SEARCH

SEARCH BY CITATION

Keywords:

  • 16S ribosomal DNA;
  • inflammatory bowel disease;
  • intestinal microbiota;
  • mucosa-associated microbiota;
  • temporal temperature gradient gel electrophoresis

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Background: The mucosa-associated microbiota, being very close to the inflammatory process associated with inflammatory bowel disease (IBD), may have a pathogenic role. We used a culture-independent method to analyze the mucosa-associated microbiota in IBD patients at various points of the distal digestive tract.

Methods: Thirty-five patients (20 with Crohn's disease, 11 with ulcerative colitis, and 4 controls) underwent colonoscopy. Biopsies (n = 126) were taken from 4 sites: the ileum, right colon, left colon, and rectum. Fecal samples were also obtained from 7 individuals. Temporal temperature gradient gel electrophoresis (TTGE) of 16S rDNA was used to evaluate dominant species diversity. TTGE profiles were compared using software that measures the degree of similarity.

Results: In a given individual, the overall similarity percentage between the 4 segments of the distal digestive tract was 94.7 ± 4.0%, regardless of clinical status. The average similarity of all profiles for a given segment was 59.3 ± 18.3% in the overall population. Dendrogram analysis showed that TTGE profiles did not cluster with clinical status. Differences were observed between the dominant fecal microbiota and the mucosa-associated microbiota of all 4 sites, with similarity percentages less than 92%.

Conclusions: These results confirm that the dominant species differ between the mucosa-associated and fecal microbiota. They also show that, in a given individual, the microbiota is relatively stable along the distal digestive tract, showing a slight evolution in dominant species diversity from the ileum to the rectum, in both healthy subjects and patients with IBD.

The endogenous microbiota plays an important role in modulating the mucosal immune response and in the pathogenesis of inflammatory bowel disease (IBD), both in animal models and in humans.1–4 The microbiota is a growing focus of interest in patients with Crohn's disease (CD) and ulcerative colitis (UC). The advent of culture-independent bacteriological methods has facilitated microbiota analysis.5–8 Several clinical and experimental studies showed that the luminal biota plays an important role in the pathogenesis of CD. For example, diversion of the fecal stream prevented postoperative recurrence of ileal CD.9 Likewise, Harper et al10 reported that reintroduction of small bowel effluent into the intestinal lumen of patients with CD treated by split ileostomy triggered inflammation, whereas reintroduction of a sterile ultrafiltrate of the effluent did not. HLA-B27/human α2 microglobulin transgenic rats and interleukin 2 (IL-2) and IL-10 knock-out mice developed spontaneous intestinal inflammation, whereas inflammation was absent or attenuated when animals were kept in a germ-free state.11–17 In a recent study, Rakoof-Nahoum et al18 reported that commensal bacteria were recognized by Toll-like receptors (TLR) under normal conditions. Interactions of commensal bacterial products with those microbial pattern recognition receptors played a critical role in resistance to epithelial injury and in intestinal homeostasis. Thus, a dysregulated interaction between bacteria and TLR may promote chronic inflammation. We and others have previously reported that the fecal microbiota of patients with IBD differed from that of healthy subjects,19–22 containing more Enterobacteriaceae and more bacteria unusually found among the dominant biota,23 and that the biodiversity of this ecological niche remained high during the disease.19 The microbiota close to the mucosa, which differs from the luminal microbiota,24 has so far received less attention, yet, it is very close to the inflammatory process.25 Culture-based studies have led to the isolation, from early and chronic ileal CD lesions, of a new pathovar of Escherichia coli, designated adherent-invasive E. coli (AIEC), that may colonize the intestinal mucosa by adhering to intestinal epithelial cells.26–28 This adherent E. coli strain is specifically associated with ileal mucosa in CD and could be involved in the initiation of the inflammatory process.29 Rayment et al30 suggested an imbalance between Bacteroides and Clostridium spp. on the one hand and Bifidobacterium and Lactobacillus spp. on the other hand in the rectal mucosa-associated microbiota of patients with active IBD, whereas Swidsinski et al,31 using both culture and culture-independent methods, reported no major difference in the composition of the mucosal microbiota between patients with IBD and controls. Bifidobacteria and peptostreptococci have also been implicated in UC.32

