- 1Department of Pathobiology and Population Sciences, The Royal Veterinary College, Hatfield, Hertfordshire, United Kingdom
- 2Department of Biology, University of Oxford, Oxford, Oxfordshire, United Kingdom
The mammalian gastrointestinal tract provides a habitat for multiple commensal and pathogenic organisms spanning all three domains of life. Both positive and negative interactions occur between gut inhabitants, with potential consequences for host health. Studies of parasite–microbiota associations in natural systems remain scarce, yet are important for understanding how parasite communities and commensal microbiota shape each other, and how these interactions influence host health. Here, we characterize associations between helminth and coccidial infections and gut microbiota profiles in a wild population of wood mice (Apodemus sylvaticus) over 3 years, using two complementary approaches. We first examined parasite–microbiota associations along the length of the gastrointestinal tract through destructive sampling. Then, in a larger non-invasive capture mark-recapture study, we assessed whether gut parasitic infections detected in feces predicted fecal microbiota diversity and composition. We found that while overall microbiota composition was not associated with infection by any common gut parasite species, microbiota richness was associated with gut parasitism in two ways: (i) infection by the trematode Corrigia vitta in the small intestine predicted higher microbiota diversity in the caecum; (ii) there was a negative relationship between gut parasite richness and fecal microbiota richness in the non-invasive study. As our results identified associations between gastrointestinal parasites and microbiota alpha diversity, a future experimental study in this tractable wild mammalian system would be valuable to definitively test the directionality of these interactions.
Introduction
It is becoming increasingly recognized that eukaryotic parasites and bacteria can interact within the host’s gastrointestinal (GI) tract, with potentially important consequences for host health and disease (Reynolds et al., 2015). For instance, parasite–microbiota interactions may help to explain how one or both types of gut inhabitant influence host nutrition (Million et al., 2017; Subramanian et al., 2014), context-dependency in gut symbiont pathogenicity (Leung et al., 2018a; Leung et al., 2018b) and may also help predict non-target effects of drug treatment programs that may impact their effectiveness (Easton et al., 2019). Gut parasites could interact with the microbiota through a wide range of mechanisms including direct physical interaction (e.g., predation; White et al., 2018), altering the physical, chemical, or nutritional environment within the gut, and/or through effects on the host immune system (Leung et al., 2018b; Loke and Harris, 2023). Therefore, parasites and gut bacteria need not co-inhabit in the same region of the GI tract to affect one another but instead may have up- or downstream effects on one another through indirect mechanisms of interaction.
Perhaps the most well-studied mechanisms of parasite–microbiota interactions are those involving the host immune system, as both gut microbes and certain parasites are known to have immunomodulatory abilities (Honda and Littman, 2016; Mcsorley et al., 2013), which together have shaped the evolution of the mammalian immune system (Gause and Maizels, 2016). These complex interactions, involving both innate and adaptive immunity, can occur in both directions, with microbiota effects on immunity affecting a parasite and vice versa. For instance, the intestinal nematode Heligmosomoides polygyrus and Lactobacillus bacteria are thought to promote each other’s persistence by inducing tolorogenic T regulatory cell (Treg) responses (Ohnmacht et al., 2015; Reynolds et al., 2014), while Lactobacillus are thought to also increase susceptibility to the nematode Trichuris muris through modulation of host Th2 responses (Dea-ayuela et al., 2008). Conversely, the gut microbiota is thought to provide some protection against infection by the apicomplexan parasite Toxoplasma gondii through priming innate immune responses, as shown in experimentally infected laboratory mice (Benson et al., 2009).
Gut parasites and the microbiota may also interact via physical changes they cause in the gut environment. For instance, laboratory-based experiments have shown that the helminth T. muris and protozoa Eimeria spp. both induce higher mucus production and alter mucus composition in the gut, which is thought to subsequently promote the growth of mucin-utilizing gut bacteria such as Clostridiales (Collier et al., 2008; Ramanan et al., 2016). Parasite-induced tissue damage can also allow for bacterial translocation across the gut epithelial barrier, with potential impacts on host health (Egan et al., 2012). Parasites and gut bacteria can even engage in direct physical interactions, as, for example, occurs during the hatching of T. muris eggs, which require the attachment of commensal bacteria to their polar egg caps (Hayes et al., 2010). Furthermore, T. muris ingest bacteria from the gut environment, restructuring cecal microbiota in a way that limits further T. muris colonization (White et al., 2018).
Whilst controlled experiments in laboratory animals can enable a better understanding of the mechanisms underlying parasite-microbiota interactions, they do not tell us how parasites and gut microbes may interact in complex natural settings. In nature, both the internal gut and external habitat of the host are highly diverse and dynamic, where hosts are exposed to diverse sets of microbes and parasites, are frequently coinfected, and experience strong selection pressures from multiple sources. As such, many more factors shape gut community variation in the wild compared to laboratory animals. Wild systems therefore provide useful model systems in which to further our fundamental understanding of within-host community ecology, particularly in well-characterized systems where experimentation is possible (Ezenwa and Jolles, 2015; Johnson et al., 2015; Knowles et al., 2013).
Only a handful of studies have, however, examined GI parasite–microbiota associations in wild mammal populations (Scotti et al., 2020) and as such we currently do not have a consistent picture of the implications gut parasite–microbe interactions may have in natural settings. The movement of laboratory mice into semi-natural enclosures, for instance, has been shown to induce parallel changes in host immune responses, gut microbiota composition, and susceptibility to gut nematode infection (Leung et al., 2018a). Another study by Kreisinger et al. (2015) examined helminth–microbiota associations along the length of the GI tract in a population of yellow-necked mice (Apodemus flavicollis) and identified a broad range of gut helminth–microbiota associations, the most pronounced occurring between the tapeworm Hymenolepis in the small intestine and the Bacteroidales family S24-7 (Muribaculaceae) in the stomach. In the closely related wood mouse (Apodemus sylvaticus), analysis of faecal samples suggested nematode infection (largely H. polygyrus) was associated with an increased abundance of Escherichia and decreased abundance of Lachnospiraceae (Maurice et al., 2015). Finally, in longitudinally sampled mouse lemurs (Microcebus rufus), Eimeria spp. infection was associated with increased gut microbial diversity, while two species of Hymenolepis had varying associations with different bacterial taxa (Aivelo and Norberg, 2018).
Inferring ecological interactions among host-associated symbionts can be done in multiple ways, using either longitudinal or cross-sectional study designs. Longitudinal studies tracking within-host covariance of symbionts over time can be more powerful than cross-sectional studies for inferring real interactions between co-infecting organisms (Fenton et al., 2014). However, this type of longitudinal monitoring requires non-invasive sampling (e.g., from fecal samples), which is not sufficient to detect covariation in parasite and microbiota at different points along the GI tract. Understanding the spatial distribution of parasites and microbes in the gut is important, as it would facilitate further insight into potential mechanisms of interaction. As such, both destructive cross-sectional and non-invasive longitudinal studies have merit in studying parasite-microbiota associations, and the use of both approaches in a single system allows for simultaneous exploration of both parasite–microbiota associations along the GI tract, as well as over time.
Here, we combine both destructive cross-sectional and non-invasive longitudinal sampling methods to examine parasite–microbiota associations in an intensively monitored wild population of wood mice (Apodemus sylvaticus). This host species provides an excellent study system for examining parasite–microbiota associations in the wild, as both their parasite and gut microbial communities have been previously well-characterized (Marsh et al., 2022; Maurice et al., 2015; Pedersen and Fenton, 2019; Sweeny et al., 2023). Seasonal dynamics in the gut microbiota of our study population were found to be driven largely by bacterial taxa in the Ruminococcaeae, Muribaculaceae, Lactobacillaceae, and Bifidobacteriaceae families (Marsh et al., 2022).
