Regional mucosa-associated bacteria from mice of the same litter were first examined by T-RFLP analysis in which the 16S rRNA gene was amplified from bacterial DNA and then cut with the restriction enzymes Msp I yielding DNA fragments of different lengths which were characteristic for different bacteria. As judged by T-RFLP dendrograms, the microbial profiles of the distal colon (DC) were distinct from those found in the proximal colon (PC) and stool samples (S), the latter two being relatively similar (representative T-RFLP dendrograms are shown in Figure 1A).

The T-RFLP profiles were further analyzed by pairwise Bray-Curtis distances to examine the similarity between samples from mice of the same litter. As shown in Figure 1B, corresponding samples collected from littermate mice clustered based on the location of sample collection. Proximal colon and stool samples with similar T-RFLP profiles grouped together. However, at a 40% distance cutoff, mucosa-associated distal colonic bacteria clustered separately from proximal colon and stool bacterial samples. To further quantify the differences of bacterial populations between proximal and distal colon, the absolute richness (the number of distinct terminal restriction fragments in each sample) of the bacterial community was calculated. The richness of the community was greater in stool samples (40.5±8.4) and the proximal colon (38.3±8.8) than that in the distal colon (28.7±6.2). The difference was significant comparing the distal colon to the other samples (n = 24, p<0.05) in all litters, but there were no significant differences in richness between the proximal colon and stool sample.

To gain insight into the specific bacterial phylotypes of the mouse proximal colon, distal colon and feces, 16S rRNA gene clone libraries were established and sequenced. The regional colonic microbiota among different litters was also examined to assess whether observed patterns in regional microbiota were random or reproducible. Two groups of conventional WT C57BL/6J mice were studied: (1) mice bred and housed in the animal care facility at the University of Chicago, and (2) mice bred and shipped from the Jackson Laboratory and maintained separately in the animal care facility. Samples from a representative mouse from each of 3 different litters of the Chicago (M1–3 (C)) and Jackson Laboratory (M1–3 (J)) groups were then harvested to construct the 16S rRNA gene libraries (Figure 2). A total of 3059 partial 16S rRNA gene sequences (average sequence length of 650 nucleotides) were obtained. 16S rRNA gene sequences were assigned into operational taxonomic units (OTUs) or phylotypes at a similarity cutoff value of 97%, such that sequences which were more than 97% similar were considered the same. Library coverage was calculated by Good's formula, and ranged from 70% to 80%, indicating that the 16S rRNA gene sequences from each library encompassed the majority of the bacterial population in the sample. Using the RDP Classifier tool to assign taxonomy to each 16S rRNA gene sequence, the majority of sequences were classified into two phyla Firmicutes and Bacteroidetes. Proteobacteria, Deferribacteres, Verrucomicrobia, Tenericutes, TM7 and a small portion of unclassified bacteria were represented at lower levels in most of the samples except for M1 (J) which had a relative dominance of Deferribacteres in the distal colon (Figure 2A). A small percentage of sequences belonging to this phylum were also found in all the samples harvested from mice obtained from the Jackson laboratory (less than 3%, Figure 2A). However, Deferribacteres could not be detected in the 1525 sequences collected from animals bred in the Chicago animal facility.

Further analysis of the partial 16S rRNA gene sequences that were identified as Deferribacteres showed that a total of 126 sequences in 9 samples could be assigned to 4 OTUs (based on 97% sequence identity). Ten colonies from each representative OTU were picked and further sequenced to obtain near full-length 16S rRNA gene sequences (accession numbers: FJ042492 to FJ042495). BLAST analysis, these sequences shared a close similarity (99%) with the previously described species Mucispirillum schaedleri which was first identified in mice from The Jackson laboratory. A phylogenetic tree was constructed from a multiple sequence alignment of 1510 bp 16S rDNA gene sequences using the tree builder tool in RDP. It included the four representative sequences from the aforementioned 4 OTUs identified in this study and 4 representative sequences available in the NCBI GenBank database described as Mucispirillum schaedler, as a control (Figure S1). The high degree of consensus and close relative phylogenetic position among these sets of sequences indicated that these 4 identified phylotypes fall into the same genus as Mucispirillum schaedleri with high confidence.

Similar to other reports, two bacterial phyla [13], Firmicutes and Bacteroidetes, were dominant in the mouse enteric microbiota. However, there was a clear difference in the relative representation of these phyla in mice bred at the two different facilities. Mice bred at The University of Chicago had a significantly lower proportion of Firmicutes and higher proportion of Bacteroidetes (F: 50.1%; B: 46.5%) compared to mice from Jackson laboratory (F: 78.4%; B: 8.9%; p<0.01). This can be further appreciated by the principal coordinate analysis (PCA) shown in Figure 2B and UniFrac P-Test (Table S1) where the separation of sample microbiota, regardless of regional location, could be seen between the two groups of mice.

