|Home | About | Journals | Submit | Contact Us | Français|
Commensal flora plays important roles in the regulation of the gene expression involved in many intestinal functions and the maintenance of immune homeostasis, as well as in the pathogenesis of inflammatory bowel diseases (IBD). The microRNAs (miRNAs), a class of small, non-coding RNAs, act as key regulators in many biological processes. The miRNAs are highly conserved among species and appear to play important roles in both innate and adaptive immunity, as they can control the differentiation of various immune cells as well as their functions. However, it is still largely unknown how microbiota regulates miRNA expression, thereby contributing to intestinal homeostasis and pathogenesis of IBD. In our current study, we found that microbiota negatively regulated intestinal miR-10a expression, in that the intestines, as well as intestinal epithelial cells and dendritic cells of specific pathogen-free (SPF) mice, expressed much lower levels of miR-10a compared to those in germ-free (GF) mice. Commensal bacteria downregulated DC miR-10a expression via TLR-TLR ligand interactions through a MyD88-dependent pathway. We identified IL-12/IL-23p40, a key molecule for innate immune responses to commensal bacteria, as a target of miR-10a. The ectopic expression of miR-10a precursor inhibited, whereas miR-10a inhibitor promoted, the expression of IL-12/IL-23p40 in DC. Mice with colitis expressing higher levels of IL-12/IL-23p40 exhibit lower levels of intestinal miR-10a compared to that in the control mice. Collectively, our data demonstrated that microbiota negatively regulates host miR-10a expression, which may contribute to the maintenance of intestinal homeostasis by targeting IL-12/IL-23p40 expression.
The intestinal mucosa is home to a massive and diverse microbiota shortly after birth(1). In the large intestine and colon, commensal bacteria can reach a density of 1012 organisms per gram and comprise more than 1,000 species, including both anaerobes and aerobes(2–4). In addition to providing benefits to their host, including the breakdown of indigestible food, the supply of energy for colonic epithelial cells, and a barrier against invasive pathogenic bacteria, the gut microbiotainstructs the postnatal maturation of gut immune defenses, which play a crucial role in maintenance of intestinal homeostasis (5–7). Commensal bacteria modulate the expression of genes involved in many intestinal functions. It has been shown that colonization by commensal bacteria, Bacteroidesthetaiotaomicron, is able to modulate the expression of host genes that participate in fundamental physiological functions(8). Colonization with segmented filamentous bacteria induces gut Th17 cell development(9), whereas Bacteroidesfragilis induces gut IL-10-producing Foxp3+ Treg cells (10, 11). Accumulating evidence from reports in multiple experimental IBD models indicates that normal intestinal microbiota is critical to the pathogenesis of inflammatory bowel diseases as well, in that intestinal inflammation only develops in mice housed in a conventional environment, but not under germ-free conditions(3, 12). However, how the normal host-commensal interaction is regulated is still largely unknown.
The recent discovery of microRNAs (miRNAs) has greatly expanded our understanding of the mechanisms that regulate gene expression(13, 14). miRNAs are small, non-protein-coding RNAs of 19–25 nucleotides that regulate gene expression by targeting mRNA in a sequence-specific manner, either by repressing translation or directly leading to cleavage of mRNA sequences(15). The miRNAs are highly conserved among species and appear to play important roles in both innate and adaptive immunity, as they can control the differentiation of various immune cells as well as their functions(16, 17). It has been shown that specific miRNAs are upregulated during the activation of innate immunity, which is able to influence innate responses to microbial and viral infections(18–20). It is now apparent that abnormal miRNA expression is a common feature of various human diseases, such as cancer, developmental abnormalities, muscular and cardiovascular disorders, and, most recently, inflammatory diseases including inflammatory bowel diseases (IBD) (21–23). However, how microbiota regulate miRNA expression and thus contribute to the maintenance of intestinal homeostasis and to IBD pathogenesis is still largely unknown. In this report, we demonstrate that microbiota negatively regulates miR-10a expression, in that the intestines, as well as intestinal epithelial cells and DCs, of mice housed under specific pathogen-free (SPF) conditions expressed much lower levels of miR-10a, compared to those in mice housed under germ-free (GF) conditions. Stimulation with TLR ligands downregulates the dendritic cell (DC) expression of miR-10a via the MyD88-dependent pathway. We have further identified IL-12/IL-23p40 as a target gene of miR-10a. Furthermore, in colitic IL-10-deficient mice which express high levels of IL-12/IL-23p40, gut miR-10a expression is much lower than that of control wild-type mice, indicating that miR-10a could negatively regulate intestinal IL-12/23p40 expression in the mice with colitis.
