corroboration and describing biological patterns in more detail, was conducted subsequent to each round of main hypothesis tests. More details on the methods and strategy for data analysis are given in Methods S1 (see Figure S4 and Table S19). Data available from the Dryad Digital Repository: http://dx.doi.org/10.5061/dryad.bk537[37] .

Nymphs and adults fed on blood infected with B. henselae at preceding stages were fed with uninfected, decomplemented, ovine blood in 2 glass feeders as described above. After 84 h of refeeding, nymphs and females attached to the skin were removed and dissected. Bartonella spp. DNA was detected every 24 h in blood from the first 48 h of attachment onward. At each time point, 3 mL of blood was removed and centrifuged for 30 min at 3,000 × g. The supernatant was aspirated, and the pellet (200 μL) was used to detect bacterial DNA. After 84 h of feeding, 10 μL of blood was used for B. henselae culture.

PCRs were performed to characterize notable bacterial species detected in tick samples through the NGS screening. The bacteria Anaplasma, Ehrlichia, Bartonella, Coxiella, Francisella and Rickettsia were targeted in the PCR amplification and sequencing. The details of all primers used for the bacteria species identification are described in Table S1.
PCRs targeting the citrate synthase gene (gltA), which amplified a 694 bp fragment, and cell division protein gene (ftsZ), which amplified a 900 bp fragment of Bartonella, were conducted in semi-nested and single PCR, respectively. For Francisella species characterization, a fraction of the T-cell epitope gene (tul4) and 16S rDNA of Francisella were targeted in a single PCR to amplify 248 bp and 1 kb fragments, respectively. Single PCRs were conducted for Rickettsia species characterization by targeting six genes: gltA gene for 580 bp, outer membrane A gene (ompA) for 542 bp, outer membrane protein B gene (ompB) for 816 bp, 17 kDa common antigen gene (htrA) for 550 bp, 16S rDNA for 1.3 kb and surface cell antigen-4 gene (Sca4) for 928 bp fragment. All PCRs for Bartonella, Francisella and Rickettsia were conducted using Ex Taq Hot Start Version (Takara Bio) in a reaction mixture of 20 µl. The conditions used in the PCR assays were as follows: 35 or 40 cycles of denaturation at 94 °C for 30 s, annealing temperature according to each respective primer set for 30 s, and extension at 72 °C for 30 s, 60 s or 90 s depending on the targeted amplicon size.
Next, to characterize the species of Anaplasma, Ehrlichia and Coxiella, we used Tks Gflex DNA Polymerase (Takara Bio) with a 25 µl reaction mixture preparation. Nested PCR was conducted for Anaplasma and Ehrlichia by amplifying a 1.3 kb fragment of 16S rDNA of Anaplasmatacea with the following conditions: initial denaturation at 95 °C for 3 min, followed by 40 cycles of denaturation step at 95 °C for 30 s, 48 or 54 °C of annealing for 30 s, and extension at 68 °C for 90 s, with a final extension at 68 °C for 5 min. A total of five genes were used for Coxiella species characterization, which included chaperone protein DnaK gene (dnaK) for 512 bp, chaperone protein GROEL gene (groEL) for 619 bp, β subunit of bacterial RNA polymerase gene (rpoB) for an estimate of 550 bp, 16S rDNA for an estimate of 1 kb and large ribosomal subunit (23S rDNA) for a 583–867 bp fragment. DNA of Coxiella was amplified with nested or semi-nested PCRs, with the following conditions: initial denaturation at 94 °C for 1 min, followed by 40 cycles of denaturation at 98 °C for 10 s, 54 or 56 °C of annealing for 15 s, and extension at 68 °C for 1 min, with a final extension at 68 °C for 5 min.
Finally, the amplicon size was verified with electrophoresis and visualized as described above. Sanger sequencing was performed on the successfully amplified samples using the BigDye Terminator version 3.1 Cycle Sequencing Kit (Applied Biosystems). The obtained sequencing products were analysed on an ABI Prism 3130X genetic analyzer (Applied Biosystems), as per the manufacturer’s instructions. The resulting sequences were assembled and trimmed using the ATGC software version 9.0.0 (GENETYX) and compared with the sequences available in the public databases using the Nucleotide Basic Local Alignment Search Tool (BLASTn) (https://blast.ncbi.nlm.nih.gov/Blast.cgi).

