|Home | About | Journals | Submit | Contact Us | Français|
The intestine exhibits profound diurnal rhythms in function and morphology, in part due to changes in enterocyte proliferation. The regulatory mechanisms behind these rhythms remain largely unknown. We hypothesized that microRNAs are involved in mediating these rhythms, and studied the role of microRNAs specifically in modulating intestinal proliferation.
Diurnal rhythmicity of microRNAs in rat jejunum was analyzed by microarrays and validated by qPCR. Temporal expression of diurnally rhythmic mir-16 was further quantified in intestinal crypts, villi, and smooth muscle using laser capture microdissection and qPCR. Morphological changes in rat jejunum were assessed by histology and proliferation by immunostaining for bromodeoxyuridine. In IEC-6 cells stably overexpressing mir-16, proliferation was assessed by cell counting and MTS assay, cell cycle progression and apoptosis by flow cytometry, and cell cycle gene expression by qPCR and immunoblotting.
mir-16 peaked 6 hours after light onset (HALO 6) with diurnal changes restricted to crypts. Crypt depth and villus height peaked at HALO 13-14 in antiphase to mir-16. Overexpression of mir-16 in IEC-6 cells suppressed specific G1/S regulators (Cyclins D1-3, Cyclin E1 and Cyclin-dependent kinase 6) and produced G1 arrest. Protein expression of these genes exhibited diurnal rhythmicity in rat jejunum, peaking between HALO 11-17 in antiphase to mir-16.
This is the first report of circadian rhythmicity of specific microRNAs in rat jejunum. Our data provide a link between anti-proliferative mir-16 and the intestinal proliferation rhythm and point to mir-16 as an important regulator of proliferation in jejunal crypts. This function may be essential to match proliferation and absorptive capacity with nutrient availability.
Circadian rhythms (24-h oscillations) play a key role in the regulation of numerous physiological functions. Circadian rhythmicity of up to 10% of gene transcripts and an even greater fraction of proteins indicate the involvement of both transcriptional and translational pathways.[1–5] Regulation at both the transcriptional and post-transcriptional level suggests a role for microRNAs in this process. MicroRNAs are non-coding RNAs able to silence numerous genes simultaneously. Bioinformatics analysis suggests that up to 30% of mammalian gene transcripts are regulated by microRNAs, short non-coding RNAs.[6–9] microRNAs suppress protein expression following recognition of complementary sequences on the 3′UTR (untranslated region) of target genes, either by inducing mRNA cleavage (which manifests as changes in mRNA levels) or inhibiting translation (manifesting as changes in protein levels).[10–12] Presence of the target sequence for each microRNA on multiple genes permits simultaneous regulation of protein expression from numerous genes by a single microRNA.[6, 13, 14] The postulated role of microRNAs in “fine-tuning” gene expression suggests they also contribute to coordinating the circadian rhythmicity of many genes and proteins.[15–18]
The intestine displays profound rhythmicity of morphology, resulting in peak absorptive function (e.g. for glucose) coinciding with maximal nutrient delivery to the bowel.[19, 20] The number of enterocytes per villus also exhibits a diurnal rhythmicity, with an increase about the time of maximal nutrient availability. Similar rhythmicity has been reported in human gastrointestinal mucosa.[22, 23] The exact pathways coordinating rhythmicity in proliferation are presently unknown.
We hypothesize that microRNAs are integral components for mediating circadian rhythms in intestinal proliferation, morphology, and function. To investigate this, we profiled microRNAs in the intestine of ad libitum fed rats using oligonucleotide arrays. The anti-proliferative microRNA mir-16 was expressed in both crypt and villus enterocytes but exhibited circadian rhythmicity only in the crypts. The cell cycle regulators Ccnd1, Ccnd2, Ccnd3, Ccne1, and Cdk6 also exhibited circadian rhythmicity but in antiphase to mir-16. An anti-proliferative role for mir-16 was supported by its ability to inhibit proliferation and decrease expression of genes involved in cell cycle regulation when over-expressed in rat IEC-6 cells. These studies point to mir-16 as a potentially important microRNA in regulating circadian rhythms in the intestine.
