Jukuri, open repository of the Natural Resources Institute Finland (Luke) All material supplied via Jukuri is protected by copyright and other intellectual property rights. Duplication or sale, in electronic or print form, of any part of the repository collections is prohibited. Making electronic or print copies of the material is permitted only for your own personal use or for educational purposes. For other purposes, this article may be used in accordance with the publisher’s terms. There may be differences between this version and the publisher’s version. You are advised to cite the publisher’s version. This is an electronic reprint of the original article. This reprint may differ from the original in pagination and typographic detail. Author(s): Jessica L. Abbate, Maxime Galan, Maria Razzauti, Tarja Sironen, Liina Voutilainen, Heikki Henttonen, Patrick Gasqui, Jean-François Cosson & Nathalie Charbonnel Title: Pathogen community composition and co-infection patterns in a wild community of rodents Year: 2024 Version: Published version Copyright: The Author(s) 2024 Rights: CC BY 4.0 Rights url: http://creativecommons.org/licenses/by/4.0/ Please cite the original version: Abbate, Jessica L.; Galan, Maxime; Razzauti, Maria; Sironen, Tarja; Voutilainen, Liina; Henttonen, Heikki; Gasqui, Patrick; Cosson, Jean-François; Charbonnel, Nathalie. Pathogen community composition and co-infection patterns in a wild community of rodents. Peer Community Journal, Volume 4 (2024), article no. e14. doi : 10.24072/pcjournal.370. https://peercommunityjournal.org/articles/10.24072/pcjournal.370/ C EN T R E MER S ENN E Peer Community Journal is a member of theCentre Mersenne for Open Scientific Publishing http://www.centre-mersenne.org/ e-ISSN 2804-3871 Peer Community Journal Section: Ecology Research article Published2024-02-07 Cite asJessica L. Abbate, Maxime Galan,Maria Razzauti, Tarja Sironen,Liina Voutilainen, HeikkiHenttonen, Patrick Gasqui,Jean-François Cosson andNathalie Charbonnel (2024)Pathogen community compositionand co-infection patterns in a wildcommunity of rodents, PeerCommunity Journal, 4: e14. Correspondencenathalie.charbonnel@inrae.fr Peer-reviewPeer reviewed andrecommended byPCI Ecology, https://doi.org/10.24072/pci. ecology.100087 This article is licensedunder the Creative CommonsAttribution 4.0 License. Pathogen community compositionand co-infection patterns in a wildcommunity of rodents Jessica L. Abbate ,1,2,3,4, Maxime Galan ,1, MariaRazzauti ,1, Tarja Sironen ,5, Liina Voutilainen ,6,Heikki Henttonen6, Patrick Gasqui7, Jean-FrançoisCosson ,†,1,8, and Nathalie Charbonnel ,1 Volume 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Abstract Rodents aremajor reservoirs of pathogens that can cause disease in humans and livestock. It istherefore important to know what pathogens naturally circulate in rodent populations, and tounderstand the factors that may influence their distribution in the wild. Here, we describe theoccurrence and distribution patterns of a range of endemic and zoonotic pathogens circulatingamong rodent communities in northern France. The community sample consisted of 713 ro-dents, including 11 host species from diverse habitats. Rodents were screened for virus expo-sure (hantaviruses, cowpox virus, Lymphocytic choriomeningitis virus, Tick-borne encephalitisvirus) using antibody assays. Bacterial communities were characterized using 16S rRNA am-plicon sequencing of splenic samples. Multiple correspondence (MCA), multiple regressionand association screening (SCN) analyses were used to determine the degree to which ex-trinsic factors (study year and site; host habitat, species, sex and age class) contributed topathogen community structure, and to identify patterns of associations between pathogenswithin hosts. We found a rich diversity of bacterial genera, with 36 known or suspected tobe pathogenic. We revealed that host species is the most important determinant of pathogencommunity composition, and that hosts that share habitats can have very different pathogencommunities. Pathogen diversity and co-infection rates also vary among host species. Aggre-gation of pathogens responsible for zoonotic diseases suggests that some rodent species maybe more important for transmission risk than others. Moreover, we detected positive asso-ciations between several pathogens, including Bartonella, Mycoplasma species, Cowpox virus(CPXV) and hantaviruses, and these patterns were generally specific to particular host species.Altogether, our results suggest that host and pathogen specificity is the most important driverof pathogen community structure, and that interspecific pathogen-pathogen associations alsodepend on host species. 1CBGP, INRAE, CIRAD, Institut Agro, IRD, Univ Montpellier, Montpellier, France, 2UMR MIVEGEC, IRD, CNRS, Univ Montpellier, Montpellier, France, 3UMR UMMISCO, IRD, UPMC, UMI, Bondy, France, 4Geomatys, Mont- pellier, France, 5Department of Virology, Faculty of Medicine, University of Helsinki, Helsinki, Finland, 6Natural Resources Institute Finland (LUKE), FI-00790 Helsinki, Finland, 7UMR EPIA, INRAE, VetAgro Sup, Theix, France,8UMR BIPAR, ENVA, INRAE, ANSES, Maisons-Alfort, France 1. Introduction Infectious diseases are among the most important global threats to biodiversity, wildlife and human health, and are associated with potential severe socioeconomic consequences (Daszak et al., 2000; Smith et al., 2006; Jones et al., 2008). Although combatting these risks is a main worldwide priority, our understanding of the processes underlying disease emergence still remains too limited for developing efficient prediction, prevention and management strategies. In humans, the majority of emerging pathogens originate as zoonoses from animal host populations in which they naturally circulate (Taylor et al., 2001; Jones et al., 2008). Thus, identifying the epidemiological features (e.g., prevalence, diversity, host specificity, geographic distribution) of zoonotic pathogen communities in their wild hosts, and the factors that influence pathogen occurrence in those communities, is as important to human health as it is to understanding the fundamentals of disease ecology (Garchitorena et al., 2017). Both extrinsic and intrinsic factors can contribute to the composition of natural pathogen communities within and between wild animal species, populations and individuals. Factors extrinsic to the hosts include geographic location, climate, periodicity of epidemic cycles and abiotic features influencing inter-specific transmission opportunities (e.g., (Harvell et al., 2002; Burthe et al., 2006; Poulin et al., 2012). Factors extrinsic to the pathogens such as host species identity, sex, age, and body condition as well as genetic and immunogenetic features have also been intensively studied (e.g., (Beldomenico et al., 2008; Streicker et al., 2010, 2013; Salvador et al., 2011; Charbonnel et al., 2014; Bordes et al., 2017)). Although less investigated, inter-specific ecological interactions (e.g., competition, facilitation) among pathogens within animal hosts are also likely to be an important intrinsic force in determining the composition of pathogen communities. Ecological interactions between free-living species are well- known to play a part in the distribution, abundance, and many other qualitative and quantitative features of populations; the application of this basic tenant of community ecology to pathogen incidence and expression of disease has become recognized as imperative for assessing both risks and potential benefits posed to human health, agriculture, wildlife, and conservation (Pedersen & Fenton, 2007). Simultaneous infection by multiple parasite species is ubiquitous in nature (Petney & Andrews, 1998; Cox, 2001; Moutailler et al., 2016), and prior infections can have lasting effects on future susceptibility via e.g., changes to host condition and behavior or through immune-mediated processes (Singer, 2010; Quiñones- Parra et al., 2016; Kumar et al., 2018; Karvonen et al., 2019). Interactions among co-circulating parasites may have important consequences for disease severity, pathogen transmission, host and pathogen evolution or co-evolution, and community-level responses to perturbations (Jolles et al., 2008; Telfer et al., 2010; Alizon et al., 2013; Seppälä & Jokela, 2016; Abbate et al., 2018). Consequences of interaction may be life-long, as exposure to pathogens circulating among juveniles have been found to be strongly associated with those experienced by adults (Fountain-Jones et al., 2019). Such interactions can also play a role in the consequences of pathogen emergence (e.g., emerging bacterial infection increasing susceptibility to an endemic virus (Beechler et al., 2015)). Henceforth, and through the advent of sequencing technologies in particular, it is now possible and essential to investigate disease emergence from a multi-host / multi-pathogen perspective (Galan et al., 2016), considering the potential influence of pathogen interactions on current and future disease distributions (Cattadori et al., 2008; Jolles et al., 2008; Budischak et al., 2015; Abbate et al., 2018). Rodent communities are relevant models for developing this community ecology approach to disease distribution and emergence. They harbor a wide variety of pathogenic taxa (Bordes et al., 2013; Pilosof et al., 2015; Koskela et al., 2016; Diagne et al., 2017) and are important reservoir hosts of agents of zoonoses that have severe implications for human health. Han et al. (2015) have revealed that about 10% of the 2277 extant rodent species are reservoirs of 66 agents of zoonoses, including viruses, bacteria, fungi, helminths, and protozoa. They also described 79 hyper-reservoir rodent species that could be infected by multiple zoonotic agents. Strong ecological interactions, such as facilitation and competition, have been shown in wild rodent populations among some of these zoonotic agents (Telfer et al., 2010), as well as between non-zoonotic agents and zoonotic agents (e.g., helminthes and bacteria (Carvalho- Pereira et al., 2019); helminthes and viruses (Guivier et al., 2014; Sweeny et al., 2020); helminthes and protozoa (Knowles et al., 2013)). In addition, rodents share a number of habitats with humans, including 2 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 urban settings, agricultural lands, and forests, providing opportunities for human-rodent contact and pathogen transmission (Davis et al., 2005). Describing the distribution and composition of natural pathogen communities in rodent populations, and determining the drivers behind pathogen associations, is imperative for understanding the risks they may pose for public health. In this study, we analyzed the pathogen communities carried by rodent communities in a rural area of northern France, a region known to be endemic for several rodent-borne diseases including nephropathia epidemica (Puumala orthohantavirus (Sauvage et al., 2002)) and borreliosis (Razzauti et al., 2015). We investigated exposure histories (via the presence of antiviral antibodies) for several viruses (hantaviruses, cowpox virus, lymphocytic choriomeningitis virus, Tick-borne encephalitis virus) and current or recent exposure to bacterial pathogens (using high-throughput 16S metabarcoding of host splenic tissue). We described the pathogens detected, their prevalence in the community and their individual distributions among host populations. We then tested the role of extrinsic factors (e.g., habitat, host species, host age) in explaining variation in pathogen distributions, and for associations (non-random co-infection frequencies) between pathogens that might indicate intrinsic drivers (e.g., competition, facilitation) of pathogen community composition. We expected that host species and habitat would be the most important factors structuring pathogen community composition because most pathogens are largely host-specific, but those sharing habitats should also share opportunities for transmission (Davies & Pedersen, 2008). After accounting for extrinsic factors, we expected to retrieve several pathogen- pathogen associations previously identified in the literature. This included i) positive associations between cowpox virus and Bartonella infections (Microtus agrestis, (Telfer et al., 2010)); ii) positive associations between distinct Mycoplasma species in mammalian hosts (Sykes et al., 2008; Tagawa et al., 2012; Fettweis et al., 2014; Volokhov et al., 2017)); iii) associations between Bartonella and hemotropic Mycoplasma species (both positive and negative associations, as well as experimental demonstration of dynamic interactions, have been described in Gerbillus andersonii (Eidelman et al., 2019)). Lastly, we also expected to find previously undescribed associations due to the large bacteria and rodent dataset included in our study. All these associations were likely to differ between host species, as differences in host specificity are also likely to be accompanied by differences in transmission dynamics and host responses to infection (Davies & Pedersen, 2008; Singer, 2010; Dallas et al., 2019). 2. Materials and methods 2.1. Study area and host sampling Rodent sampling was conducted over two years (Autumn 2010 & 2011) in rural habitats surrounding two villages (Boult-aux-Bois and Briquenay) in the Ardennes region of northern France (previously described in (Gotteland et al., 2014)). Sex and age class (based on specific body measurements and classed as ‘adult’ for sexually mature animals and ‘juvenile’ for both juveniles and sexually immature sub- adults) were recorded for each animal, a blood sample was taken for serological analyses, and animals were then euthanized using isofluorane inhalation. Spleens were taken and stored in RNAlater Stabilizing Solution (Invitrogen) at -20°C. Species captured from the two sites included (family: Cricetidae) 195 Arvicola scherman (montane water vole), 10 Microtus agrestis (field vole), 66 Microtus arvalis (common vole), 203 Myodes glareolus (bank vole); and (family: Muridae) 43 Apodemus flavicollis (yellow-necked mouse), 156 Apodemus sylvaticus (wood mouse), 32 Rattus norvegicus (brown rat). These seven focal host species were collected from traps placed in distinct landscapes (henceforth referred to as host ‘habitats’) (Supplemental Materials Figure S1): R. norvegicus were found uniquely on farms, Ar. scherman and Mi. arvalis were found almost entirely in meadows, and the five remaining species occupied both forests and hedgerows. Demographic differences between host species were observed for sex (e.g., male bias in Ap. sylvaticus; Figure S2A) and age classes (e.g., relative abundance of juveniles in Mi. arvalis and Ap. sylvaticus hosts; Figure S2B). Five Microtus subterraneus (European pine vole) and one each of three additional host species (one cricetid one echimyid and one murid) were also found in these communities, but excluded from analyses due to their rarity; these rare (non-focal) hosts and their pathogens are described in Supplemental Materials Appendix 1. Jessica L. Abbate et al. 3 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 2.2. Detecting virus exposure and bacterial infection Among the 713 rodents sampled for this study, indirect fluorescent antibody tests (IFATs; see for details (Kallio-Kokko et al., 2006)) were successfully performed on 677 animals to detect immunoglobulin G (IgG) specific to or cross-reacting with cowpox virus (CPXV, Orthopoxvirus), Puumala or Dobrava- Belgrade virus (respectively PUUV and DOBV, Orthohantavirus, collectively referred to henceforth as “hantavirus”), lymphocytic choriomeningitis virus (LCMV, Mammarenavirus), and Tick-borne encephalitis virus (TBEV, Flavivirus). We refer to these antiviral antibody tests as indicating a history of past exposure, but antibodies against hantavirus and LCMV also likely indicate continued chronic infection. In contrast, current or very recent exposure to bacterial infection was tested via 16S rRNA gene amplicon sequencing of splenic tissue, giving no indication of past exposure history. The spleen was chosen because this organ is known to filter microbial cells in mammals, allowing the detection