Furthermore, wild wood mice tend to be co-infected with multiple species of GI parasitic helminths and protozoa, which in some cases (H. polygyrus, T. muris) have been adopted as laboratory mouse models for human infections. Seasonal dynamics in parasitic infections have been documented for several of these parasite species (Abu-Madi et al., 2000; Higgs and Nowell, 2000; Montgomery and Montgomery, 1988). We, therefore, first examined whether local and/or distant associations between gut parasites and microbiota can be detected along the GI tract (occurring within or across different GI sections), which may hint at potential mechanisms of interaction. We then asked whether gut parasites and microbiota show detectable covariation over time, and in particular, whether changes in parasite infection levels could explain the seasonal microbiota dynamics previously characterized (Marsh et al., 2022).
Methods
Fieldwork
Non-invasive study
Regular trapping (approximately every 2–4 weeks) was carried out on a single 2.4 ha grid between October 2015 and October 2018 in Wytham Woods, Oxfordshire (51°46’N, 1°20’W), as previously described (Marsh et al., 2022). Briefly, sterilized small Sherman traps baited with six peanuts, a slice of apple, and sterile cotton wool for bedding were set at dusk and collected at dawn the following day. Individuals were PIT-tagged at first capture for identification and the sex, reproductive status, and body mass (g) were recorded for each capture. After tagging and measurements, mice were released at their point of capture to be subsequently followed as part of a longitudinal capture–mark–recapture (CMR) study. From the traps, up to 300 mg of feces was collected for molecular work (including gut microbiota profiling) using sterilized tweezers and stored at −80°C within 10 h of trap collection. Whenever possible, additional feces was collected, weighed, and stored in 10% formalin for later gut parasite screening using fecal flotation (see below).
Dissections
Between October 2017 and October 2018 (on the same trapping nights as the non-invasive longitudinal study), trapping was conducted at 8 additional locations each 40–300 m from the main trapping location within Wytham Woods, to perform the dissection study (see Supplementary Figure S1 for a map of grid locations). These grids were sufficiently far from the main grid that no PIT-tagged mice were ever captured there. All trapping grids were located in mixed deciduous woodland with the dominant tree species of oak, ash, and beech. Traps were pre-baited for 2–3 nights to encourage occupancy when set. Any non-target individuals and species were immediately excluded and released at their point-of-capture, as were any juvenile, pregnant, or lactating wood mice. Across 13 trapping occasions throughout the year at these additional grids, a total of 50 wood mice were humanely euthanized by intraperitoneal injection of pentobarbital followed by cervical dislocation to confirm death. Although traps were set on multiple occasions at each dissection grid, for most grids animals were only successfully captured and sampled on one occasion. The same set of morphometric measurements collected in the non-invasive CMR study (see above) were recorded (Marsh et al., 2022), and fecal samples were collected from traps and stored exactly as described above. Euthanized mice were dissected, and the GI tract was divided into five sections that were used for both microbiota and helminth infection characterization: the small intestine was divided into thirds, which approximated the duodenum, jejunum, and ileum, and the caecum and colon constituted the final two sections. For each of these five gut sections, an approx. 1 cm section of tissue and contents was removed from the middle and immediately put on dry ice before storage at −80°C for microbiota characterization. All remaining material for each section was stored in 70% ethanol for later visual examination under a dissecting microscope to quantify helminth burdens. For cestodes, absolute counts (worm burden) were not possible as scolexes (the point of attachment of an individual worm to the host) were typically not found, and therefore only their presence/absence in the whole gut was recorded using the presence of proglottids (which may have come from one or multiple individual tapeworms with unknown attachment site).
Fecal flotations
Gut parasites were detected from fecal samples using fecal flotation with sodium nitrate solution (Vetlab Supplies Ltd., specific gravity 1.2 ± 0.005). This flotation solution is expected to allow detection of most nematode and cestode species, as well as Eimeria spp. oocysts, but not trematode eggs, whose larger eggs would require the use of a flotation solution of a different density. Common pinworm (Syphacia) species are also expected to have poor detection using this method due to their mode of reproduction, in that the eggs are directly deposited around the anus and may not be present in feces (Sousa et al., 2016). Formalin-stored fecal samples were sieved and placed in 15 mL falcon tubes before a flotation solution was added to form a meniscus. A microscope cover slip was placed on top and samples spun for 10 min at 1000 rpm in a swinging bucket centrifuge. The coverslip was then removed and placed on a microscope slide for parasite identification. Each coverslip was visually divided into 12 columns for systematic counts of eggs/oocysts under a light microscope at 10x magnification. 40x magnification was used for more accurate species identification where necessary. If the egg/oocyst load was too high for accurate counts across the whole slide, counts were conducted on a smaller section (usually half) of the slide and scaled up to represent the whole slide. Eggs/oocysts per gram (EPG) of feces for each uniquely identified parasite were calculated with the following equation: (count × (12/scale))/sample weight(g).
16S rRNA gene sequencing
Genomic DNA was extracted from wood mouse fecal samples and gut tissue samples from the dissection study using Zymo Quick-DNA Fecal/Soil Microbe 96 kits according to the manufacturer’s instructions. Negative controls (water) were included in each extraction batch, with subsequent quality control tests showing no detectable DNA concentrations (using Qubit® high sensitivity assays) or amplification (visualized using gel electrophoresis). The V4 region of the bacterial 16S rRNA gene was amplified with the primers 515F/806R (Caporaso et al., 2011) as detailed previously (Marsh et al., 2022). Since small intestine samples were harder to amplify due to lower concentrations of bacterial DNA (Gu et al., 2013), PCRs for small intestine samples were optimized by eluting extractions in a smaller volume (40 μL rather than 80 μL), the DNA template in the first round PCR reaction was increased to 5.25 μL, primer concentration was increased to 20 μM, and the number of cycles was increased to 25. Library preparations followed a two-step (tailed-tag) approach with dual-indexing (D’Amore et al., 2016) and sequencing was performed on an Illumina® MiSeq with 250 bp paired-end reads. Four MiSeq runs were performed: runs 1 and 2 contained samples from the two 2 years of the non-invasive study, and runs 3 and 4 contained samples from the last year of the non-invasive study plus samples from the dissections. Full details of the microbiota amplification and sequencing methodology can be found in Marsh et al. (2022).
Bioinformatics
Raw sequence reads derived from samples in both the dissection and non-invasive studies were processed together as part of a larger set of samples (Marsh et al., 2022), using the DADA2 v1.6 pipeline in R to infer amplicon sequence variants (ASVs; Callahan et al., 2016). In brief, reads were trimmed and filtered for quality, and ASVs inferred and putative chimeras were removed before taxonomic assignment using the v128 SILVA reference database. A phyloseq object (McMurdie and Holmes, 2013) was created for further processing of the microbiota data. This included filtering taxa to remove ASVs assigned as mitochondria and chloroplast, and analysis of rarefaction and sample completeness curves in package ‘iNEXT’ to filter out samples with low total read counts (less than 8,000 reads). Putative contaminant ASVs (n = 159) belonging to 35 bacterial families (Supplementary Table S1) were identified by their prevalence in five negative extraction controls using the R package ‘decontam’ with default parameters (Davis et al., 2018). For alpha diversity analyses, ASV filtering was limited to the removal of those ASVs we could confidently assign as ‘non-gut’ taxa (mitochondria or chloroplast) or ASVs identified as contaminants by decontam, and abundance filtering was not performed as the removal of rare ASVs has a significant influence on richness estimation (Willis et al., 2017). Frequency ratio-based inference in the R package ‘breakaway’ (Willis and Bunge, 2015) was used to produce non-integer richness estimates per sample. Prior to beta diversity analyses, to guard against potential artefacts or contaminants influencing results, we performed abundance filtering in which very rare ASVs (those with less than 1 read found in less than 1% samples) were also removed from the dataset. ASV read counts were transformed to relative abundance prior to the calculation of Bray–Curtis dissimilarity (McKnight et al., 2019).
