- Open Access
Humanized microbiota mice as a model of recurrent Clostridium difficile disease
© Collins et al. 2015
Received: 12 May 2015
Accepted: 21 July 2015
Published: 20 August 2015
Clostridium difficile disease is the leading antibiotic-associated cause of diarrhea and nosocomial acquired infection in the western world. The per annum burden in the USA alone amounts to 250,000 cases with 14,000 ascribed deaths and medical costs in excess of a billion dollars. Novel models for the study of C. difficile infection are therefore pertinent.
Germ free C57BL/6 mice gavaged with a healthy human fecal microbiota maintained a stable “humanized” microbiota over multiple generations when housed under specific pathogen-free (SPF) conditions. As with mice containing a conventional microbiota, treatment with a five-antibiotic cocktail followed by a single dose of clindamycin renders the animals susceptible to C. difficile infection (CDI). Interestingly, after recovery from the initial CDI infection, a single intraperitoneal injection of clindamycin is sufficient to induce CDI relapse. Relapse of CDI can be induced up to 35 days postinfection after recovery from the initial infection, and multiple episodes of relapse can be induced.
This model enables the study of recurrent C. difficile disease in a host containing a human-derived microbiota. Probiotic treatments using human-derived microbes, either prophylactic or curative, can be tested within the model. The identification and testing of human-derived microbial communities within a humanized microbiota mouse model may enable a higher rate of successful transfer of bacteria-based treatments from the lab to human patients due to the microbes involved initiating from, and being adapted to, the human GI tract.
Clostridium difficile infection (CDI) is the principle cause of antibiotic-associated diarrhea (AAD) and nosocomial infection in the western world. Severe cases lead to pseudomembranous colitis and toxic megacolon which can require drastic surgical intervention. Recurrence occurs in approximately 25 % of cases treated with metronidazole or vancomycin with 12 % percent of patients experiencing at least two recurrences and 6 % an excess of two . The per annum burden in the USA amounts to 250,000 cases with 14,000 ascribed deaths and medical costs in excess of a billion dollars .
Disruption of the host microbiota, whose functions include the prevention and limitation of pathogen colonization and growth, is the most prominent risk factor in CDI . Recently, the recapitulation of this barrier, via fecal microbial transplantation (FMT), has shown great promise as a treatment for recurrent CDI cases, recalcitrant to other treatments . FMT, however, has not yet been approved by the FDA and there are safety concerns over the use of a complex mix of bacterial, viral, and fungal components in addition to prions and potentially unknown biologically active substances .
Numerous animal  and in vitro [7, 8] models have been used to study C. difficile biology, virulence factors, and associated disease. Until recently, the golden Syrian hamster was the most widely used model of CDI. However, hamsters rapidly succumb to disease, often becoming moribund within 48 h, and are not suitable for studies of recurrent disease. Therefore, researchers have shifted towards the use of mouse models of CDI, and a few different models for the study of C. difficile relapse/recurrence have been developed. These models rely upon repeat antibiotic treatment with or without re-infection [9, 10] or suppression of initial infection with antibiotics  to achieve a relapse-like state. To our knowledge, animal models to study CDI relapse/recurrence in the presence of a human-derived microbiota have not been described. Here, we characterize a humanized microbiota mouse (HMbmouse) model of CDI with the ability to induce relapse/recurrence of disease in otherwise healthy mice by the administration of a single dose of clindamycin. Investigation of the changes in microbiota that lead to susceptibility to CDI may facilitate a targeted approach to identifying therapeutic microbes. The benefit of this recurrence model is the potential to identify human-derived microbes or microbial communities with prophylactic or curative properties. Bacterial communities that have coevolved within their host are likely to out-compete those from an extraneous source within that niche . The use of an HMbmouse model colonized with human-derived microorganisms, un-adapted to the murine environment, overcomes this inherent problem when adding back potential therapeutic human microbes .
Results and discussion
Humanized microbiota mice
Twenty-two OTUs were present in at least four of six founder mice that were undetected in progeny, potentially indicating the inability of certain species to be transferred via the parent-offspring route
Lachnospiracea incertae sedis
Erysipelotrichace incertae sedis
To test for invasion of the HMbmice communities following multiple generations under specific pathogen-free (SPF) conditions, 16S rDNA in fecal samples from HMbmice (2.5 years following initial colonization of founder mice), were sequenced on the Illumina MiSeq platform and compared to the original pooled human fecal sample via reciprocal BLAST search . HMbMouse communities remained remarkably immutable with ~95 % of reads from the deep sequenced HMbmice matching sequences identified in the original fecal sample. Reads that could not be matched clustered into OTUs predominantly identified as Lachnospiraceae and may represent invasive species or those found below the level of detection in the original fecal material.