of a wide array of pathogenic and zoonotic agents. Funding was available to test for bacteria in just half of the animals, chosen haphazardly to equally represent all host species, study sites and years, resulting in successful analysis for 332 rodents (see Figure 1 for a breakdown of number of individuals sampled per focal host species). For each individual animal, the DNA from splenic tissue was extracted using the DNeasy Blood & Tissue kit (Qiagen) following the manufacturer recommendations. Each DNA extraction was analyzed twice independently. We followed the method described in Galan et al. (2016) to perform PCR amplification, indexing, pooling, multiplexing, de-multiplexing, taxonomic identification using the SILVA SSU Ref NR 119 database as a reference (http://www.arb-silva.de/projects/ssu-ref-nr/). Briefly, DNA samples were amplified by PCR using universal primers targeting the hyper variable region V4 of the 16S rRNA gene (251 bp) and sequencing via Illumina MiSeq. The V4 region has been proven to have reasonable taxonomic resolution at the genus level (Claesson et al., 2010). A multiplexing strategy enabled the identification of bacterial genera in each individual sample (Kozich et al., 2013). Data filtering was performed as described in Galan et al. (2016) to determine presence/absence of bacterial infections (summarized in Figure S3). Briefly, we discarded all bacterial OTUs containing fewer than 50 reads in the entire dataset and animals for which one or both individual PCR samples produced fewer than 500 reads. A bacterial OTU was considered present in an animal if the two independent PCR samples were both above a threshold number of reads, defined as the greater of either 0.012% of the total number of reads in the run for that OTU (i.e., filtering using the rate of indexing leak) or the maximum number of reads for that OTU in any negative control sample (i.e., filtering using the presence of reads in the negative controls due to contaminations) (Galan et al., 2016). We removed chimera (Ashelford et al., 2006), using the Uchime program implemented in mothur, and manually checked OTUs representing suspected chimera not identified by the program. For each OTU suspected as pathogenic, Basic Local Alignment Search Tool (BLAST) searches of the most common sequences were conducted to infer species identity where possible. Given that rare, low abundance taxa tend to show high dissimilarity between sample replicates (Smith & Peay, 2014) we assumed that only OTUs with at least 500 reads across all animals in the dataset were considered reliably detectable, allowing us to assign absent status to these OTUs in animals failing to meet the criteria for OTU presence. Thus, only OTUs for which there were at least 500 reads across all animals in the dataset (for which present and absent statuses could be assigned), and where reasonable certainty of pathogenicity could be established from the literature (see Table 1), were considered in analyses of the pathogen community. 2.3. Statistical Methods All statistical analyses were implemented in R version 3.2.2 (R Core Team, 2015). Throughout our analyses, we refer to simultaneous bacterial infections as “co-infection”, while analyses involving simultaneous presence of antiviral antibodies and bacterial infection are referred to as evaluating “co- exposure”. While we can only be sure that antiviral antibodies represent past exposure, we cannot rule out simultaneous viral and bacterial “co-infection”, particularly for viruses known to cause chronic infections. Likewise, detection of current (or very recent) bacterial infection, particularly for taxa known to cause chronic infections, cannot tell us how long the animal has carried the infection. Thus, we refrain from assuming sequence of infection for statistical tests unless specified by an a priori hypothesis from the literature. 4 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 2.3.1. Testing for extrinsic drivers of pathogen community composition across the rodent community We analyzed pathogen community composition across the whole rodent community. We use the term “pathogen community” to refer to the group of viruses and pathogenic bacteria for which we had the means to include, which was not exhaustive; thus measures of diversity are to be considered relative and not absolute. We first estimated pathogen community richness using the Shannon diversity index (alpha diversity) considering pathogenic bacterial OTUs and antiviral antibodies found in each study year, study site, habitat, host species, host sex and host age group (default options (natural logarithm Shannon index) in ‘diversity’ function from the vegan package). We evaluated a linear regression model using analysis of deviance (‘lm’ and ‘drop1’ functions from the basic stats package) to test for significance of differences in pathogen diversity due to the fixed factors listed above after first correcting for all other factors in the model (marginal error tested against the F-distribution). Post-hoc comparisons and correction for multiple tests were performed using function ‘TukeyHSD’ from package stats and ‘HSD.test’ from package agricolae to group factor levels that were not significantly different. An additional Akaike Information Criterion (AIC)-based model selection analysis was performed to assess any qualitative influence of spurious predictors on these comparisons using the ‘glmulti’ and ‘weightable’ function from package glmulti, The impact of host species diversity (Shannon diversity with Chao’s estimator correction using ‘Shannon’ in package entropart) on pathogen species diversity was tested for by correlation (‘cor.test’, package stats) across each year x site x habitat community. We next tested for differences in pathogen community composition (beta diversity) between host species, habitats, study sites, years, age classes, and sexes by applying a permutational multivariate analysis of variance (PERMANOVA) on a Bray-Curtis dissimilarity matrix (‘adonis2’ function in the vegan package). To explore how intrinsic factors (pathogen-pathogen associations) contributed to the structure of the pathogen community, we used multiple correspondence analysis (MCA) to reduce variance in presence/absence of each bacterial pathogen species and antiviral antibody, implemented with the function ‘MCA’ in the FactoMineR package and visualization tools found in the factoextra package. This produces a set of quantitative and orthogonal descriptors (dimensions) describing the pathogen community composition, revealing correlated variables. With each MCA dimension as a continuous dependent response variable, we then evaluated linear regression models using analysis of deviance with post-hoc comparisons (as detailed above) to understand how the variation described by each MCA dimension was influenced by the extrinsic factors. 2.3.2. Testing for associations between co-circulating pathogens Because we identified a large number of pathogens, the number of potential association combinations to consider was excessively high, especially with regard to the relatively small number of rodents sampled. We therefore decided to test the significance only of those associations (i) clearly suggested by the community-wide MCA or (ii) previously described in the literature: positive association between Bartonella spp. and CPXV (Telfer et al. 2010), positive associations between Mycoplasma species (Sykes et al., 2008; Tagawa et al., 2012; Fettweis et al., 2014; Volokhov et al., 2017), and both positive (Kedem et al., 2014; Eidelman et al., 2019) and negative (Cohen et al., 2015) associations between Bartonella spp. and hemotropic Mycoplasma species. Given the a priori assumption that associations would differ between host species, we analyzed each host species separately; where evidence suggested no significant differences between host species (non-significant variation in the MCA dimension among host species or non-significant host species identity x explanatory pathogen term in logistic regressions), we pooled individuals into a single analysis to gain statistical power. We tested the significance of each association using both association screening (SCN) analysis (Vaumourin et al., 2014) and multiple logistic regression analysis (GLMs, modeling the binomial ‘presence/absence’ status of each pathogen as a function of the occurrence of other pathogens) on the subset of host species in which the pathogens were found to circulate. We first performed SCN analysis, as this approach is among the most suitable for detecting pathogen associations in cross-sectional studies (Vaumourin et al. 2014). Briefly, given the prevalence of each pathogen species in the study population, SCN analysis generates a simulation-based 95% confidence envelope around the expected frequency of each possible combination of concurrent infection status (a total of 2NP combinations, where NP = the number of pathogen species) under the null hypothesis of random pathogen associations. Observed frequencies of co-infection combinations falling above or below this envelope are considered to occur Jessica L. Abbate et al. 5 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 more or less frequently, respectively, than in 95% of the random simulations. Significance of the association is given as a p-value, calculated as the number of instances in which the simulated co- infection frequency differed (above or below the upper or lower threshold, respectively) from the observed frequency divided by the total number of simulations (Vaumourin et al., 2014). The benefit of the SCN approach is a relatively high level of statistical power and the ability to identify precisely which combinations of pathogens occur outside the random expectations (Vaumourin et al., 2014). However, the SCN is sensitive to heterogeneity in the data due to extrinsic factors (e.g., host specificity, or structuring in space, time, age or sex), which can both create and mask true associations. A multiple logistic regression approach was thus also systematically employed, as it has the benefit of explicitly taking into account potentially confounding extrinsic factors. Binomial exposure (presence/absence of either bacterial infection or antiviral antibodies) to a single pathogen was set as the dependent variable with exposure to the hypothetically associated pathogen(s) treated as independent explanatory variable(s) and extrinsic factors (host sex, host age, study year, study site, and where appropriate, habitat) were specified as covariates using function ‘glm’ in the stats package with a binomial logit link. When the multiple host species were involved, we tested an interaction term (host species identity x explanatory pathogen), and either (if p < 0.05) performed separate analyses for each host species or (if p ≥ 0.05) simply added host species identity as another covariate in the model. When there was no a priori assumption concerning timing of exposure (e.g., antiviral antibody presence is more likely to affect current acute bacterial infection than the reverse), the occurrence of each pathogen involved in a given association was set as the dependent variable in reciprocal GLMs. In contrast to the MCA dimensions above, a model selection step was first performed using the ‘glmulti’ and ‘weightable’ function from package glmulti to find the best model among top-ranking models (small-sample size corrected Akaike Information Criterion (AICc) score less than 2 + lowest AICc) that retained all explanatory pathogens of interest, in an effort to limit the appearance of associations due to the inclusion of spurious predictors. Statistical significance of the association was then assessed after first correcting for all remaining covariates in the best model using the ‘drop1’ function (likelihood ratio tests via single-term deletions compared to the full model). Despite a large number of a priori hypotheses, we regarded a p-value of < 0.05 as significant due to the very low number of individuals of each host species sampled. Though conceivably important, we also did not have sufficient power to test for additional interaction terms. 2.3.3. Evaluating false discovery Given the large number of significance tests performed on this single dataset, we compiled all relevant p-values (N=77) and applied a Benjamini-Hochberg correction procedure to estimate how many of the significant results may fall within the false discovery zone (using function ‘p.adjust’ in the stats package). Where model selection was performed, p-values were taken from full models prior to model selection. Among tests of positively correlated hypotheses (e.g., pairwise tests of intrinsic pathogen- pathogen associations N=20), only one of the p-values from the pair was included in calculating the false discovery rate (selected randomly). SCN analysis results were not included, because they were also expected to be positively correlated with the logistic regressions, and because the method inherently performs correction for multiple tests. Between hypotheses, the data were often composed of different non-overlapping subsets of varying sizes, and sample sizes varied widely with some being very small. Thus, application of this procedure likely indicates a conservative (low) estimate for how many null hypotheses should truly be rejected. 3. Results 3.1. Taxonomic identification and prevalence of pathogens 3.1.1. Viral exposure The most abundant virus detected was CPXV, with 222 (32.8%) positive sera of the 677 animals tested for anti-CPXV antibodies. It was detected in all focal host species. However, significant variation in prevalence was observed between focal host species (highly prevalent (43-70%) in Ar. scherman, Mi. agrestis, and My. glareolus; Figure 1; χ2 = 119.5, df = 6, p < 10-15). 6 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Table 1 - BLAST search results for OTUs suspected of belonging to pathogenic genera. References are available in Supplemental Materials Appendix 3. Infecting species identity OTU Number Number of Reads Genbank Accession Number BLAST results (% identity) Pathogen Code Reference Pathogenic taxa, reliably detectable Bartonella spp. Otu00001 6353372 MT027154 100% Bartonella grahamii (AB426637) from wild North America rodents; 99%-100% identity to many other pathogenic Bartonella species. Bartonella Deng et al., 2012 Brevinema spp. Otu00123 5603 MT027155 97% Brevinema andersonii (NR_104855) type sequence, infectious spirochaete of short-tailed shrew and white-footed mouse in North America Brevinema Defosse et al., 1995 Candidatus Neoehrlichia mikurensis Otu00039 18358 MT027156 100% Candidatus Neoehrlichia mikurensis (KF155504) tick-borne rodent disease, opportunitstic in humans Neoehrlichia Andersson & Raberg, 2011 Mycoplasma ravipulmonis Otu00054 6086 MT027164 100% Mycoplasma ravipulmonis (AF001173), grey lung disease agent in Mus musculus; 88% M. orale from humans (LR214940) Myco10 Petterson et al., 2000 Mycoplasma spp. Otu00004 845971 MT027157 99% identity to uncultured Mycoplasma species (KU697344) from small rodents in Senegal and uncultured eubacterium (AJ292461) from Ixodes ticks; 95% (KM538694) and 94% (MK353834) identity to uncultured hemotropic Mycoplasma species in European and South American bats Myco1 Galan et al., 2016 Mycoplasma spp. Otu00003 426034 MT027158 99% Mycoplasma haemomuris (AB758439) from Rattus rattus Myco2 Conrado et al., 2016 Mycoplasma spp. Otu00006 106443 MT027159 99% uncultured Mycoplasma spp. (KP843892) from leopards in South Korea; 99% identity to uncultured Mycoplasma (KT215637) from rodents in Brazil Myco3 Goncalves et al., 2015 Mycoplasma spp. Otu00010 72724 MT027160 99% Mycoplasma coccoides comb. nov. (AY171918); 97% Candidatus Mycoplasma turicensis (KJ530704) from Indian mongoose Myco4 Conrado et al., 2015 Mycoplasma spp. Otu00005 165095 MT027161 100% Mycoplasma haemomuris-like undescribed species (KJ739312) from Rattus norvegicus Myco5 Alabi et al., 2020 Mycoplasma spp. Otu00007 92237 MT027162 99% uncultured Mycoplasma spp. (KC863983) from Micromys minutus (eurasian harvest mouse) in Hungary; 98% M. coccoides comb. nov.