The aim of this study was to analyze the mucosa-associated microbiota in patients with IBD at various sites of the colon and ileum. We chose to use temporal temperature gradient gel electrophoresis (TTGE), a culture-independent method, because it allows the dominant species diversity to be compared among samples.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Patients and Mucosal Samples

Thirty-five patients were studied (Table 1). Five patients had quiescent CD, 15 had acute CD, 5 had quiescent UC, 6 had acute UC, and the other 4 subjects were healthy controls undergoing colonoscopy for cancer surveillance in 2 cases and extraintestinal indications in 2 cases. The subjects' mean age was 39 years (age range, 18-90 yr). None of the patients had received antibiotics during the month preceding colonoscopy. Diagnosis of IBD was made in accordance with established criteria. Disease activity was assessed by the CD activity index (CDAI) for CD33 and the UC disease activity index (UCDAI) of Sutherland and Martin34 for UC, which also takes into account endoscopic lesions.

Table 1. Characteristics of Patients
 ControlsInactive CDActive CDInactive UCActive UC
  • †Results with this letter are significantly different.

  • *

    *P < 0.05 compared with all the other groups.

N451556
M/F2/21/46/91/42/4
Mean age (yr) (range)60 (55 65)*41 (30-51)37(18-90)36(26-54)44(35-49)
No. of biopsies1220581917

Inactive CD was defined by a CDAI <150, and active CD by CDAI >150. Inactive UC was defined as a UCDAI <2. Location of disease is summarized in Table 2. The protocol was approved by the local ethics committee, and informed consent was obtained from each subject before sampling.

Table 2. Location of Diseases
 CD (n)UC(n)
 2011
Location of CD  
Ileal5 
Ileo-colonic9 
Colonic2 
Segmental colitis4 
Location of UC  
Pancolitis 6
Left side colitis 5

Colonic cleansing was performed with polyethylene glycol 4000. Colonoscopy was performed with videoendoscopes. Biopsies were obtained from nonulcerated tissues using sterile single-use biopsy forceps during colonoscopy. In each subject, 1 biopsy (0.5 mg each) was taken from each of the following 4 segments: ileum, right colon, left colon, and rectum. In some cases, not all the segments were reached by colonoscopy, or the whole segment was ulcerated. Finally, a total of 126 biopsies was taken out of a tentative number of 140. There was no biopsy wash step before storage. Fecal samples were also obtained before colonic cleansing from 7 of the 35 individuals (4 with active CD and 3 with active UC). Fecal and biopsy samples were immediately frozen in liquid nitrogen and initially stored at −20 °C; they were shipped to the laboratory on dry ice within 15 days and stored at −80 °C until analysis.

Methods

DNA Isolation and 16S rDNA Amplification

Total DNA was extracted from 200-mg fecal samples, using the bead-beating method.19 The same DNA extraction method was modified for biopsy specimens to increase its efficiency. Nucleic acids were precipitated by isopropanol for 10 minutes at room temperature, followed by incubation for 15 minutes on ice and centrifugation for 30 minutes at 15,000g and 4 °C. Pellets were resuspended in 112 μL of phosphate buffer and 12 μL of potassium acetate. After the RNase treatment and DNA precipitation, nucleic acids were recovered by centrifugation at 15,000g and 4 °C for 30 minutes. The DNA pellet was finally resuspended in 30 to 100 μL of TE buffer. The concentration and integrity of nucleic acids were determined by electrophoresis on 1% agarose gel (1.25× TBE) containing ethidium bromide. DNA isolated from biopsy specimens and fecal samples was subsequently used as template to amplify the V6 to V8 regions of 16S rDNA, using primers U968-GC-F and L1401-R.35 Polymerase chain reaction (PCR) amplification was performed as described by Seksik et al.19 Several dilutions of template DNA were tested if the presence of PCR inhibitors was suspected (1 and 3 μL of crude extract or 1 μL at 10−1 dilution), and the highest PCR-positive dilutions were used for further analysis. PCR products were analyzed by electrophoresis on 1% agarose gels containing ethidium bromide to check their size (433 bp) and estimate their concentration.