All animal study protocols were prospectively approved by the Harvard Medical Area Standing Committee on Animals.
Sprague-Dawley rats (50 males, 7 weeks old) were purchased from Harlan World (Indianapolis, IN) and acclimatized to a 12:12-h light: dark photoperiod for 5 days with ad libitum access to food and water. Time is designated as Hours After Light Onset (HALO), with HALO 0 at 7 am (lights on). Rats were injected with BrdU (5-bromo-2-deoxyuridine, 50mg/kg; Sigma, St Louis, MO) 1 h before harvest to label DNA as an index of S-phase. Rats were killed at 3-h intervals over 24 h (n=6–7 per time) and jejunum harvested for microRNA microarrays, RNA and protein determination, and morphological analysis (harvest protocol detailed in Supplementary Material).
Total RNA from jejunum was extracted using the mirVana kit (Ambion; Austin, TX) and profiled on in situ hybridization arrays (Exiqon, Woburn, MA) against a reference sample consisting of RNA pooled from HALO 0 rats. Dye swaps were incorporated in the arrays to correct for any dye bias. Data were subjected to Lowess normalization and log transformed.
Expression profiles of selected microRNAs were confirmed by real-time PCR. Specific microRNAs were selected from total extracted RNA by reverse transcription using the stem-loop hybridization based microRNA reverse transcription kit and microRNA-specific primers (Taqman microRNA reverse transcription kit and Taqman microRNA assays, Applied Biosystems, Foster City, CA). microRNA expression was quantified in triplicate using the Taqman microRNA PCR primers and Taqman gene expression mastermix (Applied Biosystems). Reverse transcription and PCR were performed simultaneously on all samples to minimize differences introduced by variable reaction efficiency.
The human mir-16 gene was amplified from human genomic DNA by PCR and inserted into the MluI/ClaI sites of the tetracycline-inducible TRIPZ shRNAmir expression vector (Open Biosystems, Huntsville, AL) using restriction sites incorporated into the primers (Supplemental Table 1). A non-silencing TRIPZ inducible shRNAmir vector was used as a control (Open Biosystems). Vectors were sequenced to ensure fidelity of the microRNA sequence and insertion. Details of cell transfection are available in Supplementary Material.
IEC-6 cells were seeded in 96-well plates at a density of 1000 cells per well in triplicate. Proliferation indices were measured 48 h later using the CellTiter96 Aqueous One Solution Cell Proliferation Assay (MTS assay, Promega, Madison, WI). Cell growth rates were confirmed by cell counting in trypsinized, 48-h cultures seeded in triplicate at 104 cells/ml in 6-well dishes. All experiments were performed thrice.
For cell cycle analysis, trypsinized cells were counted and fixed overnight in 70% ethanol at −20°C. Fixed cells were collected by centrifugation at 1200rpm for 10 min at 4°C, suspended in propidium iodide (BD Biosciences, San Jose, CA) for 30mins at 37°C in darkness, and analyzed by flow cytometry (BD FACScan, BD, Franklin Lakes, NJ). Data were analyzed by ModFit (Verity, Topsham, ME). To determine apoptosis and viability, trypsinized cells were counted and stained with Annexin V-FITC (BD) and Sytox Blue (Invitrogen), respectively, and analyzed by flow cytometry (10,000 events per sample). Data were analyzed using Diva (BD).
mRNA levels of Ccnd1, Ccnd2, Ccnd3, Ccne1, Cdk4 and Cdk6 were quantified by real-time PCR as previously described (detailed in Supplementary Material) and expressed relative to B-actin. All genes had Cts within the same range, between Ct 22 and 27. Primers were custom-ordered from Invitrogen (Supplemental Table 2), with the exception of Ccnd1 mRNA which was measured using the Taqman primer-probe and gene expression Master Mix (Applied Biosystems).