(AY171918) Myco6 Conrado et al., 2015 Mycoplasma spp. Otu00012 39767 MT027163 93% uncultured Mycoplasma species (KU697341) of mice in Senegal; 92% Mycoplasma spp. (KP843892) from leopards in South Korea, 91% Mycoplasma haemomuris (AB820289) in rats Myco7 Galan et al., 2016 Mycoplasma spp. Otu00015 31528 MT027165 98% uncultured Mycoplasma spp. (KT215632) from wild rodent spleen in Brazil; 95% uncultured Mycoplasma spp. (KF713538) in little brown bats Myco8 Ricardo Gonçalves et al., 2015 Mycoplasma spp. Otu00049 40125 MT027166 96% uncultured Mycoplasma spp. from Brazilian rodents (KT215638) and S. Korean leopard (KP843892) Myco9 Ricardo Gonçalves et al., 2015 Orientia spp. Otu00111 876 MT027167 97% Orientia tsutsugamushi (KY583502) from humans in India, zoonotic Rickettsial pathogen (causes scrub typhus) Orientia Paris et al., 2013 Rickettsia spp. Otu00008 72098 MT027168 98% Rickettsia japonica (MF496166) which causes Japanese spotted fever, R. canadensis (NR_029155) & R. rhipicephali (NR_074473) type strains Rickettsia Yamamoto et al., 1992 Spiroplasma spp. Otu00093 4738 MT027169 95% uncultured Spirolasma spp. (KT983901) from Ixodes tick on a dog; 94% identity to type strain of Spiroplasma mirum (NR_121794), the agent of suckling mouse cataract disease; Spiroplasma ixodetis causes similar disease in humans. Spiroplasma Cisak et al., 2015 Pathogenic taxa, not reliably detectable Arcobacter cryaerophilus Otu00296 403 MT027170 100% Arcobacter cryaerophilus (CP032825) emerging enteropathogen in humans, zoonotic, pathogenic in rats Arcobacter Vandamme et al., 1992 Jessica L. Abbate et al. 7 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Infecting species identity OTU Number Number of Reads Genbank Accession Number BLAST results (% identity) Pathogen Code Reference Borrelia miyamotoi Otu00318 419 MT027171 100% Borrelia miyamotoi (CP010308) in humans and Ixodes, zoonotic pathogen Borrelia1 Krause et al., 2015 Borrelia spp. Otu00514 206 MT027172 96% Borrelia sp. nov “Lake Gaillard” in Peromyscus leucopus (AY536513), 95% B. hermsii (MF066892) from tick (Ornithodoros hermsi) bites in humans Borrelia2 Bunikis et al., 2005 Borrelia afzelii Otu00071 78 MT027173 100% Borrelia afzelii (CP009058) human pathogen closely related (98%) to B burgdorferii (positive control sequence) Borrelia3 Schuler et al., 2015 Leptospira spp. Otu01015 257 MT027174 100% several pathogenic Leptospira species from mammals, e.g., L. interrogans (LC474514) in humans Leptospira Vincent et al., 2019 Mycoplasma pulmonis Otu00771 255 MT027175 100% Mycoplasma pulmonis (NR_041744) chronic respiratory pathogen of mice and rats Myco0771 Piasecki et al., 2017 Mycoplasma spp. Otu04125 164 MT027176 90% Mycoplasma ravipulmonis (AF001173), grey lung disease agent in Mus musculus; 80% M. phocidae from California sea lions (DQ521594) Myco4125 Petterson et al., 2000 Eukaryotic family Sarcocystidae Otu00056 8684 Not submitted* 97% similar to plastid small ribosomal unit of Hyaloklossia lieberkuehni (AF297120), a parasitic protazoa of European green frog; 96% Neospora caninum (MK770339) & Sarcocystis muris (AF255924); 95% Toxoplasma gondii (TGU28056) Sarcocystidae1 Calarco & Ellis, 2020 Eukaryotic family Sarcocystidae Otu00191 3678 Not submitted* 92% Neospora caninum (MK770339) parasite; 90% Toxoplasma gondii (U87145) zoonotic pathogen Sarcocystidae2 Calarco & Ellis, 2020 Eukaryotic family Sarcocystidae Otu00254 1219 Not submitted* 98% Sarcocystis muris (AF255924) coccidian parasite first found in mice Sarcocystidae3 Orosz, 2015 Uncertain pathogenicity, reliably detectable Corynebacterium xerosis Otu00050 1853 MT027177 100% Corynebacterium xerosis (MH141477), only opportunistic infections identified Corynebacterium Bernard, 2012 Dietzia spp. Otu00102 2626 MT027178 100% Dietzia spp. e.g., D. aurantiaca (MK25331); common contaminant; opportunistic in humans; thought to out-compete Trypanosomes Dietzia Kampfer et al., 2012 Helicobacter spp. Otu00013 34894 MT027179 96% homology to Helicobacter suncus (AB006147) isolated from shrews with chronic gastritis; 95% identity to type speciment for H. mustelae (NR_029169) which causes gastritis in ferrets; but could be normal gut flora Helico1 Goto et al., 1998 Helicobacter spp. Otu00025 8702 MT027180 97% similar to Helicobacter trogontum (AY686609) and H. suncus (AB006147), both enterohepatic Helicobacter spp. associated with intestinal diseases Helico2 Goto et al., 1998 Helicobacter spp. Otu00087 2303 MT027181 99% identical to Helicobacter aurati (NR_025124.1), a pathogen of Syrian hamsters; 98% identical to H. fennelliae (GQ867176), a human pathogen Helico3 Patterson et al., 2000 Helicobacter spp. Otu00128 1178 MT027182 99% Helicobacter winghamensis (AF363063), associated with gastroenteritis in humans ; however, minor sequences were 100% identical to H. rodentium (AY631957) which is only associated to gastritis in rodents when coinfecting with other Helicobacter strains Helico4 Melito et al., 2001 Neisseria spp. Otu00612 780 MT027183 97% uncultured Neisseria spp. associated with human prostatitis (HM080767) and cataracts (MG696979), but undistinguishable from environmental samples and healthy flora (e.g., JF139578) Neisseria1 Genbank, unpublished Pasteurellaceae Otu00129 1430 MT027184 100% uncultured bacterium (MN095269) of mouse oral flora; 99% Muribacter muris (KP278064) of unknown pathogenicity, water fowl pathogen Avibacterium gallinarum (AF487729), and cattle respiratory disease agent Mannheimia haemolytica (CP017491) Pasteurella1 Nicklas et al., 2015 Pasteurellaceae Otu00203 521 MT027185 99% Aggregatibacter aphrophilus (LR134327) and Haemophilus parainfluenzae (CP035368) opportunistic pathogens but otherwise part of normal flora Pasteurella2 Genbank, unpublished 8 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Infecting species identity OTU Number Number of Reads Genbank Accession Number BLAST results (% identity) Pathogen Code Reference Rickettsiella spp. Otu00187 592 MT027186 99%-100% identity to several endosymbionts of insects, eg. uncultured Diplorickettsia spp. in sand flies (KX363696), Rickettsiella spp. in Ixodes ticks (KP994859); 99% identity to Rickettsiella agriotidis (HQ640943) pathogen of wireworms Rickettsiella Duron et al., 2015 Streptococcus spp. Otu00115 1681 MT027187 99% Streptococcus hyointestinalis from intestines of swine (KR819489) Streptococcus Genbank, unpublished Yersinia spp. Otu00041 7420 MT027188 A heterogeneous OTU some major sequences 100% Yersinia spp. and Serratia spp., including pathogenic zoonotic bacteria (e.g., Y. pestis NR_025160) and non-pathogenic endosymbionts of plants (NR_157762) ; some major sequences 100% Pantoea agglomerans (MN515098) opportunists Yersinia Kim et al., 2003 Uncertain pathogenicity, not reliably detectable Fusobacterium spp. Otu00791 108 MT027189 100% Fusobacterium ulcerans (CP028105) from tropical foot ulcers in humans; but also 100% identity with other fecal isolates of unknown pathogenicity in mammals (e.g., F. varium LR134390) Fusobacterium Genbank, unpublished Neisseria spp. Otu00454 148 MT027190 98% uncultured microbiota of bat mating organs (KY300287), 98% Simonsiella muelleri commensal from human saliva (AF328145); 97% Kingella kingae (MF073277) pathogen in humans Neisseria2 Genbank, unpublished Treponema spp. Otu00235 348 MT027191 93% uncultured rumen Treponema spp. (AB537611) Treponema Bekele et al., 2011 Williamsia spp. Otu00614 274 MT027192 100% Williamsia phyllospherae (MG205541) and Williamsia maris (NR_024671), closely related to opportunistic pathogens in humans Williamsia Genbank, unpublished * No reference specimen for identification of Eukaryote Anti-hantavirus antibodies were detected in 16 animals (2.4%), and were significantly structured among host species (with exposure highest in Mi. arvalis (9.7%), R. norvegicus (3.3%) and My. glareolus (3.1%); Figure 1; χ2 = 19.4, df = 6, p = 0. 0036). Anti-LCMV antibodies were detected in two Mi. arvalis individuals (Figure 1). No animals were positive for anti-TBE antibodies. 3.1.2. Bacterial pathogens Out of 952 bacterial OTUs represented by at least 50 reads in the dataset, 498 were considered positive in at least one animal after data filtering (presented in Supplemental Materials Table S1). Two OTUs (00024 & 00037) identified as Bartonella with low bootstrap values (74 and 92 respectively) appeared to represent chimeric sequences between the two highly amplified genera (Bartonella and Mycoplasma) in co-infected samples. Two OTUs (00009 & 00117) which were unclassified but which had a large number of reads in positive animals were also found to represent chimeric sequences between the two genera, despite high bootstrap values (100). Three additional chimeric Mycoplasma OTUs with under 500 reads were also excluded (OTUs 00076, 00159, and 00316). Two OTUs (00002 & 00059) were found to be redundant with OTUs Myco1 and Myco3, respectively, and two more (00134 & 00220) were chimera between Myco OTUs. These 11 OTUs were manually removed from the database, and are not included in Table S1. We identified 43 OTUs belonging to bacterial genera with members known or thought to be pathogenic in mammals (Table 1). After BLAST queries, we found 16 of these OTUs (representing 7 distinct genera) which could be considered as reliably detectable pathogens in the focal host species (Figure 1). An additional 24 OTUs were considered potentially pathogenic but excluded from analyses because they were only observed in rare host species, because presence-absence could not be reliably established due to a low total number of reads (<500 in the dataset, e.g., Borrelia spp. and Leptospira spp.), because we could not rule out contamination by natural sources of non-pathogenic flora during dissection (e.g., Helicobacter spp., Streptococcus spp.) or by known contaminants of sequencing reagents (e.g., Williamsia spp.; (Salter et al., 2014)), or because their identity to a pathogenic species was uncertain due to insufficient genetic variation at the 16S rRNA locus (e.g., Yersinia spp.) (Table 1). We also identified three OTUs belonging to the eukaryotic family Sarcocystidae (98% sequence similarity to the Jessica L. Abbate et al. 9 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 coccidian parasite Sarcocystis muris); though each OTU was represented by >500 reads, there are currently no data on the reliability of this method for detection (Table 1). Individual infection status for each of these OTUs is given in Table S2. Figure 1 - Pathogen occurrence across the rodent species community. The 16 reliably detectable pathogenic OTUs included Bartonella spp., 10 Mycoplasma spp. OTUs, Rickettsia canadensis, “Candidatus Neoehrlichia mikurensis”, Orientia spp., Brevinema andersonii, and Spiroplasma spp. Phylogenetic analysis including published sequences from BLAST queries revealed that the 10 Mycoplasma spp. OTUs belonged to three distinct species: Myco. haemomuris (Myco1-3,5,7-9), Myco. coccoides (Myco4 and Mco6), and “Candidatus Myco. ravipulmonis” (Myco10) (Figure S4). In general, these bacterial infections were present in all but 30 of the 332 animals tested (91.0% prevalent), and were not concentrated in any particular focal host species (χ2 = 9.7, df = 6, p = 0.139). Prevalence of each pathogen in each focal host species is presented in Figure 1. 3.2. Extrinsic drivers of pathogen community diversity and composition within rodent community 3.2.1. Analyses of pathogen diversity We found evidence for the co-circulation of between 3 (in Mi. agrestis) and 12 (in My. glareolus) pathogen taxa per host species across the rodent community (Figure 1). Using multiple regression analysis on the Shannon diversity index, we found that marginal mean pathogen diversity differed significantly between host species (F5,69 = 6.86, p < 10-4) and habitats (F2,69 = 4.97, p = 0.0096), and it was significantly higher in adults than in juveniles (F1,69 = 21.43, p < 10-4). Pathogen diversity did not, however, differ between study sites, years, or host sexes (Figure 2, Table S3). Post-hoc Tukey tests showed that after correcting for all other factors in the model, meadow habitats had higher diversity of pathogen exposure than forest habitats, and the diversity of pathogen communities in host species fell along a continuum between My. glareolus (high) and Ar. scherman (low) extremes (Figure 2, Table S3). The results were qualitatively identical when non-significant predictors were removed from the model following model selection (Figure S5). Host species diversity in each community (year x site x habitat) was positively correlated with pathogen diversity (r = 0.62, t = 2.5, df = 10, p = 0.032). To understand the relative pathogen diversity of R. norvegicus hosts, not included in the above analysis because they were entirely confounded with farm habitats, we analyzed a modified model −0.20.00.20.40.60.81.01.2 0.0 0.2 0.4 0.6 0.8 1.01 185 2 8 3 59 82 105 1 22 21 40 111 30 Hantavirus - 2 - - - 6 5 1 - - - 1 - 1 CPXV - 80 1 6 - 11 49 53 1 6 1 3 7 4 LCMV - - - - - 2 - - - - - - - - 0 70 2 6 3 47 27 35 0 20 16 16 54 30 Bartonella - 57 2 4 2 41 23 19 - 18 14 15 44 3 Myco1 - - - 4 3 38 3 4 - 1 2 - - - Myco2 - - - - - - 1 1 - 1 1 3 9 27 Myco3 - - - - - 9 9 12 - - - - - - Myco5 - - - - - - - 1 - - - - - 16 Myco7 - - - - - - - - - 1 2 - 2 5 Myco8 - - - - - 9 - - - - - - - - Myco9 - - - - 1 3 - 3 - - - - - - Myco4 - - - - - - - - - - - 2 16 - Myco6 - 1 - - - - 1 2 - 1 1 - - 10 Myco10 - - - - - - - - - - - - - 6 Rickettsia - 2 - - - 4 - - - - - - - 2 Neoehrlichia - 2 - - - v 1 3 - - - - 2 - Orientia - 2 - - - 6 - - - - - - - - Brevinema - - - - - - 2 - - - - - - - Spiroplasma - - - - - - 1 1 - - - - - - Mycoplasma haemomuris Viruses (antibodies) Mycoplasma coccoides Candidatus M. ravipulmonis Meadow Farm Landscape of trapping location Prevalence 90% <1% 20% 50% Cricetidae hosts Muridae hosts A rv ic ol a sc he rm an M ic ro tu s ag re st is M ic ro tu s ar va lis M yo de s gl ar eo lu s A po de m us fla vi co lli s A po de m us sy lv at ic us Ra tt us no rv eg ic us Forest Hedgerow Number tested for bacteria Number tested for antibodies Beyond typical host habitat 10 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 excluding host habitat and including all seven focal host species. Post-hoc Tukey tests from this model showed that R. norvegicus hosts had the second most diverse pathogen community (Table S4). Figure 2 - Extrinsic drivers of pathogen diversity in a rodent species community. Differences in Shannon diversity index was tested on marginal means for each factor in the multiple regression model. Different letters signify statistically significant differences at p < 0.05, with post-hoc Tukey adjustments for multi-level factors. We also found an enormous amount of both bacterial co-infections and concurrent history of viral exposures (Figures 3A, 3B). The percentage of animals co-infected with two or more reliably detectable pathogenic bacterial OTUs among all those infected in each host species ranged between 84.4% (in Mi. arvalis) and 10.5% (in Ar. scherman). This co-infection frequency was significantly correlated with the diversity (Shannon Index) of bacteria circulating in each rodent species (Figure 3C; analysis of deviance Pseudo-R2 = 0.69, p < 10-11, calculated using logistic regression weighted by the number of infected animals per species). Bacterial co-infections were more frequent than expected in Mi. arvalis, and less frequent than expected in My. glareolus (according to Cook’s Distance, Figure S6A). Results were similar when co-occurrence of antiviral antibodies was considered along with bacterial OTU exposure (Figure 3D; Pseudo-R2 = 0.65, p < 10-7). While Mi. arvalis had both more bacterial co-infections and slightly more pathogen co-exposures than expected based on pathogen diversity, other outliers differed between the two measures (Figure S6B): My. glareolus co-exposure frequencies were not lower than expected, and both Apodemus species had lower than expected co-exposures. Host species diversity in each community did not correlate with bacterial co-infection (r = 0.33, t = 1.11, df = 10, p = 0.29) or pathogen co-exposure (r = 0.022, t = 0.071, df = 10, p = 0.95) frequencies. We also found an enormous amount of both bacterial co-infections and concurrent history of viral exposures (Figures 3A, 3B). The percentage of animals co-infected with two or more reliably detectable pathogenic bacterial OTUs among all those infected in each host species ranged between 84.4% (in Mi. arvalis) and 10.5% (in Ar. scherman). This co-infection frequency was significantly correlated with the -0.55 -0.5 -0.45 -0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 A r. sc he rm an M i. ag re st is M i. ar va lis M y. g la re ol us A p. fl av ic ol lis A p. s yl va tic us Pa th og en d iv er si ty (m ar gi na l m ea n Sh an no n In de x) bc bc bc -0.55 -0.5 -0.45 -0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 Fo re st H ed ge ro w s M ea do w s b a ab -0.55 -0.5 -0.45 -0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 A du lt Ju ve ni le a b A. Host species B. Habitat C. Host age ab a c Jessica L. Abbate et al. 11 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 diversity (Shannon Index) of bacteria circulating in each rodent species (Figure 3C; analysis of deviance Pseudo-R2 = 0.69, p < 10-11, calculated using logistic regression weighted by the number of infected animals per species). Bacterial co-infections were more frequent than expected in Mi. arvalis, and less frequent than expected in My. glareolus (according to Cook’s Distance, Figure S6A). Results were similar when co-occurrence of antiviral antibodies was considered along with bacterial OTU exposure (Figure 3D; Pseudo-R2 = 0.65, p < 10-7). While Mi. arvalis had both more bacterial co-infections and slightly more pathogen co-exposures than expected based on pathogen diversity, other outliers differed between the two measures (Figure S6B): My. glareolus co-exposure frequencies were not lower than expected, and both Apodemus species had lower than expected co-exposures. Host species diversity in each community did not correlate with bacterial co-infection (r = 0.33, t = 1.11, df = 10, p = 0.29) or pathogen co-exposure (r = 0.022, t = 0.071, df = 10, p = 0.95) frequencies. Figure 3 - Bacterial co-infection and co-exposure patterns among host species. 