Temporal Temperature Gradient Gel Electrophoresis

As previously described,19,36 electrophoresis was run for 20 hours in a Dcode Universal Mutation Detection System (Bio-Rad, Paris, France) at a fixed voltage corresponding to 64 mA, an initial temperature of 66 °C, and a ramp rate of 0.2 °C/h. Gels were stained in a solution of SYBR Green I (nucleic acid gel stain, Roche Diagnostics, Mannheim, Germany), and fluorescence was read on a Storm device (Molecular Dynamics, Amersham Biosciences, Freiburg, Germany).

Reading of TTGE Profiles and Statistical Analysis

TTGE profiles were compared using Gel Compar II software (Applied Maths, Kortrigk, Belgium). The analysis took into account the number of bands, their position on the gel, and their intensity. This software translates each TTGE profile into a densitometric curve, drawing a peak for each band (the area under the peak being proportional to the intensity of the band). A threshold area value was used to remove small peaks on the densitometric curves (these can be detected purely as a result of the amount of DNA applied to the gel). Only profiles with at least 3 bands were taken into account and, whenever possible, only gels obtained in the same run were compared. A marker consisting of a PCR amplicon mix of 7 cloned rDNAs from different bacterial species was used to normalize the profiles.19,36 During this step, the gel strips were stretched or shrunk so that the assigned bands on the reference patterns matched their corresponding reference positions. Similarity coefficients (Pearson correlation method) were calculated for each pair of profiles, yielding a similarity matrix. A dendrogram was constructed from this matrix by using an unweighted pair group method using arithmetic averages (UPGMA) algorithm. To define a positive similarity threshold, biopsy specimens from the same intestinal segment of a given individual were analyzed twice, with independent DNA extractions and PCR-TTGE. Similarity percentages between the mucosa-associated microbiota in these paired samples (3 paired samples from 3 individuals) were calculated with Gel Compar II software. When compared 2 × 2, the average similarity percentage was 92.0%. Taking into account methodological bias, this value of 92% was used as the positive similarity threshold. Median values of different samples were compared using the Wilcoxon test (P < 0.05). We focused on potential differences in the microbiota composition (1) between the different segments of the distal digestive tract, (2) between individuals in a given diagnostic group, and (3) between fecal samples and biopsy specimens. Statistical analyses were performed for intra- or interindividual comparisons using paired or unpaired Student's t-test.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Evolution of Dominant Species Diversity of the Mucosa-associated Microbiota Along the Distal Digestive Tract of a Given Individual

DNA was obtained from all samples, and 120 TTGE profiles with more than 3 bands could be analyzed. Electrophoretic patterns of mucosa-associated dominant microbiota are shown in Figure 1 for 1 subject from each diagnostic group. Mean similarity percentages between the 4 segments of the distal digestive tract (pairwise comparisons) ranged from 90.6 ± 7.2% (ileum versus rectum) to 96.3 ± 1.2% (left colon versus rectum; Table 3). The overall similarity percentage between biopsy specimens from all intestinal segments from a given individual was 94.7 ± 4.0%. Mean similarity percentages were highest between adjacent segments (mean > 95.5%). In the inactive CD group, the rectal mucosa-associated microbiota was significantly closer to the left colon one than to the ileal one. Moreover, the right colon mucosa-associated microbiota was closer to the left colon than to the rectal microbiota. In the active CD group, the right colon mucosa-associated microbiota was closer to the left colon than to the rectal one, and the mucosa-associated microbiota of the rectum was more similar to the left colon one than to the right colon microbiota. This indicated a moderate but significant evolution of the mucosa-associated microbiota along the intestine from rectum to ileum (Figure 2). A similar trend was noticed for the other groups, but it was not statistically significant. Ileal and left colon mucosa-associated microbiota were significantly more similar in the inactive CD group than in the active CD or inactive UC groups. This was also observed when comparing right and left colon microbiota.