Protein expression of Ccnd1, Ccnd2, Ccnd3, Ccne1, Cdk4 and Cdk6 was measured in total lysates from jejunal mucosal scrapings or IEC-6 cell lysates as previously described, and detailed in Supplementary Material.
Sections of jejunum were fixed overnight in 10% formalin, then orientated and embedded in paraffin blocks, cut at 7μm thickness, mounted and stained with haematoxylin and eosin. Crypt depth, villus height, villus width, crypt enterocyte width, villus enterocyte width, and number of enterocytes per crypt were measured by a blinded observer under light microscopy (Olympus BX50; Center Valley, PA) at 100x or 400x magnification. Only samples displaying a single layer of enterocytes and villi with a visible central lacteal were included in the analysis (5–10 measurements per rat). For measurement of rhythmicity of proliferation, blocks of jejunum were cut at 7μm and sections incubated with anti-BrdU primary antibody (Sigma), biotinylated secondary antibody, and visualized using the avidin-biotin-peroxidase complex method with diaminobenzidine tetrahydrochloride as the chromogen. Sections were counterstained with haematoxylin and eosin to facilitate counting of BrdU-negative nuclei.
Sections of jejunum from rats killed at HALO 6 and HALO 18, the respective circadian peak and trough of mir-16 expression, were embedded in OCT compound over dry ice and isopentane. Sections (9μm) were cut from the fresh frozen specimens and stained with Histogene staining solution (Molecular Devices, Sunnyvale, CA). Crypts (all cells in the lower half), villi (all cells in the top half), or smooth muscle were isolated by laser capture microdissection (Veritas Microdissection System, Molecular Devices). Total RNA was extracted from each section (RNAqueous RNA extraction kit, Ambion, Austin, TX) and subjected to microRNA reverse transcription and real-time PCR as described above for quantification of mir-16 expression in each fraction.
Data are presented as means ± SE. Graphical analysis was performed using GraphPad Prism (San Diego, CA). microRNAs exhibiting a 2-fold or greater difference between any two timepoints were selected for further analysis, and a false discovery rate (q-value) of <0.05 was considered significant. Circadian rhythmicity of microRNAs, gene and protein expression and morphological changes in rat tissue was determined by cross-sectional analysis and assuming a 24-h period as described previously, using the Cosinor procedure which is freely available online. [25–27] The acrophase (time of peak expression), mesor (rhythm-adjusted mean), amplitude of rhythmicity, and significance of fit to a 24-h period (as indicated by p<0.05) for each gene was abstracted from the program. ANOVA (analysis of variance) with post-hoc Tukey’s multiple comparisons test was used to identify significant differences across the 3 intestinal fractions at each timepoint. T-tests were used to compare the effects of mir-16 overexpression with control cells in the in vitro experiments.
Of 238 microRNAs tested on in situ hybridization arrays, 13 microRNAs exhibited ≥ 2-fold difference between peak and trough values (range 2.0- to 3.4-fold; q<0.05), 8 of which are conserved among human, mouse and rat and were therefore selected for further evaluation. Real-time PCR (qPCR) confirmed circadian rhythmicity for mir-16, mir-20a and mir-141 as determined by the cosinor procedure, with a 24-hour periodicity. Peak expression of these three microRNAs occurred between HALO 4 and 6, corresponding to the lights-on fasting period (Supplementary Table 3, Figure 1A–C). Two of these are reportedly involved in proliferation: mir-20a is pro-proliferative and mir-16 is anti-proliferative.[28–34] Intestinal villus height and cell number have been shown to peak in anticipation of maximal nutrient intake in previous studies. Because anti-proliferative mir-16 began to wane late in the light phase, when intestinal proliferation has been shown to increase, we selected this microRNA for further study and designed experiments to ascertain its role in the rhythm of intestinal proliferation.