In summary, HMbmice are distinguishable from CMbmice by sequencing, are resistant to widespread invasion by murine species, and remain stable over multiple generations when housed under SPF conditions.
HMbMouse C. difficile infection model
A Kruskal-Wallis test on spore count data identified a significant difference in VPI10463 spore numbers 24 h post challenge between antibiotic pretreatment groups (χ 2 = 12:91, p = 0.002). Post hoc Mann-Whitney tests with Holm correction established significant differences between groups 3 day (below level of detection) and 5 day (mean = 8.00 × 104 spores g−1 feces), p = 0.008, and groups 4 day (mean = 2.78 × 103 spores g−1 feces) and 5 day, p = 0.023. After 48 h, only mice in the 3- and 4-day antibiotic pretreatment groups survived with no significant difference in spore counts (Fig. 6). Mice challenged with CD3017 did not have a significant difference in spore production between antibiotic pretreatment groups (Fig. 7). However, spore levels were higher in mice that became moribund compared to those that recovered (mean = 4.61 × 106 vs mean = 1.41 × 106 spores g−1 feces), though this difference was not statistically significant due to a large standard deviation from the mean. Across all time points sampled, CD3017 consistently produced more spores than VPI10463 and continued producing detectable spore levels for a longer period of time. CD3017 spores were observable 14 days post challenge (mean across all CD3017 groups day 14 = 4.42 × 103 spores g−1 feces), whereas VPI10463 spores were undetectable by day 11. The VPI10463 strain is known to be a poor spore former both in vitro  and in vivo . Conversely, increased spore production has been proposed as a mechanism for the spread of NAP1/B1/027 strains , though this is controversial with a wide range of sporulation efficiencies seen in vitro .
Recurrent model of CDI
C. difficile spores in feces were below the level of detection at the time of relapse induction (as determined by plating). One possibility is that spores adhered to the coat of HMbmice (cages were changed with sterile replacements weekly), triggering relapse upon ingestion in post clindamycin-susceptible animals. Another hypothesis is that vegetative cells or spores remained associated, at low levels, within the murine gastrointestinal (GI) tract. Within 24 h of triggering CDI relapse, spores were detected at levels ≥107 g−1 feces, where they remained for ~5 days before waning to undetectable levels (Fig. 9). This rapid increase suggests that vegetative cells may already be present within the intestine, allowing for rapid growth and spore production when conditions permit. An association of C. difficile and the gut mucosa/epithelium has been observed previously in both animal  and in vitro experiments .
Effect of antibiotics and C. difficile infection on the HMbmouse microbiota
The use of antibiotics can significantly reduce or ablate sensitive species while simultaneously allowing resistant opportunists to bloom . To examine this effect in HMbmice, fecal samples were sequenced by Illumina MiSeq (average reads per sample 35,505, sub-sampled to an even depth of 23,057) from three time points: base composition, prior to antibiotic insult; post-antibiotics, immediately following the five-antibiotic cocktail and clindamycin IP injection; recovered, following challenge with C. difficile and subsequent recovery from CDI, 14–17 days post cessation of antibiotics.
Pre- vs post-antibiotic treated microbiota
To examine the effect of antibiotics on the HMbmouse microbiome in more detail, data were subset into pre-antibiotic and immediately post-antibiotic cessation samples. Differentially abundant OTUs between the two groups were detected using negative binomial generalized linear models with DESeq2 in R. Originally designed to detect differentially abundant gene expression from RNA-Seq data, this package has been shown to work well with 16S sequencing count data . p values were corrected for multiple inference using the Benjamini-Hochberg false discovery rate (FDR) procedure and a strict adjusted alpha cutoff value of 0.01 set as criteria for inclusion. Two hundred and eighteen OTUs were found to have significantly different abundances. One hundred and ninety-eight OTUs decreased while 20 OTUs increased post-antibiotic use (see Additional file 2: Table S1). Of those OTUs that decreased, 44.4 % belonged to the Lachnospiraceae, 16.7 % to the Ruminococcaceae, 9.1 % Bacteroidaceae, and the remaining 29.8 % were split across 12 family groups or unclassified. This wide-ranging reduction of taxa can be explained by the fact that the five-antibiotic cocktail possesses a broad spectrum of antimicrobial activity, targeting both Gram positive and negative species via multiple modes of action.