However, despite the differences in microbiota among the various litters, regional differences between the mucosa-associated microbiota of distal colon versus proximal colon and feces remained consistent. This is shown in Figure 2C and Figure S2 where the T-RFLP results of two different litters are compared. While the microbiota differed substantially between the litters, the distal colon microbiota consistently clustered separately from the proximal colon and feces which were similar to each other. Further analysis of the differences of microbiota between mouse proximal and distal colon did not reveal a specific pattern of microbial communities shared among litters. However, limited diversity was invariably found in distal colon of all mice by 16S rRNA gene library cloning and sequencing as demonstrated by T-RFLP. The library cloning showed that the distal colon had significantly lower numbers of OTUs compared to the proximal colon and stool sample (Figure 2C, p = 0.001). This conclusion was also supported by rarefaction analysis in each individual mouse (Figure S3).

Toll-like receptors (TLRs) play a key role in innate immunity by regulating regional host immune responses, and, in doing so, we believe affects assemblage of mucosa-associated microbiota. We therefore examined the mucosal expression of TLR1, TLR2, TLR3, TLR4 and TLR5 in proximal and distal colon using real-time PCR. These TLRs were selected because they have been reported to be expressed in colonic epithelial cells [14], [15]. As shown in Figure S4, transcriptional expression of those TLRs was compared between mouse proximal and distal colon in both SPF mice and GF mice. Increased expression of TLRs was found in conventional WT mice compared to their GF controls. However, only TLR2 and TLR4 were found to be differentially expressed in the colon of SPF C57BL/6J mice (Figure 3). Both RT-PCR and Western blot analysis showed that TLR2 was highly expressed in proximal colon while TLR4 was highly expressed in mouse distal colon (Figure 3). In contrast, TLR2 and TLR4 were minimally expressed in the colonic mucosa of GF mice by Western blot analysis (Figure 3C). The regional differences in TLRs in the mouse colon are therefore likely dependent upon the regional differences of mucosa-associated bacteria. This analysis was from mucosal scrapings that may contain many cell types that express TLRs, therefore, to determine if the differential expression of TLR2 and TLR4 was specific to the colonic epithelium, laser capture and microdissection were performed to harvest colonic epithelial cells from proximal and distal colon in both SPF and GF mice. After RNA extraction from colonic epithelial cells, transcriptional expression of TLR2 and TLR4 in colonic epithelial cells of GF mice were detected by reverse transcriptional PCR and gel electrophoresis (data not shown here). However, the differential transcriptional expression of TLR2 and TLR4 along the proximal-distal axis of the colon was only found in microdissected epithelial cells of SPF mice but not in those of GF mice by real-time PCR (Figure 4). Thus, the responses in TLR expression to the enteric microbiota of whole mucosa are reflected by changes in colonic epithelial cells in these regions.

To determine the role of enteric microbiota in regulating regional TLR2 and TLR4 expression, microbiota transplantation (transfaunation) experiments were performed using GF C57Bl/6J mice. Microbiota from the cecum of conventional SPF WT mice were prepared as previously described [16] and transferred into GF recipient mice by oral gavage (200 ul cecal content in sterile saline buffer). Three weeks post-colonization, GF control mice and GF mice transfaunated were terminated. Samples from stool and proximal and distal colons were harvested for bacterial analysis. By T-RFLP analysis, both stool and mucosa-associated microbiota were established in GF mice with microbiota transplantation. T-RFLP profiles from stool, proximal and distal colon in GF recipient mice clustered tightly with the profiles of corresponding site samples from donor mice (Figure 5A). As in the SPF donors, the regional mucosa-associated microbiota of proximal and distal colon was significantly different. The mucosa-associated microbiota or the distal colon had a much lower diversity compared with that in proximal colon (Figure 5B). These findings suggest that regional host factors determine the assemblage of regional mucosa-associated microbiota.

Following transfaunation, the regional pattern of TLR2 and TLR4 expression in colonic mucosa was restored, as assessed by real-time PCR (data not shown) and Western blot analysis (Figure 5C). These results support the notion that the mucosal expression TLR2 and TLR4 in murine colon is dependent on regional mucosa-associated microbiota which is important in maintaining the gut innate immune function of the host.