To compare mir-16 expression levels in crypt, villus and smooth muscle, these cell types were isolated by laser capture microdissection at HALO 6 and 18, the respective mir-16 peak and nadir. At HALO 18, expression was not significantly different across all three cell types (Fig. 1D; p=0.97). However, mir-16 expression was 3.2-fold higher in crypts at HALO 6 vs. HALO 18 (Fig. 1D; p=0.003) while it was not detectably different in villi or smooth muscle. Thus, mir-16 rhythmicity appears restricted to crypts, the proliferative compartment of the intestinal mucosa.
To determine the effect of mir-16 on enterocyte proliferation, mir-16 was overexpressed in rat IEC-6 cells, a cell line derived from intestinal crypts. Stable transfection of IEC-6 cells with the mir-16 expression vector led to a 2.1-fold increase in mir-16 expression vs. the control (p=0.03, Figure 2A). This modest difference, comparable to the peak/trough difference observed in mir-16 expression on a diurnal basis, had a profound effect on cell proliferation. At 48 h after plating, the proliferation rate was decreased 76% vs. control cells as measured by the MTS assay (4.2-fold difference; p=0.006, Fig 2B) and by 80% as measured by cell counts (5.2-fold difference; p=0.02, Fig 2C). Over-expression of mir-16 also led a significantly larger fraction of cells in G1 compared to control as revealed by flow cytometry (65% vs. 24%, p<0.001, Fig 2D). This result indicates that proliferation was curbed by arresting enterocytes in G1 rather than the reported effect of mir-16 on apoptosis.[31, 36] The lack of increase in apoptosis in IEC-6 cells over-expressing mir-16 (p=0.63, Figure 2E) substantiates this conclusion. These results point to an effect of mir-16 on the cell cycle in enterocytes, specifically regulators of the G1/S transition.
To identify specific mir-16 targets involved in reducing proliferation in enterocytes, the microRNA target prediction algorithm Targetscan was interrogated for the presence of mir-16 binding sequences in the 3′UTRs of G1/S regulatory genes. Potential mir-16 targets in both rat and human included Cyclin D1 (Ccnd1), cyclin D2 (Ccnd2), cyclin D3 (Ccnd3), cyclin E (Ccne1) and cyclin-dependent kinase 6 (Cdk6). These are all known to regulate the G1/S transition and were therefore examined for responsiveness to mir-16. Cyclin-dependent kinase 4 (Cdk4), a G1 regulator lacking a mir-16 target site in its mRNA 3′UTR, was included as a negative control.
Overexpression of mir-16 significantly decreased protein levels of Ccnd1, Ccnd2, Ccnd3, Ccne1 and Cdk6 in IEC-6 cells compared to the non-silencing control (levels 0.3–0.5 that of controls, p<0.05, Table 1, Supplemental Figure 1). mir-16 appeared to affect translation of Ccnd1, Ccnd3 and Ccne1 rather than mRNA cleavage because mRNA levels did not change detectably (p=0.660, 0.151 and 0.181 respectively, Table 1). In contrast, reduction of Ccnd2 and Cdk6 mRNAs by 75% (p=0.002) and 58% (p=0.001), respectively (Table 1) indicated that mir-16 overexpression primarily affected transcription and/or mRNA stability of these regulators. Our data point to one or more of these G1/S proteins as mir-16-regulated mediators on cell cycle progression. As expected, neither Cdk4 mRNA (p=0.591) or protein (p=0.223) levels were altered detectably by mir-16 overexpression (Table 1). These results confirm that Cdk4 is not a mir-16 target and indicate that mir-16 overexpression does not exert non-specific effects on cell cycle proteins.
