2.1 Mouse intestinal crypt isolation and organoid culture

All animal procedures were approved by the University of Kentucky Institutional Animal Care and Use Committee. Intestinal crypts were isolated as previously described.²⁰ Briefly, mouse small intestine (SI) was opened longitudinally, washed with ice-cold PBS to remove the luminal contents and cut into 2–4 mm pieces. The intestinal fragments were incubated in ice-cold PBS containing 10 mM EDTA for 60 min at 4°C. Crypts were released by shaking with ice-cold PBS. Washing in ice-cold PBS was repeated until most of the crypts were released, as determined by microscopic analysis. Crypt suspensions were passed through a 70 μm cell strainer and centrifuged at 300 × g for 5 min. Isolated crypts were mixed with Matrigel (Corning, Corning, NY, 356231) and cultured in IntestiCult Organoid Growth Medium (Stemcell Technologies, Vancouver, BC, Catalogue # 06005) as described previously.²² To determine the role of miR-27a-3p mimic in the intestine, mouse SI organoids were infected with lentivirus expressing miR-27a-3p mimic (Sigma, St. Louis, MO, MLMIR0237) or miRNA mimic NC (Sigma, NCLMIR002). The infected organoids were cultured in the presence of puromycin (2 μg/mL) to select for organoid cells with stable overexpression of miR-27a-3p mimic as described.²³

2.2 Human duodenum epithelial cell isolation and organoid culture

Histologically normal human duodenal samples were obtained from patients at the University of Kentucky undergoing pancreatic resection for cancer. Collection of all patient materials for this study was obtained following patient consent and approved by the University of Kentucky Institutional Review Board (IRB # 48678). Samples from the operating room were immediately placed in ice-cold PBS. The intestinal fragments were incubated in ice-cold PBS containing 10 mM EDTA for 60 min at 4°C. After incubation, crypts were released by shaking with ice-cold PBS. Crypt suspensions were collected and passed through a 100 μm cell strainer and centrifuged at 300 × g for 5 min. Isolated crypts were mixed with Matrigel (Corning) and cultured in IntestiCult Organoid Growth Medium (Human, Stem Cell Technologies, Catalogue #100–0340). To determine the role of miR-27a-3p, mouse and human organoids were incubated with either miR-27a-3p miRCURY LNA miRNA inhibitor or miRNA negative control oligos (Qiagen, Germantown, MD).

2.3 Western blot analysis

Western blotting was performed as we have previously described.²⁴ Total protein was resolved on a 10% polyacrylamide gel and transferred to polyvinylidene fluoride membranes. Membranes were incubated for 1 h at room temperature in blotting solution and then incubated in primary antibody diluted in 5% bovine serum albumin solution containing 0.02% sodium azide overnight at 4°C. Antibodies to Cyclin D1 (ab134175) from Abcam, Boston, MA, OLFM4 (#39141), Na,K-ATPase (#3010 CST), CDX2 (#3977, CST), PCNA (sc-56, Santa Cruz, Dallas, TX), VILLIN (sc-7672, Santa Cruz), β-catenin (610,154, BD Transduction Laboratories, Franklin Lakes, NJ), active β-catenin (#8814 CST) and β-actin (A1978, Sigma) were used and following blotting with a horseradish peroxidase-conjugated secondary antibody, protein expression was visualised using an enhanced chemiluminescence (ECL) detection system.

2.4 Quantitative real time RT-PCR analysis

Total RNA was extracted using a RNeasy Mini Kit (Qiagen) and treated with DNase (RQ1, Promega, Madison, WI). Synthesis of cDNA was performed using reagents in the TaqMan Reverse Transcription Reagents Kit (Applied Biosystems, Foster City, CA). TaqMan probe and primers for human and mouse IAP, FABP1, VILLIN, KRT20, Na,K-ATPase, SI, CDX2, LGR5, ASCL2, OLFM4 and GAPDH were purchased from Thermo Fisher Scientific (Waltham, MA). All values were normalised to GAPDH levels. To determine the expression of mature miRNA species, total RNA was extracted using a mirVana miRNA Isolation Kit (Thermo Fisher Scientific). cDNA for miRNA gene expression analysis was synthesised using a TaqMan Advanced miRNA cDNA Synthesis Kit (Thermo Fisher Scientific). hsa-miR-27a-3p Assay ID: 478384 mir, mm-miR-27a-3p Assay ID: mmu478384_mirand U6 snRNA Assay ID: 001973 purchased from Thermo Fisher Scientific. For miRNA expression levels, all values were normalised to U6 levels. Quantitative real time RT-PCR analysis was performed with an Applied Biosystems Prism 7000HT Sequence Detection System using TaqMan universal PCR master mix as we have described previously.^{20, 25}

