Fig 1.
Systematically sampled canine fecal sites in Flagstaff, AZ.
A) The blue dot on the inset map shows the location of Flagstaff, USA. B) Clostridium difficile positive fecal samples occurred throughout the sampling area (red dots represent toxigenic isolates and yellow dots represent non-toxigenic isolates), as did C. difficile negative samples (black dots). The area shaded in light grey indicates Flagstaff city limits. This map was created using ArcGIS software by Esri.
Fig 2.
Clostridium difficile detection and sequence typing workflow.
The prevalence and overall diversity of C. difficile in Flagstaff canines was determined by TaqMan PCR and multilocus sequence typing (MLST) [27]. A) The outcome of TaqMan PCR, culturing, and MLST from two canine sampling sources. MLST was performed on extractions from pure culture or directly from the fecal enrichment extraction (FEE), when culturing was unsuccessful (see text). B) Total outcomes from both sampling sources.
Fig 3.
Whole-genome sequencing (WGS) maximum-likelihood phylogenetic tree containing 54 Clostridium difficile canine isolates from Flagstaff, USA.
The total number of single nucleotide polymorphisms (SNPs) within the four most prominent sequence type (ST) groups from our study are reported. Additionally, we provided an average number of SNPs within and between canine hosts for each of these four ST groups. WGS phylogeny was rooted with reference strain CD630 [41]. The scale bar represents SNPs and the consistency index is reported.
Fig 4.
Sequence type 3 (ST3) whole-genome sequencing sub-clade tree including all ST3 isolates from this study (highlighted in red) plus publically available non-toxigenic ST3 isolates.
By including global representatives of the non-toxigenic ST3 clade we see that there is geographically distinct fine-scale diversity within identical STs isolated from canines in Flagstaff, USA. Identical STs from within a single host were highly similar (0–2 SNPs), consistent with a single source of infection. The only exception to this was sample DGF_0205, which revealed 12 SNPs between isolates. This result is indicative of multiple colonizing strains that happened to be of the same ST (see text). The maximum-likelihood phylogeny was rooted with reference strain CD630 [41]. The scale bar represents SNPs and the consistency index is reported.
Fig 5.
Sequence type 15 (ST15) whole-genome sequencing sub-clade tree including all ST15 isolates from this study (highlighted in red) plus publically available ST15 isolates.
By including global representatives of ST15 we see that there is geographically distinct fine-scale diversity within identical STs isolated from canines in Flagstaff, USA. Identical STs from within a single host were highly similar (0–2 SNPs), consistent with a single source of infection. The maximum-likelihood phylogeny was rooted with reference strain CD630 [41]. The scale bar represents SNPs and the consistency index is reported.
Fig 6.
Sequence type 2 (ST2) and 110 whole-genome sequencing sub-clade tree including all ST2 and 110 isolates from this study (highlighted in red) plus publically available ST2 and 110 isolates.
By including global representatives of this ST group we see that much of the global fine-scale diversity within these STs (1580/4495 SNPs or 35%) was represented in isolates from canines in Flagstaff, USA. The maximum-likelihood phylogeny was rooted with reference strain CD630 [41]. The scale bar represents SNPs and the consistency index is reported.
Fig 7.
Sequence type 42 (ST42) whole-genome sequencing sub-clade tree including all ST42 isolates from this study (highlighted in red) plus publically available ST42 isolates.
By including global representatives of ST42 we see that there is geographically distinct fine-scale diversity within identical STs isolated from canines in Flagstaff, USA. Identical STs from within a single host were highly similar (0–2 SNPs), consistent with a single source of infection. The maximum-likelihood phylogeny was rooted with reference strain CD630 [41]. The scale bar represents SNPs and the consistency index is reported.
Table 1.
All sequence types (STs) identified in this study and their association with human disease.
Fig 8.
Multilocus sequence typing (MLST) neighbor-joining tree.
This tree was constructed using the concatenated sequences of all available Clostridium difficile sequence types (STs) from www.pubmlst.org/cdifficile with the 12 STs from clade 1 found in Flagstaff canines highlighted in red font [17]. No other clade was represented by the STs identified in this study. The tree was rooted using ST204 from clade C-I. Clade descriptions are the same as previously described [26].
Table 2.
Clostridium difficile co-colonizations observed in canines.