3.2.2. Analyses of pathogen community composition Many pathogen taxa were found only in a single host species (Mycoplasma haemomuris OTU Myco8, “Candidatus Mycoplasma ravipulmonis” (Myco10), Brevinema spp., Spiroplasma spp., LCMV), and each C_Ar_sche C_Mi_arva M_Ap_flav M_Ra_norv 0 10 20 30 40 50 60 C_Ar_sche C_Mi_arva M_Ap_flav M_Ra_norv 0 10 20 30 40 50 60 70 0.5 1.0 1.5 2.0 0. 0 0. 2 0. 4 0. 6 0. 8 1. 0 Relative number of co−circulating pathogen taxa (Shannon Index) % in div idu als co −e xp os ed (a m on g all e xp os ed a nim als ) 0.5 1.0 1.5 2.0 0. 0 0. 2 0. 4 0. 6 0. 8 1. 0 Relative number of co−circulating bacterial taxa (Shannon Index) % in div idu als co inf ec te d (a m on g all in fe cte d an im als ) N um be r o f a ni m al s 0 1 2 3 4 5 Host species Ar. scherman Mi.agrestis Mi.arv lis My.glareolus Ap.sylvaticus Ap.flavicollis R.norvegicus N um be r o f a ni m al s 0 1 2 3 4 5 Host species Ar. scherman Mi.agrestis Mi.arvalis My.glareolus Ap.sylvaticus Ap.flavicollis R.norvegicus Pr op or tio n an im al s co -in fe ct ed b y ≥ 2 ba ct er ia l t ax a (a m on g al l i nf ec te d an im al s) Relative number of co-circulating bacterial taxa (Shannon Diversity Index) A. Number of bacterial infections per animal B. Number of pathogen co-exposures per animal C. D. Pr op or tio n an im al s ex po se d to ≥ 2 pa th og en s (a m on g al l e xp os ed a ni m al s) Relative number of pathogen co- exposures (Shannon Diversity Index) Mi. agrestis Ar. scherman Mi. arvalis My. glareolus R. norvegicus Ap. sylvaticus Ap. flavicollis Mi. agrestis Ar. scherman Mi. arvalis My. glareolus R. norvegicus Ap. sylvaticus Ap. flavicollis Pseudo R2 = 0.65 p < 0.0001 Pseudo R2 = 0.69 p < 0.0001 12 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 host species had a unique combination of co-circulating pathogens (Figure 1). In order to best identify extrinsic and intrinsic factors potentially driving the composition of pathogen communities within the rodent community, we reduced the dataset to limit biases. We excluded R. norvegicus individuals due to competing a priori hypotheses that host species and habitat would be important factors (as this host species was the only one found in farm habitats, confounding these two variables; but see MCA results when R. norvegicus was included in Figures S7, S8). Likewise, we excluded pathogens that occurred only in one habitat type of one host species (not including R. norvegicus). Two additional individuals were excluded due to missing sex and age information. Analyses were performed on the remaining 280 individuals from six host species and their 14 pathogens (Bartonella, Myco1, Myco2, Myco3, Myco4, Myco6, Myco7, Myco9, Rickettsia, Neoehrlichia, Orientia, Spiroplasma, and antibodies against CPXV and hantaviruses) (Figure 4). Figure 4 - Results of multiple correspondence analysis (MCA) for pathogen community composition in rodents (excluding R. norvegicus) are described by (A) the contribution of each dimension to the overall variance in the data, (B) variable correlations with the first two dimensions of the MCA, and (C) variable contributions to each orthogonal MCA dimension. Horizontal line in (A) represents the percent variance expected due to chance (100/14) = 7.14%. Overall, pathogen species composition was significantly structured by host species identity (F5,252 = 16.23, p = 0.001) and habitat type (F2,252 = 2.51, p = 0.024; Table S5). Out of 14 orthogonal dimensions returned by the MCA, the first two captured 23.0% of the variation in pathogen and antibody occurrence, and the first seven explained a cumulative 63.4% of the total variance (Figure 4A). Further dimensions captured less variance than would be expected if all dimensions contributed equally to overall inertia in the data. Dimension 1 (MCA Dim1; explaining 13.1% of the variation in pathogen community and loading heavily with the presence of Myco1, Myco3 and anti-hantavirus antibodies) differed significantly between host species (F5,268 = 23.83, p < 0.0001) and age classes (F1,268 = 6.27, p = 0.013; Table S6). Dimension 2 (MCA Dim2; explaining 9.9% of the variation in pathogen community and primarily describing the occurrence of Bartonella) was also structured significantly by host species (F5,268 = 3.89, p = 0.002; Table S6). While these first two dimensions varied by host species (Figure S10A) and host age class (Figure S10B), variance in host habitats (Figure S10C) was not significant after accounting for the other factors. Host species was the most consistently important extrinsic driver of pathogen community composition, ● 0.0 0.1 0.2 0.3 0.4 0. 0 0. 1 0. 2 0. 3 Dim 1 (13.11%) Di m 2 (9 .9 3% ) Bartonella Myco1_Mh Myco2_Mh Myco3_Mh Myco7_Mh Myco9_Mh Myco4_Mc Myco6_Mc Rickettsia Neoehrlichia Orientia Spiroplasma CPXV_Abs Hantavirus_Abs ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 5.19 10.39 15.58 20.77 25.96 31.16 36.35 41.54 46.73 51.93 Di m 1 Di m 2 Di m 3 Di m 4 Di m 5 Di m 6 Di m 7 Di m 8 Di m 9 Di m 1 0 Di m 1 1 Di m 1 2 Di m 1 3 Di m 1 4 Bartonella_0 Bartonella_1 Myco1_Mh_0 Myco1_Mh_1 Myco2_Mh_0 Myco2_Mh_1 Myco3_Mh_0 Myco3_Mh_1 Myco7_Mh_0 Myco7_Mh_1 Myco9_Mh_0 Myco9_Mh_1 Myco4_Mc_0 Myco4_Mc_1 Myco6_Mc_0 Myco6_Mc_1 Rickettsia_0 Rickettsia_1 Neoehrlichia_0 Neoehrlichia_1 Orientia_0 Orientia_1 Spiroplasma_0 Spiroplasma_1 CPXV_Abs_0 CPXV_Abs_1 Hantavirus_Abs_0 Hantavirus_Abs_1 13.1% 9.9% 9% 8.7% 7.7% 7.6% 7.4% 6.8% 6.3% 5.9% 5.1% 4.7% 4.6% 3.2% 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 MCA Dimensions % of ex pla ine d v ari an ce s Variances − MCA 1 2 3 4 5 6 7 8 9 10 11 12 13 14 C) A) B) % e xp la in ed v ar ia nc es 50 % 40% 30% 20% 10% 0% MCA Dimension MCA Dim 1 (13.1%) M C A D im 2 (9 .9 % ) 10 5 0 Jessica L. Abbate et al. 13 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 significantly explaining variation captured in six of the first seven dimensions, MCA Dim1 – MCA Dim7 (except for MCA Dim 5; Table S6). 3.3. Associations between pathogens 3.3.1. Validation of the associations detected by MCA We applied SCN and GLM analyses to further characterize patterns detected using MCA. Strong and relatively equal loading of MCA Dim1 with Myco1, Myco3, and anti-hantavirus antibody presence indicated that these three pathogens were positively associated with one-another. Indeed, the six animals with anti-hantavirus antibodies were found exclusively in animals infected with Myco3, and Myco1 was found in 2/3 of hantavirus-exposed animals but in just 1/3 of those without hantavirus exposure. The MCA also revealed significant differences among host species and host age classes for Dim 1; hantavirus and Myco3 only circulated in two host species (Mi. arvalis and My. glareolus) and 34 of those 35 occurrences were in adults. To exclude positive associations arising from mutual host specificity and age-related accumulation of exposure probability, we focused our analyses on the dataset restricted to adults of the two host species in which all three pathogens co-circulated (Mi. arvalis and My. glareolus). Individual SCN analyses performed on adults of each host species revealed no associations (Table S7A). However, since values of MCA Dim 1 did not differ between the two host species (according to post-hoc tests given in Table S6), we also ran a single SCN analysis on the pooled data from adults of both species to improve statistical power (Table S7A). According to this pooled SCN analysis, the three- way co-occurrence of Myco1, Myco3 and anti-hantavirus antibodies was significantly more frequent than would be expected by random chance (p = 0.008), with a trend for anti-hantavirus antibodies occurring by themselves more rarely than expected (sitting on the lower bound at zero; p = 0.13; Table S7A, Figure S11). We also investigated this association using GLM. However, given the small number of hantavirus exposures and perfect association with Myco3 infection, there was insufficient statistical power to explicitly test for an association between all three pathogens and extrinsic factors. We therefore ran three reciprocal GLM models on the restricted dataset, one for each pathogen as a function of extrinsic factors to control for heterogeneous host groups (host species, host sex, study site, habitat, and year sampled) and exposure to the two other pathogens (Table S7B). The number of extrinsic variables in each model was reduced using model selection (results in Figure S12). These models showed that there remained significant unexplained positive associations between hantavirus exposure and Myco1 infection (anti-hantavirus antibodies ~ Myco1: χ2 = 5.80, p = 0.016) and between hantavirus exposure and Myco3 infection (Myco3 ~ anti-hantavirus antibodies: χ2 = 13.66, p < 0.001), but that there was no evidence of direct association between Myco1 and Myco3 infections (Myco1 ~ Myco3: χ2 = 0.01, p = 0.94; Myco3 ~ Myco1: χ2 < 0.37, p = 0.54; Figure 5A). The third MCA dimension also presented a clear hypothesis with sufficient statistical power to be tested. MCA Dim3 was characterized by co-variation in Myco2 (Myco. haemomuris) and Myco4 (Myco. coccoides) infections suggesting a positive association between members of these two Mycoplasma species. Myco2 and Myco4 OTUs co-circulated only in Ap. sylvaticus hosts, thus we limited our analysis to this host species. There was no significant association between the two OTUs detected by SCN analysis (Table S8A), and after correcting for extrinsic factors remaining in the models after model selection (Figure S13), there remained only a non-significant trend (Myco2 ~ Myco4: χ2 = 2.79, p = 0.12; Myco4 ~ Myco2: χ2 = 2.34, p = 0.13; Table S8B) for a positive association between the two OTUs. No additional associations with sufficient variance for statistical tests were clearly suggested by the MCA analysis. 3.3.2. Validation of associations described in the literature We tested the a priori hypothesis that seropositivity to CPXV would be positively associated with Bartonella infection, previously detected in Mi. agrestis (Telfer et al., 2010). The whole dataset was considered as these two pathogens co-circulated in all host species (Figure 1). SCN analyses performed independently for each host species revealed no associations (Table S9A), and the MCA results suggested that pooling data across host species would be inappropriate. After correcting for extrinsic factors using GLM, we found reciprocal evidence for a positive association in Ar. scherman hosts (Bartonella ~ anti- CPXV antibodies: c2 = 5.07, p = 0.024; anti-CPXV antibodies ~ Bartonella: c2 = 5.07, p = 0.024; Figure 5B), but not in any other host species (Figure S14; Table S9B). It is of note that there were only eight Mi. 14 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 agrestis individuals, rendering statistical power to test for the association while controlling for extrinsic factors insufficient in this host species where the association was previously described. While prevalence of both pathogens in Mi. agrestis was relatively high compared to other host species, one of the two animals without Bartonella infection was positive for anti-CPXV antibodies, also precluding evidence for a within-species trend. Figure 5 - Associations between pathogens in a community of rodents. Association hypotheses were generated by multiple correspondence analysis (A) or previously noted in the literature (B, C, D). Only associations supported by significant statistical tests (p < 0.05) are illustrated. Red arrows represent positive associations, blue arrows represent negative associations. We next focused on the potential associations between OTUs identified as belonging to two different species of hemotropic Mycoplasma, Myco. haemomuris (HM) and Myco. coccoides (HC), within the four host species in which they both circulated (My. glareolus, Ap. flavicollis, Ap. sylvaticus, R. norvegicus; Figure 1; Figure S4). We found no significant associations using independent SCN analyses for each host species (Table S10A). However, after controlling for extrinsic factors using GLM, a significant positive association was detected (HM ~ HC: c2 = 9.5, p = 0.0021; HC ~ HM: c2 = 9.59, p = 0.002), and did not differ between host species (non-significant interaction term between host species by explanatory pathogen occurrence in each reciprocal model, Figure S15; Table S10B; Figure 5C). We note that only one R. norvegicus animal was uninfected with Myco. haemomuris, and that animal also had no Myco. coccoides infection; thus the trend for the association in this host species was also positive but lacked sufficient variance for independent statistical analysis. Finally, we tested for associations between Bartonella spp. and hemotropic Mycoplasma species, grouping the occurrence of different OTUs of the latter (Myco1 – Myco9) into a single presence-absence variable. There was no association detected by SCN analyses (Table S11A), and marginal evidence that any association may differ by host species after correcting for extrinsic factors using GLM with model selection (Figure S16; Table S11B). After controlling for extrinsic factors using independent GLMs for each host species (where possible), we found a negative association between the two pathogen groups only in My. glareolus hosts (Bartonella ~ Mycoplasma: c2 = 4.14, p = 0.042; Mycoplasma ~ Bartonella: c2 = 6.59, p = 0.010, Table S11B, Figure 5D). 