Statistical analyses
All statistical analyses were performed in the R software package (version 4.3.1; R Core Team, 2023). Bayesian linear regression models were performed with the R package ‘brms’ and model fit was assessed by leave-one-out cross-validation with the R package ‘loo’ (Bürkner, 2017; Vehtari et al., 2017). Terms were interpreted as ‘significant’ if the Bayesian 95% credible intervals did not overlap with 0. Estimated richness per sample was used to test for associations between parasite infection and microbiota alpha diversity. Bray–Curtis dissimilarity (variation in presence and abundance of taxa among samples) was used to test for associations between parasite infection and microbiota beta diversity. Correlations among covariates were examined in each dataset prior to modelling (Supplementary Figure S2).
Dissection study analyses
To explore the effects of parasite infection on microbiota alpha diversity, we used a series of Bayesian mixed models with estimated microbiota richness as the response variable fitted with Gaussian model family. First, to test whether overall infection status for each helminth (presence/absence across the entire GI tract) predicted gut microbiota richness, we modelled gut section-specific microbiota richness as a function of overall helminth infection status, and infection status*gut section interaction terms. Mouse ID was included as a random factor to control for multiple samples per individual, and both MiSeq sequencing run and read depth (z-transformed) were included as technical covariates. To more explicitly test whether helminths might affect the microbiota either up- or downstream of their infection site, we then tested whether microbiota richness was predicted by the presence of each helminth and if this depended on whether helminths were in the same gut section (local) to that from which the richness estimate was derived. To do this, we constructed a series of models, one per gut section, predicting gut-section-specific microbiota richness with the local (in the same gut section) and non-local (summed across all other sections) presence of each helminth species. Covariates included trapping location (/date), MiSeq run and (z-transformed) read depth. Most trapping locations only had data from a single trapping occasion, so trapping location was confounded with trapping date. We therefore included trapping location as a covariate in our models to control for both spatial and temporal variation. Some parasites were never found in a particular gut section (e.g., H. polygyrus was never found in the caecum or colon), and in these cases, a local infection term for that parasite was not included when modelling that section’s microbiota richness. For Hymenolepis sp., only presence/absence throughout the entire gut was used as a predictor as scolexes were not found and therefore we could not infer this helminth’s exact infection site or burden.
To assess whether helminth infection predicted microbiota composition, we used partial redundancy analysis (RDA, on Hellinger-transformed community data; Legendre and De Cáceres, 2013). To test for overall effects of gut parasite infection status (presence/absence) on microbiota composition, an RDA was performed including all samples with mouse ID as a condition term to control for repeated measures per mouse. The gut section was included as a factor together with interactions term per parasite species with gut section. Read depth and MiSeq run were again included as covariates. To test whether helminth infection status predicted microbiota composition in each gut section and how this depended on whether those helminths were in the same section (local) or a different one to the sampled microbial community, we ran separate RDAs predicting microbiota composition in each gut section, including local (in the same gut section) and non-local (summed across all other sections) presence of each helminth species as predictors. Covariates included trapping location (which is equivalent to trapping day as mice were only caught at each location once), MiSeq run and (z-transformed) read depth. We corrected for multiple testing across gut-section-specific models using false discovery rate (FDR). The significance of the overall model and marginal significance of each term was tested by permutations (n = 999).
To validate microbiota profiles obtained from fecal samples (as used in the non-invasive study), we used the dissection dataset to compare microbiota alpha and beta diversity estimates from feces with those obtained from different gut sections in the same mice. A Bayesian mixed model was used to assess the differences in microbial richness between fecal samples and gut section samples, with trap location, MiSeq run, and read depth as covariates, and animal ID as a random term (n = 6 samples per mouse). Compositional differences in microbial composition between fecal and gut samples of the same mice were modelled with permutational multivariate analyses of variance (PERMANOVA) with the gut section as the main predictor; MiSeq run, trap location, and read depth as covariates; and individual ID as the strata (n = 6 samples per mouse). To test whether fecal microbiota profiles were representative of the lower GI tract of an individual, we tested whether the compositional similarity between the colon and feces within an individual was greater than the similarity among the same sample type (colon–colon and feces–feces) across different individuals caught on the same day using permutational Wilcoxon tests with 1,000 permutations on pairwise Bray–Curtis dissimilarities. To directly test how similar beta diversity patterns among individuals were when using either fecal samples or colon samples, we performed a Procrustes analysis on Bray–Curtis dissimilarity.
Non-invasive study analyses
To test whether binary infection status for each parasite predicted fecal microbial alpha diversity (estimated richness), we used a Bayesian linear mixed model with Gaussian model family structure. We then constructed a model to test whether the overall richness of parasites detected in feces predicted fecal microbiota richness (multiple infections with 2 or more parasites were grouped). Both models included collection date, MiSeq run, and read depth as covariates and individual ID as a random factor to control for repeated measures for some mice. A redundancy analyses (RDA) on Hellinger-transformed data (similar to that used to analyze the dissection dataset) was used to test whether gut parasite burdens detected in feces predicted fecal microbiota composition. Parasite burdens were fitted as log-transformed variables (log(1 + egg/oocyst count)), mouse ID was included as a conditional term, and read depth and MiSeq run were included as covariates. Significance was tested by permutations (n = 999). A Mantel test was used to test whether fecal parasite community dissimilarity (variation in parasite presence and abundance among fecal samples as detected via flotation) correlated with fecal microbiota dissimilarity (variation in microbiota presence and abundance among fecal samples as detected via 16S sequencing), using Bray–Curtis dissimilarities for both parasites and microbes and including only samples in which at least one parasite was detected (n = 142 samples, 999 permutations).
To examine whether previously documented seasonal shifts in the gut microbiota (Marsh et al., 2022) might be influenced by gut parasitic infection status, we used a generalized additive mixed model (GAMM, R package ‘mgcv’) to model values from the first axis of a principle coordinate analysis (PCoA) performed on Bray–Curtis dissimilarity (PC1 values) as a function of both season (day of the year fitted as a cyclic cubic spline term) and parasite burden variables (z-scores of eggs/oocysts per gram). PC1 values were taken from an ordination performed on a larger longitudinal set of 448 fecal samples (of which the 223 samples analyzed here are a subset, Supplementary Figure S10), as we previously showed that values on this axis show a consistent pattern of seasonal variation (Marsh et al., 2022). Year and MiSeq run were included as fixed factors, read depth as a covariate, and individual ID as a random factor to control for repeated measures.
