SEARCH

SEARCH BY CITATION

Keywords:

  • pouches;
  • pouchitis;
  • microbiota;
  • high throughput sequencing

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  7. Supporting Information

Background:

Chronic pouchitis is an important long-term complication following ileal pouch–anal anastomosis for ulcerative colitis. Antibiotic administration reduces symptoms of pouchitis, indicating that bacteria have a role in pathogenesis. The aim of the research was to investigate the bacterial content of pouches using nucleic acid-based methods.

Methods:

Stool microbiota of 17 patients with normal pouches (NP), 17 patients with pouchitis (CP) utilizing samples collected from each patient when antibiotic-treated (CP-on, asymptomatic) and when untreated (CP-off, symptomatic), and 14 familial adenomatous polyposis (FAP) patients were analyzed by high-throughput sequencing, fluorescence in situ hybridization technologies, and quantitative polymerase chain reaction (qPCR).

Results:

Fluorescence in situ hybridization analysis revealed an expanded phylogenetic gap in NP and CP-off patients relative to FAP. Antibiotic treatment reduced the gap in CP stool. The phylogenetic gap of CP-off patients was due to members of the bacterial families Caulobacteriaceae, Sphingomonadaceae, Comamonadaceae, Peptostreptococcaceae, and Clostridiaceae. There was a greater diversity of phylotypes of Clostridiaceae in CP-off subjects. The phylogenetic gap of NP stool was enriched by Ruminococcaceae and Bifidobacteriaceae. CP stool microbiota had reduced diversity relative to NP and FAP stool due largely to a reduction in Lachnospiraceae/Insertae Sedis XIV/clostridial cluster IV groups.

Conclusions:

Bacterial groups within the expanded phylogenetic gap of pouch patients may have roles in the pathogenesis of pouchitis. Further research concerning the physiology of cultured members of these groups will be necessary to explain their specific roles. Members of the Lachnospiraceae, Incertae Sedis XIV, and clostridial cluster IV could be useful biomarkers of pouch health. (Inflamm Bowel Dis 2011;)

Chronic pouchitis is the most common cause of troublesome, long-term functional disturbance for patients with ileo–anal pouches.1 Pelvic pouches show varying degrees of chronic mucosal inflammation and associated changes, but usually function well.2–4 In some patients, however, acute inflammation and clinical pouchitis develop by a process that possibly parallels ulcerative colitis (UC).5 Pouchitis is unusual in patients with pouches formed because of familial adenomatous polyposis (FAP), but occurs in 30%–50% of patients who have UC.6, 7 In pouchitis it is likely that there is an abnormal interaction between the immune system of UC patients and as yet unidentified factors associated with the pouch contents. Empirical success in the treatment of pouchitis with antibiotics points to bacteria as the likely trigger that activates the mucosal immune system.8 We used a variety of DNA-based analytical methods to determine the bacterial composition of pouch stool because some of the bacteria inhabiting the human bowel have not yet been cultivated under laboratory conditions.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  7. Supporting Information

Patients

Normal pouch (NP) patients who had UC were recruited in Otago, New Zealand, using the Otago Audit database at Dunedin Hospital. Twenty-two patients were recruited. Inclusion criteria for patients with NP created for UC were the absence of documented episodes of pouchitis and absence of symptoms of troublesome pouch dysfunction. Patients with pouchitis (CP) or FAP were identified using information contained in the Mount Sinai Hospital Inflammatory Bowel Diseases database, Toronto, Canada. Patients were first contacted by mail and then by telephone for recruitment. Twenty patients with a pelvic pouch formed for FAP and 20 patients with a pouch formed for UC but who were experiencing chronic (18 subjects) or relapsing (two subjects) pouchitis were recruited. Chronic pouchitis was defined as symptoms of pouchitis requiring long-term medication for symptom control. Pouchitis had to be proven by endoscopy and histology when examined at least once while symptomatic, and the Pouchitis Disease Activity Index (PDAI) had to be ≥7.0 when symptomatic.9 Relapsing pouchitis was defined as three or more documented episodes of pouchitis with symptoms and endoscopic evidence on three occasions and a PDAI of ≥7.0. In addition, to add rigor beyond the PDAI of significant pouch inflammation, all patients in the CP group had to have had previous biopsies showing an acute histology score of ≥4.0 using the scoring system of Moskowitz et al.10 All CP patients had been scoped within the last 5 years to exclude Crohn's disease, cuffitis, and other possible causes of pouch dysfunction as part of a follow-up study of these patients. Patients were not re-endoscoped specifically for this study because the diagnosis of pouchitis was considered to be robust. Each subject was asked to collect three stool samples with each sample collected 4 weeks apart. The patients also completed a questionnaire to obtain information about current pouch function. For patients on long-term antibiotics (ciprofloxacin, ceftin, cefuroxime, or metronidazole) pouch function was recorded while on their usual medication. A summary of these details is given in Table 1. Patients on continuous antibiotic treatment for chronic pouchitis were asked to stop the antibiotics for 10 days prior to collection of two of the three stool samples. The 10-day period was chosen to provide a minimal period without treatment and consequent symptoms, and because regeneration of microbiotas usually occurs relatively rapidly once antibiotic treatment has ceased. Patients were asked not to provide samples within 10 days of being on antibiotics for any other reason. Approval for the study was obtained from the Mount Sinai Hospital Research Ethics Board and from the Otago Regional Ethics Committee.

Table 1. Summary of Patient Details
GroupPouch Design (n)Mean Age of Pouch (Range)Clinical Component of Pouchitis Disease Activity Index (Maximum = 6.0) at Time of Sample CollectionMean Pouch Frequency per 24 HoursUrgency (Defer Passing Stool < 5 Min) (n)Long-term Antibiotics (n)
Familial adenomatous polyposis (n = 14)11 J6.7 years (2–17 years)1.48.011
3 S
Pouchitis(n = 17)14 J12.1 years (3–20 years)1.6 (on treatment)8.7 (on treatment)215
3 S
Normal pouches (n = 17)15 J5.4 years (1–13 years)1.47.010
2 W

The stool samples were frozen soon after collection and stored at −80°C prior to bacteriological analysis. Several subjects did not deliver all three samples and some samples were not suitable for analysis. Samples suitable for analysis were obtained from 17 NP, 14 FAP, and 17 CP patients.