Archived specimens from the Tropical Infectious Diseases Research and Education Centre (TIDREC), Universiti Malaya were utilized in this study. They consisted of tissues of small mammals from two sampling sites, viz. at UM Plantations Sdn. Bhd., Johor (an oil palm plantation) and Kampung Tumbuh Hangat, Perak (oil palm plantation bordering paddy fields and human settlements). These samples were collected at different times between December 2018 and December 2019 [21 (link)]. Ethical approval was obtained from the Universiti Malaya Institutional Animal Care and Use Committee (G8/01082018/24052018-01/R) and permission to conduct the study at Kampung Tumbuh Hangat, Perak was granted by the Department of Orang Asli Development (JAKOA), Malaysia (JAKOA/PP.30.052Jld13 (32)). Approval for small mammal trapping was also received from the University of Liverpool’s Animal Welfare and Ethics Review Body with reference no. AWC0127.
All small mammals captured were initially identified using morphological analysis [22 (link)]. Subsequently, tree shrew and rodent DNA barcoding was performed on DNA extracted from their spleens and other organs. Extracted rodent and tree shrew DNA was subjected to a polymerase chain reaction (PCR) targeting the cytochrome c oxidase I (COI) gene to determine the rodent and tree shrew species group [23 ]. The organs were stored at −80 °C immediately after harvesting and the extracted DNAs were aliquoted into three tubes to avoid multiple freeze-thawing. The primers used are listed in Table 1. Positive controls used were genomic DNAs of O. tsutsugamushi strain UT176 received from University of Liverpool, United Kingdom, and Rickettsia roultii strain established from a tick cell line in TIDREC. Long oligo DNAs were synthesized for the positive controls of Borrelia spp. and Bartonella spp. The positive control fragments of the flagellin gene, flaB and the citrate synthase gene, gltA were obtained from Borrelia burgdorferi NC001318.1 (501 bp) and Bartonella quintana NC005955 (410 bp), respectively. Nuclease-free water was the negative control used in PCR protocols.
The remaining COI amplicons (approximately 20 µℓ each) were purified and subsequently sequenced (Apical Scientific Sdn. Bhd., Seri Kembangan, Malaysia). The DNA sequences obtained were trimmed and compared to those available in GenBank using the Basic Local Alignment Search Tool (BLAST). Each identified species was deposited into the GenBank accordingly.

About 10 mg of each spleen tissue of the rodents and tree shrews was subjected to DNA extraction following the NucleoSpin^® Tissue Extraction Kit (Macherey-Nagel, Düren, Germany) protocol. The extracted genomic DNA was utilized to amplify genes specific for O. tsutsugamushi, Borrelia spp., Bartonella spp. and Rickettsia spp. The types of surface antigen 47 kDa gene TSA47 specific to O. tsutsugamushi [24 (link)] and flaB specific to the Borrelia spp. [25 (link)] were amplified according to previously published protocols. The detection of Bartonella spp. and Rickettsia spp. followed two different PCR protocols that target gltA [26 (link),27 (link),28 (link)]. Primers used in the present study are listed in Table 1.
The PCR-positive DNA samples for O. tsutsugamushi and Borrelia spp. were further subjected to multi-locus sequence typing (MLST) following the protocols for Borrelia spp. [29 (link)] and O. tsutsugamushi [30 (link)]. These protocols are available at their respective PubMLST databases (https://pubmlst.org/organisms/borrelia-spp (accessed on 13 October 2021) and https://pubmlst.org/organisms/orientia-tsutsugamushi (accessed on 13 October 2021). All obtained amplicons were purified and subsequently sequenced in both directions by a third party (Apical Scientific Sdn. Bhd., Malaysia). The DNA sequences obtained were trimmed and compared to those available in GenBank and PubMLST.