3.4. Evaluating false discovery Benjamini-Hochberg correction of p-values from hypothesis tests throughout the study suggested that those above ~0.01 may lie above the false discovery cutoff for statistical significance (Figure S17), and that null hypotheses rejected with smaller p-values have been rejected with confidence. R. norvegicus A. Myco1 anti-Hantavirus antibodies Myco3 My. glareolus Mi. arvalis B. anti-CPXV antibodies Bartonella spp. Ar. scherman C. Mycoplasma coccoides (Myco4,6) Mycoplasma haemomuris (Myco1,2,3,5,7,8,9) D. Mycoplasma coccoides Mycoplasma haemomuris (Myco1 – Myco9) Bartonella spp. My. glareolus My. glareolus Ap. flavicollis Ap. sylvaticus +"+" +" +" #" +" #"Positive association Negative association Jessica L. Abbate et al. 15 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 4. Discussion Rodents have long been recognized as important reservoirs of infectious agents, with a high transmission potential to humans and domestic animals (Kruse et al., 2004). Europe is identified as a hotspot of rodent reservoir diversity and one third of rodent species are considered hyper-reservoirs, carrying up to 11 zoonotic agents (Han et al., 2015). Nevertheless, associations between these pathogens have still only rarely been investigated (but see, for example, studies from field voles in the UK (Telfer et al., 2010) and in Poland (Pawelczyk et al., 2004), gerbils in Israel (Cohen et al., 2015), across a rodent community in North America (Dallas et al., 2019), and co-infection frequencies of zoonotic pathogens from rodents in Croatia (Tadin et al., 2012)). In this study, we confirmed that rodent communities in northern France may harbor a large diversity of potential zoonotic pathogens, with at least 10 bacterial genera and antibodies against at least four genera of viruses. Some of these pathogens have already been reported in the study region or in geographic proximity, including viruses (Orthohantavirus, Orthopoxvirus, Mammarenavirus (Charbonnel et al., 2008; Salvador et al., 2011)) and bacteria (e.g., Bartonella, Mycoplasma, Rickettsia, “Candidatus Neoehrlichia”, Orientia, Spiroplasma, Treponema, Leptospira, Borrelia, Neisseria, Pasteurella; see (Vayssier-Taussat et al., 2012; Razzauti et al., 2015)). A previously undetected relative of the putatively pathogenic spirochaete Brevinema andersonii that infects short-tailed shrews and white-footed mice in North America (Defosse et al., 1995) was among our findings, and TBEV is not known to circulate this far east (Lindquist & Vapalahti, 2008). The high prevalence of anti-hantavirus antibodies in Mi. arvalis is likely explained by cross-reactivity between the anti-PUUV antibodies used in our assay and those elicited against the related Tula orthohantavirus (TULA) virus common to European voles (Deter et al., 2007; Tegshduuren et al., 2010). Three zoonotic pathogens were particularly prevalent: Orthopoxvirus, Bartonella spp., and Mycoplasma spp. The wide range of hosts with anti-Orthopoxvirus antibodies corroborates prior evidence that cowpox virus could be widespread in European rodents, particularly voles (Bennett et al., 1997; Essbauer et al., 2010; Forbes et al., 2014). An astounding 77% of all individuals in the study were infected by Bartonella spp., a diverse group of hemotrophs known to commonly infect rodents and other mammals (Breitschwerdt & Kordick, 2000; Bai et al., 2009) and which have also been implicated in both zoonotic and human-specific disease (Iralu et al., 2006; Breitschwerdt, 2014; Vayssier-Taussat et al., 2016). We could not assess the specific diversity of Bartonella spp. circulating in these rodent communities because accurate resolution in this genus requires additional genetic markers (Matar et al., 1999; Guy et al., 2013). Hemotropic and pneumotropic Mycoplasma spp. were also highly prevalent across all host species, though surprisingly nearly absent from Ar. scherman despite expectations from related Ar. terrestris (Villette et al., 2017). These Mycoplasma species are also known pathogens of humans and rodents (Harwick et al., 1972; Baker, 1998). Here, we found two distinct hemotropic Mycoplasma species (Myco. haemomuris and Myco. coccoides) and the pneumotropic Mycoplasma species Myco. pulmonis and “Candidatus Myco. ravipulmonis”. The former two are both hemotropic mycoplasmas responsible for vector-transmitted infectious anaemia of wild mice, rats, and other rodent species (Neimark et al., 2001, 2005; Messick, 2004). In contrast, Myco. pulmonis and “Candidatus Myco. ravipulmonis” cause respiratory infections, are more closely related to other pneumotropic mycoplasmas, and “Candidatus Myco. ravipulmonis” has only ever before been described in laboratory mice (formerly termed Grey Lung virus (Andrews & Glover, 1945; Neimark et al., 1998; Graham & Schoeb, 2011; Piasecki et al., 2017)). Our results also corroborated the status of hyper-reservoir (more than two zoonotic pathogens carried by a reservoir species) for all seven of the focal rodent species studied here (Han et al., 2015). Even the rare host species Mi. subterraneous also carried two potentially zoonotic pathogens (Bartonella spp. and Brevinema spp.; Appendix 1). Overall, we found a high variability in the number of pathogens circulating in each species despite correction for sampling effort, with low levels observed in Apodemus species and Arvicola scherman, and high levels detected in Mi. arvalis, My. glareolus, and R. norvegicus. While physiology, genetics, and behavior can contribute to the number of pathogen species able to infect a given host species, larger geographic range size is highly correlated with higher pathogen species diversity (Morand, 2015); this explanation matches the pattern among hosts in the communities sampled 16 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 here (i.e., Ar. scherman and Apodemus spp. have small geographic ranges compared to those of Mi. arvalis, My. glareolus, and R. norvegicus). Several studies have emphasized the influence of host habitat specialization on parasite species richness, low habitat specialization being associated with both high species richness of macro- and micro- parasites (e.g., (Morand & Bordes, 2015)). Our results did not fully corroborate this association; while the grassland-specific Ar. scherman had the lowest pathogen diversity and the multi-habitat spanning My. glareolus had the highest pathogen diversity, entirely farm-dwelling R. norvegicus had high pathogen diversity nearly equal to that of My. glareolus, and the two Apodemus hosts (neither with significantly higher pathogen diversity than Ar. scherman) were found across both meadows and hedgerows. Instead, we found that more diverse host species communities hosted more diverse pathogen communities. However, the implications of that result are complex because while exposure to diverse (i.e., potentially novel) pathogens is a risk for disease emergence, diverse host species communities are thought to keep individual pathogen prevalence low due to the dilution effect – which should limit risk of zoonoses (Keesing et al., 2010). The search for factors that drive parasite species richness, diversity and community composition has been at the core of numerous studies (Poulin, 1995; Poulin & Morand, 2000; Nunn et al., 2003; Mouillot et al., 2005; Krasnov et al., 2010; Sallinen et al., 2020). Here, we emphasized that both pathogen diversity and community composition was mainly structured by host species identity, despite both shared habitats and shared pathogen taxa. Pathogen beta diversity was also structured by habitat, which could result from particular environmental suitability (e.g., for vectors) or opportunities for cross-species transmission. We found no evidence that any specific pathogen-pathogen associations were likely to be as important as host species identity in determining pathogen distributions across the community of rodents. The strong influence of host characteristics (Cohen et al., 2015) and host species identity (Dallas et al., 2019) on pathogen community composition has recently been described in comparison to intrinsic pathogen-pathogen associations in other rodents. Moreover, the pathogen community composition provided a unique signature for each rodent species, even among those most closely related (e.g., Ap. flavicolis and Ap. sylvaticus). This result is in line with the conclusions of meta-analyses showing that phylogeny, over other host traits, has a minimal impact on pathogen diversity in rodent species (Luis et al., 2013; Guy et al., 2019). The importance of host species identity in shaping pathogen community composition may not stem from strict host-pathogen specificity, as most pathogens were found to infect multiple host species – a broad result echoed across animal communities (Taylor et al., 2001; Woolhouse et al., 2001; Cleaveland et al., 2001; Pedersen et al., 2005; Streicker et al., 2013). However, we might be cautious as more precise molecular analyses are necessary to test whether different species of a bacteria genus or divergent populations of the same bacteria species may circulate independently in different rodent host species, with little or no transmission. For example, two genera seemed to be largely shared among the rodent species studied here, Bartonella and Mycoplasma. But previous studies have shown strong host- specificity when considering the genetic variants of Bartonella (Buffet et al., 2013; Withenshaw et al., 2016; Brook et al., 2017). Evidence in the literature for host specificity of Mycoplasma species has led to a mix of conclusions (Pitcher & Nicholas, 2005), as cases of cross-species transmission are commonly reported – particularly in humans – despite a general consensus that most species are highly host- specific. We found that some Mycoplasma taxa were dominant contributors to prevalence in a single host species, and that when shared, they were shared with just a few other specific host species. Rare infections in unexpected host species (e.g., Myco6 in Ar. scherman and Myco1 in Ap. flavicollis) were represented by fewer sequence reads compared to positive samples in host species where they were more prevalent, suggesting a low potential for amplification and sustained transmission from these occasional hosts (Figure S4). On the other hand, while the Cricetidae appeared to be susceptible only – with rare exception – to taxa within the Myco. haemomuris group, host species in the Muridae family were susceptible to all three distinct Mycoplasma species detected. The biggest exception to this pattern was that three of 62 Myodes glareolus (sister to all other sampled Cricetidae in the study) animals were found to be infected by both hemotropic Mycoplasma species. These results both support the observation that cross-species transmission naturally occurs among wild rodents and suggest that the degree of host specificity may be driven by both host and pathogen factors. Jessica L. Abbate et al. 17 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Concurrent exposure to multiple pathogens within individuals was also frequent, as high as 89 % (in Mi. arvalis hosts), in line with recent studies that have shown that co-infections by multiple pathogens are common in natural populations (e.g., in mammals, birds, amphibians, ticks, humans (Telfer et al., 2010; Griffiths et al., 2011; Moutailler et al., 2016; Clark et al., 2016; Stutz et al., 2018)). Variation in the frequency of pathogen co-exposure was highly correlated to the diversity of pathogens circulating in each host species, suggesting the dominance of a random process of pathogen exposure for each individual. However, there were a few intriguing outliers: My. glareolus hosts were less co-infected than expected based on diversity of bacterial taxa, but not when viral antibodies were included; conversely, Ar. scherman hosts were more co-exposed when viruses were considered, but not when only bacteria were considered; and Mi. arvalis hosts had consistently higher proportions of co-exposures whether viruses were or were not considered along with bacteria. The non-random grouping of pathogen exposures within individuals (as in Mi. arvalis) may result from heterogeneity in extrinsic transmission, environmental, or susceptibility factors (Cattadori et al., 2006; Swanson et al., 2006; Beldomenico et al., 2008; Fenton et al., 2010; Beldomenico & Begon, 2010) or from intrinsic interactions between pathogens (e.g., facilitation mediated by hosts immune response). Differences in the pattern of co-exposure frequencies when including or excluding antiviral antibodies (as with My. glareolus and Ar. scherman) could result from different mechanisms (e.g., bacterial manipulation of innate immunity (Diacovich & Gorvel, 2010)) affecting pathogen community assemblage. However, a lack of deviance from the expected co-exposure frequency does not exclude the possibility that both extrinsic and intrinsic processes may be occurring. We found evidence in support of three previously identified pathogen-pathogen associations (positive association between Myco. haemomuris and Myco. coccoides infections; positive association between Bartonella spp. infection and the presence of anti-CPXV antibodies; negative association between Bartonella spp. and hemotropic Mycoplasma spp. infections) and characterized one set of associations not previously described (positive associations between the presence of anti-hantavirus antibodies and infections by two specific Myco. haemomuris OTUs) – each in a unique subset of host species. Mycoplasma spp. blood infections are likely transmitted through bites of blood-sucking arthropod vectors (Volokhov et al., 2017), meaning vectors could prefer some individuals over others (Malmqvist et al., 2004). Positive associations detected between Myco. haemomuris and Myco. coccoides could also result from similarities in rodent susceptibility. Indeed Mycoplasma spp. infection can lead to acute or chronic infection, and the establishment of chronic bacteremia seems to occur in immunosuppressed or immunocompromised individuals (Cohen et al., 2018). Co-infections with multiple Mycoplasma spp. might therefore be more likely to be detected in these immunocompromised rodents with chronic infections. The existence of chronic infections might also lead to additional co-infections and positive associations as a result of disease-induced changes in population dynamics, immune system function, or through direct pathogen-pathogen interactions (Fenton, 2008; Aivelo & Norberg, 2018; Fountain-Jones et al., 2019). Whether through the accumulation of exposure probabilities or increased susceptibility, the previously-undocumented positive association we found here between Myco. haemomuris OTUs (Myco1 and Myco3) and anti-hantavirus antibodies may similarly be explained by the chronic nature of both Mycoplasma spp. and hantavirus infections in rodents (e.g., for Puumala hantavirus in bank voles (Yanagihara et al., 1985; Meyer & Schmaljohn, 2000; Vaheri et al., 2013)). This positive association was found in both host species where the majority of hantavirus exposures occurred (Microtus arvalis and Myodes glareolus), consistent with the generality of association between Mycoplasma species across host taxa detailed above, suggesting the intrinsic ecology of these pathogens contributes to shaping variation in the pathogen community. Curiously, we found no evidence for direct associations between OTUs of the same Mycoplasma species, thus facilitation interactions are unlikely to explain the high diversity of Mycoplasma taxa both within and between host species. Infections by Bartonella species are also known to often result in subclinical and persistent bacteremia in mammals, including rodents (Birtles et al., 2001; Kosoy et al., 2004). The positive association detected in Ar. scherman between Bartonella spp. and anti-CPXV antibodies might therefore be explained by, for example, joint accumulation of both chronic bacterial infections and long-lived antiviral antibodies used to test for prior exposure to relatively short-lived CPXV infections. However, if the same processes governing association of the chronic infections described above were at play here, 18 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 we would have expected to find both pathogens implicated in positive associations (i) with other chronic infections, and (ii) across host species given their ubiquitous prevalence. While the failure to recover the association in Mi. agrestis (previously described (Telfer et al., 2010)) was likely due to low statistical power, the lack of a general pattern in other host species despite adequate sampling suggests a more specific, and potentially immune-mediated, ecological process between these two pathogens. Indeed, pox virus infections, including CPXV, have been shown to induce immunomodulation that increases host susceptibility