Table 3. Average Intraindividual Site-to-site Similarity Coefficients (% ± SD) of the Noninflamed Mucosa-associated Microbiota of 30 Patients
 1 Versus RCI Versus LCI Versus RRC Versus LCRC Versus RLC Versus R
  • Mean similarity percentages were calculated using the Pearson correlation method (Gel Compar II software).

  • I, Ileum; RC, right colon; LC, left colon; r, rectum; n, number of biopsies.

  • *, ‡

    *Horizontal comparisons between intersegment similarities in each group of patients using a paired Student′s t test were significantly different (P < 0.05).

  • †Horizontal comparisons between intersegment similarities in each group of patients using a paired Student′s t test were significantly different (P < 0.05).

  • §

    †Horizontal comparisons between intersegment similarities in each group of patients using a paired Student′s t test were significantly different (P < 0.05).

  • §, ∥

    §.Vertical comparisons of intersegment similarities between the different clinical groups using a Student′s t test were significantly different (P < 0.05).

  • ∥, ¶, **

    .Vertical comparisons of intersegment similarities between the different clinical groups using a Student′s t test were significantly different (P < 0.05).

  • .Vertical comparisons of intersegment similarities between the different clinical groups using a Student′s t test were significantly different (P < 0.05).

  • **

    **.Vertical comparisons of intersegment similarities between the different clinical groups using a Student′s t test were significantly different (P < 0.05).

Controls95.687.379.996.5 ± 1.593.6 ± 3.897.7 ± 1.4
 (n = 2)(n = 2)(n = 2)(n = 6)(n = 6)(n = 8)
Inactive CD96.4 ± 2.396.5 ± 2.8§95.6 ± 2.797.7 ± 3.3***96.3 ± 3.9*97.3 ± 2.9
 (n = 10)(n - 10)(n = 8)(n = 10)(n = 8)(a = 8)
Active CD95.1 ± 1.993.6 ± 3.0§93.0 ± 4.495.1 ± 2.4*91.8 ± 5.2*95.0 ± 3.0
 (n = 10)(n = 10)(n = 12)(n = 18)(n = 18)(n = 22)
Inactive UC94.7 ± 1.991.4 ± 1.1193.9 ± 0.994.2 ± 2.7**92.0 ± 5.196.4 ± 2.2
 (n = 4)(n=4)(n = 4)(n = 10)(n = 10)(n = 10)
Active UC99.398.297.199.298.296.9 ± 1.8
 (n = 2)(n = 2)(n = 2)(n = 2)(n = 2)(n = 4)
All95.5 ± 0.792.2* ± 3.990.6 ± 7.295.9 ± 1.593.4 ± 2.196.3* ± 1.2
thumbnail image

Figure 1. TTGE profiles of the V6 to V8 regions of 16S rDNA from the ileo-colonic mucosa-associated microbiota of 5 patients in different diagnostic groups (healthy controls, inactive CD, active CD, inactive UC, and active UC). I, ileum; RC, right colon; LC, left colon; R, rectum. Similarity percentages were calculated for each pair of profiles using the Pearson correlation method (Gel Compar II software).