Results
Gut parasites and their infection location
Overall, four types of parasitic nematode were detected in this population: Heligmosomoides polygyrus, Syphacia spp., Aoncotheca sp., and Trichuris muris. Additionally, one cestode (Hymenolepis sp.), one trematode (Corrigia vitta), and Eimeria spp. protozoa were also detected (Figure 1A). Differences in our ability to detect these parasites by dissection and fecal flotation were apparent for several species (Figure 1A). For example, Syphacia pinworm were rarely detected in feces but were common in dissections, C. vitta was only detected in dissections (likely because their eggs were too heavy to float with our flotation solution), while Aoncotheca sp., T. muris and Eimera spp. were only detected in fecal flotation but not in dissections. These two diagnostics methods therefore had variable, and species-specific sensitivity. Aoncotheca sp. may have been present in the stomach lining, which was not inspected during dissections. Of the dissected mice, 90% harbored at least one parasite, with 36% co-infected by more than one parasite and most co-infected individuals (83%) showing co-infection within a gut section. Fecal flotations for the non-invasive study showed a lower overall infection rate, with 66% faecal samples containing at least one type of parasite and 24% containing two or more (mean = 1.02 ± 0.064 parasites per sample, range 0–4).
Figure 1. Summary of gut parasites found in the wood mice of Wytham woods. (A) Parasite prevalence and 95% confidence intervals for all six species of parasite detected, according to dataset/diagnostic method. The non-invasive dataset was collected through fecal flotations on 221 fecal samples collected from live-trapped mice over a 3-year period (2015–18), and the dissection dataset included 50 mice destructively sampled in 2017–2018 from a nearby area within the same woodland. (B) Detection of gut parasites in the dissection study across different sections of the GI tract (duodenum, jejunum, ileum, caecum, and colon). Worm burdens of Syphacia spp., H. polygyrus and C. vitta are shown in relation to the y-axis on the left-hand side, while the prevalence of Hymenolepis sp. proglottids per gut section is shown in relation to the y-axis on the right-hand side (as burdens could not be quantified for this species). Mean and standard error of worm burdens are shown, as well as prevalence (95% CI’s) of Hymenolepis sp.
Of the four helminths detected internally, H. polygyrus was the most localized and found in the duodenum and jejunum, with similar burdens in each (Figure 1B). Hymenolepis sp. burdens could not be measured as we did not find scolexes, though their proglottids were mostly found in the jejunum (Figure 1B). The trematode C. vitta was most abundant in the duodenum, but also occasionally found further along the GI tract (Figure 1B). Syphacia spp. were found all along the GI tract but burdens were higher in the small intestine than the caecum and colon (Figure 1B). We identified two species of Syphacia, with S. stroma mostly found in the small intestine and S. frederici largely confined to the caecum and colon, consistent with previous studies on these species (Stewart et al., 2018), though both species were detected to some extent in all gut sections (Supplementary Figure S3). As Syphacia were too abundant to identify every individual to species level, they were grouped as Syphacia spp. for subsequent statistical analyses. There were no strong correlations in the counts of each parasite across gut sections (Supplementary Figure S4) or among parasites within each gut section (Supplementary Figure S5).
Local and distant effects of helminth infection on the gut microbiota
Estimated microbiota richness varied across gut sections (Figure 2A and Supplementary Figure S6), with richness highest in the large intestine (especially the caecum), and lower in the small intestine (Supplementary Table S2). However, average microbiota richness was not predicted by infection status for Syphacia spp., Hymenolepis sp., H. polygyrus, or C. vitta within any gut section or across the whole gut (no gut section: parasite presence interaction terms were significant; Supplementary Table S2). Furthermore, helminth infection status did not predict duodenal or colonic microbial richness, irrespective of whether parasites were present in the same gut section or not (Supplementary Figure S7). However, estimated microbiota richness was higher in both the caecum [78.59 more ASVs, 95% CI 11.06 to 146.34] and ileum [102.52 more ASVs, 95% CI 47.47 to 158.12] when C. vitta was present in other, mostly upstream, gut sections (Supplementary Figure S7). Microbial richness in the jejunum was lower when Hymenolepis sp. was detected anywhere in the whole gut (61.04 fewer ASVs, 95% CI 10.60 to 112.47, Supplementary Figure S7). For those helminths whose variation in abundance allowed (Syphacia spp. and H. polygyrus), the inclusion of a logged helminth burden variable instead of presence/absence did not change the results (parameter estimate 95% CIs still included 0).
Figure 2. Richness and composition of the microbiota along the gastrointestinal tract. The richness and composition of the gut microbiota in different sections of the gastrointestinal (GI) tract were estimated from a dissection study of wild wood mice (n = 50). Samples were taken from the 5 gut sections (duodenum, jejunum, ileum, caecum, and colon) and fecal samples from the trap of the same mice for comparison. (A) Posterior means and 95% credible intervals from a Bayesian mixed model assessing variation in community richness along the GI tract and feces, with raw data points shown. (B) The family-level composition of the bacterial community along the GI tract and feces. Relative abundance of the top 5 most abundant families is shown, with less abundant families grouped as ‘other’.
Although microbiota composition varied among gut sections (Supplementary Table S3 and Figure 2B), it was not predicted by overall infection status with any of the four helminth parasites found in the dissection study (Supplementary Table S3). Similarly, neither local nor non-local helminth presence predicted microbiota composition within any one gut section (Supplementary Table S3), and this remained unchanged when including the logged burden of Syphacia spp. and H. polygyrus instead of presence/absence (p > 0.05).
Gut microbiota composition and comparison across sample types
The most abundant bacterial families in the dissection dataset were Lactobacillaceae (52.92%), Lachnospiraceae (17.72%), and Muribaculaceae (17.25%). The small intestine showed lower richness and was dominated by Lactobacillaceae compared to the large intestine, where Muribaculaceae and Lachnospiraceae were more abundant (Figure 2).
Before analyzing associations between parasites and the gut microbiota as assessed from faecal data in the non-invasive study, we first tested how well fecal microbiota profiles reflected internal microbiota profiles among dissected mice. Fecal microbiota richness was similar to that observed in the caecum and colon and higher than in the small intestine (Figure 2A and Supplementary Figure S6). Fecal microbiota composition was broadly similar to that of the colon at family level, while the caecum showed higher relative abundances of Lachnospiraceae (Figure 2B). At the ASV level, microbiota composition varied according to sample type within individuals [PERMANOVA; Gut section F(5,267) = 11.93, p = 0.001, marginal R2 = 0.158]. However, differences in mean dissimilarity, dispersion, or both could contribute to this result as sample types varied in compositional dispersion [F(5,277) = 19.15, p = 0.001].
As fecal samples appeared similar to the microbial composition of the colon (Figure 2B), we further assessed whether, despite being collected non-invasively, fecal samples can provide a reliable individual-specific gut microbiota profile. To do this, we tested whether within-individual microbiota similarity between the colon and feces outweighed microbiota similarity among samples of the same type (colon or feces) collected from different individuals. This was indeed the case with significantly lower Bray–Curtis dissimilarity between colon and fecal samples taken from the same individual than among samples of the same type (colon or feces) taken from different individuals (permutational Wilcoxon tests, colon-colon vs. colon–feces; U = 15,672, p < 0.001, feces–feces vs. colon–feces; U = 2,416, p < 0.001, Supplementary Figure S8). Individual ID also explained much more variance in microbiota composition than sample type in a PERMANOVA including paired colon and fecal samples for each individual [Gut section; F(1,88) = 2.73, p = 0.001, marginal R2 = 0.014, Animal ID; F(49,88) = 3.08, p = 0.001, marginal R2 = 0.766]. Differences in group dispersions were significant for both animal ID [F(49,39) = 4.69, p = 0.001] and sample type [F(1,87) = 6.16, p = 0.014], with colon samples showing higher dispersion than fecal samples. Procrustes analysis showed a high concordance in microbiota beta diversity patterns analyzed using fecal and internal colon samples (m122 = 0.839, p = 0.001; Supplementary Figure S8), indicating that non-invasive fecal samples provide a good representation of the beta diversity patterns that would be obtained using invasive colonic samples.