to other parasites (Johnston & McFadden, 2003). These interactions could be of variable intensities according to the rodent species considered, due to potential differences in impacts of CPXV infection on immunity among host species, or to the influence of other infections not examined here on host immune responses during pox infections (e.g., helminths (Cattadori et al., 2007), protozoa (Telfer et al., 2010)). Furthermore, Bartonella spp. infection was negatively associated with Mycoplasma spp. infections in My. glareolus, corroborating negative interactions reported in co-infection experiments in gerbils (Eidelman et al., 2019). This association may therefore originate from an interaction mediated by specific (immune) genetic features of My. glareolus, and not ecological conditions as proposed by Eidelman et al. (2019). The antagonistic and host-specific nature of this association lends further support to the interpretation that Bartonella spp. infections do not behave in ways similar to other chronic infections in the community. However, few studies have investigated the robustness of within-host interactions across different host species (e.g., (Lello et al., 2018)), and this question deserves further investigation. Our results suggest that intrinsic ecological interactions could help shape the composition of the pathogen community within hosts. However, this suggestion provides only a hypothesis that requires further investigation. Interpretation of associations can be misleading, as they may arise from unmeasured co-factors such as exposure to shared transmission routes, and may even run counter to the underlying ecological process (Fenton et al., 2014). The associations we found here were not visible (or even misleading, in the case of a 3-way interaction between hantavirus, Myco1 and Myco3), for instance, when ignoring extrinsic factors using the SCN analysis, despite the increased statistical power it offered. Our decision to screen only the spleen means we could have missed evidence of exposure to pathogens that can only be found by screening the liver, kidney or brain (Mangombi et al., 2021). Evidence for interactions between pathogens within hosts initially came from laboratory studies (e.g., in the development of vaccines, reviewed in (Casadevall & Pirofski, 2000)), and until recently, many studies conducted in the wild could not detect such interactions (e.g., (Behnke, 2008)). Developments in statistical approaches have contributed to improve sampling designs and analyses, in particular by better controlling for confounding factors, enabling the detection of associations resulting from these within- host interactions (e.g., (Lello et al., 2004; Telfer et al., 2010; Galen et al., 2019)). A structural equation modeling approach, for instance, could be applied to identify potential influence of unmeasured (latent) variables (Carver et al., 2015). Experiments conducted in semi-controlled environments have been used to confirm the importance of interactions suggested by associations (e.g., (Knowles et al., 2013)). Both facilitation mediated by immune responses (e.g., (Ezenwa et al., 2010)) and competition mediated by shared resources (e.g., (Brown, 1986; Budischak et al., 2018)) have been emphasized. There remain additional important limits to the interpretation of snapshot observational studies from wild populations such as ours. For instance, they cannot provide information about the sequence or duration of infection, although these features strongly affect the outcome of within-host interactions (Eidelman et al., 2019). Moreover, both the 16S metabarcoding approach and serological antibody tests can only be interpreted in terms of presence/absence of exposure to pathogens, although co-infection may rather impact parasite abundance (e.g., (Thumbi et al., 2013; Gorsich et al., 2014)). Other extrinsic factors, such as seasonal variation in pathogen community composition, could also impact both interpretation and year-round generality of our results due to adherence to autumn sampling dates (Maurice et al., 2015; Villette et al., 2020). Lastly, we also acknowledge several caveats to consider with our methods. We removed animals from which fewer than 500 reads were amplified in one or both bacterial metabarcoding PCR replicates. While 16 of these samples removed were due to random failure of PCR amplification from just one of the two replicates, 12 of the animals had poor amplification in both PCR replicates. In the absence of an internal positive control, e.g., a spike-in standard (Zemb et al., 2020), we were unable to verify whether a lack of reads was due to poor DNA extraction or a true lack of infections. Although this has a risk of artificially inflating prevalence rates by selectively removing Jessica L. Abbate et al. 19 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 uninfected individuals, it is unlikely to have had a qualitative effect on our results. Similarly, limiting our analyses to OTUs with 500 reads or more in the entire dataset may select against detection of very rare or low-burden infections. We also removed many OTUs corresponding to bacteria normally occurring in external or internal microbiomes of healthy animals, some of which were represented by a high abundance of reads in positive animals. This was due mainly to the fact that 16S data cannot often distinguish between pathogenic and commensal taxa of many such genera. We know that, for instance, Helicobacter species are naturally found in the digestive tract, but can also cause pathogenic infections. Parasitism can affect host microbiome composition (Gaulke et al., 2019), and this in turn can have impacts on host health and disease susceptibility (reviewed in (Murall et al., 2017; Rosshart et al., 2017)). Thus, our choice to ignore OTUs corresponding to microbes typical of healthy flora contributes to the problem of missing data, such as information on intestinal helminth infections or other viruses, which may explain or alter the associations we were able to detect. Furthermore, the evaluation of diversity measures (e.g., Shannon diversity index) based only on a selection of taxa violates the assumption that all species are represented in the sample; thus, patterns of diversity could also be influenced by missing data. These caveats are common problems for disease surveillance and community ecology studies, irrespective of the diagnostic methods, and it is difficult to speculate about their overall impacts on the present study. Finally, it is well-understood that this bias towards detection of common pathogens and difficulty in interpreting evidence for the absence of a pathogen in a given individual or population can make testing for negative associations driven by antagonistic ecological interactions incredibly difficult, if not impossible (Weiss et al., 2016; Cougoul et al., 2019). 5. Conclusions Our results add to a growing number of studies finding that (i) rodents host many important zoonotic human pathogens and (ii) pathogen communities are shaped primarily by host species identity. We also detected a number of previously undescribed associations among pathogens within these rodent communities, and we also confirmed previously identified associations, sometimes in other rodent species than those in which they were previously described. These associations can be considered in the future as hypotheses for pathogen-pathogen interactions within rodent hosts, and that participate in shaping the community of pathogens in rodent communities. Long-term survey and experimental studies are now required to confirm these interactions and understand the mechanisms underlying the patterns of co-infection detected. In addition to these biological results, we have identified several methodological caveats, with regard to both pathogen and association detection, that deserve further investigation to improve our ability to make robust inference of pathogen interactions. 6. Acknowledgements We would like to thank all collaborators and research assistants that previously helped with the field work (Yannick Chaval, Cécile Gotteland, Marie-Lazarine Poulle). None of the rodent species investigated here has protected status (see list of the International Union for Conservation of Nature). All procedures and methods were carried out in accordance with relevant regulations and official guidelines from the American Society of Mammalogists. All protocols presented here were realized with prior explicit agreement from relevant institutional committee (Centre de Biologie pour la Gestion des Populations (CBGP): 34 169 003). We thank Hélène Vignes of the Cirad Genotyping platform for her assistance with the MiSeq sequencing, Sylvain Piry, Alexandre Dehne-Garcia and Marie Pagès for their help with the bioinformatic analysis, and Gael Kergoat for helpful comments and proofreading. We also thank the reviewers and recommenders at PCI Ecology for their careful reviews, insights, and patience. Version 4 of this preprint has been peer-reviewed and recommended by Peer Community In Ecology (PCI Ecology) (https://doi.org/10.24072/pci.ecology.100071; Blanchet, 2021). 20 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 7. Funding This study was partly funded by the INRA Metaprogram MEM Hantagulumic, and by the European programme FP7-261504 EDENext. The manuscript is registered with the EDENext Steering Committee as EDENext409. 8. Conflict of Interest Disclosure The authors declare that they comply with the PCI rule of having no financial conflicts of interest in relation to the content of the article. The authors declare the following non-financial conflict of interest: JLA and NC are PCI Ecology recommenders. 9. Data Accessibility Supplementary data deposited in Zenodo (https://doi.org/10.5281/zenodo.7092812; Abbate et al., 2022) include the following 16S metabarcoding data: (i) raw sequence reads (fastq format), (ii) raw output files generated by the mothur program (iii) raw abundance table and (iv) filtered occurrence table, as well as (v) scripts and data files for statistical analyses. Items iii-v are also provided in Supplemental Materials Appendix 2 to directly accompany this publication. References Abbate JL, Galan M, Razzauti M, Sironen T, Voutilainen L, Henttonen H, Gasqui P, Cosson J-F, Charbonnel N (2022). Data from: Pathogen community composition and co-infection patterns in a wild community of rodents [Data set]. In Peer Community in Ecology. Zenodo. https://doi.org/10.5281/zenodo.7092812 Abbate JL, Ezenwa VO, Guégan J-F, Choisy M, Nacher M, Roche B (2018) Disentangling complex parasite interactions: Protection against cerebral malaria by one helminth species is jeopardized by co- infection with another. PLOS Neglected Tropical Diseases, 12, 1–13. https://doi.org/10.1371/journal.pntd.0006483 Aivelo T, Norberg A (2018) Parasite–microbiota interactions potentially affect intestinal communities in wild mammals. Journal of Animal Ecology, 87, 438–447. https://doi.org/10.1111/1365-2656.12708 Alizon S, de Roode JC, Michalakis Y (2013) Multiple infections and the evolution of virulence. Ecology Letters, 16, 556–567. https://doi.org/10.1111/ele.12076 Andrews CH, Glover RE (1945) Grey Lung virus: an agent pathogenic for mice and other rodents. British Journal of Experimental Pathology, 26, 379–387. Bai Y, Kosoy MY, Lerdthusnee K, Peruski LF, Richardson JH (2009) Prevalence and genetic heterogeneity of Bartonella strains cultured from rodents from 17 provinces in Thailand. American Journal of Tropical Medicine and Hygiene, 81, 811–816. https://doi.org/10.4269/ajtmh.2009.09-0294 Baker DG (1998) Natural Pathogens of Laboratory Mice , Rats , and Rabbits and Their Effects on Research. Clinical Microbiology Reviews, 11, 231–266. https://doi.org/10.1128/cmr.11.2.231 Beechler BR, Manore CA, Reininghaus B, O’Neal D, Gorsich EE, Ezenwa VO, Jolles AE (2015) Enemies and turncoats: Bovine tuberculosis exposes pathogenic potential of rift valley fever virus in a common host, African buffalo (Syncerus caffer). Proceedings of the Royal Society B: Biological Sciences, 282, 20142942. https://doi.org/10.1098/rspb.2014.2942 Behnke JM (2008) Structure in parasite component communities in wild rodents: predictability, stability, associations and interactions .... or pure randomness? Parasitology, 135, 751–766. https://doi.org/DOI: 10.1017/S0031182008000334 Beldomenico PM, Begon M (2010) Disease spread, susceptibility and infection intensity: vicious circles? Trends in Ecology & Evolution, 25, 21–27. https://doi.org/10.1016/j.tree.2009.06.015 Beldomenico PM, Telfer S, Gebert S, Lukomski L, Bennett M, Begon M (2008) Poor condition and infection: a vicious circle in natural populations. Proceedings. Biological sciences, 275, 1753–1759. https://doi.org/10.1098/rspb.2008.0147 Jessica L. Abbate et al. 21 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Bennett M, Crouch AJ, Begon M, Duffy B, Feore S, Gaskell RM, Kelly DF, McCracken CM, Vicary L, Baxby D (1997) Cowpox in British voles and mice. Journal of Comparative Pathology, 116, 35–44. https://doi.org/https://doi.org/10.1016/S0021-9975(97)80041-2 Birtles RJ, Hazel SM, Bennett M, Bown K, Raoult D, Begon M (2001) Longitudinal monitoring of the dynamics of infections due to Bartonella species in UK woodland rodents. Epidemiology and infection, 126, 323–329. https://doi.org/10.1017/s095026880100526x Blanchet S (2021) “Hidden” natural selection and the evolution of body size in harvested stocks. Peer Community in Ecology, 1, 100071. https://doi.org/10.24072/pci.ecology.100071 Bordes F, Caron A, Blasdell K, de Garine-Wichatitsky M, Morand S (2017) Forecasting potential emergence of zoonotic diseases in South-East Asia: network analysis identifies key rodent hosts. Journal of Applied Ecology, 54, 691–700. https://doi.org/10.1111/1365-2664.12804 Bordes F, Herbreteau V, Dupuy S, Chaval Y, Tran A, Morand S (2013) The diversity of microparasites of rodents: a comparative analysis that helps in identifying rodent-borne rich habitats in Southeast Asia. Infection ecology & epidemiology, 3. https://doi.org/10.3402/iee.v3i0.20178 Breitschwerdt EB (2014) Bartonellosis: One health perspectives for an emerging infectious disease. ILAR Journal, 55, 46–58. https://doi.org/10.1093/ilar/ilu015 Breitschwerdt EB, Kordick DL (2000) Bartonella infection in animals: Carriership, reservoir potential, pathogenicity, and zoonotic potential for human infection. Clinical Microbiology Reviews, 13, 428– 438. https://doi.org/10.1128/CMR.13.3.428-438.2000 Brook CE, Bai Y, Yu EO, Ranaivoson HC, Shin H, Dobson AP, Metcalf CJE, Kosoy MY, Dittmar K (2017) Elucidating transmission dynamics and host-parasite-vector relationships for rodent-borne Bartonella spp. in Madagascar. Epidemics, 20, 56–66. https://doi.org/10.1016/j.epidem.2017.03.004 Brown AF (1986) Evidence for density-dependent establishment and survival of Pomphorhynchus laevis (Müller, 1776) (Acanthocephala) in laboratory-infected Salmo gairdneri Richardson and its bearing on wild populations in Leuciscus cephalus (L.). Journal of Fish Biology, 28, 659–669. https://doi.org/10.1111/j.1095-8649.1986.tb05201.x Budischak SA, Sakamoto K, Megow LC, Cummings KR, Urban JF, Ezenwa VO (2015) Resource limitation alters the consequences of co-infection for both hosts and parasites. International Journal for Parasitology, 45, 455–463. https://doi.org/https://doi.org/10.1016/j.ijpara.2015.02.005 Budischak SA, Wiria AE, Hamid F, Wammes LJ, Kaisar MMM, van Lieshout L, Sartono E, Supali T, Yazdanbakhsh M, Graham AL (2018) Competing for blood: the ecology of parasite resource competition in human malaria–helminth co-infections. Ecology Letters, 21, 536–545. https://doi.org/10.1111/ele.12919 Buffet J-P, Pisanu B, Brisse S, Roussel S, Félix B, Halos L, Chapuis J-L, Vayssier-Taussat M (2013) Deciphering Bartonella Diversity, Recombination, and Host Specificity in a Rodent Community. PLOS ONE, 8, e68956. https://doi.org/10.1371/journal.pone.0068956 Burthe S, Telfer S, Lambin X, Bennett M, Carslake D, Smith A, Begon M (2006) Cowpox virus infection in natural field vole Microtus agrestis populations: delayed density dependence and individual risk. Journal of Animal Ecology, 