Download figure to PowerPoint

thumbnail image

Figure 2. Evolution of dominant species diversity of the mucosa-associated microbiota along the intestine from rectum to ileum. Each histogram bar indicates the mean similarity percentage between rectal mucosa-associated microbiota and mucosa-associated microbiota of other locations. Horizontal lanes indicate statistical comparisons between the mean similarity percentages. *, P < 0.05.

Download figure to PowerPoint

Comparisons of the Dominant Mucosa-associated Microbiota Between Subjects

Biopsy specimens were compared among the different subjects. The results for left colonic biopsy specimens from healthy controls and patients with inactive CD, active CD, inactive UC, and active UC are shown in Figure 3. These results were consistent for the different intestinal segments (ileum, right colon, left colon, and rectum), and we only show the results for the left colon, for which the number of interpretable biopsy specimens was the highest. Electrophoresis showed marked differences in the diversity of dominant mucosa-associated bacterial species among individuals. For the left colon, average similarity was 67.3% overall (Table 4), 76.6 ± 10.2% in controls, 66.6 ± 13.7% in patients with CD patients, and 69.3 ± 13.2% in patients with UC. The similarity between left colon mucosa-associated microbiota was significantly higher among healthy individuals than among patients with CD. The same was observed for the rectal mucosa-associated microbiota, which had a higher similarity index in healthy controls than in patients with CD or UC.

Table 4. Average Interindividual Similarity Coefficients (%) of the Noninflamed Mucosa-associated Microbiota of the 30 Patients
 IRCLCRAll
  1. Results are expressed as average percentages (%) ± SD.

  2. *†‡§∥¶Horizontal comparisons between the 4 intestinal segments, in each clinical groups using a paired Student′s t test were significantly different (P < 0.05).

  3. **††‡‡Vertical comparisons between the 3 different groups, for each intestinal segment using a Students t test were significantly different((P < 0.05).

ControlsND50.7 ± 26.5*† (n = 4)76.6 ± 10.2***(n = 4)78.4 ± 10.3†††††(n = 4)68.6 ± 20.9
CD59.2 ± 17.5† (n= 1550.9 ±19.7‡; (n == 17)66.6 ± 13.7‡§** (n= 20)53.5 ± 18.5§‡‡( n =20)57.6 ± 18.4
UC69.9 ±14.6 (n = 6)60.7 ± 14.4 (n = 8)69.3 ± 13.2 (n= 10)59.3 ± 13.2†‡ (n = 11)64.8 ± 13.8
All60.1 ± 17.852.5§ ± 19.967.3§; ± 13.856.0; ± 18.6 
thumbnail image

Figure 3. TTGE profiles (V6 to V8 regions of 16S rDNA) of the mucosa-associated microbiota in biopsy specimens analyzed by Gel Compar II software. Examples of comparison of the left colon mucosa-associated microbiota of 2 patients from each diagnostic group (healthy controls, inactive CD, active CD, inactive UC, and active UC).

Download figure to PowerPoint

Average similarity coefficients were significantly higher between biopsy specimens from the left colon (67.3 ± 13.8%) than from the right colon (52.5 ± 19.9%) or rectum (56.1 ± 18.6%; Table 4). The same significant differences were found in the CD group, where the left colon microbiota was more similar between patients compared with the other locations. In the control group only, the mucosa-associated microbiota of the right colon was significantly more different between individuals than left colon or rectal microbiota (P < 0.05). No significant differences were found between segments for the UC patients. Dendrogram analysis showed that the TTGE profiles did not cluster with clinical status (UPGMA dendrogram not shown).

Comparison of the Mucosa-associated and Fecal Microbiota

The dominant fecal microbiota was compared with the dominant mucosa-associated microbiota of the different segments in the 7 subjects who provided fecal samples before colonoscopy (4 with active CD and 3 with active UC). Differences were observed at each site, with similarity percentages below the defined positive similarity threshold of 92%. Average similarity percentages between fecal and biopsy specimens ranged from 79.9 ± 6.0% between ileum and feces to 88.2 ± 6.6% between rectum and feces (Table 5), but no statistical differences were observed. The mucosa-associated microbiota was significantly more similar to the fecal microbiota in patients with active CD than in patients with active UC (89.2 ± 4.3% and 80.2 ± 6.4%, respectively). This difference was mostly because of the mucosa-associated microbiota of the right colon, which was significantly closer to the fecal microbiota in patients with active CD than in patients with active UC.