Proposed mechanisms of colonization resistance in respect to C. difficile include competition for sparse resources within a niche; microbially produced short-chain fatty acids, specifically butyrate; and the conversion of primary bile acids into secondary acids via 7α-dehydroxylating bacteria. Overall, 69.7 % of the OTUs significantly decreased by antibiotic treatment belonged to the Clostridia, potentially freeing a niche for C. difficile to exploit. Butyrate possesses anti-inflammatory properties, provides the preferred energy source in colonocytes, and can directly inhibit C. difficile growth in vitro . Members of the Lachnospiraceae and Ruminococcaceae are known to be primary butyrate producers , so their significant reduction likely leads to reduced butyrate production, increasing susceptibility to CDI. Primary bile acids, such as taurocholic acid, are known to induce C. difficile germination while secondary bile acids can exert inhibitory effects on germination or kill vegetative cells . The majority of known bile acid 7α-dehydroxylating bacteria belong to Clostridium cluster XIVa , of which six OTUs were significantly reduced following antibiotics. Removal of 7α-dehydroxylating bacteria can alter the proportions of primary and secondary bile acids, switching the environment from inhibitory to stimulatory in respect to C. difficile spore germination [29, 30].
Of the OTUs that increased, the largest increase (base mean = 159, log2 fold change = 8.48) belonged to an Enterococcaceae which are known to bloom in humans undergoing certain antibiotic treatments [31, 32] and have been shown to correlate positively with CDI . These blooms may be incidental or may have a synergistic effect on disease severity.
Microbiota in moribund vs recovered mice
Differentially abundant OTUs in moribund vs recovered HMbmice communities
Clostridium sensu stricto
Recovery of HMbmouse microbiota following antibiotic treatment and C. difficile infection
Comparison of alpha measurements between pre-antibiotic and recovered HMbmouse communities
Fecal microbial transplantation (FMT) as a treatment for C. difficile disease is not a contemporary idea, having been reported in scientific literature for over 50 years . Recently, the use of FMT for the treatment of recurrent CDI has burgeoned with a reported efficacy of >90 % . Despite this, CDI remains the most prevalent nosocomal disease in the developed world. There are tangible risks associated with administering an unknown microbiota to a patient. Increased use of FMT will bring with it a heightened likelihood of seeing complications such as infection, allergic response, or long-term unsuspected sequelae such as weight gain . An alternative to using the mélange of bacteria, fungi, and virons present in FMT samples is to create a defined cocktail of well characterized, sequenced, bacteria that is shown to be efficacious in its treatment or prevention of CDI. This approach has been successfully applied in conventional mouse models using indigenous murine microbes  and to ameliorate disease using specific human derived microbiota .
The use of HMbmice as a model of both CDI and recurrent CDI in combination with sequencing technology and advanced bacterial isolation techniques provides a resource for the identification of beneficial microbes and the means to test them among other human-derived microbes.
Humanized microbiota mice
Germ-free C57BL/6 mice were administered, via oral gavage, a human-derived fecal microbiota from a pooled fecal slurry of 12 healthy adults as described in Robinson et al.  and classified as HMbmice. HMbmice derived by gavage were housed and bred in gnotobiotic isolators. Progeny and descendants of the founder HMbmice acquired their microbiota via maternal transfer and were maintained under SPF conditions. To limit the introduction of invasive species, gloves were changed between handling conventional microbiota mice (CMbmice) and all chow, water, and bedding autoclaved prior to use. All descendant HMbmouse samples were taken from mice aged 6–10 weeks from offspring of founder mice at time points 4, 9, and 11 months post initial gavage of fecal material. A separate sample was taken from mice 2.5 years, and multiple generations, out for comparison to the original starting fecal material. The animal use protocol was approved by the Animal Ethics Committee of Baylor College of Medicine (Protocol no. AN-6675).