75, 1416–1425. https://doi.org/10.1111/j.1365-2656.2006.01166.x Carvalho-Pereira TSA, Souza FN, Santos LR do N, Pedra GG, Minter A, Bahiense TC, Reis MG, Ko AI, Childs JE, da Silva EM, Costa F, Begon M (2019) Coinfection modifies carriage of enzootic and zoonotic parasites in Norway rats from an urban slum. Ecosphere, 10, e02887. https://doi.org/10.1002/ecs2.2887 Carver S, Beatty JA, Troyer RM, Harris RL, Stutzman-Rodriguez K, Barrs VR, Chan CC, Tasker S, Lappin MR, Vandewoude S (2015) Closing the gap on causal processes of infection risk from cross-sectional data: Structural equation models to understand infection and co-infection. Parasites and Vectors, 8. https://doi.org/10.1186/s13071-015-1274-7 Casadevall A, Pirofski LA (2000) Host-pathogen interactions: basic concepts of microbial commensalism, colonization, infection, and disease. Infection and immunity, 68, 6511–6518. https://doi.org/10.1128/iai.68.12.6511-6518.2000 Cattadori IM, Albert R, Boag B (2007) Variation in host susceptibility and infectiousness generated by co- infection: the myxoma–Trichostrongylus retortaeformis case in wild rabbits. Journal of The Royal Society Interface, 4, 831–840. https://doi.org/10.1098/rsif.2007.1075 22 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Cattadori IM, Boag B, Hudson PJ (2008) Parasite co-infection and interaction as drivers of host heterogeneity. International Journal for Parasitology, 38, 371–380. https://doi.org/10.1016/j.ijpara.2007.08.004 Cattadori IM, Haukisalmi V, Henttonen H, Hudson PJ (2006) Transmission ecology and the structure of parasite communities in small mammals BT - Micromammals and Macroparasites: From Evolutionary Ecology to Management. In: (eds Morand S, Krasnov BR, Poulin R), pp. 349–369. Springer Japan, Tokyo. https://doi.org/10.1007/978-4-431-36025-4_18 Charbonnel N, Deter J, Chaval Y, Laakkonen J, Henttonen H, Voutilainen L, Vapalahti O, Vaheri A, Morand S, Cosson J-F (2008) Serological Evidence of Viruses Naturally Associated with the Montane Water Vole (Arvicola scherman) in Eastern France. Vector-Borne and Zoonotic Diseases, 8, 763–768. https://doi.org/10.1089/vbz.2007.0167 Charbonnel N, Pagès M, Sironen T, Henttonen H, Vapalahti O, Mustonen J, Vaheri A (2014) Immunogenetic factors affecting susceptibility of humans and rodents to hantaviruses and the clinical course of hantaviral disease in humans. Viruses, 6, 2214–2241. https://doi.org/10.3390/v6052214 Claesson MJ, Wang Q, O’Sullivan O, Greene-Diniz R, Cole JR, Ross RP, O’Toole PW (2010) Comparison of two next-generation sequencing technologies for resolving highly complex microbiota composition using tandem variable 16S rRNA gene regions. Nucleic Acids Research, 38, e200-213. https://doi.org/10.1093/nar/gkq873 Clark NJ, Wells K, Dimitrov D, Clegg SM (2016) Co-infections and environmental conditions drive the distributions of blood parasites in wild birds. Journal of Animal Ecology, 85, 1461–1470. https://doi.org/10.1111/1365-2656.12578 Cleaveland S, Laurenson MK, Taylor LH (2001) Diseases of humans and their domestic mammals: pathogen characteristics, host range and the risk of emergence. Philosophical transactions of the Royal Society of London. Series B, Biological sciences, 356, 991–999. https://doi.org/10.1098/rstb.2001.0889 Cohen C, Einav M, Hawlena H (2015) Path analyses of cross-sectional and longitudinal data suggest that variability in natural communities of blood-associated parasites is derived from host characteristics and not interspecific interactions. Parasites & vectors, 8, 429. https://doi.org/10.1186/s13071-015- 1029-5 Cohen C, Shemesh M, Garrido M, Messika I, Einav M, Khokhlova I, Tasker S, Hawlena H (2018) Haemoplasmas in wild rodents: Routes of transmission and infection dynamics. Molecular Ecology, 27, 3714–3726. https://doi.org/10.1111/mec.14826 Cougoul A, Bailly X, Vourc’h G, Gasqui P (2019) Rarity of microbial species: In search of reliable associations. PLOS ONE, 14, e0200458. https://doi.org/10.1371/journal.pone.0200458 Cox FEG (2001) Concomitant infections , parasites and immune responses. Parasitology, 122, S23–S38. https://doi.org/10.1017/s003118200001698x Dallas TA, Laine A-L, Ovaskainen O (2019) Detecting parasite associations within multi-species host and parasite communities. Proceedings of the Royal Society B: Biological Sciences, 286, 20191109. https://doi.org/10.1098/rspb.2019.1109 Daszak P, Cunningham AA, Hyatt AD (2000) Emerging Infectious Diseases of Wildlife - Threats to Biodiversity and Human Health. Science, 287, 443–449. https://doi.org/10.1126/science.287.5452.443 Davies TJ, Pedersen AB (2008) Phylogeny and geography predict pathogen community similarity in wild primates and humans. Proceedings. Biological sciences / The Royal Society, 275, 1695–701. https://doi.org/10.1098/rspb.2008.0284 Davis S, Calvet E, Leirs, H (2005) Fluctuating Rodent Populations and Risk to Humans from Rodent-Borne Zoonoses. Vector-Borne and Zoonotic Diseases, 5, 305–314. https://doi.org/10.1089/vbz.2005.5.305 Defosse DL, Johnson RC, Paster BJ, Dewhirst FE, Fraser GJ (1995) Brevinema andersonii gen. nov., sp. nov., an infectious spirochete isolated from the short-tailed shrew (Blarina brevicauda) and the white- footed mouse (Peromyscus leucopus). International journal of systematic bacteriology, 45, 78–84. https://doi.org/10.1099/00207713-45-1-78 Deter J, Chaval Y, Galan M, Gauffre B, Morand S, Henttonen H, Laakkonen J, Voutilainen L, Charbonnel N, Cosson J-F (2007) Kinship, dispersal and hantavirus transmission in bank and common voles. Archives of Virology, 153, 435. https://doi.org/10.1007/s00705-007-0005-6 Jessica L. Abbate et al. 23 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Diacovich L, Gorvel J-P (2010) Bacterial manipulation of innate immunity to promote infection. Nature Reviews Microbiology, 8, 117–128. https://doi.org/10.1038/nrmicro2295 Diagne C, Galan M, Tamisier L, d’Ambrosio J, Dalecky A, Bâ K, Kane M, Niang Y, Diallo M, Sow A, Gauthier P, Tatard C, Loiseau A, Piry S, Sembène M, Cosson J-F, Charbonnel N, Brouat C (2017) Ecological and sanitary impacts of bacterial communities associated to biological invasions in African commensal rodent communities. Scientific reports, 7, 14995. https://doi.org/10.1038/s41598-017-14880-1 Eidelman A, Cohen C, Navarro-Castilla Á, Filler S, Gutiérrez R, Bar-Shira E, Shahar N, Garrido M, Halle S, Romach Y, Barja I, Tasker S, Harrus S, Friedman A, Hawlena H (2019) The dynamics between limited- term and lifelong coinfecting bacterial parasites in wild rodent hosts. The Journal of Experimental Biology, 222, jeb203562. https://doi.org/10.1242/jeb.203562 Essbauer S, Pfeffer M, Meyer H (2010) Zoonotic poxviruses. Veterinary Microbiology, 140, 229–236. https://doi.org/10.1016/j.vetmic.2009.08.026 Ezenwa VO, Etienne RS, Luikart G, Beja-Pereira A, Jolles AE (2010) Hidden Consequences of Living in a Wormy World: Nematode-Induced Immune Suppression Facilitates Tuberculosis Invasion in African Buffalo. The American Naturalist, 176, 613–624. https://doi.org/10.1086/656496 Fenton A (2008) Worms and germs: the population dynamic consequences of microparasite- macroparasite co-infection. Parasitology, 135, 1545–1560. https://doi.org/10.1017/S003118200700025X Fenton A, Knowles SCL, Petchey OL, Pedersen AB (2014) The reliability of observational approaches for detecting interspecific parasite interactions: Comparison with experimental results. International Journal for Parasitology, 44, 437–445. https://doi.org/10.1016/j.ijpara.2014.03.001 Fenton A, Viney ME, Lello J (2010) Detecting interspecific macroparasite interactions from ecological data: Patterns and process. Ecology Letters, 13, 606–615. https://doi.org/10.1111/j.1461- 0248.2010.01458.x Fettweis JM, Serrano MG, Huang B, Brooks JP, Glascock AL, Sheth NU, Consortium VM, Strauss 3rd JF, Jefferson KK, Buck GA (2014) An emerging mycoplasma associated with trichomoniasis, vaginal infection and disease. PloS one, 9, e110943–e110943. https://doi.org/10.1371/journal.pone.0110943 Forbes KM, Voutilainen L, Jääskeläinen A, Sironen T, Kinnunen PM, Stuart P, Vapalahti O, Henttonen H, Huitu O (2014) Serological survey of rodent-borne viruses in Finnish field voles. Vector borne and zoonotic diseases (Larchmont, N.Y.), 14, 278–283. https://doi.org/10.1089/vbz.2013.1526 Fountain-Jones NM, Packer C, Jacquot M, Blanchet FG, Terio K, Craft ME (2019) Endemic infection can shape exposure to novel pathogens: Pathogen co-occurrence networks in the Serengeti lions. Ecology Letters, 22, 904–913. https://doi.org/10.1111/ele.13250 Galan M, Razzauti M, Bard E, Bernard M, Brouat C, Charbonnel N, Dehne-Garcia A, Loiseau A, Tatard C, Tamisier L, Vayssier-Taussat M, Vignes H, Cosson J-F (2016) 16S rRNA Amplicon Sequencing for Epidemiological Surveys of Bacteria in Wildlife. mSystems, 1, e00032-16. https://doi.org/10.1128/mSystems.00032-16 Galen SC, Speer KA, Perkins SL (2019) Evolutionary lability of host associations promotes phylogenetic overdispersion of co-infecting blood parasites. Journal of Animal Ecology, 88, 1936–1949. https://doi.org/10.1111/1365-2656.13089 Garchitorena A, Sokolow SH, Roche B, Ngonghala CN, Jocque M, Lund A, Barry M, Mordecai EA, Daily GC, Jones JH, Bonds MH, Leo GA De, Gue JF (2017) Disease ecology , health and the environment : a framework to account for ecological and socio-economic drivers in the control of neglected tropical diseases. Philosophical Transactions of the Royal Society B, 372, e20160128. https://doi.org/10.1098/rstb.2016.0128 Gaulke CA, Martins ML, Watral VG, Humphreys IR, Spagnoli ST, Kent ML, Sharpton TJ (2019) A longitudinal assessment of host-microbe-parasite interactions resolves the zebrafish gut microbiome’s link to Pseudocapillaria tomentosa infection and pathology. Microbiome, 7, 10. https://doi.org/10.1186/s40168-019-0622-9 Gorsich EE, Ezenwa VO, Jolles AE (2014) Nematode-coccidia parasite co-infections in African buffalo: Epidemiology and associations with host condition and pregnancy. International Journal for Parasitology: Parasites and Wildlife, 3, 124–134. https://doi.org/10.1016/j.ijppaw.2014.05.003 24 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Gotteland C, Chaval Y, Villena I, Galan M, Geers R, Aubert D, Poulle M-L, Charbonnel N, Gilot-Fromont E (2014) Species or local environment, what determines the infection of rodents by Toxoplasma gondii? Parasitology, 141, 259–268. https://doi.org/10.1017/S0031182013001522 Graham JE, Schoeb TR (2011) Mycoplasma pulmonis in Rats. Journal of Exotic Pet Medicine, 20, 270–276. https://doi.org/10.1053/j.jepm.2011.07.004 Griffiths EC, Pedersen AB, Fenton A, Petchey OL, Sheffield S (2011) The nature and consequences of coinfection in humans. The Journal of infection, 63, 200–6. https://doi.org/10.1016/j.jinf.2011.06.005 Guivier E, Galan M, Henttonen H, Cosson J-F, Charbonnel N (2014) Landscape features and helminth co- infection shape bank vole immunoheterogeneity, with consequences for Puumala virus epidemiology. Heredity, 112, 274–281. https://doi.org/10.1038/hdy.2013.103 Guy L, Nystedt B, Toft C, Zaremba-Niedzwiedzka K, Berglund EC, Granberg F, Näslund K, Eriksson AS, Andersson SGE (2013) A Gene Transfer Agent and a Dynamic Repertoire of Secretion Systems Hold the Keys to the Explosive Radiation of the Emerging Pathogen Bartonella. PLoS Genetics, 9, e1003393. https://doi.org/10.1371/journal.pgen.1003393 Guy C, Thiagavel J, Mideo N, Ratcliffe JM (2019) Phylogeny matters: revisiting ‘a comparison of bats and rodents as reservoirs of zoonotic viruses.’ Royal Society Open Science, 6, 181182. https://doi.org/10.1098/rsos.181182 Han BA, Schmidt JP, Bowden SE, Drake JM (2015) Rodent reservoirs of future zoonotic diseases. Proceedings of the National Academy of Sciences, 112, 7039–7044. https://doi.org/10.1073/pnas.1501598112 Harvell CD, Mitchell CE, Ward JR, Altizer S, Dobson AP, Ostfeld RS, Samuel MD (2002) Climate warming and disease risks for terrestrial and marine biota. Science, 296, 2158–2162. https://doi.org/10.1126/science.1063699 Harwick HJ, Kalmanson GM, Guze LB (1972) Human diseases associated with Mycoplasmas. California Medicine, 116, 1–7. https://europepmc.org/article/pmc/pmc1518391 Iralu J, Bai Y, Crook L, Tempest B, Simpson G, McKenzie T, Koster F (2006) Rodent-associated Bartonella febrile illness, southwestern United States. Emerging Infectious Diseases, 12, 1081–1086. https://doi.org/10.3201/eid1207.040397 Johnston JB, McFadden G (2003) Poxvirus Immunomodulatory Strategies: Current Perspectives. Journal of Virology, 77, 6093 LP – 6100. https://doi.org/10.1128/JVI.77.11.6093-6100.2003 Jolles AE, Ezenwa VO, Etienne RS, Turner WC, Olff H (2008) Interactions between macroparasites and microparasites drive infection patterns in free-ranging african buffalo. Ecology, 89, 2239–2250. https://doi.org/10.1890/07-0995.1 Jones KE, Patel NG, Levy M a, Storeygard A, Balk D, Gittleman JL, Daszak P (2008) Global trends in emerging infectious diseases. Nature, 451, 990–3. https://doi.org/10.1038/nature06536 Kallio-Kokko H, Laakkonen J, Rizzoli A, Tagliapietra V, Cattadori I, Perkins SE, Hudson PJ, Cristofolini A, Versini W, Vapalahti O, Vaheri A, Henttonen H (2006) Hantavirus and arenavirus antibody prevalence in rodents and humans in Trentino, Northern Italy. Epidemiology and Infection, 134, 830. https://doi.org/10.1017/S0950268805005431 Karvonen A, Jokela J, Laine A-L (2019) Importance of Sequence and Timing in Parasite Coinfections. Trends in Parasitology, 35, 109–118. https://doi.org/10.1016/j.pt.2018.11.007 Kedem H, Cohen C, Messika I, Einav M, Pilosof S, Hawlena H (2014) Multiple effects of host-species diversity on coexisting host-specific and host-opportunistic microbes. Ecology, 95, 1173–1183. https://doi.org/10.1890/13-0678.1 Keesing F, Belden LK, Daszak P, Dobson A, Harvell CD, Holt RD, Hudson P, Jolles A, Jones KE, Mitchell CE, Myers SS, Bogich T, Ostfeld RS (2010) Impacts of biodiversity on the emergence and transmission of infectious diseases. Nature, 468, 647–52. https://doi.org/10.1038/nature09575 Knowles SCL, Fenton A, Petchey OL, Jones TR, Barber R, Pedersen AB (2013) Stability of within-host- parasite communities in a wild mammal system. Proceedings. Biological sciences, 280, 20130598. https://doi.org/10.1098/rspb.2013.0598 Koskela KA, Kalin-Mänttäri L, Hemmilä H, Smura T, Kinnunen PM, Niemimaa J, Henttonen H, Nikkari S (2016) Metagenomic Evaluation of Bacteria from Voles. Vector-Borne and Zoonotic Diseases, 17, 123– 133. https://doi.org/10.1089/vbz.2016.1969 Jessica L. Abbate et al. 25 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Kosoy M, Mandel E, Green D, Marston E, Jones D, Childs J (2004) Prospective Studies of Bartonella of Rodents. Part II. Diverse Infections in a Single Rodent Community. Vector-Borne and Zoonotic Diseases, 4, 296–305. https://doi.org/10.1089/vbz.2004.4.296 Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD (2013) Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the miseq illumina sequencing platform. Applied and Environmental Microbiology, 79, 5112–5120. https://doi.org/10.1128/AEM.01043-13 Krasnov BR, Mouillot D, Shenbrot GI, Khokhlova IS, Poulin R (2010) Deconstructing spatial patterns in species composition of ectoparasite communities: the relative contribution of host composition, environmental variables and geography. Global Ecology and Biogeography, 19, 515–526. https://doi.org/10.1111/j.1466-8238.2010.00529.x Kruse H, Kirkemo AM, Handeland K (2004) Wildlife as source of zoonotic infections. Emerging Infectious Diseases, 10, 2067–2072. https://doi.org/10.3201/eid1012.040707 Kumar N, Sharma S, Barua S, Tripathi BN, Rouse T (2018) Virological and Immunological Outcomes of Coinfections. Clinical Microbiology Reviews, 31, e00111-17. https://doi.org/10.1128/CMR.00111-17 Lello J, Boag B, Fenton A, Stevenson IR, Hudson PJ (2004) Competition and mutualism among the gut helminths of a mammalian host. Nature, 428, 840–844. https://doi.org/10.1038/nature02490 Lello J, McClure SJ, Tyrrell K, Viney ME (2018) Predicting the effects of parasite co-infection across species boundaries. Proceedings. Biological sciences, 285, 20172610. https://doi.org/10.1098/rspb.2017.2610 Lindquist L, Vapalahti O (2008) Tick-borne encephalitis. The Lancet, 