Analysis of the Stool Microbiota

Oligonucleotide Probes and Fluorescence In Situ Hybridization (FISH)

Details of the oligonucleotide probes (Eub338, Erec482, Clep866/cp, Bac303, Bif164, Ato291, Lab158, Enter1432) that were used can be found in the publications of Wallner et al,11 Amann et al,12 Seksik et al,13 Franks et al,14 Lay et al,15 Manz et al,16 Langendijk et al,17 Harmsen et al,18, 19 and Sghir et al.20 Control probes were covalently linked at the 5′ end either to fluorescein isothiocyanate (FITC) or to the sulfoindocyanine dye indodicarbocyanine (Cy5). The probes specific for particular bacterial groups were linked at the 5′ end to Cy5. Competitor oligonucleotides were unlabeled. All of the probes were purchased from MWG Biotech (Bangalore, India). FISH/flow cytometry was carried out as described previously.21 The proportions (%) of the seven bacterial groups detected by FISH probes were summed and the sums were subtracted from 100%. The value obtained gave the proportion of the microbiota not detected by the seven probes (referred to as the “phylogenetic gap”).

Total Bacterial Counts

The EUB 338 probe was used to obtain total counts of bacteria. Unlabeled beads with a diameter of 6 μm (Flow Check Undyed Particle, Polysciences, Warrington, PA) were added to each sample to give a final concentration of 9.0 × 104 beads per mL and were used as an internal standard to calibrate the sample volume. The ratio of EUB 338-labeled cells to the number of beads was used to determine the bacterial cell count per gram of sample.

Fluorescence Activated Cell Sorting (FACS)

The aim of the procedure was to separate bacterial cells that did not hybridize any of the set of seven group-specific phylogenetic probes (members of the phylogenetic gap). The sorted bacteria could then be identified by sequencing their 16S rRNA genes. FISH was carried out as described above except that, in the same microfuge tube, 50 μL of bacterial cells suspended in hybridization buffer was hybridized with the set of phylogenetic probes including EUB338 (4 ng/μL final concentration). Ten tubes per specimen were used and the cells were pooled after the final washing step. FACS was carried out using a FACSAria cell sorter (Becton Dickinson, San Jose, CA) equipped with 488 nm (blue) and 633 nm (red) lasers. The nozzle size was 70 μm and “high-pressure sort” was selected (high-throughput sorting). The “sort” parameters were applied according to the manufacturer's instructions (Becton Dickinson). Targeted events (EUB338-FITC mono-labeled bacterial cells) were identified by drawing gates around the population of interest in the density plot representing red fluorescence (CY5) versus green fluorescence (FITC). The sorting procedure involved two steps: a first sort on yield following by a second sort on purity. About 106 cells were recovered with a purity higher than 90%. Analysis was carried out using FACSDiva software (Becton Dickinson). DNA was extracted from EUB338-FITC mono-labeled cells by a previously described method22 and was used as template in polymerase chain reaction (PCR) to amplify ≈500 bp of the 16S rRNA gene (nucleotides 11–539 in the Escherichia coli gene). Amplicons were used to prepare clone libraries in E. coli DH5α using PCR cloning vector pGEM-T Easy (Promega, Madison, WI) as previously reported.23 Fifty clones from libraries prepared from the microbiota of four CP patients (CP8, CP9, CP13, CP16) in the presence or absence of antibiotic treatment were sequenced (Macrogen, Korea). Sequences were compared to those in the NCBI database using the BLASTn algorithm.24

High-throughput Sequencing (HTS) of 16S rRNA Genes

DNA was extracted from stool using a previously described bead beating/phenol extraction method.22 DNA from CP patients (three timepoints), FAP patients (one timepoint), and NP patients (one timepoint) were individually barcoded for sequencing. A region comprising the V1–V3 regions of the bacterial 16S rRNA gene was amplified using a two-step protocol similar to that described by Dowd et al.25 First-round PCR was carried out for 15 cycles using the 8fAll (GRGTTYGATYMTGGCTCAG)/HDA2 (GTATTACCGCGGCTGCTGGCAC) primer set under the following conditions: 94°C for 1 min, 57°C for 1 min, 72°C for 1 min, with a final extension step of 72°C for 5 min. This product was diluted 1/5 with PCR grade water and 1 μL was used as template in a 20 μL secondary PCR. The secondary PCR was carried out for 30 cycles using the 8fAll primer with the 454 sequencing Lib-A adapter sequence A (CGTATCGCCTCCCTCGCGCCATCAGGRGTTYGATYMTGGCTCAG) and the HDA2 primer with the 454 sequencing Lib-A adapter sequence B plus a 10 base barcode, shown as N's, (CTATGCGCCTTGCCAGCCCGCTCAGNNNNNNNNNNGTATTACCGCGGCTGCTGGCAC) using conditions identical to the primary PCR. Products were cleaned using Qiagen PCR clean up columns (Qiagen, Hilden, Germany) and quantified using a Nanodrop 1000 spectrometer. Equivalent quantities of PCR product from each sample were pooled, and the pooled DNA was recleaned through a Qiagen PCR clean up column, quantified and sent to Macrogen (Korea) for unidirectional sequencing from the reverse primer on the Roche-454 Genome Sequencer using Titanium chemistry.