C57BL/6 (B6) mice, B6.IL-10−/− mice, B6.MyD88−/− mice, and B6.RAG−/− mice were obtained from The Jackson Laboratory and maintained in the animal facilities of the University of Alabama at Birmingham(UAB) and the University of Texas Medical Branch (UTMB). Germ-freeB6mice were derived by hysterectomy and maintained in Trexler-type isolators, according to standard gnotobiotic techniques in which germfreeSW mice (Taconic) are used as foster mothers(24). Isolators were monitored for contamination monthly by examination of Gram-stained films of fresh fecal samples, aerobic and anaerobic bacterial and fungal cultures of fresh fecal samples, and swabs of water bottle sipper tubes and isolator interiors. We used 8-to 10-wk-old female mice in these experiments. All experiments were reviewed and approved by the Institutional Animal Care and Use Committees of UAB and UTMB.
RPMI 1640, DMEM, HEPES, penicillin-streptomycin, FBS, 2-mercaptoethanol, L-glutamine, and sodium pyruvate were purchased from LifeTechnologies (Carlsbad, CA). GM-CSF, anti-CD11c, anti-CD11b, anti-CD80, anti-CD86 were purchased from BD Biosciences (San Diego, CA). Restriction endonucleases, T4 DNA ligase, psiCHECK-2 vector, dual-luciferase reporter system were from Promega (Madison, WI). Lipofectamine™ 2000 and TRIzol® were purchased from Invitrogen (Carlsbad, CA). CollegenaseIV was obtained from Sigma-Aldrich (St. Louis, MO). TaqMan® microRNA reverse transcription kits and TaqMan® gene expression assays were from Applied Biosystems (Carlsbad, CA). miR-10a precursors and inhibitors were purchased from Ambion (Carlsbad, CA). Nucleotides were synthesized by Fisher Scientific (Pittsburgh, PA).
Bone marrow cells were isolated as described previously. Briefly, bone marrow cells were suspended at 2.5 × 105/ml in complete RPMI 1640 media containing 10% heat-inactivated FCS (Atlanta Biologicals, Lawrenceville, GA), 25 mM HEPES buffer, 2 mM sodium pyruvate, 50 mM 2-mercaptoethanol, 100 IU/ml Penicillin, and 100 μg/ml Streptomycin (CellgroMediatech, Manassas, VA). The cells were cultured in the presence of 20 ng/ml GM-CSF (R&D Systems, Minneapolis, MN) in 6-well plates at 37°C in 5% CO2 in humid air. On day 8, BMDCs were harvested and used as described in text.
To isolate intestinal epithelial cells, the intestines were washed and cut into small pieces. Then the latter were incubated with calcium- and magnesium-free HBSS supplemented with 2% FBS and 5 mM EDTA (Sigma-Aldrich) on a magnetic stirrer at 37°C for 30 min. The liberated cells were collected by passage through a stainless steel sieve. The isolated cells were pooled together and epithelial cells separated on a 20/75% discontinuous Percoll gradient (Pharmacia). To isolate lamina propria DCs and T and B cells, after removal of epithelial cells and intraepithelial lymphocytes, the intestinal tissues were incubated with RPMI 1640 containing 5% FBS and 0.5 mg/ml collagenase type IV (Sigma) for 30 min at 37°C with stirring. The liberated cells were collected by passage through a stainless steel sieve. Then isolated cells were pooled together and separated on a 40/75% discontinuous Percoll gradient (Pharmacia). The cell yield was typically ~2 × 106 lymphocytes per mouse with >90% cell viability. DCs, T cells and B cells were further isolated by MACS for which we used CD11c-beads, CD3-beads, and B220-beads. The cell purity was normally around 95% for each cell population.