Mouse fecal samples were collected in sterile microfuge tubes and stored at −80 °C until needed. DNA was extracted by bead beating and modified extraction and cleanup with the Qiagen DNEasy Tissue Kit as described previously . For temporal analysis of the HMbmice, the V3–V5 region of the 16S rRNA gene was amplified by PCR using the 357F/962R primers with unique barcodes designed by the Human Microbiome Project  and sequenced using a 454 GS Junior (Roche Diagnostics). Analysis of microbial communities following antibiotic administration and C. difficile challenge experiments was carried out by amplifying the V4 region of the 16S rRNA gene with primers F515/R806, using a dual indexing approach (4 forward primer; 96 reverse primer) and subsequent sequencing by Illumina MiSeq. The 96-indexed R806 primers used were previously described in Caparaso et al.  (806rbc0–806rbc96). The indexed F515 primers were essentially as described in Kozich et al. , except that we generated four barcodes that balanced the nucleotide composition at each position (atcgatgg, tcacgaca, ggtatctc, and cagtcgat) in place of those described in Kozich et al.
PCR reactions were carried out in triplicate. Each reaction contained 4 μl of diluted template, 1X Phusion High-Fidelity Buffer (New England Biolabs), 200 μM dNTPs (Promega or Invitrogen), 10 nM primers, 0.2 units of Phusion DNA Polymerase (New England Biolabs), and PCR grade water to a final volume of 20 μl. The amplification cycle consisted of an initial denaturation at 98 °C for 30 s, followed by 30 cycles of 10 s at 98 °C, 20 s at 51 °C, and 1 min at 72 °C. Replicates were pooled and cleaned using AMPure beads as previously described . DNA sample concentrations were determined with Quant-iT (Life Technologies) and pooled at equimolar ratios. Prior to sequencing, the quality of pooled DNA was assessed by analysis on a Bioanalyzer High Sensitivity DNA Kit (Agilent).
Sequence data analysis
Sequence data were processed using platform-specific (454/MiSeq) pipelines in Mothur v.1.34.3  based upon published SOPs [44, 45]. Alignment was achieved using the Silva 16S rRNA gene reference database. Chimeric sequences and any classifying as chloroplast, mitochondria, Archaea, or Eukaryota were removed after identification by the Mothur implementation of uchime or mothur-formatted ribosomal database project (version 9) classifier, respectively. Sequences were clustered by the average-neighbor method into operational taxonomic units (OTUs) of ≥97 % sequence identity. Analysis of OTU data were performed in R  using the phyloseq , DESeq2 , and indicspecies  packages. Sequence data are available via the Sequence Read Archive (SRA) accession numbers SRP061089 and SRP061088.
S obs = the number of species in the sample, F 1 = the number of singletons (single occurrence within a sample) and F 2 = the number of doubletons.
pi = proportion of Sobs made up of the ith species.
ni = the number of individuals in the ith OTU and N = the total number of individuals in the community.
The Bray-Curtis dissimilarity matrix was used to describe differences in microbial community structure. Patterns of similarity among the different HMb/CMbmouse fecal communities were examined using principal coordinates analysis (PCoA). Robustness of results was tested by recomputing output using alternative dissimilarity measures (Jaccard, thetayc) and ordinations (non-metric multi-dimensional scaling (NMDS)). No significant changes were observed (data not shown).
Spore preparation and enumeration
Individual C. difficile colonies were picked into pre-reduced BHIS media and propagated anaerobically at 37 °C overnight. Spores were cultivated by spread plating overnight BHIS cultures onto BHIS medium and incubating anaerobically at 37 °C for 5 days. Spores were scraped from plates, resuspended in sterile water, and heated to 65 °C for 30 min to kill vegetative cells. Viable spores were enumerated by plating serial dilutions on BHIS plates supplemented with 0.1 % taurocholic acid. Spore preparations were diluted in sterile water to a final concentration of 5 × 106 spores ml−1 and maintained at 4 °C until used. To enumerate C. difficile spores from HMbmouse feces, fecal pellets were weighed, homogenized in 500 μl sterile water, and incubated at 65 °C for 30 min to kill vegetative cells. Following heat killing, spores were serially diluted and spot plated in duplicate onto selective TCCFA plates containing 0.1 % taurocholic acid.