Diurnal rhythmicity in intestinal proliferation is likely to be mediated by an underlying diurnal rhythmicity in cell cycle proteins.[21, 38] Moreover, involvement of mir-16 in the jejunal mucosa cell cycle via suppression of these proteins as suggested by the IEC-6 studies would likely be evidenced by a corresponding displacement of their rhythms from mir-16. To these ends, we examined the temporal protein expression patterns for the 5 mir-16 targets as well as Cdk4 in jejunum. All six proteins exhibited diurnal rhythmicity with a 24-hour period, with acrophases (expression peaks) falling between HALO 11 and HALO 17 (p<0.05, Figure 3A–E; Table 2, Supplemental Figure 2) and nadirs between HALO 3 and 6. These temporal patterns would be expected for targets suppressed by mir-16 with its peak expression at HALO 6. Ccnd2, Ccnd3 and Cdk4 displayed rhythmicity at the transcriptional level (p=0.011, 0.00018, and 0.00015, respectively; Figure 4B–C, Table 2). Ccnd1 and Ccne1 mRNAs exhibited temporal changes but these did not qualify as significant circadian rhythms, in keeping with the lack of response at an mRNA level with mir-16 overexpression in vitro. In contrast, Cdk6 did not display diurnal rhythmicity of transcription in vivo (p=0.77, Figure 4F) despite its transcriptional responsiveness to mir-16 overexpression in IEC-6 cells.
To define the relationship of proliferation to the cyclin expression rhythm, we assessed the temporal patterns of DNA synthesis (S-phase) and crypt-villus morphology. The number of cells in S-phase, as measured by BrdU labeling, peaked at HALO 5 (p<0.001, Supplemental Figure 3A). Crypt cell number peaked several hours later at HALO 12 (p=0.001, Supplemental Figure 3B), followed by crypt depth and villus height at HALO 13 and HALO 14, respectively (p=0.005 and 0.043, Supplemental Figure 3C and D). Enterocyte number per 100μm of villus increased modestly in anticipation of nutrient arrival but significant rhythmicity was not achieved (p=0.099, Supplemental Figure 3E). Cell width exhibited circadian rhythmicity in crypts with a peak at HALO 15 (p=0.033, Supplemental Figure 3F) but not in villi (p=0.217). Overall these data demonstrate that a combination of cell proliferation and hypertrophy produced the observed changes in crypt and villus morphology (Table 3).
This study is the first to profile microRNA expression in rat jejunum as well as to establish rhythmic expression of specific microRNAs. In particular, our data supports a role for the anti-proliferative microRNA mir-16 in the intestinal proliferation rhythm. In support of this, we have shown that mir-16 expression peaks at HALO 6, coincident with the troughs in villus height and in crypt depth and cell number. mir-16 rhythmicity was also restricted to intestinal crypts, the primary site of proliferation. The anti-proliferative effect of mir16 was confirmed in vitro, where mir-16 inhibited proliferation of IEC-6 enterocytes, and suppressed expression of 5 key G1/S regulators—Ccnd1, Ccnd2, Ccnd3, Ccne1 and Cdk6. Finally, protein abundances of all five G1/S regulators presumably targeted by mir-16 as well as the non-target Cdk4 exhibit diurnal rhythmicity in rat jejunum in antiphase to mir-16. These coordinated responses point to mir-16 as an important regulator of proliferation in jejunal crypts. This function may be essential to coordinate intestinal circadian rhythms, serving to optimally match proliferation and absorptive capacity with nutrient availability.
Circadian rhythmicity of microRNA expression has been shown to regulate cell behavior and gene expression. In the suprachiasmatic nucleus, rhythmic expression of mir-219 and mir-132 mediate photic entrainment of circadian clock activity. Similarly, depletion of mir-122 in liver disrupted the circadian rhythmicity of numerous transcripts regulating metabolism.[18, 39] In the retina, 12 microRNAs display circadian rhythmicity of which two – mir-96 and mir-182 – were shown to mediate rhythmic expression of the Adcy6 (adenylate cyclase type 6) gene. Here we highlight another potential role for microRNAs as regulators of intestinal circadian rhythms. Interestingly, the 1.8- to 3.2-fold amplitude changes we observed in intestinal microRNAs are consistent with the 1.25- to 3-fold changes observed in the retina.[15–17] Three microRNAs, mir-16, mir-20a and mir-141 were shown to exhibit circadian rhythmicity in this study, however the limited amount of tissue obtained from laser capture microdissection restricted us to the examination of only mir-16 expression at HALO 6 and 18. Further studies are necessary to determine the rhythmicity of the remaining microRNAs in the individual intestinal fractions at circadian timepoints, particularly for mir-20a which is known to have a pro-proliferative function and may therefore contribute to the regulation of rhythmicity of intestinal proliferation.