Sequences were processed using a combination of methods from both the QIIME v. 1.2.126 and RDP Pyrosequencing pipeline27 packages. Sequences were excluded from analysis if they were <250 or >550 bases in length, had an average quality score <25, contained one or more ambiguous bases, had >1 mismatch with the sequencing primer, or had a homopolymer run >6. Following splitting into barcoded samples and initial quality filtering, the sequences were passed through the QIIME pipeline using default parameters, including chimera checking. After quality screening, an average of 6094 ± 669 (range 1128–22,235) sequences per barcoded sample were recovered for downstream analysis. Thus, a total of ≈412,423 sequences were obtained from CP patients for phylogenetic analysis, 33,860 sequences from NP patients, and 29,079 sequences from FAP subjects. Species level taxonomy was obtained by filtering Operational Taxonomic Unit (OTU) tables, containing taxonomic data generated using the RDP classifier, at a genus level, extracting representative sequences and using BLAST to identify species level matches within the NCBI database. Bootstrapped neighbor-joining trees were generated using the RDP Tree builder software, incorporating reference strain sequences from the RDP database.

Real-time Quantitative PCR to Quantify C. perfringens and Clostridium difficile in Stool

Real-time quantitative PCR was carried out using an ABI 7500 Fast system in MicroAmp Fast Optical 96-Well plates with Optical adhesive film (Applied Biosystems, Foster City, CA). Primers and 5′ nuclease probe sequences targeting the C. perfringens alpha toxin (phospholipase C) gene were those previously described by Amar et al.28 and were purchased from Invitrogen (La Jolla, CA) and Applied Biosystems, respectively. Primers targeting the C. difficile 16S rRNA gene were those previously described by Rintillä et al29 and were purchased from Invitrogen. All reactions were carried out in a final volume of 20 μL containing 1× TaqMan Fast Universal PCR Mastermix (C. perfringens) or 1× Fast SYBR Green PCR Mastermix (C. difficile) (Applied Biosystems), 300 nM of each primer, and 100 nM fluorescent probe. Template DNA was extracted from stool using a previously described method22 and 5 μL was added to the mastermix. The thermocycling profile consisted of an initial activation of the polymerase at 95°C for 30 seconds, followed by 40 cycles of 95°C for 5 seconds, and 60°C for 30 seconds. Fluorescence levels were measured after a 60°C annealing/extension step. Standard curves were generated using genomic DNA extracted from C. perfringens ATCC 13124T and C. difficile ATCC 9689T using the Qiagen DNeasy blood and tissue kit and following the Gram Positive Bacteria protocol. A melt curve was generated to analyze product specificity. The standard DNA was quantified spectrophotometrically using a Nanodrop 1000 (Thermo Scientific, Pittsburgh, PA) and diluted in 10-fold steps from 5 × 106 to 5 × 101 genomes/reaction calculated on the basis of 10 copies (C. difficile 16S rRNA gene) or one copy (C. perfringens phospholipase C gene) of the target gene per genome (NCBI). All reactions were carried out in duplicate and were run twice on separate plates.

Statistical Analysis

The Mann–Whitney test, Fisher's exact test, and statistical methods associated with 16S rRNA analysis software packages were used.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  7. Supporting Information

Evidence of Marked Differences Between the Composition of the Stool Microbiota of CP, FAP, and NP Patients

Beta diversity analysis of 16S rRNA gene sequences obtained by HTS, in which groups of samples were compared using the unifrac metric, showed that CP microbiota compared with themselves were significantly more similar than CP microbiota compared with either FAP or NP stool. Community differences between CP and NP groups were greatest, while FAP patients were no more similar to each other than they were to other groups. A comparison of CP-on and CP-off antibiotic groups showed that there was more similarity within the groups than between the groups. Patients were more similar to themselves across sampling times than they were to other patients (Fig. 1a–c). Therefore, the CP stool microbiota was somewhat individualistic, dissimilar from that of FAP and NP patients, and was altered by antibiotic treatment.

thumbnail image

Figure 1. Comparison of bacterial community structure (average unweighted Unifrac distance) between: (a) CP patients on antibiotics, off antibiotics, and between the two groups; (b) CP patients compared with themselves, FP, and NP patients (c) between individuals over time (within) or between patients (values are mean ± SEM).

Download figure to PowerPoint

Alpha diversity analysis was carried out on rarefied OTU tables (Fig. 2). The FAP and NP microbiota were more biodiverse (Chao1 predictions of 637 and 735 OTU) than the CP group (<200 OTU; Shannon index of alpha diversity CP-off 3.63, CP-on 2.94, FAP 5.58, NP 5.65; P < 0.0001). The CP-on microbiota was less biodiverse than the CP-off antibiotic group (P < 0.0001). The FAP and NP groups did not differ in diversity. Thus, CP was characterized by a simplified microbiota composition. Lachnospiraceae formed much greater proportions, and were more biodiverse (Shannon index of alpha diversity CP-off 2.06, CP-on 2.34, FAP 4.31, NP 4.96; P < 0.003) in the stool microbiota of NP and FAP compared to CP patients (Table 2, Fig. 3; Supporting Fig. S1).

thumbnail image

Figure 2. Observed species (alpha rarefaction) curves for bacterial communities in the CP (on or off antibiotic), FP, and NP groups. Curves are based on 1–2000 sequences per dataset and show mean ± SEM.

Download figure to PowerPoint

thumbnail image

Figure 3. Observed species (alpha rarefaction) curves for Lachnospiraceae family members in the CP (on or off antibiotic), FP, and NP groups. Curves are based on 1–500 sequences per dataset and show mean ± SEM.

Download figure to PowerPoint

Table 2. Comparison of Proportions of 16S rRNA Gene Sequences of the 17 Most Highly Represented Families in Pouch Stool
Bacterial FamilyCP-offCP-onFAPNP
  • a

    Mean % (SEM).

  • b

    Comparison to CP-off P < 0.05.

  • c

    Comparison to CP-on P < 0.05.

  • d

    Comparison to both CP-off and CP-on P < 0.05.