To test whether the differential expression of TLR2 and TLR4 in the colonic mucosa is determined by the regional distribution of mucosa-associated microbiota, rectal enemas were performed using cecal microbiota which had been shown to be very similar to the mucosa-associated microbiota in the proximal colon. After twice daily treatments for 3 consecutive days, we observed the induction of TLR2 expression in the distal colon by exposure of the mucosal surface with the cecal slurry, but not with sterile saline (Figure 6). A major difference between the cecal slurry and luminal pellets of stool in the distal colon is that the former is dispersed and more likely to directly interact with the mucosa. TLR2 expression was increased by more than 3-fold compared to controls by Western blot analysis. In contrast, we did not detect changes in distal colonic TLR4 expression in either group (Figure 6). These data suggest that the mucosa-associated microbiota of the distal colon was still present, but the augmentation of microbial signals from exposure to proximal microbiota was sufficient to induce TLR2 expression as well. Thus, the distal colonic mucosa is not restricted in its ability to mount a TLR2 response and that regional TLR2 and TLR4 expression is most likely determined by the regional heterogeneity in mucosa-associated microbiota.

Male, conventional wild type (WT) C57BL/6J mice (8–10 w/o) were maintained in a 12-h light:dark cycle and allowed free access to food and water. Male, germ-free WT C57BL/6J mice (8–10 w/o) were maintained under the same conditions in the gnotobiotic facility of the Digestive Disease Research Core Center (DDRCC) at the University of Chicago. The germ-free (GF) colonies are routinely tested for microbes and parasites by the facility's staff and by our lab to ensure germ-free conditions. Two groups of conventional WT mice were studied. One was bred and housed in the animal care facilities of the University of Chicago under specific pathogen free conditions (SPF), while the other was obtained from the Jackson Laboratories and housed separately in the animal care facility. All groups of mice were allowed ad-libitum access to Harlan Teklad 7012 (SPF) or its autoclaved equivalent NIH 31 (GF) chow. All animal studies were approved by the Institutional Animal Care and Use Committee of the University of Chicago (permit number 72007).

Colonic samples at 1 cm proximal to the colonic-anal junction and distal to the cecal-colonic junction, respectively, were harvested for analysis. These sites will be referred to as distal and proximal colon, respectively. For bacterial population analysis, approximately 0.5 cm intact proximal colon and distal colon devoid of macroscopic luminal content were collected with minimal manipulation. Additional samples of the luminal contents of the cecum and colon were obtained as stool samples. All samples were kept on ice until completion of the tissue harvest and then immediately processed.

Samples were first homogenized in 1 ml extraction buffer [50 mM Tris (pH 7.4), 100 mM EDTA (pH 8.0), 400 mM NaCl, 0.5% SDS] containing 20 ul proteinase K (20 mg/ml). 500 ul of a slurry of 0.1-mm-diameter zirconia/silica beads (BioSpec Products, Bartlesville, OK) were added into the extraction tubes and a Mini-Beadbeater-8k Cell Disrupter (BioSpec Products) set on 5 min was used to lyse the microbial cells. After overnight incubation at 55°C, standard DNA extraction with phenol:chloroform:isoamyl alcohol, and precipitation with ethanol was performed. Isolated DNA was dissolved in TE buffer and stored at −80°C.

Bacterial populations in the samples were analyzed by 16S rRNA gene sequence-based T-RFLP as described previously [26]. Briefly, 16S rRNA gene sequences were amplified from DNA templates using broad-range primers 8F (5′-AGAGTT TGATCCTGGCTCAG-3′) labeled with 6′ carboxyfluorescein (6-FAM) and 1492R (5′-GGTTACCTTGTTACGACTT-3′) for the bacterial domain. PCR products were verified by electrophoresis of aliquots of PCR mixtures (8 ul) in 1.0% agarose and purified by precipitation. Aliquots of purified PCR products were digested by 20 U Msp I (New England Biolabs Inc.) and subsequently subjected to capillary electrophoresis using the Applied Biosystems DNA sequencer 3130.

Restriction-digest fragment abundance was determined using GeneMapper software (Applied Biosystems). Raw electropherograms were analyzed for artifacts, such as electrical anomalies, optical cross-talk between the capillaries, baseline drift, fluorescence of non-FAM-labeled contaminants, and distortions of the sizing ladder. Terminal restriction fragment data generated by GeneMapper were filtered and binned by the method developed by Abdo et al [27]. Based on the normalized T-RFLP profile, the number and height of peaks were treated as number and abundance of bacterial phylotypes represented in samples as described previously [26]. Pairwise Bray-Curtis distances [28] were calculated to examine the relationship between communities using the software package MEGA (http://www.megasoftware.net).

Unlabeled PCR primers 8F and 1492R were used to amplify 16S rRNA gene sequences from the samples using the same protocol as those for T-RFLP analysis. PCR products were purified by QIAquick gel extraction kit (Qiagen, Valencia, CA) and cloned into pCR-2.1-TOPO® vectors (Invitrogen, Carlsbad, CA) using the TOPO-TA Cloning Kit according to the manufacturer's instructions. From each library, 200 colonies were picked randomly and processed for DNA sequencing using 8F as the primer.