Several observations from our studies merit further discussion. First, a modest increase of mir-16 in IEC-6 cells, similar to the diurnal change in jejunum, almost completely arrested growth in these cells. mir-16 has been suggested to act as a tumour suppressor gene in prostate: mir16 is frequently downregulated in advanced prostate cancer and mir16 knockdown in prostate cancer cells promotes proliferation and invasiveness. Similarly, mir-16 expression is reduced in squamous cell carcinomas and adenocarcinomas of the lung, and mir-16 overexpression in lung cancer cell lines induces cell cycle arrest. Our findings reveal that the anti-proliferative function of mir-16 serves an important physiological role in normal tissues. We note that, in contrast to its lack of effect on IEC-6 cell apoptosis, mir-16 was shown to increase apoptosis in leukaemic cell lines, gastric cancer cells and prostate cancer via downregulation of pro-survival protein BCL2 (B-cell lymphoma 2).[31, 36, 40, 41] This apparent discrepancy in our observations, may in fact be due to different properties of BCL2 pathways in the small intestine; while Bcl2 is expressed in enterocytes, it may perform different functions in this tissue. Indeed, ablation of Bcl2 in mice increases the apoptosis rate in the colon but not the small intestine.[42, 43]
Second, in IEC-6 enterocytes mir-16 suppressed levels of several cell cycle proteins involved in the G1/S transition concomitantly with G1 arrest. In normal cell cycle progression, D-type cyclins (Ccnd1, Ccnd2 and Ccnd3) complex with cyclin-dependent kinases (Cdk4 and Cdk6) during G1 to phosphorylate and thereby inactivate the retinoblastoma protein pRb, in turn activating cell cycle proteins (including Ccne1 and its complex Cdk2) vital for entering S-phase.[44, 45] Upregulation of mir-16 expression suppressed expression of Ccnd1, Ccnd2, Ccnd3, Ccne1 and Cdk6 in vitro, thereby corroborating existing evidence that small changes in microRNA expression alter cellular phenotypes by downregulating multiple components of single pathways.[13, 46, 47] In vivo, we found that G1 proteins Ccnd1 and Ccnd2 peaked at HALO 12, while the remaining D-type cyclin family member Ccnd3 peaked later at HALO 17. These findings are consistent with reported differences in the relative timing of D cyclins in various cell types, as well as differential regulation and a degree of functional redundancy.[48, 49] We were unable to definitively corroborate rhythms of mir-16 in the crypt with rhythms of cell cycle proteins in the crypt due to the small amount of tissue obtained from laser capture microdissection, however previous studies have demonstrated that in the intestine the D-type cyclins and cyclin-dependent kinases are most strongly expressed in intestinal crypts. Our study showed peak S-phase at HALO 5, indicating a G1/S duration of approximately 12 to 17 hours, in agreement with previous studies showing a long G1/S and short G2/M period in the small intestine.[51–53] The 63% change in cell labeling we observed at HALO 6 vs. HALO 15 is also similar to the 30–60% increase at HALO 3 in murine jejunum reported by Scheving et al.[38, 54] The rhythmicity in proliferation translated to rhythmicity in morphological parameters in the jejunum. The large number of crypts and villi across the length of the intestine suggests that these small changes are likely to result in a large change in absorptive surface area over the diurnal period. Examination of these morphological parameters in the terminal ileum and corroboration of these measurements with mir-16 expression in the ileum may reveal new insights into the regulation of mir-16.