Total RNA isolation was performed by using the TRIzol method (Invitrogen, Carlsbad, CA), according to the manufacturer’s instructions. Total RNA labeling and hybridization on miRNA microarrays were performed as previously described. Briefly, we biotin labeled 5 μg of total RNA of each testing sample by reverse transcription with a 5′ biotin end-labeled random octomeroligo primer. Biotin-labeled cDNA was hybridized on an miRNA microarray chip (The Ohio State University, Ver. 4.0), containing 1700 miRNA gene-specific oligo probes derived from 474 human miRNA and 373 mouse microRNA genes (http://microrna.sanger.ac.uk/cgi-bin/sequences/browse.pl; accessed Nov 2006) and printed on arrays in duplicate. Hybridization signals were detected by biotin binding of a streptavidin-Alexa647 conjugate on an Axon Scanner 4000B (Axon Instruments, Union City, CA). The images were quantified by GENEPIX 6.0 software (Axon Instruments, Union City, CA). The microarray data were deposited in ArrayExpress (http://www.ebi.ac.uk/arrayexpress), accession numbers as: E-MEXP-3406.
Total RNA was extracted with TriZol reagent and followed by cDNA synthesis with Superscript reverse transcriptase (Invitrogen, Carlsbad, CA). Quantitative PCR reactions were performed by using TaqMan® Gene Expression Assays for miR-10a and IL-12/IL-23p40 (Appliedbiosystems, Union City) on a Bio-Rad iCycler (Bio-Rad, Hercules, CA) and all data were normalized to Gapdh mRNA expression.
The dual-luciferase psiCHECK-miR-10a and psiCHECK-IL-12/IL-23p40 vectors were constructed by synthesizing the candidate seed sequences in these genes located in the 3′-UTR and inserting the annealing products into the psiCHECK-2 vector (Promega, Madison, WI) by using Not I (GCGGCCGC) and Xho I (CTCGAG) restriction endonucleases. For mutant constructs psiCHECK-mmiR-10a and psiCHECK-mIL-12/IL-23p40, three-base pair mutations were introduced into the seed sequences. The nucleotide sequences of constructed plasmids were confirmed by enzyme digestion and DNA sequencing.
Murine macrophage RAW264.7 cells were cultured in complete Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 10 mM HEPES, 100 units/mL penicillin/streptomycin, 2 mM glutamine, 1 mM sodium pyruvate, and 50 mM b-mercaptoethanol at 37°C in 5% CO2/95% air under humidified conditions. For reporter assays, cells were seeded in 24-well plates (1.5×106/well) and transfected with one of the following: 0.8 mg recombinant dual-luciferase vectors alone, vectors plus 30-nM hairpin precursors, or 30-nM inhibitors, with Lipofectamine™ 2000 reagent (Invitrogen, Carlsbad, CA). Luciferase assays were performed 24 hours later by using the Dual-Luciferase reporter system. The renilla and firefly luciferase signals were measured on the VeritasMicroplateLuminometer (Promega, Madison, WI).
Levels of significance were determined by Student’s t test. P values of < 0.05 were considered statistically significant.