Parasite–microbiota associations in the non-invasive study
Of the longitudinal non-invasive samples, we analyzed 221 samples from 105 mice captured between 2015 and 2018 that had paired fecal microbiota profiles and parasite fecal flotation data. These mice were sampled between 1 and 8 times, with 54 mice sampled more than once and an average of 2.1 samples per mouse. Fecal samples had a mean of 44,828 reads after filtering (± 979 standard error), with an average of 207 (± 4.2 standard error) ASVs per sample. The dominant bacterial families were Muribaculaceae (37.83%), Lactobacillaceae (23.37%), and Lachnospiraceae (18.26%), with similar overall relative abundances to those found in fecal samples from the dissection study (Supplementary Table S4).
Microbiota richness was not predicted by infection status for any single parasite, though individuals infected with T. muris showed a tendency towards lower fecal microbiota diversity (posterior mean = −17.74, 95% CI -36.95 to 1.09; Supplementary Figure S9). While microbiota richness was not strongly associated with infection by any one parasite, it was negatively associated with the total number of parasite species detected in a fecal sample (posterior mean = −16.72, 95% CI –27.31 to −6.32, n = 221 mice; Figure 3). For comparison, fecal microbial richness in the dissection dataset was not found to be significantly predicted by the total number of parasite species detected, though the sample size was much smaller for this analysis (posterior mean = 0.85, 95% CI –18.42 to −20.33, n = 42 mice, Supplementary Table S5).
Figure 3. Fecal microbiota richness decreases with increased gut parasite richness. Fecal flotations and 16S sequencing were conducted to profile gut parasite burdens and the gut microbiota of wood mice from a longitudinal, non-invasive study (n = 221). A Bayesian mixed model was used to assess the relationship between the number of parasite species (0, 1, or > = 2) detected per fecal sample and the richness of the microbial community, after controlling for MiSeq run, read depth, and repeated measures per individual. Raw data points (jittered) as well as the conditional effect of parasite richness and its 95% credible intervals (shaded area) from the model are shown.
Microbiota composition was not associated with the burden of any one parasite detected in feces, though Eimeria infection showed a near significant association [F(1, 102) = 1.73, p = 0.065; Supplementary Table S6]. There was also no association between overall parasite community composition and microbiota composition (Mantel test r = 0.019, p = 0.126, n = 148).
Seasonal dynamics of parasites and the fecal microbiota
The burden of Eimeria spp. oocysts and T. muris eggs in feces both significantly predicted microbiota PC1 (values on the seasonally shifting primary axis of a Bray–Curtis PCoA, from Marsh et al., 2022; Supplementary Figure S10) in a GAMM modelling seasonality in this variable (Supplementary Table S7). Despite this, the explanatory power of the seasonal (day of year) term in this model did not change appreciably when parasite variables were included in the model (day of year adjusted R2 = 0.485 vs. R2 = 0.462 in models including and excluding parasite variables respectively), and neither T. muris nor Eimeria. spp. showed any significant pattern of seasonal variation in burdens [T. muris GAMM, s(day of year) F = 0.00, p = 0.458; Eimeria spp. GAMM, s(day of year) F = 0.00, p = 0.847; Supplementary Figure S11], suggesting changes in gut parasitic infections cannot explain previously documented seasonal variation in the gut microbiota in this population.
Discussion
Interactions between gut parasites and the microbiota are known to occur through a variety of mechanisms and can have important implications for host health and disease (Leung et al., 2018b). Much of this mechanistic understanding has been gained from laboratory-based studies often using mice and model helminths, with few observational studies on wild mammalian populations conducted so far. Here, we combined two approaches to examine correlational evidence for parasite-microbiota interactions in a natural population of wild wood mice. A dissection study examined local and distant parasite–microbiota associations along the GI tract, and a longitudinal, non-invasive study examined the relative importance of parasite infection as a potential driver of microbiota community dynamics. Overall, we found that natural parasite infections in our study population showed some association with gut microbiota alpha diversity, but we found very little evidence for parasite associations with composition (beta diversity).
No strong associations between parasite infection and microbiota composition
We found no associations between overall fecal microbiota composition and infection by any of the parasites detected, though Eimeria spp. infection and T. muris infection did predict the major compositional axis of variation (PC1), independent of previously documented seasonal variation in this variable (Supplementary Table S7). Our finding of a lack of association between parasitic infections and overall microbiota composition differs from previous studies on related host species. For instance, Kreisinger et al. (2015) detected a variety of local and distant associations between gut helminths and microbiota beta diversity along the GI tract in a closely related mouse species (Apodemus flavicollis), the strongest being between infection with the tapeworm Hymenolepis spp. and increased Bacteroidetes S24-7 (Muribaculaceae) in the stomach. A previous non-invasive study on wood mice also detected associations between nematode infection (largely H. polygyrus) increased relative abundance of a Lachnospiraceae genus and decreased relative abundance of Escherichia in feces (Maurice et al., 2015). Several differences exist between the studies which could explain these discrepancies. Notably, wood mice in the current study harbored comparatively few parasites overall (Behnke et al., 1999; Kreisinger et al., 2015), and the stomach region, where effects were detected previously in Apodemus (Kreisinger et al., 2015), was not included in this study.
In the longitudinal non-invasive study, infection with Eimeria spp. and to a lesser extent, T. muris was associated with variation in the first axis of a PCA (PC1), which explained 12.89% of the variation in fecal microbiota composition and has previously been shown to display reproducible seasonal dynamics (Marsh et al., 2022; Maurice et al., 2015). However, these parasite effects were independent of seasonal variation in PC1, and neither T. muris nor Eimeria spp. showed consistent seasonal patterns in our 3-year dataset. These two observations together that infections by these parasites are very unlikely to be key drivers of the previously documented repeatable seasonal shifts in gut microbiota composition among UK wood mice.
Microbiota richness is associated with specific parasite infections and coinfections
We found stronger evidence for associations between parasite infection and microbiota alpha diversity (richness) in both the dissection and non-invasive datasets. Microbial richness in the ileum and caecum was positively associated with the abundance of the trematode C. vitta elsewhere in the GI tract, suggesting an indirect mechanism of interaction. C. vitta was mostly found in the duodenum and more rarely further downstream. In fact, this is a parasite of the interlobary canals of the pancreas where in heavy infections it can disrupt the flow of pancreatic juices into the GI tract (Lewis et al., 2021; Magee et al., 1993). This may provide a potential mechanism underlying the association between C. vitta infection and increased bacterial diversity in the caecum, as pancreatic juices show antimicrobial activity to prevent overgrowth of pathogenic bacteria (Zhang et al., 2022) and bile has myriad impacts on the microbiota (Collins et al., 2023) that could influence diversity if bile flow into the gut was affected by C. vitta infection. We also detected a weaker negative association between infection by the tapeworm Hymenolepis sp. and microbial diversity in the jejunum. Previous studies have detected associations between Hymenolepis sp. infection and the abundance of specific bacterial taxa (Kreisinger et al., 2015; Aivelo and Norberg, 2018), yet it remains unclear what the mechanism underlying such associations might be. Infection with the nematode T. muris was associated with a (non-significant) reduction in fecal microbial richness in the non-invasive study. Although T. muris was not detected in the dissection study and therefore we cannot say if this association occurred locally or not, this finding is consistent with two previous T. muris experimental infection studies, which detected an infection-driven decline in microbiota diversity in lab mice (Holm et al., 2015; Houlden et al., 2015).