The 16S rRNA gene sequences were analyzed as described previously [26]. Briefly, raw sequence data were processed using the RDP pipeline server at the Ribosomal Database Project II (RDP-II) website (http://rdp.cme.msu.edu/pipeline) by base-calling, quality-trimming and alignment. Potential chimeric sequences were checked using the SimRank 2.7 package available through the RDP and excluded [29]. The classifier analysis tool of RDP-II and NEBI BLAST tool were used to assign 16S rRNA sequences to the taxonomical hierarchy at different levels. The program DOTUR with the furthest neighbor algorithm was used to group sequences into operational taxonomical units (OTUs) or phylotypes which represented the number of 16S rRNA sequence similarity groupings. A 97% cutoff value was used such that sequences with more than 97% similarity were considered the same. For principal coordinate analysis (PCA), all 16S rRNA gene sequences were imported into the ARB software package and aligned into a phylogenetic tree which was used to perform clustering analysis using online UniFrac without abundance weighting [30]. The P-Test in the UniFrac was performed to determine whether each sample was significantly different from others. All sequences were deposited in the GenBank nucleotide sequence databases under the accession numbers HQ318942-HQ322000.

Total RNA was extracted from mouse colonic mucosal scraping by Trizol (Invitrogen, Grand Island, NY) according to the manufacturer's instructions. Complementary DNA was synthesized using SuperScript II (Invitrogen) and random hexanucleotide primers. The forward and reverse primers to amplify TLRs were listed in Table 1. Real-time PCR was performed in an iCycler (Bio-Rad, Hercules, CA) using iQSYBR Green PCR Supermix (Bio-Rad). A two-step quantification cycling protocol was used. The Ct value is defined as the cycle number at which the fluorescence crosses a fixed threshold above the baseline. As a relative quantification, fold changes were measured using the ΔΔCt method as described previously [31].

The portions of proximal and distal colons were excised, embedded in OCT compound (Sakura Finetechnical, Tokyo, Japan), frozen on dry ice, and the tissue blocks were stored at −80°C. Frozen sections were cut to 6 µm thickness, mounted on positively charged MembraneSlides (PEN-Membrane 2.0 um, Leica), stained with the H&E method for LCM, and then air dried. LCM was performed using a Leica AS LMD system to harvest the mucosal epithelial cells. The microdissected samples were carefully collected into a PCR tube. RNA was extracted by using a PicoPure RNA Isolation Kit (Arcturus, Mountain View, CA) according to the manufacturer's instructions. The RNA was converted into cDNA using SuperScript II and random hexanucleotide primers (Invitrogen). Both regular PCR and real-time PCR were performed to detect the expression of TLRs in the cDNA samples.

Mouse mucosal scraping samples from proximal and distal colons were homogenized in cell lysis buffer (Cell Signaling Technology, Danvers, MA). Protein concentration determination and immunoblotting were performed as previously described [31]. Briefly, twenty micrograms of protein were separated by SDS-PAGE, transferred to polyvinylidene difluoride membranes and blotted with primary antibodies for TLR2, TLR4, and actin (Cell Signaling Technology, Danvers, MA). The specificity of the anti-TLR2 and –TLR4 antibodies was confirmed by Western blot comparison of proteins harvested from colonic mucosa from TLR2- and TLR4-knockout (KO) mice (Figure S5). Quantification of images was performed by scanning densitometry using NIH Image J 1.54 software (National Institutes of Health, Bethesda, MD).

Germ-free WT C57Bl/6J mice (8–10 w/o) were gavaged orally with cecal luminal contents harvested from C57Bl/6J donor mice (transfaunation) housed in the SPF facility (1 donor and 3 recipients; 2 independent experiments). Donor mice were terminated at day 0 and recipient mice 21 days after transfaunation. Cecal contents and colonic samples from both donor and recipient mice were collected for microbiota analysis as described previously [16].

Littermate conventional C57Bl/6J mice were randomly divided into a control or treatment group (n = 3 each). One additional mouse from the same litter was terminated at day 0 as the donor and the cecal content were collected, re-suspended in 4 ml sterile saline and stored in −20°C. Each morning and afternoon for 3 consecutive days, mice were anesthetized by an intraperitoneal injection of 200 µl of a mixture of 10 mg/ml of ketamin (LIoyd, Shenandoah, IA) and 1 mg/ml of xylazine (Abbott laboratories, Chicago, IL). Mice then received a rectal instillation of 200 µl of saline or 200 ul of a cecal slurry in saline by 2.4 cm steel cannula for control mice or treatment mice, respectively. Mice were kept in a head down vertical position for 20 min before returning to their cages to optimize retention of the enema solution. Following the last treatment on day 3, all mice were sacrificed and mucosal scrapings from the distal colon collected for analysis.

All data were expressed as mean ± SD. Student's t-test and one-way ANOVA were used to test the significance of differences between groups or samples. Statistical significance was set at p<0.05.