Although miRNAs have been implicated in regulation of both innate and adaptive immune responses, there is limited information on microbiota regulation of miRNA expression of the intestinal mucosal system. To determine whether there is a unique miRNA expression profile in the intestines and whether microbiota regulates such miRNA expression, RNA was isolated from the intestines and spleen of B6 mice housed under SPF or GF conditions. miRNAmicroarray profiling of approximately 373 mouse microRNA genes was performed. Several miRNAs, including miR-10a, were differentially expressed between the spleen and intestines. As shown in Figure 1A, miR-10a was highly expressed in the intestines, but at very low levels in the spleen of GF mice, indicating that the miR-10a was predominantly expressed on the intestines. The intestinal tract is the home of extremely diverse and dense commensal bacteria that are normally non-pathogenic in an immunocompetent host, whereas the spleen remains in a sterile condition. Microbiota have been shown to exert a great effect on hosts and, most particularly, on the intestines. In our study, intestinal miR-10a expression was decreased in mice under SPF conditions compared to that in the GF mice (Figure 1A). Recolonization of the GF mice with normal flora led to a decreased intestinal miR-10a expression (Figure 1B). These data collectively indicated that commensal bacteria could down-regulate intestinal miR-10a expression. Notably, miR-10a was expressed predominantly in the intestines but not in the spleen, even in SPF mice.
Positioned at the interface between the intestinal lumen and the host mucosal immune system, epithelial cells and DCs form the first line that senses microbiota, and commensal bacteria modify the mucosal immune response mainly by regulating the mucosal innate response. A recent report demonstrated that miR-10a affects the pro-inflammatory phenotype in athero-susceptible endothelium by regulating NFkB activation. When we compared miR-10a expression by the spleen and the intestines of RAG−/− mice housed under SPF and GF conditions, similar to those of wild-type mice, miR-10a was highly expressed in the intestines, but at very low levels in the spleens of GF RAG−/− mice. Furthermore, intestinal miR-10a expression was decreased in SPF RAG−/− mice compared to that in GF RAG−/− mice (Figure 1C). We then investigated whether miR-10a expression varied between wild-type B6 mice housed under SPF conditions and GF conditions by examining different populations of intestinal cells by real-time PCR. In GF mice, intestinal epithelial cells (IEC) expressed the highest levels of miR-10a, while lamina propria (LP) DC miR-10a expression was lower than that of IEC, but significantly higher than in LP T cells and LP B cells. Furthermore, miR-10a expression by IECs and DCs was decreased under SPF conditions (Figure 1D). In contrast, the expression of control miR-329 by IECs and DCs of GF mice was at a level comparable to that of SPF mice (Figure 1E). Notably, miR-10a expression of LPDC was much higher than that of splenic DC (Figure 1F). Collectively, these data indicated that intestinal IECs and DCs preferably express miR-10a.
To determine whether commensal bacterial stimulation inhibited DC miR-10a expression, we generated bone marrow-derived DC (BMDC) from B6 mice by culturing BM cells with GM-CSF for 8 days. BMDC expressed miR-10a at a level slightly higher than that of splenic DC but lower than intestinal DC (Supplemental Figure 1). BMDC were stimulated with lysates of E. coli isolated from the intestinal lumen of B6 mice, as well as flagellated A4 commensal bacteria which produce immunodominant commensal antigen CBir1 flagellin(25). miR-10a expression was determined by real-time PCR 24 hrs later. As shown in Figure 2A, BMDC miR-10a expression was downregulated by treatment with E. coli and A4 bacteria. Among many bacterial products, TLR ligands stimulate DC via interaction with TLRs. To determine the roles of TLR ligands in E. coli and A4 bacteria downregulation of DC miR-10a expression, we treated BMDC with various ligands for TLR1/2 (PAM3 CSK), TLR4 (LPS), TLR5 (FliC), TLR9 (CpG ODN), and NOD2 (MDP). miR-10 expression was determined by real-time PCR. As shown in Figure 2B, DC miR-10a expression was downregulated by ligands for TLR1/2, TLR4, TLR5, TLR9, and NOD2. Collectively, these data demonstrated that microbiota-derived TLR ligands inhibited DC miR-10a expression.