In addition to these specific associations with individual parasite species, an overall association between coinfection with multiple parasites (mostly helminths) and reduced microbial diversity was found in the non-invasive study (Figure 3). The relationship between co-infection with multiple species of parasites and the microbiota has rarely been considered, though Cooper et al. (2013) found reduced human fecal microbiota diversity in mixed infections of Trichuris trichiura and Ascaris lumbricoides compared to single species infections, and a recent meta-analysis also identified a trend towards reduced gut microbial diversity in infected vs. uninfected rodents (Scotti et al., 2020). The overall negative association of gut parasitism with microbiota diversity found here could arise through various mechanisms, including the secretion of antimicrobial compounds by gut parasites (Cotton et al., 2012), increased susceptibility to parasitic infection of mice with less diverse gut microbiota (Buffie and Pamer, 2013), or both. Since our data are purely correlational, further study would be needed to determine whether these associations reflect causal relationships, and if so what their direction and underlying mechanisms might be.
Combining methodological approaches can provide valuable insight
We provided validation of trap-collected feces as a representative, non-invasive sample of the microbiota in the large intestine which captures reliable individual-specific gut microbiota profiles. Colon samples were more variable among individuals than fecal samples, suggesting a slight loss of this individual specificity with non-invasive methods. This is consistent with previous studies which have highlighted that signatures of host evolution are better retained when studying intestinal mucosal samples vs. fecal samples (Ingala et al., 2018), and that the fecal microbiota represents a subset of the endogenous microbiota (Kohl et al., 2015). Despite this slight loss of individuality, fecal samples were still highly representative of individual differences in gut microbiota composition, with individual identity explaining the majority of variation among fecal microbiota samples.
Study to understand how common gut parasitic infectious might shape the gut microbiota and vice versa is still very much in its infancy, and synthesis and consensus across studies are yet to emerge (Cortés et al., 2019; Scotti et al., 2020). Some study designs are likely to yield more fruitful inference than others in the future study in this area. Longitudinal study designs where individuals are repeatedly sampled over time and key covariates are controlled for should offer more power than cross-sectional studies to detect important interactive effects between parasites and the gut microbiota. They can also enable a better understanding of specific ecological interactions among gut inhabitants, by investigating the temporal covariation in abundance among pairs of taxa to infer positive or negative associations (Roche et al., 2023). Additionally, experimental approaches (e.g., parasite treatment experiments) in tractable wild systems could increase our understanding of causal relationships further (Davidson et al., 2020; Fenton et al., 2014). Such an experimental approach was recently used effectively to test how helminthic infection altered the gut microbiota in African buffalo, revealing complex co-infection-dependent impacts (Sabey et al., 2021). Future studies of this nature in other tractable systems like rodents would be highly valuable to understand what underpins patterns of association between parasites and the microbiota such as those detected here, as well as what implications they may have for host physiology, immunity, and ultimately fitness.
Data availability statement
All raw sequence data associated with this study is deposited in the European Nucleotide Archive under the accession PRJEB49639. In addition, the processed ASV count and taxonomy tables along with the metadata for the longitudinal and dissection datasets are available as supplementary material. The R code used for the analyses presented here are also available as Supplementary material.
Ethics statement
The animal study was approved by the RVC Clinical Research Ethical Review Board (URN 2017 1721-2) and University of Oxford Department of Zoology Animal Welfare and Ethical Review Board (APA/1/5/ZOO/NASPA/Knowles/Wood mouse). The study was conducted in accordance with the local legislation and institutional requirements.
Author contributions
KM: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Validation, Visualization, Writing – original draft, Writing – review & editing. AR: Data curation, Investigation, Writing – review & editing. JPW: Supervision, Writing – review & editing. SCLK: Conceptualization, Funding acquisition, Resources, Supervision, Writing – review & editing.
Funding
The author(s) declare that financial support was received for the research, authorship, and/or publication of this article. This research was funded in whole, or in part, by the UKRI NE.L011867/1. For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. KM was supported by a Royal Veterinary College (RVC) studentship (supervised by SCLK and JPW). SCLK was also supported by funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement n° 851550).
Acknowledgments
We would like to sincerely thank Jerzy Behnke for training and help in helminth identification.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb.2024.1440427/full#supplementary-material
References
Abu-Madi, M. A., Behnke, J. M., Lewis, J. W., and Gilbert, F. S. (2000). Seasonal and site specific variation in the component community structure of intestinal helminths in Apodemus sylvaticus from three contrasting habitats in south-East England. J. Helminthol. 74, 7–15. doi: 10.1017/S0022149X00700022
Aivelo, T., and Norberg, A. (2018). Parasite–microbiota interactions potentially affect intestinal communities in wild mammals. J. Anim. Ecol. 87, 438–447. doi: 10.1111/1365-2656.12708
Al-Menayes, J. (2014). The relationship between mobile social media use and academic performance in university students. New Media and Mass Communication, 25, 23–29.