Establishment of robust C. difficile infection model in HMbmice
To produce a robust infection model in HMbmice, a number of antibiotic pretreatment regimes, known to work well in CMbmice, were tested including the following: cefoperazone (0.5 mg ml−1) in sterile drinking water for 10 days  and clindamycin (0.25 mg ml−1) in drinking water for 7 days . However, only the use of a five-antibiotic cocktail developed by Chen et al.  produced reproducible results in HMbmice. HMbMice were administered kanamycin (0.4 mg ml−1), gentamicin (0.035 mg ml−1), colistin (850 U ml−1), metronidazole (0.215 mg ml−1), and vancomycin (0.045 mg ml−1) ad libitum in drinking water for 3–5 days followed 24 h later by IP injection of clindamycin (10 mg kg−1). Twenty-four hours post IP injection, mice were challenged, via oral gavage, with 5 × 105 spores of either the high toxin producing strain VPI10463 or the clinically relevant NAP1/B1/027 ribotype strain CD3017. Symptoms were similar to those observed in humans: diarrhea, weight loss, and histologic damage.
HMbMice were administered the five-antibiotic cocktail for 4 days ad libitum in drinking water followed by IP clindamycin injection. Twenty-four hours later, mice were challenged with 5 × 104 CD3017 spores (tenfold lower compared to prior experiments was used to minimize mouse mortality while maintaining CDI symptoms). Following recovery from initial CDI, relapse was triggered by administering a further single dose of clindamycin (10 mg kg−1) up to 4 weeks post recovery from initial infection.
Histological sections of distal colon were scored blindly by an independent researcher utilizing the scoring system outlined in Theriot et al. . Briefly, three criteria: edema, cellular infiltration, and epithelial damage were individually scored from 0–4 (where a score of 0 indicates no pathology) and the values summed to give an overall disease severity score.
- Kwon JH, Olsen MA, Dubberke ER. The morbidity, mortality, and costs associated with clostridium difficile infection. Infect Dis Clin North Am. 2015;29(1):123–34.PubMedView ArticleGoogle Scholar
- Antibiotic Resistance Threats in the United States, 2013. http://www.cdc.gov/drugresistance/threat-report-2013/.
- Britton RA, Young VB. Interaction between the intestinal microbiota and host in clostridium difficile colonization resistance. Trends Microbiol. 2012;20(7):313–9.PubMed CentralPubMedView ArticleGoogle Scholar
- Gough E, Shaikh H, Manges AR. Systematic review of intestinal microbiota transplantation (fecal bacteriotherapy) for recurrent clostridium difficile infection. Clin Infect Dis. 2011;53(10):994–1002.PubMedView ArticleGoogle Scholar
- Moayyedi P, Marshall JK, Yuan Y, Hunt R. Canadian association of gastroenterology position statement: fecal microbiota transplant therapy. Can J Gastroenterol Hepatol. 2014;28(2):66.PubMed CentralPubMedGoogle Scholar
- Hutton ML, Mackin KE, Chakravorty A, Lyras D. Small animal models for the study of clostridium difficile disease pathogenesis. FEMS Microbiol Lett. 2014;352(2):140–9.PubMedView ArticleGoogle Scholar
- Baines SD, Freeman J, Wilcox MH. Effects of piperacillin/tazobactam on clostridium difficile growth and toxin production in a human gut model. J Antimicrob Chemother. 2005;55(6):974–82.PubMedView ArticleGoogle Scholar
- Robinson CD, Auchtung JM, Collins J, Britton R. Epidemic clostridium difficile strains demonstrate increased competitive fitness over non-epidemic isolates. Infect Immun. 2014;82(7):2815–2825.Google Scholar
- Chen X, Katchar K, Goldsmith JD, Nanthakumar N, Cheknis A, Gerding DN, et al. A mouse model of clostridium difficile-associated disease. Gastroenterology. 2008;135(6):1984–92.PubMedView ArticleGoogle Scholar
- Sun X, Wang H, Zhang Y, Chen K, Davis B, Feng H. Mouse relapse model of clostridium difficile infection. Infect Immun. 2011;79(7):2856–64.PubMed CentralPubMedView ArticleGoogle Scholar