It has been shown that the TLR-TLR ligand interacts through MyD88 (26–29). To determine the role of MyD88 pathway in microflora-driven downregulation of miR-10a expression, we determined miR-10a expression by LPDC obtained from wild-type and MyD88−/− B6 mice. As shown in Figure 2C, MyD88 deficiency led to an increase of miR-10a expression by LPDC compared to that in wild-type mice under SPF conditions. Consistently, treatment with lysates of E. coli and A4 bacteria as well as a variety of TLR ligands inhibited wild-type BMDC miR-10a expression, but did not downregulate MyD88−/− BMDC miR-10a expression, demonstrating that the MyD88 is involved in commensal bacterial downregulation of DC miR-10a expression (Figure 2B).
Activation of NFκB has been implicated in many functional aspects of TLR-TLR ligand interaction(26, 27). To determine whether the NFκB pathway is involved in commensal bacteria and TLR ligand-downregulation of DC miR-10a expression, we treated BMDC with LPS in the presence or absence of NFκB inhibitor, Bay 11–7082, for 24 hrs. As shown in Figure 2D, addition of NFκB inhibitor alone increased the base line expression of BMDC miR-10a. Furthermore, a blockade of NFκB activation reversed the inhibition of DC miR-10a expression by LPS. These data indicated that NFκB activation negatively regulates miR-10a expression, and that commensal bacteria and their TLR ligand down-regulation of DC miR-10a expression requires NFκB activation.
MicroRNAs mainly function as negative regulators by binding with their target genes and blocking mRNA transcription or protein expression. A computer-based microRNAs target detection program was used to predict the potential target genes of miR-10a. IL-12/IL-23p40 was predicted as one of the candidate genes of miR-10a by various programs of miRNA target gene prediction. To determine if IL-12/IL-23p40 geneis indeed the target of miR-10a, we constructed dual-luciferase reporter vectors containing the predicted seed sequence in the 3′-UTR of IL-12/IL-23p40, as well as the accordingly mutant vectors in which 3 random nucleotide mutations were introduced into the seed sequences (Figure 3A). The empty vector psiCHECK-2 and the vector containing the theoretical miR-10a seed sequence were used as negative and positive controls, respectively. All of these vectors were used to transfect a murine RAW264.7 macrophage cell line alone or to co-transfect with the synthesized precursor (mimics) or inhibitor of miR-10a, and the Renilla luciferase (Rluc) was normalized to firefly luciferase (Fluc). As shown in Figure 3B, when compared with the vector-alone transfected group, the co-transfection of the miR-10a precursor remarkably repressed the activity of Renilla luciferase containing the seed sequence in the 3′-UTR of IL-12/IL-23p40 (p<0.001). Moreover, a greater than 33% reduction of luciferase activity was observed, but the co-transfection of miR-10a inhibitor up-regulated the Renilla luciferase about 43% in IL-12/IL-23p40 (p<0.05). These data indicated that miR-10a is constitutively expressed at a very low level in the RAW264.7 cell line, and its function can be strengthened by precursors, but weakened by inhibitors. To verify the specific effect of miR-10a on IL-12/IL-23p40, a 3-nucleotide mutation was introduced into the theoretical miR-10a binding site of the vector; no significant difference in Renilla luciferase activity was found when RAW264.7 cells were transfected with the mutated vector alone or co-transfected with miR-10a precursor or inhibitor (Figure 3C). We observed similar results when these vectors were used to transfect HEK293A cells (data not shown). Collectively, these data indicate that miR-10 specifically down-regulates IL-12/IL-23p40 expression.