Behnke, J., Lewis, J., Zain, M., and Gilbert, F. (1999). Helminth infections in Apodemus sylvaticus in southern England: interactive effects of host age, sex and year on the prevalence and abundance of infections. J. Helminthol. 73, 31–44. doi: 10.1017/S0022149X00700162
Benson, A., Pifer, R., Behrendt, C. L., Hooper, L. V., and Yarovinsky, F. (2009). Gut commensal Bacteria direct a protective immune response against toxoplasma gondii. Cell Host and Microbe 6, 187–196. doi: 10.1016/j.chom.2009.06.005
Buffie, C. G., and Pamer, E. G. (2013). Microbiota-mediated colonization resistance against intestinal pathogens. Nat. Rev. Immunol. 13, 790–801. doi: 10.1038/nri3535
Bürkner, P. (2017). Brms: an R package for Bayesian multilevel models using Stan. J. Stat. Softw. 80, 1–28. doi: 10.18637/jss.v080.i01
Callahan, B. J., McMurdie, P. J., Rosen, M. J., Han, A. W., Johnson, A. J. A., and Holmes, S. P. (2016). DADA2: high-resolution sample inference from Illumina amplicon data. Nat. Methods 13, 581–583. doi: 10.1038/nmeth.3869
Caporaso, J. G., Lauber, C. L., Costello, E. K., Berg-Lyons, D., Gonzalez, A., Stombaugh, J., et al. (2011). Moving pictures of the human microbiome. Genome Biol. 12:R50. doi: 10.1186/gb-2011-12-5-r50
Collier, C. T., Hofacre, C. L., Payne, A. M., Anderson, D. B., and Kaiser, P. (2008). Coccidia-induced mucogenesis promotes the onset of necrotic enteritis by supporting Clostridium perfringens growth. Vet Immunol Immunopathol. 122, 104–115. doi: 10.1016/j.vetimm.2007.10.014
Collins, S. L., Stine, J. G., Bisanz, J. E., Okafor, C. D., and Patterson, A. D. (2023). Bile acids and the gut microbiota: metabolic interactions and impacts on disease. Nat. Rev. Microbiol. 21, 236–247. doi: 10.1038/s41579-022-00805-x
Cooper, P., Walker, A. W., Reyes, J., Chico, M., Salter, S. J., Vaca, M., et al. (2013). Patent human infections with the whipworm, Trichuris trichiura, are not associated with alterations in the Faecal microbiota. PLoS One 8:e76573. doi: 10.1371/journal.pone.0076573
Cortés, A., Peachey, L. E., Jenkins, T. P., Scotti, R., and Cantacessi, C. (2019). Helminths and microbes within the vertebrate gut – not all studies are created equal. Parasitology 146, 1371–1378. doi: 10.1017/S003118201900088X
Cotton, S., Donnelly, S., Robinson, M. W., John, P., and Thivierge, K. (2012). Defense peptides secreted by helminth pathogens: antimicrobial and/or immunomodulator molecules? Front. Immunol. 3:269. doi: 10.3389/fimmu.2012.00269
D’Amore, R., Ijaz, U. Z., Schirmer, M., Kenny, J. G., Gregory, R., Darby, A. C., et al. (2016). A comprehensive benchmarking study of protocols and sequencing platforms for 16S rRNA community profiling. BMC Genomics 17, 20–55. doi: 10.1186/s12864-015-2194-9
Davidson, G. L., Raulo, A., and Knowles, S. C. L. (2020). Identifying microbiome-mediated behaviour in wild vertebrates. Trends Ecol. Evol. 35, 972–980. doi: 10.1016/j.tree.2020.06.014
Davis, N. M., Proctor, D., Holmes, S. P., Relman, D. A., and Callahan, B. J. (2018). Simple statistical identification and removal of contaminant sequences in marker-gene and metagenomics data. Microbiome 6, 226–214. doi: 10.1186/s40168-018-0605-2
Dea-ayuela, M. A., Rama-iñiguez, S., and Bolás-fernandez, F. (2008). Enhanced susceptibility to Trichuris muris infection of B10Br mice treated with the probiotic Lactobacillus casei. Int. Immunopharmacol. 8, 28–35. doi: 10.1016/j.intimp.2007.10.003
Easton, A. V., Quiñones, M., Vujkovic-cvijin, I., Oliveira, R. G., Kepha, S., Odiere, M. R., et al. (2019). The impact of anthelmintic treatment on human gut microbiota based on cross-sectional and pre- and Postdeworming comparisons in Western Kenya. MBio 10:e00519. doi: 10.1128/mBio.00519-19
Egan, C. E., Cohen, S. B., and Denkers, E. Y. (2012). Insights into inflammatory bowel disease using toxoplasma gondii as an infectious trigger. Immunol. Cell Biol. 90, 668–675. doi: 10.1038/icb.2011.93
Ezenwa, V. O., and Jolles, A. E. (2015). Opposite effects of anthelmintic treatment on microbial infection at individual versus population scales. Science 347, 175–177. doi: 10.1126/science.1261714
Fenton, A., Knowles, S. C. L., Petchey, O. L., and Pedersen, A. B. (2014). The reliability of observational approaches for detecting interspecific parasite interactions: comparison with experimental results. Int. J. Parasitol. 44, 437–445. doi: 10.1016/j.ijpara.2014.03.001
Gause, W. C., and Maizels, R. M. (2016). Macrobiota — helminths as active participants and partners of the microbiota in host intestinal homeostasis. Curr. Opin. Microbiol. 32, 14–18. doi: 10.1016/j.mib.2016.04.004
Gu, S., Chen, D., Zhang, J., Lv, X., Wang, K., Duan, L., et al. (2013). Bacterial community mapping of the mouse gastrointestinal tract. PLoS One 8:e74957. doi: 10.1371/journal.pone.0074957
Hayes, K. S., Bancroft, A. J., Goldrick, M., Portsmouth, C., Roberts, I. S., and Grencis, R. K. (2010). Exploitation of the intestinal microflora by the parasitic nematode Trichuris muris. Science 328, 1391–1394. doi: 10.1126/science.1187703
Higgs, S., and Nowell, F. (2000). Population biology of Eimeria (Protozoa: Apicomplexa) in Apodemus sylvaticus: a capture/recapture study. Parasitology 120, 355–363. doi: 10.1017/S0031182099005545
Holm, J. B., Sorobetea, D., Kiilerich, P., and Ramayo-caldas, Y. (2015). Chronic Trichuris muris infection decreases diversity of the intestinal microbiota and concomitantly increases the abundance of lactobacilli. PLoS One 10:e0125495. doi: 10.1371/journal.pone.0125495
Honda, K., and Littman, D. R. (2016). The microbiota in adaptive immune homeostasis and disease. Nature 535, 75–84. doi: 10.1038/nature18848
Houlden, A., Hayes, K. S., Bancroft, A. J., Worthington, J. J., Wang, P., Grencis, R. K., et al. (2015). Chronic Trichuris muris infection in C57BL / 6 mice causes significant changes in host microbiota and metabolome: effects reversed by pathogen clearance. PLoS One 10:e0125945. doi: 10.1371/journal.pone.0125945
Ingala, M. R., Simmons, N. B., Wultsch, C., Krampis, K., Speer, K. A., and Perkins, S. L. (2018). Comparing microbiome sampling methods in a wild mammal: fecal and intestinal samples record different signals of host ecology, evolution. Front. Microbiol. 9, 1–13. doi: 10.3389/fmicb.2018.00803
Johnson, P. T. J., De Roode, J. C., and Fenton, A. (2015). Why infectious disease research needs community ecology. Science, 349:1259504. doi: 10.1126/science.1259504
Knowles, S. L., Fenton, A., Petchey, O. L., Jones, T. R., Barber, R., and Pedersen, A. B. (2013). Stability of within-host-parasite communities in a wild mammal system. Proc. R. Soc. B 280:20130598. doi: 10.1098/rspb.2013.0598
Kohl, K. D., Luong, K., and Dearing, M. D. (2015). Validating the use of trap-collected feces for studying the gut microbiota of a small mammal (Neotoma lepida). J. Mammal. 96, 90–93. doi: 10.1093/jmammal/gyu008
Kreisinger, J., Bastien, G., Hauffe, H. C., Marchesi, J., and Perkins, S. E. (2015). Interactions between multiple helminths and the gut microbiota in wild rodents. Philos. Transac. R. Soc. B Biol. Sci. 370:20140295. doi: 10.1098/rstb.2014.0295