Caulobacteraceae15.08a,c (3.37)5.83 (3.25)0.00d (0.00)0.00d (0.00)
Enterobacteriaceae9.81 (3.24)8.79 (4.19)3.82d (3.12)0.03d (0.02)
Sphingomonadaceae7.04c (2.12)2.02 (1.15)0.02d (0.01)0.01d (0.01)
Moraxellaceae4.85 (1.09)1.90 (0.96)0.06b (0.02)0.00d (0.00)
Comamonadaceae3.70c (0.84)1.50 (0.83)0.06b (0.03)0.03d (0.02)
Streptococcaceae12.30 (4.27)11.19 (3.90)1.54c (0.41)10.37 (5.43)
Lachnospiraceae6.93 (1.94)9.78 (3.85)33.61d (6.36)21.86d (4.07)
Peptostreptococcaceae8.58 c (2.40)0.60 (0.49)2.24 (0.96)4.20 (1.29)
Enterococcaceae7.19c (2.81)35.11 (8.44)0.00d (0.00)0.00d (0.00)
Clostridiaceae5.28c (1.52)1.13 (0.64)10.35 (5.81)3.47c (1.02)
Erysipelotrichaceae2.62 (0.89)1.56 (1.36)6.22d (1.76)6.25c (3.85)
Lactobacilliaceae1.60c (0.69)4.84 (1.87)0.01d (0.01)0.01d (0.01)
Veillonellaceae1.18 (0.43)0.19 (0.06)0.07d (0.04)0.31 (0.12)
Ruminococcaceae0.09 (0.04)0.08 (0.04)1.95b (0.94)5.40d (1.50)
Coriobacteriaceae0.91 (0.61)0.02 (0.01)3.50d (0.93)2.21d (0.51)
Bifidobacteriaceae1.34 (0.82)0.96 (0.78)8.51d (4.85)18.99d (4.79)
Incertae Sedis XIV5.27 (1.90)8.19 (3.64)21.74d (5.11)20.91d (4.28)

Total bacterial populations in CP stool were lower than those detected in the stool of FAP and NP subjects (mean log10 per gram 7.5 [standard error 0.1] compared to 8.2 [0.1] and 8.4 [0.5], respectively; Mann–Whitney P < 0.0001). Antibiotic administration did not reduce the total number of bacteria in CP pouch samples (7.4 [0.1]; P > 0.05).

FISH analysis showed that the phylogenetic gap (the proportion of the microbiota that did not react with the seven standard FISH probes; usually about 25% for human feces) was greater in the microbiota of CP-off and NP patients compared with that of the FAP samples (Fig. 4; P < 0.0001). Antibiotic administration reduced the size of the CP phylogenetic gap (Fig. 4; P < 0.0001). It was important therefore to determine the bacterial composition of the phylogenetic gap by comparison of the microbiota from antibiotic-treated and untreated CP patients. Differences in microbiota composition between the two patient groups might reveal etiological agents of pouchitis.

thumbnail image

Figure 4. Comparison of the size of phylogenetic gaps (the proportion of the microbiota that did not react with the seven standard FISH probes). Means and SEM are shown.

Download figure to PowerPoint

Bacterial Groups Responsible for the Expanded Phylogenetic Gap of CP Patient Stool

Phylum level analysis of 16S rRNA gene sequences showed that the microbiota of all patient groups was dominated by Firmicutes (Fig. 5). The CP-off microbiota had fewer Firmicutes sequences (53.5%) than CP-on (74.5%, P = 0.012). This trend was reversed for the phylum Proteobacteria, with a significantly higher percentage of sequences in the CP-off group (42.9%) than the CP-on group (21.7%, P = 0.005).

thumbnail image

Figure 5. Relative contribution of each of the four predominant phyla to the stool bacterial populations in CP (on or off antibiotic), FP, and NP patients. Means and SEM are shown.

Download figure to PowerPoint

The 17 most highly represented bacterial families encompassed >94% of all sequences in each study group. Comparisons of these families showed that the Caulobacteriaceae, Sphingomonadaceae, Comamonadaceae, Peptostreptococcaceae, and Clostridiaceae were reduced in the CP-on group compared to the CP-off group (Table 2). Therefore, the expanded phylogenetic gap observed in FISH/FC studies of CP-off stool was due to members of these bacterial families. Lactobacillaceae and Enterococcaceae formed a larger proportion of 16S rRNA gene sequences in CP-on microbiota, whereas Bifidobacteriaceae and Ruminococcaceae sequences were more common in NP microbiota (Table 2).

Evidence for Reduced Populations of Bacteria Belonging to the Lachnospiraceae, Insertae Sedis XIV, and Clostridial Cluster IV in the Stool of CP Patients

Two methodologies revealed that simplification of the microbiota composition in CP patient stool was associated with a reduction in the proportions of members of clostridial cluster IV, the Lachnospiraceae, and Incertae Sedis XIV. Members of clostridial cluster IV were not detected by FISH/FC in the stool of CP patients but formed 1.4% (SEM 0.8) and 3.2% (0.8), respectively, of NP and FAP stool microbiota. Faecalibacterium prausnitzii sequences, a species within clostridial cluster IV, were detected in NP (average 1.6% of total 16S rRNA gene sequences) and FAP (0.6%) subjects but were rarely detected in CP patients (CP-off = 0.039%; CP-on = 0.005%). Lachnospiraceae and Insertae Sedis XIV formed significantly (P < 0.05) greater proportions of the stool microbiota of NP and FAP stool compared to CP patients (Table 2). These differences were largely due to a decrease in sequences from members of the genera Ruminococcus, Dorea, Clostridium, and Eubacterium in CP patients (Table 3; Fig. S1).

Table 3. Comparison of Proportions of 16S rRNA Gene Sequences of the Four Most Highly Represented Lachnospiraceae Genera in Pouch Stool
Lachnospiraceae GenusCP-offCP-onFPNP
  • a

    Mean % (SEM).

  • b

    Comparison to CP-off P < 0.05.

  • c

    Some Ruminococcus species are currently included in the Ruminococcaceae, others in the Lachnospiraceae.