371, 1861–1871. https://doi.org/10.1016/S0140-6736(08)60800-4 Luis AD, Hayman DTS, O’Shea TJ, Cryan PM, Gilbert AT, Pulliam JRC, Mills JN, Timonin ME, Willis CKR, Cunningham AA, Fooks AR, Rupprecht CE, Wood JLN, Webb CT (2013) A comparison of bats and rodents as reservoirs of zoonotic viruses: are bats special? Proceedings. Biological sciences, 280, 20122753. https://doi.org/10.1098/rspb.2012.2753 Malmqvist B, Strasevicius D, Hellgren O, Adler PH, Bensch S (2004) Vertebrate host specificity of wild– caught blackflies revealed by mitochondrial DNA in blood. Proceedings of the Royal Society of London. Series B: Biological Sciences, 271, S152–S155. https://doi.org/10.1098/rsbl.2003.0120 Mangombi JB, N’dilimabaka N, Lekana-Douki J-B, Banga O, Maghendji-Nzondo S, Bourgarel M, Leroy E, Fenollar F, Mediannikov O (2021) First investigation of pathogenic bacteria, protozoa and viruses in rodents and shrews in context of forest-savannah-urban areas interface in the city of Franceville (Gabon). PLOS ONE, 16, e0248244. https://doi.org/10.1371/journal.pone.0248244 Matar GM, Koehler JE, Malcolm G, Lambert-Fair MA, Tappero J, Hunter SB, Swaminathan B (1999) Identification of Bartonella species directly in clinical specimens by PCR-restriction fragment length polymorphism analysis of a 16S rRNA gene fragment. Journal of Clinical Microbiology, 37, 4045–4047. https://doi.org/10.1128/jcm.37.12.4045-4047.1999 Maurice CF, Cl Knowles S, Ladau J, Pollard KS, Fenton A, Pedersen AB, Turnbaugh PJ (2015) Marked seasonal variation in the wild mouse gut microbiota. ISME Journal, 9, 2423–2434. https://doi.org/10.1038/ismej.2015.53 Messick JB (2004) Hemotrophic mycoplasmas (hemoplasmas): a review and new insights into pathogenic potential. Veterinary clinical pathology / American Society for Veterinary Clinical Pathology, 33, 2–13. https://doi.org/10.1111/j.1939-165X.2004.tb00342.x Meyer BJ, Schmaljohn CS (2000) Persistent hantavirus infections: characteristics and mechanisms. Trends in Microbiology, 8, 61–67. https://doi.org/10.1016/S0966-842X(99)01658-3 Morand S (2015) (macro-) Evolutionary ecology of parasite diversity: From determinants of parasite species richness to host diversification. International journal for parasitology. Parasites and wildlife, 4, 80–87. https://doi.org/10.1016/j.ijppaw.2015.01.001 Morand S, Bordes F (2015) Parasite diversity of disease-bearing rodents of Southeast Asia: habitat determinants and effects on sexual size dimorphism and life-traits. Frontiers in Ecology and Evolution, 3, 1. https://doi.org/10.3389/fevo.2015.00110 Mouillot D, Šimková A, Morand S, Poulin R (2005) Parasite species coexistence and limiting similarity: a multiscale look at phylogenetic, functional and reproductive distances. Oecologia, 146, 269–278. https://doi.org/10.1007/s00442-005-0194-1 26 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Moutailler S, Moro CV, Vaumourin E, Michelet L (2016) Co-infection of Ticks : The Rule Rather Than the Exception. PLoS Neglected Tropical Diseases, 10, e0004539. https://doi.org/10.1371/journal.pntd.0004539 Murall CL, Abbate JL, Puelma Touzel M, Allen-Vercoe E, Alizon S, Froissart R, McCann K (2017) Invasions of Host-Associated Microbiome Networks. Elsevier Ltd. https://doi.org/10.1016/bs.aecr.2016.11.002 Neimark H, Johansson KE, Rikihisa Y, Tully JG (2001) Proposal to transfer some members of the genera Haemobartonella and Eperythrozoon to the genus Mycoplasma with descriptions of “Candidatus Mycoplasma haemofelis”, “Candidatus Mycoplasma haemomuris”, “Candidatus Mycoplasma haemosuis” a. International Journal of Systematic and Evolutionary Microbiology, 51, 891–899. https://doi.org/10.1099/00207713-51-3-891 Neimark H, Mitchelmore D, Leach RH (1998) An approach to characterizing uncultivated prokaryotes: the Grey Lung agent and proposal of a Candidatus taxon for the organism, “Candidatus Mycoplasma ravipulmonis”. International journal of systematic bacteriology, 48 Pt 2, 389–94. https://doi.org/10.1099/00207713-48-2-389 Neimark H, Peters W, Robinson BL, Stewart LB (2005) Phylogenetic analysis and description of Eperythrozoon coccoides, proposal to transfer to the genus Mycoplasma as Mycoplasma coccoides comb. nov. and request for an opinion. International Journal of Systematic and Evolutionary Microbiology, 55, 1385–1391. https://doi.org/10.1099/ijs.0.63437-0 Nunn CL, Altizer S, Jones KE, Sechrest W (2003) Comparative Tests of Parasite Species Richness in Primates. The American Naturalist, 162, 597–614. https://doi.org/10.1086/378721 Pawelczyk A, Bajer A, Behnke JM, Gilbert FS, Sinski E (2004) Factors affecting the component community structure of haemoparasites in common voles (Microtus arvalis) from the Mazury Lake District region of Poland. Parasitology Research, 92, 270–284. https://doi.org/10.1007/s00436-003-1040-1 Pedersen AB, Altizer S, Poss M, Cunningham AA, Nunn CL (2005) Patterns of host specificity and transmission among parasites of wild primates. International Journal for Parasitology, 35, 647–657. https://doi.org/10.1016/j.ijpara.2005.01.005 Pedersen AB, Fenton A (2007) Emphasizing the ecology in parasite community ecology. Trends in Ecology and Evolution, 22, 133–139. https://doi.org/10.1016/j.tree.2006.11.005 Petney TN, Andrews RH (1998) Multiparasite communities in animals and humans: frequency, structure and pathogenic significance. International Journal for Parasitology, 28, 377–393. https://doi.org/10.1016/S0020-7519(97)00189-6 Piasecki T, Chrzastek K, Kasprzykowska U (2017) Mycoplasma pulmonis of Rodents as a Possible Human Pathogen. Vector-Borne and Zoonotic Diseases, 17, 475–477. https://doi.org/10.1089/vbz.2016.2104 Pilosof S, Morand S, Krasnov BR, Nunn CL (2015) Potential Parasite Transmission in Multi-Host Networks Based on Parasite Sharing. PLOS ONE, 10, 1–19. https://doi.org/10.1371/journal.pone.0117909 Pitcher DG, Nicholas RAJ (2005) Mycoplasma host specificity: Fact or fiction? Veterinary Journal, 170, 300–306. https://doi.org/10.1016/j.tvjl.2004.08.011 Poulin R (1995) Phylogeny, ecology, and the richness of parasite communities in vertebrates. Ecological Monographs, 65, 283–302. https://doi.org/10.2307/2937061 Poulin R, Blanar CA, Thieltges DW, Marcogliese DJ (2012) Scaling up from epidemiology to biogeography: local infection patterns predict geographical distribution in fish parasites. Journal of Biogeography, 39, 1157–1166. https://doi.org/10.1111/j.1365-2699.2011.02667.x Poulin R, Morand S (2000) The Diversity of Parasites. The Quarterly Review of Biology, 75, 277–293. https://doi.org/10.1086/393500 Quiñones-Parra SM, Clemens EB, Wang Z, Croom HA, Kedzierski L, McVernon J, Vijaykrishna D, Kedzierska K (2016) A Role of Influenza Virus Exposure History in Determining Pandemic Susceptibility and CD8 + T Cell Responses . Journal of Virology, 90, 6936–6947. https://doi.org/10.1128/jvi.00349-16 R Core Team (2015) R: A language and environment for statistical computing. Razzauti M, Galan M, Bernard M, Maman S, Klopp C, Charbonnel N, Vayssier-Taussat M, Eloit M, Cosson J-F (2015) A Comparison between Transcriptome Sequencing and 16S Metagenomics for Detection of Bacterial Pathogens in Wildlife. PLOS Neglected Tropical Diseases, 9, 1–21. https://doi.org/10.1371/journal.pntd.0003929 Rosshart SP, Vassallo BG, Angeletti D, Hutchinson DS, Morgan AP, Takeda K, Hickman HD, McCulloch JA, Badger JH, Ajami NJ, Trinchieri G, Pardo-Manuel de Villena F, Yewdell JW, Rehermann B (2017) Wild Jessica L. Abbate et al. 27 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Mouse Gut Microbiota Promotes Host Fitness and Improves Disease Resistance. Cell, 171, 1015- 1028.e13. https://doi.org/10.1016/j.cell.2017.09.016 Sallinen S, Norberg A, Susi H, Laine AL (2020) Intraspecific host variation plays a key role in virus community assembly. Nature Communications, 11, 1–11. https://doi.org/10.1038/s41467-020-19273- z Salter SJ, Cox MJ, Turek EM, Calus ST, Cookson WO, Moffatt MF, Turner P, Parkhill J, Loman NJ, Walker AW (2014) Reagent and laboratory contamination can critically impact sequence-based microbiome analyses. BMC Biology, 12, 87. https://doi.org/10.1186/s12915-014-0087-z Salvador AR, Guivier E, Xuéreb A, Chaval Y, Cadet P, Poulle M-L, Sironen T, Voutilainen L, Henttonen H, Cosson J-F, Charbonnel N (2011) Concomitant influence of helminth infection and landscape on the distribution of Puumala hantavirus in its reservoir, Myodes glareolus. BMC Microbiology, 11, 30. https://doi.org/10.1186/1471-2180-11-30 Sauvage F, Penalba C, Vuillaume P, Boue F, Coudrier D, Pontier D, Artois M (2002) Puumala hantavirus infection in humans and in the reservoir host, Ardennes region, France. Emerging infectious diseases, 8, 1509–1511. https://doi.org/10.3201/eid0812.010518 Seppälä O, Jokela J (2016) Do Coinfections Maintain Genetic Variation in Parasites? Trends in parasitology, 32, 930–938. https://doi.org/10.1016/j.pt.2016.08.010 Singer M (2010) Pathogen-pathogen interaction: A syndemic model of complex biosocial processes in disease. Virulence, 1, 10–18. https://doi.org/10.4161/viru.1.1.9933 Smith DP, Peay KG (2014) Sequence Depth, Not PCR Replication, Improves Ecological Inference from Next Generation DNA Sequencing. PLOS ONE, 9, e90234. https://doi.org/10.1371/journal.pone.0090234 Smith KF, Sax DOVF, Lafferty KD (2006) Evidence for the Role of Infectious Disease in Species Extinction and Endangerment. Conservation Biology, 20, 1349–1357. https://doi.org/10.1111/j.1523- 1739.2006.00524.x Streicker DG, Fenton A, Pedersen AB (2013) Differential sources of host species heterogeneity influence the transmission and control of multihost parasites. Ecology Letters, 16, 975–984. https://doi.org/10.1111/ele.12122 Streicker DG, Turmelle AS, Vonhof MJ, Kuzmin I V, McCracken GF, Rupprecht CE (2010) Host phylogeny constrains cross-species emergence and establishment of rabies virus in bats. Science (New York, N.Y.), 329, 676–9. https://doi.org/10.1126/science.1188836 Stutz WE, Blaustein AR, Briggs CJ, Hoverman JT, Rohr JR, Johnson PTJ (2018) Using multi-response models to investigate pathogen coinfections across scales: Insights from emerging diseases of amphibians. Methods in Ecology and Evolution, 9, 1109–1120. https://doi.org/10.1111/2041-210X.12938 Swanson SJ, Neitzel D, Reed KD, Belongia EA (2006) Coinfections acquired from ixodes ticks. Clinical microbiology reviews, 19, 708–727. https://doi.org/10.1128/CMR.00011-06 Sweeny AR, Thomason CA, Carbajal EA, Hansen CB, Graham AL, Pedersen AB (2020) Experimental parasite community perturbation reveals associations between Sin Nombre virus and gastrointestinal nematodes in a rodent reservoir host: SNV-nematode associations in the wild. Biology Letters, 16, 20200604. https://doi.org/10.1098/rsbl.2020.0604 Sykes JE, Terry JC, Lindsay LL, Owens SD (2008) Prevalences of various hemoplasma species among cats in the United States with possible hemoplasmosis. Journal of the American Veterinary Medical Association, 232, 372–379. https://doi.org/10.2460/javma.232.3.372 Tadin A, Turk N, Korva M, Margaletić J, Beck R, Vucelja M, Habuš J, Svoboda P, Zupanc TA, Henttonen H, Markotić A (2012) Multiple co-infections of rodents with hantaviruses, Leptospira, and Babesia in Croatia. Vector borne and zoonotic diseases (Larchmont, N.Y.), 12, 388–392. https://doi.org/10.1089/vbz.2011.0632 Tagawa M, Takeuchi T, Fujisawa T, Konno Y, Yamamoto S, Matsumoto K, Yokoyama N, Inokuma H (2012) A Clinical Case of Severe Anemia in a Sheep Coinfected with Mycoplasma ovis and ‘Candidatus Mycoplasma haemovis’ in Hokkaido, Japan. Journal of Veterinary Medical Science, 74, 99–102. https://doi.org/10.1292/jvms.11-0296 Taylor LH, Latham SM, Woolhouse MEJ (2001) Risk factors for human disease emergence. Philosophical Transactions of the Royal Society London B, 356, 983–989. https://doi.org/10.1098/rstb.2001.0888 Tegshduuren E, Yoshimatsu K, Taruishi M, Endo R, Shimizu K, Koma T, Yasuda SP, Kariwa H, Arikawa J, Ishihara C (2010) Different cross-reactivity of human and rodent sera to Tula virus and Puumala virus. 28 Jessica L. Abbate et al. Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370 Comparative Immunology, Microbiology and Infectious Diseases, 33, e67–e73. https://doi.org/10.1016/j.cimid.2010.01.001 Telfer S, Lambin X, Birtles R, Beldomenico P, Burthe S, Paterson S, Begon M (2010) Species Interactions in a Parasite Community Drive Infection Risk in a Wildlife Population. Science, 330, 243–246. https://doi.org/10.1126/science.1190333 Thumbi SM, De. C. Bronsvoort BM, Poole EJ, Kiara H, Toye P, Ndila M, Conradie I, Jennings A, Handel IG, Coetzer JAW, Hanotte O, Woolhouse MEJ (2013) Parasite co-infections show synergistic and antagonistic interactions on growth performance of East African zebu cattle under one year. Parasitology, 140, 1789–1798. https://doi.org/10.1017/S0031182013001261 Vaheri A, Strandin T, Hepojoki J, Sironen T, Henttonen H, Mäkelä S, Mustonen J (2013) Uncovering the mysteries of hantavirus infections. Nature Reviews Microbiology, 11, 539–550. https://doi.org/10.1038/nrmicro3066 Vaumourin E, Vourc’h G, Telfer S, Lambin X, Salih D, Seitzer U, Morand S, Charbonnel N, Vayssier-Taussat M, Gasqui P (2014) To be or not to be associated: power study of four statistical modeling approaches to identify parasite associations in cross-sectional studies. Frontiers in Cellular and Infection Microbiology, 4, 1–11. https://doi.org/10.3389/fcimb.2014.00062 Vayssier-Taussat M, Moutailler S, Femenia F, Raymond P, Croce O, La Scola B, Fournier PE, Raoult D (2016) Identification of novel zoonotic activity of Bartonella spp., France. Emerging Infectious Diseases, 22, 457–462. https://doi.org/10.3201/eid2203.150269 Vayssier-Taussat M, Le Rhun D, Buffet J-P, Maaoui N, Galan M, Guivier E, Charbonnel N, Cosson J-F (2012) Candidatus Neoehrlichia mikurensis in bank voles, France. Emerging infectious diseases, 18, 2063– 2065. https://doi.org/10.3201/eid1812.120846 Villette P, Afonso E, Couval G, Levret A, Galan M, Goydadin AC, Cosson JF, Giraudoux P (2020) Spatio- temporal trends in richness and persistence of bacterial communities in decline-phase water vole populations. Scientific Reports, 10. https://doi.org/10.1038/s41598-020-66107-5 Villette P, Afonso E, Couval G, Levret A, Galan M, Tatard C, Cosson JF, Giradoux P (2017) Consequences of organ choice in describing bacterial pathogen assemblages in a rodent population. Epidemiology and Infection, 145, 3070–3075. https://doi.org/10.1017/S0950268817001893 Volokhov D V, Hwang J, Chizhikov VE, Danaceau H, Gottdenker NL (2017) Prevalence, Genotype Richness, and Coinfection Patterns of Hemotropic Mycoplasmas in Raccoons (Procyon lotor) on Environmentally Protected and Urbanized Barrier Islands. Applied and environmental microbiology, 83, e00211-17. https://doi.org/10.1128/AEM.00211-17 Weiss S, Van Treuren W, Lozupone C, Faust K, Friedman J, Deng Y, Xia LC, Xu ZZ, Ursell L, Alm EJ, Birmingham A, Cram JA, Fuhrman JA, Raes J, Sun F, Zhou J, Knight R (2016) Correlation detection strategies in microbial data sets vary widely in sensitivity and precision. The ISME Journal, 10, 1669– 1681. https://doi.org/10.1038/ismej.2015.235 Withenshaw SM, Devevey G, Pedersen AB, Fenton A (2016) Multihost Bartonella parasites display covert host specificity even when transmitted by generalist vectors. The Journal of animal ecology, 85, 1442– 1452. https://doi.org/10.1111/1365-2656.12568 Woolhouse MEJ, Taylor LH, Haydon DT (2001) Population Biology of Multihost Pathogens. Science, 292, 1109–1112. https://doi.org/10.1126/science.1059026 Yanagihara R, Amyx HL, Gajdusek DC (1985) Experimental infection with Puumala virus, the etiologic agent of nephropathia epidemica, in bank voles (Clethrionomys glareolus). Journal of virology, 55, 34– 38. https://doi.org/10.1128/jvi.55.1.34-38.1985 Zemb O, Achard CS, Hamelin J, De Almeida M-L, Gabinaud B, Cauquil L, Verschuren LMG, Godon J-J (2020) Absolute quantitation of microbes using 16S rRNA gene metabarcoding: A rapid normalization of relative abundances by quantitative PCR targeting a 16S rRNA gene spike-in standard. MicrobiologyOpen, n/a, e977. https://doi.org/10.1002/mbo3.977 Jessica L. Abbate et al. 29 Peer Community Journal, Vol. 4 (2024), article e14 https://doi.org/10.24072/pcjournal.370