Legendre, P., and De Cáceres, M. (2013). Beta diversity as the variance of community data: dissimilarity coefficients and partitioning. Ecol. Lett. 16, 951–963. doi: 10.1111/ele.12141
Leung, J. M., Budischak, S. A., The, H. C., Hansen, C., Bowcutt, R., Neill, R., et al. (2018a). Rapid environmental effects on gut nematode susceptibility in rewilded mice. PLoS Biol. 16:e2004108. doi: 10.1371/journal.pbio.2004108
Leung, J. M., Graham, A. L., and Knowles, S. C. L. (2018b). Parasite-microbiota interactions with the vertebrate gut: synthesis through an ecological Lens. Front. Microbiol. 9:843. doi: 10.3389/fmicb.2018.00843
Lewis, J. W., Morley, N. J., and Behnke, J. M. (2021). Spatial interactions between two nematode species along the intestine of the wood mouse Apodemus sylvaticus from woodland and grassland sites in southern England. J. Helminthol. 95:e57. doi: 10.1017/S0022149X21000468
Loke, P., and Harris, N. L. (2023). Networking between helminths, microbes, and mammals. Cell Host Microbe 31, 464–471. doi: 10.1016/j.chom.2023.02.008
Magee, C. A., Cahir, M., Halton, D. W., Johnston, C. F., and Shaw, C. (1993). Cytochemical observations on the nervous system of adult Corrigia vitta. J. Helminthol. 67, 189–199. doi: 10.1017/S0022149X00013122
Marsh, K. J., Raulo, A. M., Brouard, M., Troitsky, T., English, H. M., Allen, B., et al. (2022). Synchronous seasonality in the gut microbiota of wild mouse populations. Front. Microbiol. 13:809735. doi: 10.3389/fmicb.2022.809735
Maurice, C. F., Cl Knowles, S., Ladau, J., Pollard, K. S., Fenton, A., Pedersen, A. B., et al. (2015). Marked seasonal variation in the wild mouse gut microbiota. ISME J. 9, 2423–2434. doi: 10.1038/ismej.2015.53
McKnight, D. T., Schwarzkopf, L., Zenger, K. R., Alford, R. A., Huerlimann, R., and Bower, D. S. (2019). Methods for normalizing microbiome data: an ecological perspective. Methods Ecol. Evol. 10, 389–400. doi: 10.1111/2041-210x.13115
McMurdie, P. J., and Holmes, S. (2013). Phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS One 8:e61217. doi: 10.1371/journal.pone.0061217
Mcsorley, H. J., Hewitson, J. P., and Maizels, R. M. (2013). Immunomodulation by helminth parasites: defining mechanisms and mediators. Int. J. Parasitol. 43, 301–310. doi: 10.1016/j.ijpara.2012.11.011
Million, M., Diallo, A., and Raoult, D. (2017). Gut microbiota and malnutrition. Microb. Pathog. 106, 127–138. doi: 10.1016/j.micpath.2016.02.003
Montgomery, S. S. J., and Montgomery, W. I. (1988). Cyclic and non-cyclic dynamics in populations of the helminth parasites of wood mice, Apodemus sylvaticus. J. Helminthol. 62, 78–90. doi: 10.1017/S0022149X00011263
Ohnmacht, C., Park, J., Cording, S., Wing, J. B., Atarashi, K., Obata, Y., et al. (2015). The microbiota regulates type 2 immunity through ROR γ t T cells. Science 349, 989–993. doi: 10.1126/science.aac4263
Pedersen, A. B., and Fenton, A. (2019). “Wild rodents as a natural model to study within-host parasite interactions” in Wildlife disease ecology; linking theory to data and application. eds. K. Wilson, A. Fenton and D. Tompkins (Cambridge: Cambridge University Press), 58–90.
Ramanan, D., Bowcutt, R., Lee, S. C., Tang, M. S., Kurtz, Z. D., Ding, Y., et al. (2016). Helminth infection promotes colonization resistance via type 2 immunity. Science 352, 608–612. doi: 10.1126/science.aaf3229
R Core Team (2023). _R: A Language and Environment for Statistical Computing_. R Foundation for Statistical Computing, Vienna, Austria. Available at: https://www.R-project.org/. R version 4.3.1
Reynolds, L. A., Finlay, B. B., and Maizels, R. M. (2015). Cohabitation in the intestine: interactions among helminth parasites, bacterial microbiota, and host immunity. J. Immunol. 195, 4059–4066. doi: 10.4049/jimmunol.1501432
Reynolds, L. A., Smith, K. A., Filbey, K. J., Harcus, Y., Hewitson, J. P., Redpath, S. A., et al. (2014). Commensal-pathogen interactions in the intestinal tract. Gut Microbes 5, 522–532. doi: 10.4161/gmic.32155
Roche, K. E., Bjork, J. R., Dasari, M. R., Grieneisen, L., Jansen, D., Gould, T. J., et al. (2023). Universal gut microbial relationships in the gut microbiome of wild baboons. eLife 12:e83152. doi: 10.7554/eLife.83152
Sabey, K. A., Song, S. J., Jolles, A., Knight, R., and Ezenwa, V. O. (2021). Coinfection and infection duration shape how pathogens affect the African buffalo gut microbiota. ISME J. 15, 1359–1371. doi: 10.1038/s41396-020-00855-0
Scotti, R., Southern, S., Boinett, C., Jenkins, T. P., Cortés, A., and Cantacessi, C. (2020). MICHELINdb: a web-based tool for mining of helminth-microbiota interaction datasets, and a meta-analysis of current research. Microbiome 8:10. doi: 10.1186/s40168-019-0782-7
Sousa, J. E. N., Carvalho, E. F. G., Levenhagen, M. A., and Chaves, L. A. (2016). Diagnosis of the pinworm Syphacia muris in the Wistar rat Rattus norvegicus. Journal of Helminthology, 90, 117–120. doi: 10.1017/S0022149X14000753
Stewart, A., Lowe, A. N. N., Smales, L., Bajer, A., Bradley, J. A. N., Nik, D. D. W. U. Ż., et al. (2018). Parasitic nematodes of the genus Syphacia Seurat, 1916 infecting Muridae in the British Isles, and the peculiar case of Syphacia frederici. Parasitology 145, 269–280. doi: 10.1017/S0031182017001470
Subramanian, S., Huq, S., Yatsunenko, T., Haque, R., Mahfuz, M., Alam, M. A., et al. (2014). Persistent gut microbiota immaturity in malnourished Bangladeshi children. Nature 510, 417–421. doi: 10.1038/nature13421
Sweeny, AR, Lemon, H, Ibrahim, A, Watt, KA, Wilson, K, Childs, DZ, et al. (2023). A mixed-model approach for estimating drivers of microbiota community composition and differential taxonomic abundance. Msystems, 8:e00040–23.
Vehtari, A., Gelman, A., and Gabry, J. (2017). Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Stat. Comput. 27, 1413–1432. doi: 10.1007/s11222-016-9696-4
White, E. C., Houlden, A., Bancroft, A. J., Hayes, K. S., Goldrick, M., Grencis, R. K., et al. (2018). Manipulation of host and parasite microbiotas: survival strategies during chronic nematode infection. Sci. Adv. 4:eaap7399. doi: 10.1126/sciadv.aap7399
Willis, A., and Bunge, J. (2015). Estimating diversity via frequency ratios. Biometrics 71, 1042–1049. doi: 10.1111/biom.12332
Willis, A., Bunge, J., and Whitman, T. (2017). Improved detection of changes in species richness in high diversity microbial communities. J. R. Statistical Soc. 66, 963–977. doi: 10.1111/rssc.12206
Keywords: microbiota, parasite, wildlife, observational, rodent
Citation: Marsh KJ, Raulo AR, Webster JP and Knowles SCL (2024) Parasite–gut microbiota associations in wild wood mice (Apodemus sylvaticus). Front. Microbiol. 15:1440427. doi: 10.3389/fmicb.2024.1440427
Edited by:
Raquel Xavier, Centro de Investigacao em Biodiversidade e Recursos Geneticos (CIBIO-InBIO), PortugalReviewed by:
Hadas Hawlena, Ben-Gurion University of the Negev, IsraelSarah E. Perkins, Cardiff University, United Kingdom
Copyright © 2024 Marsh, Raulo, Webster and Knowles. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Sarah C. L. Knowles, sarah.knowles@biology.ox.ac.uk