  • d

    Comparison to both CP-off and CP-on P < 0.05.

Dorea0.25a (0.16)0.75 (0.42)3.25b (1.04)2.62d (0.61)
Clostridium0.28 (0.17)2.03 (0.98)2.75b (0.74)3.02d (0.91)
Eubacterium0.84 (0.40)0.39 (0.28)6.40d (1.57)4.71d (1.00)
Ruminococcusc3.94 (1.42)5.96 (2.71)14.66d (5.12)6.31d (1.71)

Investigation of a Possible Etiological Role for C. perfringens in Pouchitis

Enrichment of unknown bacteria comprising the phylogenetic gap by cell sorting showed that for patients CP8, CP9, and CP16 most of the cloned genes represented Enterobacteriaceae. For patient CP13, in the absence of antibiotic treatment, 50% of clones from each of two samples represented C. perfringens. In contrast, C. perfringens was not represented in the sample obtained during antibiotic treatment. This result indicated that C. perfringens might have an important association with pouchitis in some patients. This possibility was investigated further by HTS analysis of the stool microbiota. Sequences from the Clostridiaceae comprised a higher proportion of total sequences in the CP-off and FAP groups (Table 2). Of particular note was the plethora of clostridial OTU, many of which did not represent known species, in CP-off microbiota (Fig. S2a). The collection of clostridial OTU was much simpler in CP-on, FAP, and NP microbiota (Fig. S2b–d). C. perfringens sequences were most abundant in the CP-off and FAP groups (Table 4). Quantitative PCR measurements showed that the prevalence of C. perfringens in stool samples differed between CP-off and FAP groups, but not with NP stool (Fig. 6). The numbers of C. perfringens when detected in stool were similar between CP and FAP groups (CP: median log10 genome equivalents per gram 5.9, range 5.7–8.9; FAP: median 6.5, range 5.5–7.8) but were lower in NP patients (4.9, range 4.9–6.9; P = 0.01). Therefore, C. perfringens is commonly associated with pouches.

thumbnail image

Figure 6. Prevalence of C. perfringens in the stool of patients based on results of qPCR detection. Fisher exact test P-values are shown.

Download figure to PowerPoint

Table 4. Comparison of Proportions of 16 S rRNA Gene Sequences of the Three Most Highly Represented Clostridial Species in Pouch Stool
Clostridial SpeciesCP-offCP-onFAPNP
  • a

    Mean % (SEM).

  • b

    Comparison to CP-off P < 0.05.

  • c

    Comparison to CP-on P < 0.05.

  • dComparison to both CP-off and CP-on P < 0.05.

C. butyricum0.37a (0.22)0.02 (0.01)0.00b (0.00)0.00b (0.00)
C. disporicum1.01 (0.38)0.09 (0.05)3.03 (1.35)2.06c (0.75)
C. perfringens3.51c (1.41)0.06 (0.04)5.04 (3.20)0.83 (0.51)

C. difficile Was Rarely Detected in Stool

Clostridium difficile was detected in two CP patients in the absence of antibiotic treatment, one NP patient, and one FAP patient (mean log10 genome equivalents per gram range 4.1 to 6.0).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  7. Supporting Information

We have analyzed the bacterial content (microbiota) present in the stool of FAP patients and UC patients with normal pouches (NP) and chronic or relapsing pouchitis (CP). We assumed that the immune system of FAP patients was not abnormally reactive to the pouch microbiota. Therefore, we postulated that if the composition of the microbiota were the same in FAP and CP pouches, then immunological factors would be predominant in the etiology of pouchitis. Another strength of including FAP pouches is that they have less mucosal inflammation. This limits a potential confounding factor in comparing normal UC pouches and pouches with pouchitis, as some “normal” pouches created for UC may be considered to have “mild” or “subclinical” pouchitis. There is certainly ongoing inflammation at a mucosal level.3 If the composition of the microbiota differed between FAP, NP, or CP pouches, then the nature of the bacterial inhabitants of CP pouches would be implicated as having a key role in the etiology of the pouchitis. Further, by comparing the CP microbiota in the absence (CP-off) or presence (CP-on) of antibiotic administration, a correlation between particular types of bacteria and pouchitis might be revealed.

The results of FISH analysis showed that bacteria not commonly present in human feces, nor in the stool of FAP patients, comprised about 50% of the stool microbiota of CP-off patients. Antibiotic treatment reduced the proportion of these unknown bacteria in the stool of pouchitis patients. Therefore, chronic or recurrent pouchitis was found to be associated with a microbiota that contained bacteria not commonly associated with human feces or FAP pouches. The uncommon bacteria that formed large proportions of the CP-off, but not of CP-on, microbiota were found to be members of the Caulobacteraceae, Sphingomonadaceae, Comamonadaceae, Peptostreptococcaceae, and Clostridiaceae. At least some of these groups may therefore be linked to the pathogenesis of pouchitis. Of additional note was the large diversity of clostridial OTU in CP-off microbiota, which presumably denotes particularly favorable conditions in the pouch for expansion of clostridial populations. NP stool also had an expanded phylogenetic gap relative to FAP subjects. In contrast to the CP-off microbiota, Ruminococcaceae, and particularly Bifidobacteriaceae, seemed responsible for the phylogenetic gap of NP stool. This knowledge may be useful in developing strategies to colonize newly formed pouches with innocuous bacteria.