To determine the role of miR-10a in DC production of IL-12/IL-23p40, we first investigated BMDC expression of IL-12/IL-23p40stimulated by commensal bacteria and their TLR ligands. As previously reported, commensal E coli, PAM3CSK, LPS, and FliC stimulated BMDC IL-12/IL-23p40 expression at various levels (Figure 4A). We then transfected BMDC with miR-10a mimics to obtain DC ectopic expression of miR-10a. BMDCs transfected with control miRNA served as a negative control. At 24 h later, BMDCs were stimulated with 100 ng/ml LPS, and the expression of IL-12/IL-23p40 was analyzed by real-time PCR. The overexpression of miR-10a mimics significantly reduced LPS-induced DC expression of IL-12/IL-23p40 mRNA (Figure 4B). To determine whether miR-10a also inhibited DC production of IL-12/IL-23p40, as well as that of IL-12p70 and IL-23 protein, we measured IL-12/IL-23p40, IL-12p70 and IL-23 in culture supernatants. The ectopic expression of miR-10a inhibited DC production of IL-12/IL-23p40 (Figure 4C) as well as that of IL-12p70 and IL-23 (Figure 4D and E). These data collectively indicated that miR-10a inhibits DC production of IL-12/IL-23p40, thus also inhibiting IL-12p70 and IL-23. Considering the requirement of IL-12/IL-23 in the induction of Th1 and Th17 cell differentiation, these results demonstrate that miR-10a acts as a negative regulator of both innate and adaptive immune responses to microbiota and could play a role in the regulation of intestinal immune homeostasis and the pathogenesis of IBD.
It has been shown that there is more commensal bacteria translocation through the intestinal epithelium in coliticrather than normal mice, and, as a result, the colitic animals are more subject to an enhanced stimulation by the commensal bacteria. To investigate whether miR-10a was differentially expressed in experimental colitis, we assessed the expression of IL-12/IL-23p40 and miR-10a in IL-10-deficient mice which have developed severe colitis. As shown previously, inflamed intestinal tissues of colitic mice produced more IL-12/IL-23p40 compared to those of normal mice (Figures 5A). Such increased intestinal IL-12/IL-23p40 was mainly produced by intestinal LPDC, as intestinal LPDC from colitic mice also produced more IL-12/IL-23p40 compared to that from normal mice when stimulated with A4 bacteria (Figures 5B). Consistently, intestinal LPDC from colitic mice also produced more IL-17p70 and IL-23 (Figure 5C). Interestingly, although miR-10a expression levels were very low in the spleens of both colitic mice and normal mice, intestinal miR-10a expression was further decreased in colitic IL-10−/− mice compared to that in normal mice (Figure 5D). miR-10a expression by IEC and LPDC from colitic IL-10−/− mice was also lower compared to that in wild-type mice (Figure 5E). This effect was not due to a lack of direct effect of IL-10, as IL-10 had no effect on miR-10a expression on DC (Figure 5F).
Multiple levels of regulatory mechanisms control host intestinal immune homeostasis to microbiota(3, 30, 31). A coordinated interplay between commensal microbiota and mucosal immune responses is reciprocally regulated by each other to maintain the host intestinal immune homeostasis. Microbiotaare not only important in shaping the development and function of mucosal immune systems(32–34), but also regulate inflammation, as they are crucial for the induction of inflammatory bowel diseases, as well as experimental arthritis and experimental autoimmune encephalomyelitis(35–38). However, the mechanisms involved in the regulation by microbiota of mucosal immunity are still not well understood. In the present study, we found that microbiotadownregulated mucosal DC miR-10a expression through TLR-TLR ligand interaction in a MyD88-dependent manner. miR-10a inhibited DC production of IL-12/IL-23p40, and thus could control the host innate response to microbiota, thereby contributing to the maintenance of intestinal homeostasis.