- Lawley TD, Clare S, Walker AW, Stares MD, Connor TR, Raisen C, et al. Targeted restoration of the intestinal microbiota with a simple, defined bacteriotherapy resolves relapsing clostridium difficile disease in mice. PLoS Pathog. 2012;8(10):1002995.View ArticleGoogle Scholar
- Oh PL, Benson AK, Peterson DA, Patil PB, Moriyama EN, Roos S, et al. Diversification of the gut symbiont lactobacillus reuteri as a result of host-driven evolution. ISME J. 2010;4(3):377–87.PubMedView ArticleGoogle Scholar
- Seedorf H, Gri_n NW, Ridaura VK, Reyes A, Cheng J, Rey FE, et al. Bacteria from diverse habitats colonize and compete in the mouse gut. Cell. 2014;159(2):253–66.PubMedView ArticleGoogle Scholar
- Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215(3):403–10.PubMedView ArticleGoogle Scholar
- Small J. Fatal enterocolitis in hamsters given lincomycin hydrochloride. Lab Anim Care. 1968;18(4):411–20.PubMedGoogle Scholar
- Best EL, Freeman J, Wilcox MH. Models for the study of clostridium difficile infection. Gut Microbes. 2012;3(2):145–67.PubMed CentralPubMedView ArticleGoogle Scholar
- Mukherjee K, Karlsson S, Burman LG, Åkerlund T. Proteins released during high toxin production in clostridium difficile. Microbiology. 2002;148(7):2245–53.PubMedGoogle Scholar
- Åkerlund T, Svenungsson B, Lagergren Å, Burman LG. Correlation of disease severity with fecal toxin levels in patients with clostridium difficile-associated diarrhea and distribution of pcr ribotypes and toxin yields in vitro of corresponding isolates. J Clin Microbiol. 2006;44(2):353–8.PubMed CentralPubMedView ArticleGoogle Scholar
- Merrigan M, Venugopal A, Mallozzi M, Roxas B, Viswanathan V, Johnson S, et al. Human hypervirulent clostridium difficile strains exhibit increased sporulation as well as robust toxin production. J Bacteriol. 2010;192(19):4904–11.PubMed CentralPubMedView ArticleGoogle Scholar
- Burns DA, Heeg D, Cartman ST, Minton NP. Reconsidering the sporulation characteristics of hypervirulent clostridium difficile bi/nap1/027. PLoS One. 2011;6(9):24894.View ArticleGoogle Scholar
- Borriello S, Welch A, Barclay FE, Davies HA. Mucosal association by clostridium difficile in the hamster gastrointestinal tract. J Med Microbiol. 1988;25(3):191–6.PubMedView ArticleGoogle Scholar
- Calabi E, Calabi F, Phillips AD, Fairweather NF. Binding of clostridium difficile surface layer proteins to gastrointestinal tissues. Infect Immun. 2002;70(10):5770–8.PubMed CentralPubMedView ArticleGoogle Scholar
- Lozupone CA, Stombaugh JI, Gordon JI, Jansson JK, Knight R. Diversity, stability and resilience of the human gut microbiota. Nature. 2012;489(7415):220–30.PubMed CentralPubMedView ArticleGoogle Scholar
- McMurdie PJ, Holmes S. Waste not, want not: why rarefying microbiome data is inadmissible. PLoS Comput Biol. 2014;10(4):1003531.View ArticleGoogle Scholar
- Rolfe RD. Role of volatile fatty acids in colonization resistance to clostridium difficile. Infect Immun. 1984;45(1):185–91.PubMed CentralPubMedGoogle Scholar
- Antharam VC, Li EC, Ishmael A, Sharma A, Mai V, Rand KH, et al. Intestinal dysbiosis and depletion of butyrogenic bacteria in clostridium difficile infection and nosocomial diarrhea. J Clin Microbiol. 2013;51(9):2884–92.PubMed CentralPubMedView ArticleGoogle Scholar
- Sorg JA, Sonenshein AL. Inhibiting the initiation of clostridium difficile spore germination using analogs of chenodeoxycholic acid, a bile acid. J Bacteriol. 2010;192(19):4983–90.PubMed CentralPubMedView ArticleGoogle Scholar
- Ridlon JM, Kang D-J, Hylemon PB. Bile salt biotransformations by human intestinal bacteria. J Lipid Res. 2006;47(2):241–59.PubMedView ArticleGoogle Scholar
- Giel JL, Sorg JA, Sonenshein AL, Zhu J. Metabolism of bile salts in mice influences spore germination in clostridium difficile. PLoS One. 2010;5(1):8740.View ArticleGoogle Scholar
- Buffe CG, Jarchum I, Equinda M, Lipuma L, Gobourne A, Viale A, et al. Profound alterations of intestinal microbiota following a single dose of clindamycin results in sustained susceptibility to clostridium difficile-induced colitis. Infect Immun. 2012;80(1):62–73.View ArticleGoogle Scholar