Clostridium perfringens (comprising about 30% of the total bacterial community) was detected by qPCR (targeting the phospholipase gene sequence) in symptomatic pouches of 10 out of 17 (58.8%) CP-off patients. This species was not detected by qPCR in the same pouches when asymptomatic. However, 16S rRNA gene sequences of C. perfringens were detected in asymptomatic pouches, albeit as very small proportions of the total microbiota. The discrepancy between the results obtained by the two assay methods probably reflects their relative sensitivities of detection. The number of target sequences per cell differed greatly between assays. C. perfringens was detected in some normal pouches by qPCR, but in numbers about 30-fold lower than in pouchitis. Other authors have reported the detection of C. perfringens in pouchitis samples from UC patients.30–32 Taken together, the results suggest that C. perfringens may have an etiological role in pouchitis in some patients. Nevertheless, C. perfringens was present in similar numbers in CP-off and FAP stool, although of significantly lower prevalence among patients in the latter. There is thus an intriguing relationship between C. perfringens, a well-known opportunist pathogen, and pouchitis. Future research could compare the relative toxicity of isolates of these bacteria obtained from FAP and CP-off patients because CP isolates might have higher virulence than those normally detected in feces of healthy subjects. This would be reminiscent of the detection of adherent, invasive E. coli strains in patients with ileal CD that are more virulent than commensal strains.33

Clostridium difficile was rarely detected in our subjects. We did not obtain evidence, therefore, to invoke C. difficile as a common cause of pouchitis, although it is clear that in certain circumstances, it can have devastating effects on pouch patients.34

Duffy et al35 and Ohge et al36 reported the detection of sulfate-reducing bacteria in pouches, especially in patients with pouchitis. However, we did not detect any sequences representative of sulfate-reducing bacteria in the stool microbiota.

There is a growing consensus that inflammatory bowel diseases (IBDs), including pouchitis, are associated with a decrease in the number of phylotypes (reduced biodiversity) comprising the stool microbiota.37–43 The results of our study clearly support this consensus. The results of studies to date show a quantitative and a qualitative (biodiversity) reduction in representation of the Firmicutes phylum, and particularly clostridial cluster IV members in the stool of CD patients. This phylogenetic group contains several butyrate-producing bacteria, such as Faecalibacterium prausnitzii. Butyrate and other short chain fatty acids are believed to be important sources of energy for colonic epithelial cells and may have antiinflammatory properties, as well as improving barrier function of the bowel epithelium. Hence, the decrease in butyrate-producing bacteria in the colon or pouch might have an overall detrimental effect on the colonic mucosa.44–48F. prausnitzii was not detected in CP stool microbiota, but was detected at low levels in FAP and NP stool. Sequencing results were consistent with FISH results targeting these same bacteria (clostridial cluster IV). Moreover, members of the Lachnospiraceae, some of which produce butyric acid, were depleted in CP microbiota (average 6.93% CP-off sequences, 9.78% CP-on) relative to FAP and NP (average 33.61% and 21.86%, respectively). It is not clear from other studies whether the reduced biodiversity initiates IBD, perpetuates the diseases, or is a result of the diseases. NP and FAP pouches had similar proportions of Lachnospiraceae and clostridial cluster IV, indicating that the proportions of these bacterial groups in the microbiota might anyway be used as a biomarker of pouch health.

VSL#3 is a probiotic that contains viable lactobacilli, bifidobacteria, and streptococci. Kuehbacher et al49 carried out a double-blind, randomized, placebo-controlled trial to study the impact of VSL#3 on the microbiota of chronic pouchitis patients in remission induced by antibiotics. The authors focused mainly on the Enterobacteriaceae group. However, Lactobacillus and Bifidobacterium clone libraries generated from the VSL#3 group displayed a diverse spectrum of species in comparison with the two other experimental groups (pretreatment remission, n = 15; placebo group, n = 5). Some of these species were those included in the probiotic preparation. CP-on subjects in our study had larger proportions of Lactobacillaceae compared to CP-off patients. Bifidobacteriaceae were present in a higher proportion in NP stool compared to CP pouches. Therefore, pouches formed in UC patients provide a suitable habitat for Bifidobacteriaceae. Why they are in reduced proportions in CP stool poses another interesting topic for investigation. Since both Lactobacillaceae and Bifidobacteriaceae have the capacity to populate pouches and are associated with absence of inflammation, future research could investigate the physiological characteristics of these kinds of bacteria in relation to the pouch habitat. This could lead to interventions that would result in the population of newly constructed pouches with these bacteria, certain species of which may be more suitable for colonization of pouches than others.