Table 5. Average Similarity Percentages Between the Fecal and Mucosa-associated Microbiota
 Feces Compared with
Patients1RCLCRAll
  1. Mean similarity percentages were calculated using tin: Pearson correlation method (Gel Compar II software).

  2. Results are given as mean intraindividual similarity percentages ± SD.

  3. n, number of patients; sec Table 3 for other abbreviations.

  4. *‡Average similarity values with the same letters were significantly different using a paired Student′s t test (P < 0.05).

Active CD (n = 4)82.7±5.0 (n = 2)86.5 ± 1.2† (n = 2)90.8 ± 1.3 (n = 4)92.1 ± 3.0 (n †4)89.2 ±4.3*
Active UC ( n =3)74.3 ( n =1)76.5 ±1.1† ( n =2)81.8 ±8.4 ( n =3)82.9 ± 6.8 ( n =3)80.2 ±6.4*
All79.9 ± 6.081.5 ±5.887.0 ± 6.988.2 ±6.6 

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

This study confirms that the dominant mucosa-associated microbiota differs from the dominant fecal microbiota. Furthermore, it shows that, in a given individual, the microbiota is relatively stable along the distal digestive tract, showing a slight evolution in dominant species diversity from the ileum to the rectum, in both healthy subjects and patients with active or inactive CD.

The method used here-denaturing gel electrophoresis-is capable of separating bacterial sequences with the same size but different thermal stability.19,35 Because 16S rDNA from different bacterial species have different nucleotide sequences in variable regions, their thermal stability is also different. In theory, sequences differing by a single base can be separated by this method. When applied to complex microbial communities, this method gives profiles corresponding to all the dominant bacterial species present in the sample. This study did not aim to determine the composition of the mucosa-associated microbiota in terms of bacterial genera or species but assessed for each individual the biodiversity along the digestive tract. The method used was very efficient for comparing the biodiversity of complex bacterial communities. TTGE profiles were obtained with 120 of the 126 biopsy specimens tested in this study. Five of the 6 biopsy specimens that yielded less than 3 bands after PCR were from patients with acute phase IBD. This PCR amplification failure could be caused by a reduced bacterial biomass or the presence of PCR inhibitors. Previous studies have shown an increased mucosal bacterial load in IBD patients relative to healthy controls,31,37 supporting the second explanation. However, the differences in mucosal biomass between published results may partly be caused by different sample preparations. Indeed, a significant part of the intestinal microbiota may be associated with mucus,37 and some authors thoroughly wash biopsy specimens before analysis.31 In our study, no biopsy wash was performed to take into account the bacteria of the mucus layer. Furthermore, a wash step for biopsies could have a detrimental effect on anaerobic bacteria, thus underestimating the microbiota diversity. The patients in our control group were older than our IBD patients (60 versus 39 yr). Indeed, full colonoscopy is rarely indicated in young subjects without IBD. Several studies38–41 have shown that the intestinal microbiota differs in elderly subjects more than 65 years of age, and precise knowledge is presently lacking concerning potential changes in adults more than 40 years of age. The effects of dietary factors such as consumption of prebiotics or fermented milks containing probiotics on the mucosal microbiota are currently assessed by many research teams in the hope to modify the course of IBD.1,42 We did not study dietary habits, which probably differed between our patient groups. Indeed, this study focused on the stability of the microbiota between different parts of the distal intestinal tract in an individual but did not intend to identify modulating factors.