- Pérez-Cobas AE, Gosalbes MJ, Friedrichs A, Knecht H, Artacho A, Eismann K, et al. Gut microbiota disturbance during antibiotic therapy: a multi-omic approach. Gut. 2013;62(11):1591–601.PubMed CentralPubMedView ArticleGoogle Scholar
- Stecher B, Maier L, Hardt W-D. ‘Blooming’ in the gut: how dysbiosis might contribute to pathogen evolution. Nat Rev Microbiol. 2013;11(4):277–84.PubMedView ArticleGoogle Scholar
- Schubert AM, Rogers MA, Ring C, Mogle J, Petrosino JP, Young VB, et al. Microbiome data distinguish patients with clostridium difficile infection and non-C. difficile-associated diarrhea from healthy controls. MBio. 2014;5(3):01021–14.View ArticleGoogle Scholar
- De Cáceres M, Legendre P, Moretti M. Improving indicator species analysis by combining groups of sites. Oikos. 2010;119(10):1674–84.View ArticleGoogle Scholar
- Van den Abbeele P, Belzer C, Goossens M, Kleerebezem M, De Vos WM, Thas O, et al. Butyrate-producing clostridium cluster xiva species specifically colonize mucins in an in vitro gut model. ISME J. 2013;7(5):949–61.PubMed CentralPubMedView ArticleGoogle Scholar
- Buffe CG, Bucci V, Stein RR, McKenney PT, Ling L, Gobourne A, et al. Precision microbiome reconstitution restores bile acid mediated resistance to clostridium difficile. Nature. 2015;517(7533):205–8.View ArticleGoogle Scholar
- Jernberg C, Löfmark S, Edlund C, Jansson JK. Long-term ecological impacts of antibiotic administration on the human intestinal microbiota. ISME J. 2007;1(1):56–66.PubMedView ArticleGoogle Scholar
- Eiseman B, Silen W, Bascom G, Kauvar A. Fecal enema as an adjunct in the treatment of pseudomembranous enterocolitis. Surgery. 1958;44(5):854–9.PubMedGoogle Scholar
- Alang N, Kelly CR. Weight gain after fecal microbiota transplantation. In: Open Forum Infectious Diseases, vol. 2, p. 004. Oxford: Oxford University Press; 2015.Google Scholar
- Consortium HMP et al. Structure, function and diversity of the healthy human microbiome. Nature. 2012;486(7402):207–14.View ArticleGoogle Scholar
- Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Huntley J, Fierer N, et al. Ultra-high-throughput microbial community analysis on the illumina hiseq and miseq platforms. ISME J. 2012;6(8):1621–4.PubMed CentralPubMedView ArticleGoogle Scholar
- Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the miseq illumine sequencing platform. Appl Environ Microbiol. 2013;79(17):5112–20.PubMed CentralPubMedView ArticleGoogle Scholar
- Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, et al. Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol. 2009;75(23):7537–41.PubMed CentralPubMedView ArticleGoogle Scholar
- Mothur: 454 SOP. Accessed Dec 2014. http://www.mothur.org/wiki/454_SOP.
- Mothur: MiSeq SOP. Accessed Dec 2014. http://www.mothur.org/wiki/MiSeq_SOP.
- R.C. Team. R: a language and environment for statistical computing. 2014.Google Scholar
- McMurdie PJ, Holmes S. Phyloseq: an r package for reproducible interactive analysis and graphics of microbiome census data. PloS One. 2013;8(4):61217.View ArticleGoogle Scholar
- Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for rna-seq data with deseq2. Genome Biol. 2014;15(12):550.PubMed CentralPubMedView ArticleGoogle Scholar
- Cáceres MD, Legendre P. Associations between species and groups of sites: indices and statistical inference. Ecology. 2009;90(12):3566–74.PubMedView ArticleGoogle Scholar
- Theriot CM, Koumpouras CC, Carlson PE, Bergin II, Aronoff DM, Young VB. Cefoperazone-treated mice as an experimental platform to assess differential virulence of clostridium difficile strains. Gut Microbes. 2011;2(6):326–34.PubMed CentralPubMedView ArticleGoogle Scholar
- Deakin LJ, Clare S, Fagan RP, Dawson LF, Pickard DJ, West MR, et al. The clostridium difficile spo0a gene is a persistence and transmission factor. Infect Immun. 2012;80(8):2704–11.PubMed CentralPubMedView ArticleGoogle Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.