As stated earlier, we hypothesized that if the composition of the microbiota differed between NP/FAP and CP pouches, then the nature of the bacterial inhabitants of CP pouches would be implicated as having a key role in the etiology of the pouchitis. Our detailed analysis of the bacterial content of ileo–anal pouches showed that differences in microbiota composition clearly existed between patient groups. The principal outcome was slightly incongruous: CP pouches had a microbiota of reduced biodiversity but expanded phylogenetic gap. In other words, our data showed that the CP microbiota was composed of less than 200 OTU representing uncommon fecal inhabitants. Further, by comparing the CP microbiota in the absence (CP-off) or presence (CP-on) of antibiotic administration, we hoped to associate particular types of bacteria with pouchitis. Possible associations exist but there is clearly a need to build on our detailed phylogenetic study by investigating the physiology of cultured bacteria representative of the phylogenetic groups that we have identified, and by testing them for potential to cause inflammation in experimental animals. This will help to explain the bacteriology of pouchitis in functional, ecological, and pathological terms.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  7. Supporting Information
  • 1
    Sandborn WJ. Pouchitis following ileal pouch-anal anastomosis: definition, pathogenesis, and treatment. Gastroenterology. 1994; 107: 18561860.
  • 2
    Fruin AB, El-Zammer O, Stucchi AF, et al. Colonic metaplasia in the ileal pouch is associated with inflammation and is not the result of long-term adaptation. J Gastrointest Surg. 2003; 7: 246253.
  • 3
    Thompson-Fawcett MW, Marcus V, Redston M, et al. Risk of dysplasia in long-term ileal pouches and pouches with chronic pouchitis. Gastroenterology. 2001; 121: 275281.
  • 4
    Thompson-Fawcett MW, Marcus VA, Redston M, et al. Adenomatous polyps develop commonly in the ileal pouch of patients with familial adenomatous polyposis. Dis Colon Rectum. 2001; 44: 347353.
  • 5
    Thompson-Fawcett MW. Pouchitis and pouch dysfunction. In: Satsangi J, Sutherland LR, eds. Inflammatory Bowel Diseases. Amsterdam: Elsevier, 2003; 567585.
  • 6
    Achkar JP, Al-Haddad M, Lashner B, et al. Differentiating risk factors for acute and chronic pouchitis. Clin Gastroenterol Hepatol. 2005; 3: 6066.
  • 7
    Ambroze WL Jr, Dozois RR, Pemberton JH, et al. Familial adenomatous polyposis: results following ileal pouch-anal anastomosis and ileorectostomy. Dis Colon Rectum. 1992; 35: 1215.
  • 8
    Madden MV, McIntyre AS, Nicholls RJ. Double-blind crossover trial of metronidazole versus placebo in chronic unremitting pouchitis. Dig Dis Sci. 1994; 39: 11931196.
  • 9
    Sandborn WJ, Tremaine WJ, Batts KP, et al. Pouchitis after ileal pouch-anal anastomosis: a Pouchitis Disease Activity Index. Mayo Clin Proc. 1994; 69: 409415.
  • 10
    Moskowitz RL, Shepherd NA, Nicholls RJ. An assessment of inflammation in the reservoir after restorative proctocolectomy with ileoanal ileal reservoir. Int J Colorectal Dis. 1986; 1: 167174.
  • 11
    Wallner G, Amann R, Beisker W. Optimizing fluorescent in situ hybridization with rRNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry. 1993; 14: 136143.
  • 12
    Amann RI, Krumholz L, Stahl DA. Fluorescent-oligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. J Bacteriol. 1990; 172: 762770.
  • 13
    Seksik P, Rigottier-Gois L, Gramet G, et al. Alterations of the dominant faecal bacterial groups in patients with Crohn's disease of the colon. Gut. 2003; 52: 237242.
  • 14
    Franks AH, Harmsen HJ, Raangs GC, et al. Variations of bacterial populations in human feces measured by fluorescent in situ hybridization with group-specific 16S rRNA-targeted oligonucleotide probes. Appl Environ Microbiol. 1998; 64: 33363345.
  • 15
    Lay C, Sutren M, Rochet V, et al. Design and validation of 16S rRNA probes to enumerate members of the Clostridium leptum subgroup in human faecal microbiota. Environ Microbiol. 2005; 7: 933946.
  • 16
    Manz W, Amann R, Ludwig W, et al. Application of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria of the phylum cytophaga-flavobacter-bacteroides in the natural environment. Microbiology. 1996; 142: 10971106.
  • 17
    Langendijk PS, Schut F, Jansen GJ, et al. Quantitative fluorescence in situ hybridization of Bifidobacterium spp. with genus-specific 16S rRNA-targeted probes and its application in fecal samples. Appl Environ Microbiol. 1995; 61: 30693075.
  • 18
    Harmsen HJ, Raangs GC, He T, et al. Extensive set of 16S rRNA-based probes for detection of bacteria in human feces. Appl Environ Microbiol. 2002; 68: 29822990.
  • 19
    Harmsen HJ, Wildeboer-Veloo AC, Grijpstra J, et al. Development of 16S rRNA-based probes for the Coriobacterium group and the Atopobium cluster and their application for enumeration of Coriobacteriaceae in human feces from volunteers of different age groups. Appl Environ Microbiol. 2000; 66: 45234527.
  • 20
    Sghir A, Gramet G, Suau A, et al. Quantification of bacterial groups within human fecal flora by oligonucleotide probe hybridization. Appl Environ Microbiol. 2000; 66: 22632266.
  • 21
    Lay C, Rigottier-Gois L, Holmstrom K, et al. Colonic microbiota signatures across five northern European countries. Appl Environ Microbiol. 2005; 71: 41534155.
  • 22
    Walter J, Tannock GW, Tilsala-Timisjarvi A, et al. Detection and identification of gastrointestinal Lactobacillus species by using denaturing gradient gel electrophoresis and species-specific PCR primers. Appl Environ Microbiol. 2000; 66: 297303.
  • 23
    Tannock GW, Munro K, Bibiloni R, et al. Impact of consumption of oligosaccharide-containing biscuits on the fecal microbiota of humans. Appl Environ Microbiol. 2004; 70: 21292136.
  • 24
    Altschul SF, Gish W, Miller W, et al. Basic local alignment search tool. J Molec Biol. 1990; 215: 403410.
  • 25
    Dowd SE, Callaway TR, Wolcott RD, et al. Evaluation of the bacterial diversity in the feces of cattle using 16S rDNA bacterial tag-encoded FLX amplicon pyrosequencing (bTEFAP). BMC Microbiol. 2008; 8: 125132.
  • 26
    Caporaso JG, Kuczynski J, Stombaugh J, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods. 2010; 7: 335336.
  • 27
    Cole JR, Wang Q, Cardenas E, et al. The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucl Acids Res. 2009; 37(Database Issue): D141D145.
  • 28
    Amar CF, East CL, Grant KA, et al. Detection of viral, bacterial, and parasitological RNA or DNA of nine intestinal pathogens in fecal samples archived as part of the English infectious intestinal disease study: assessment of the stability of target nucleic acid. Diagn Mol Pathol. 2005; 14: 9096.
  • 29
    Rintillä T, Kassinen A, Malinen E, et al. Development of an extensive set of 16S rDNA-targeted primers for quantification of pathogenic and indigenous bacteria in faecal samples by real-time PCR. J Appl Microbiol. 2004; 97: 11661177.
  • 30
    Ruseler van Embden JG, Schouten WR, van Lieshout LM. Pouchitis: result of microbial imbalance? Gut. 1994; 35: 658664.
  • 31
    Gosselink MP, Schouten WR, van Lieshout LM, et al. Eradication of pathogenic bacteria and restoration of normal pouch flora: comparison of metronidazole and ciprofloxacin in the treatment of pouchitis. Dis Colon Rectum. 2004; 47: 15191525.
  • 32
    Falk A, Olsson C, Ahrne S, et al. Ileal pelvic pouch microbiota from two former ulcerative colitis patients, analysed by DNA-based methods, were unstable over time and showed the presence of Clostridium perfringens. Scand J Gastroenterol. 2007; 42: 973985.
  • 33
    Darfeuille-Michaud A, Neut C, Barnich N, et al. Presence of adherent Escherichia coli strains in ileal mucosa of patients with Crohn's disease. Gastroenterology. 1998; 115: 14051413.
  • 34
    Shen BO, Jiang ZD, Fazio VW, et al. Clostridium difficile infection in patients with ileal pouch-anal anastomosis. Clin Gastroenterol Hepatol. 2008; 6: 782788.
  • 35
    Duffy M, O'Mahony L, Coffey JC, et al. Sulfate-reducing bacteria colonize pouches formed for ulcerative colitis but not for familial adenomatous polyposis. Dis Colon Rectum. 2002; 45: 384388.
  • 36
    Ohge H, Furne JK, Springfield J, et al. Association between fecal hydrogen sulfide production and pouchitis. Dis Colon Rectum. 2005; 48: 469475.
  • 37
    Komanduri S, Gillevet PM, Sikaroodi M, et al. Dysbiosis in pouchitis: evidence of unique microfloral patterns in pouch inflammation. Clin Gastroenterol Hepatol. 2007; 5: 352360.
  • 38
    Lim M, Adams JD, Wilcox M, et al. An assessment of bacterial dysbiosis in pouchitis using terminal restriction fragment length polymorphisms of 16S ribosomal DNA from pouch effluent microbiota. Dis Colon Rectum. 2009; 52: 14921500.
  • 39
    Frank DN, St Amand AL, Feldman RA, et al. Molecular-phylogenetic characterization of microbial community imbalances in human inflammatory bowel diseases. Proc Natl Acad Sci U S A. 2007; 104: 1378013785.
  • 40
    Sokol H, Pigneur B, Watterlot L, et al. Faecalibacterium prausnitzii is an anti-inflammatory commensal bacterium identified by gut microbiota analysis of Crohn's disease patients. Proc Natl Acad Sci U S A. 2008; 105: 1673116736.
  • 41
    McLaughlin, SD, Walker AW, et al. The bacteriology of pouchitis: a molecular phylogenetic analysis using 16S rRNA gene cloning and sequencing. Ann Surg. 2010; 252: 9098.
  • 42
    Zella GC, Hait EJ, Glavan T, et al. Distinct microbiome in pouchitis compared to healthy pouches in ulcerative colitis and familial adenomatous polyposis. Inflamm Bowel Dis. 2011; 17: 10921100.
  • 43
    Joosens M, Huys G, Cnockaert M, et al. Dysbiosis of the faecal microbiota in patients with Crohn's disease and their unaffected relatives. Gut. 2011; 60: 631637.
  • 44
    Nasmyth DG, Godwin PG, Dixon MF, et al. Ileal ecology after pouch-anal anastomosis or ileostomy. A study of mucosal morphology, fecal bacteriology, fecal volatile fatty acids, and their interrelationship. Gastroenterology. 1989; 96: 817824.
  • 45
    Sagar PM, Taylor BA, Godwin P, et al. Acute pouchitis and deficiencies of fuel. Dis Colon Rectum. 1995; 38: 488493.
  • 46
    Bohmig GA, Krieger PM, Saemann MD, et al. N-butyrate downregulates the stimulatory function of peripheral blood-derived antigen-presenting cells: a potential mechanism for modulating T-cell responses by short-chain fatty acids. Immunology. 1997; 92: 234243.
  • 47
    Segain JP, Raingeard de la Bletiere D, Bourreille A, et al. Butyrate inhibits inflammatory responses through NFkappaB inhibition: implications for Crohn's disease. Gut. 2000; 47: 397403.
  • 48
    Thibault R, Blachier F, Darcy-Vrillon B, et al. Butyrate utilization by the colonic mucosa in inflammatory bowel diseases: a transport deficiency. Inflamm Bowel Dis. 2010; 16: 684695.
  • 49
    Kuehbacher T, Ott SJ, Helwig U, et al. Bacterial and fungal microbiota in relation to probiotic therapy (VSL#3) in pouchitis. Gut. 2006; 55: 833841.