In a given individual, most of the bands obtained with the samples from the ileum, right colon, left colon, and rectum were common with similar intensities. This indicates that essentially the same bacterial species predominate from the ileal to the rectal mucosa. Similar results have been obtained in the colon by Zoetendal et al.24 The ecological conditions in the intestinal lumen differ greatly between the ileum, right colon, and distal colon. We and others have previously reported that the luminal biota also differs greatly between these parts of the distal digestive tract.43,44 Therefore our results strongly suggest that the composition of the mucosa-associated microbiota is more strongly influenced by host factors than by environmental conditions. The mucosa-associated microbiota might have a more important pathogenic role than the luminal microbiota in some intestinal diseases, because it is closer to epithelial and immune cells. Increased adherence of commensal bacteria to inflamed mucosa may enhance the exposure of the mucosal immune system to intestinal bacteria or bacterial components, resulting in sustained inflammation.37 In this study, we also found very few regional differences in the mucosa-associated microbiota in patients with IBD, whatever their clinical status. These differences were mostly observed in the inactive and active CD groups. We observed a trend in the dominant bacterial species from the rectum to the ileum. The rectal microbiota was more similar to the left colon one than to the right colon or ileal one. Moreover, the microbiota associated to the right colon mucosa was closer to the left colon one than to the rectal one in both CD groups. The same tendency was observed for the control group, but the number of comparisons was too small to be statistically analyzed.

The mucosa-associated microbiota and fecal microbiota appeared different. This confirms the result obtained by Zoetendal et al24 in healthy individuals, and the significance was also confirmed despite the low number of patients (n = 7). Moreover, the mucosa-associated microbiota was significantly more similar to the fecal microbiota in patients with active CD than in those with active UC (P < 0.05). This may suggest that bacterial agents present in the luminal biota have a more important pathogenic role in CD than in UC.

Our results also showed that the mucosa-associated microbiota was specific to each individual. This is consistent with results reported for fecal samples.19,35 The average similarity percentage among the 30 patients studied was 59.3 ± 18.3%. Previously, Mangin et al23 studied the microbial species composing the fecal microflora of patients with CD using a molecular inventory method. They showed numerous uncommon clones in the feces of those patients. This would lead to an increase in interindividual variability, which is consistent with our observations in the patients with IBD. Indeed, we found the microbiota associated to the left colon to be more similar between healthy individuals than between CD patients. In addition, the rectal mucosa-associated microbiota was also more similar in healthy controls than between CD or UC patients. In contrast, dendrogram analysis did not separate the profiles according to clinical status, suggesting that the dominant mucosal microbiota is not specifically altered by IBD, regardless of disease activity. If the mucosa-associated microbiota is indeed involved in pathogenesis of IBD, we identified no dominant bacterial species specific to either CD or UC. Moreover, the high complexity of the dominant mucosa-associated microbiota, as reflected by TTGE profiles, does not suggest that a single group of bacteria is involved. Culture-based studies have also shown marked interindividual variability of the microbiota in terms of both total counts and genera present in rectal biopsy specimens from patients with UC.45,46

In conclusion, we observed, both in individuals with IBD and in healthy controls, a remarkably stable dominant mucosa-associated microbiota from the ileum to the rectum. A slight trend toward higher similarities was noticed between adjacent segments. No particular pattern of dominant mucosa-associated microbiota was linked to the disease status or location. This suggests that microbial factors possibly involved in the pathogenesis of IBD are likely to be complex, in keeping with the concept of dysbiosis.47 Our work emphasized that the mucosa-associated microbiota differed from the fecal microbiota, and this should be taken into account in future studies on the role of the endogenous microbiota in IBD. In addition, because the mucosa-associated microbiota seems to be specific of an individual, it would be highly relevant to compare, for each patient with CD, ulcerated mucosa to nonulcerated mucosa, the latter being considered as an internal control.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

We thank the volunteers for participation in this study and Nelly Boulay and Michèle Sérézat for technical support.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References