Several miRNAs have been implicated in the regulation of the innate immune response to bacterial stimulation. In response to TLR ligands and proinflammatory cytokines, macrophage and DC express high levels of miR-147, miR-21 and miR-9 (18, 20). miR-155 expression is induced in bone marrow-derived macrophages in response to various TLR ligands(39). Certain TLR ligands and the cytoplasmic sensor retinoic acid-inducible gene I (RIG-I) induce miR-146 expression in a N F-κB-dependent manner(40). Interestingly, those miRNAs can negatively regulate the activation of inflammatory pathways in innate cells. miR-146 has been reported to directly inhibit several signaling molecules downstream of the TLRs, including IL-1R-associated kinase 1 (IRAK1), IRAK2 and TNFR-associated factor 6 (TRAF6), all of which are key mediators in inflammation (18, 40). Thus miR-146a suppresses the inflammatory pathway mediated by Toll-like receptors by repressing the translation of the IRAK1(18, 40). On the other hand, a decrease in certain miRNAs expression on innate cells following TLR ligand stimulation has also been reported. Macrophage expression of miR-125b and let-7i is decreased in response to LPS stimulation(41, 42). However, it is still unclear how the microbiota regulates miRNAs, and thereby contributes to the maintenance of intestinal homeostasis. In this report, we have demonstrated that the IEC and LPDC expression of miR-10a was much lower in mice housed under conventional SPF conditions when compared to those maintained in germ-free conditions, and these findings may imply the down-regulation of miR-10a expression by microbiota. Indeed, treatment of BMDC with the commensal bacteria E. coli and A4 bacteria, as well as with various TLR ligands, greatly inhibited miR-10a expression in a MyD88-dependent manner, as MyD88 deficiency abrogated inhibition of miR-10a by microflora and their TLR ligands. Although these data indicates that TLR/MyD88 signaling is sufficient, we still do not know whether this is the only or predominant pathway negatively regulating miR-10a in vivo, and, thus, other pathways may be involved as well. It is also intriguing that the inhibition of the NFκB pathway reversed TLR ligand inhibition of miR-10a (Figure 2). However, it is still not clear whether such regulation is mediated directly by NFκB binding of miR-10a or indirectly through interactions with other pathways, and much more work need to be done to pin down the NFκB pathway assessment of miR-10a expression.
miR-10a has been implicated in the regulation of development and pathogenesis of various tumors. In a recent report, miR-10a expression was shown to promote metastatic behavior of pancreatic tumor cells, and that repression of miR-10a was sufficient to inhibit invasion and metastasis formation. Thus, miR-10a was demonstrated to serve as a key mediator of metastatic behavior in pancreatic cancer. Our data indicated that miR-10a was predominantly expressed in the intestines, and thus could play a key role in the regulation of a host response to a huge microbiota challenge. Interestingly, one of the target genes of miR-10a was IL-12/IL-23p40, as co-expressing miR-10a precursor inhibited IL-12/IL-23p40 expression, whereas co-expressing miR-10a inhibitor promoted IL-12/IL-23p40 expression. IL-12 (p35/p40) and IL-23 (p19/p40) share the same p40 subunit and have emerged as key molecules in the regulation of both innate and adaptive immune responses (43, 44). Many studies have addressed the roles of IL-12 and IL-23 in the regulation of intestinal homeostasis, as well as in the pathogenesis of IBD (45–50). Thus microfloradownregulation of miR-10a allowing for expression of IL-12 and/or IL-23 could set up a “low inflammatory environment” in the intestines, which is important in promoting intestinal immune homeostasis through increasing epithelial barrier function and protective immunity under steady-state. On the other hand, IL-12 and/or IL-23 production stimulated by microflora through inhibition of miR-10a could contribute to the progression of intestinal inflammation under inflammatory conditions. Indeed, incolitic IL-10−/− mice, miR-10a expression was decreased, whereas IL-12 and IL-23 production was increased in the intestines, as well as in IECs and LPDCs compared to levels in normal mice, demonstrating a possible relationship between miR-10a regulation of IL-12 and IL-23 and development of colitis. This may be further indicative of the inhibition of IL-12/IL-23p40 by miR-10 in response to a high level of commensal bacterial stimulation in coliticmice, however, more work is needed to understand its contribution to the pathogenesis of IBD.
In summary, our data demonstrated that microbiota regulate DC expression of IL-12/IL-23p40 through the inhibition of miR-10a. Through regulation of innate cell IL-12/IL-23p40 expression in response to microbiota stimulation, miR-10a mediates the host response to microbiota, and thus could be an important mediator in the maintenance of host immune homeostasis as well as in the pathogenesis of IBD. Manipulation of miR-10a expression could provide a new avenue into therapeutics for IBD.
This work was supported by research grants from NIH DK079918, AI083484, Digestive Diseases Research Development Center (grant DK064400), John Sealy Memorial Endowment Fund, and a start-up fund from The University of Texas Medical Branch.