Supporting Information

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  7. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IBD_21936_sm_SuppFigS1A.eps249KSupporting Information Figure S1. Weighted Neighbor Joining trees with bootstrap support (100 repetitions) showing sequences identified as belonging to the ‘Lachnospiraceae’. Each leaf node indicates an OTU or a reference strain (RDP database). a, CP patients off antibiotics; b, CP patients on antibiotics; c, FAP patients; d, NP patients. Trees were generated using the RDP Treebuilder software. Trees show representative OTU, not the prevalence of OTU sequences.
IBD_21936_sm_SuppFigS1B.eps1324KSupporting Information Figure S1.
IBD_21936_sm_SuppFigS1C.eps1535KSupporting Information Figure S1.
IBD_21936_sm_SuppFigS1D.eps1496KSupporting Information Figure S1.
IBD_21936_sm_SuppFigS2A.eps1474KSupporting Information Figure S2. Weighted Neighbor Joining trees with bootstrap support (100 repetitions) showing sequences identified as belonging to the ‘Clostridia’. Each leaf node indicates an OTU or a reference strain (RDP database). a, CP patients off antibiotics; b, CP patients on antibiotics; c, FAP patients; d, NP patients. Trees were generated using the RDP Treebuilder software. Trees show representative OTU, not the prevalence of OTU sequences.
IBD_21936_sm_SuppFigS2B.eps1604KSupporting Information Figure S2.
IBD_21936_sm_SuppFigS2C.eps1547KSupporting Information Figure S2.
IBD_21936_sm_SuppFigS2D.eps1410KSupporting Information Figure S2.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.