Our data show that mir-16 is able to affect translation of Ccnd1, Ccnd3 and Ccne1 without affecting mRNA expression, corroborating previous data showing microRNAs are able to suppress protein levels independent of mRNA expression. This was also demonstrated by our data in vivo; Ccnd1 and Ccne1 showed rhythmicity only at the protein level. This is in keeping with previous data showing that almost half of the proteins demonstrating circadian rhythmicity in the mouse liver lack a corresponding cycling transcript. Together with our findings this suggests the possibility that the rhythmic protein expression in jejunum in our study may be produced solely by miRNAs, whether by mir-16 alone or in combination with others. Cell-type specificity of mir-16 rhythmicity, such as seen in the intestinal crypts in our study, would then lead to consequent rhythmicity of target proteins. Cell cycle proteins are known to have a relatively short half-life, which is likely to facilitate regulation of these proteins by rhythmicity in microRNA expression and allow increased responsiveness to other stimuli that may accelerate or arrest the cell cycle.
Regulation of gene expression by microRNAs is a complex process, with the potential for each to target many related or unrelated genes and for responsive genes to be regulated by multiple microRNAs. In the case of the cell cycle, microRNAs let-7a, mir-34a, mir-192 and mir-215 have been shown, like mir-16, to arrest cells in G1, while mir-106b and mir-221 accelerate G1/S progression by suppressing the cyclin-dependent kinase inhibitors p21 and p27, respectively.[56–61] Factors other than microRNAs are also clearly important in cuing the intestinal proliferation rhythm. For instance, clock gene Period 2 (Per2) regulates proliferation in peripheral tissues via cell cycle genes c-Myc, Cyclin A, Mdm-2 and Gadd45α, as well as the mir-16 target Ccnd1. Ultimately, proliferation rhythms likely result from combined inputs of circadian clock components, other transcription factors and rhythmic microRNAs. The ability of non-microRNA transcriptional regulators such as clock genes to regulate rhythmicity of proliferation may explain rhythmicity in Cdk4, a cell cycle gene not regulated by mir-16, and the lack of transcriptional rhythmicity in Cdk6 in vivo despite responsiveness to mir-16 overexpression in vitro. Generation of knockout mice lacking mir-16 will be invaluable in defining its functions and dissecting these regulatory pathways.
Finally, a broader implication can be drawn from our study. The behavior of mir-16 reveals another potential route for linking proliferation to nutrient availability, which cues the intestinal rhythms. Rhythmic mir-16 expression in crypt cells could be initiated by luminal nutrients directly or via neuro-hormonal pathways. In either case, proliferation may be a key early component to expand the mucosal surface area in the anticipatory diurnal increases in absorptive capacities for glucose, peptides, and other nutrients.[19, 63]
In summary, we show for the first time rhythmicity of microRNA expression in the intestine, and anti-proliferative effects of the diurnally expressed mir-16 in untransformed enterocytes in vitro. We hypothesize that rhythmicity of mir-16 in jejunum may act to mediate the rhythmicity in intestinal proliferation and coordinate the proliferative response with nutrient availability to optimize intestinal absorption and function.
The authors gratefully acknowledge the excellent technical assistance of Jan Rounds and Roger Lis in the experimental procedures.
Grant support: NIH - 5 R01 DK047326 (SWA), ADA - 7-05-RA-121 (DBR), Harvard Clinical Nutrition Research Center - P30-DK040561 (AT), Nutricia Research Foundation (AB) and Berkeley Fellowship (ATS).
Disclosures: All authors confirm no conflicts of interest.
Copyright licence statement: The data in this manuscript was presented as a “Poster of Distinction” at Digestive Diseases Week 2009, as a platform presentation at Gastro 2009, and published in abstract form only in the supplementary issue of Gastroenterology and Gut (less than 400 words). The authors attest that this manuscript is not currently under consideration by any other journal.
Transcript profiling: The raw microarray data have been submitted to GEO with the accession number: GSE20603 This is a GEO-generate link that the reviewers can use to access the data: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=txgtdamsseqmufo&acc=GSE20603
Author contributions to manuscript: AB, SWA, DBR and AT designed the studies; AB and ATS conducted the research; AB, PJP and JMD analyzed the data; AB, DBR and AT wrote the paper. All authors read and approved the final manuscript.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.