2.5 IF and IHC staining

Organoids grown in Matrigel were harvested in cold phosphate-buffered saline (PBS) followed by centrifugation at 300 × g for 5 min. Organoids were fixed with 4% formaldehyde at room temperature for 1 h, centrifuged at 300 × g for 5 min to remove the 4% formaldehyde and resuspended in HISTOGEL (Thermo Scientific; HG-4000-012). Paraffin-embedded organoid sections were processed for routine immunofluorescence (IF) or immunohistochemistry (IHC) staining. DAPI (Electron Microscopy Sciences, Hatfield, PA; 17,989–20) was used as a counterstain to detect nuclei. Images were acquired with a confocal microscope (Nikon ECLIPSE Ti2) at 40× magnification. IHC staining was performed as we have described previously.²⁶ Tissues were processed for routine IF or IHC staining using the following antibodies: anti-KRT20 (Abcam; ab854), anti-Na,K-ATPase (Cell Signalling; #3010), anti-VILLIN (Santa Cruz Biotechnology, sc-58,897) and anti-FABP1 (Cell Signalling Technology, Danvers, MA; #13368) and cyclin D1 (Abcam; ab134175). Negative controls (including no primary antibody or isotype-matched mouse immunoglobulin G) were used in each assessment. Cyclin D1-positive cells were identified and quantitated using the HALO image analysis platform. Image contrast was adjusted for visualisation purposes and quantification was always applied on raw, non-adjusted images.

2.6 Alkaline phosphatase activity assay

Organoids grown in Matrigel were harvested in cold PBS followed by centrifugation. Organoid cell lysates were used to determine IAP activity by a commercially available kit (Sigma, AP0100) as we have described.^{21, 27}

2.7 Statistical Analysis

Bar graphs were generated to represent mean ± SD for each cell culture condition. Relative levels of mRNA and IAP activity were calculated based on mean levels in the non-targeting control (NC) group. Statistical tests were performed using two-sample t-test for comparisons between NTC versus miR-27a-3p inhibitor or analysis of variance with contrast statements for pairwise testing of multiple groups of NC, miR-27a-3p+/-XAV mimics. Multiple testing was addressed using adjusted p-values from the Holm's method. Homoscedasticity of variance assumptions for the parametric tests were assessed and appropriate data transformation was employed if indicated. p-values <0.05 were considered statistically significant.

3.1 Repression of miR-27a-3p promotes intestinal enterocyte differentiation in mouse intestinal organoids

Elevated miR-27a expression has been found in the inflamed intestinal mucosa,^{8, 9} however, the role of miR-27a in the maintenance of intestinal homeostasis is not known. We have determined the expression pattern of miR-27a-3p in mouse intestine. A gradient increase in miR-27a-3p levels was found in the intestinal crypts isolated from duodenum, jejunum and ileum (Figure S1).To determine the role of miR-27a-3p in intestinal IECs, mouse small intestinal (SI) organoids were incubated with inhibitor NC oligos or miR-27a-3p inhibitor to repress endogenous miR-27a-3p and intestinal differentiation was assessed. Incubation of IECs with miR-27a-3p inhibitor repressed the endogenous miR-27a-3p expression as expected (Figure 1A). Importantly, treatment with the miR-27 inhibitor significantly increased intestinal enterocyte differentiation as noted by increased IAP activity (Figure 1B) and mRNA expression of enterocyte markers IAP, VILLIN, FABP1 and sucrase-isomaltase (Figure 1C). The increased protein expression of VILLIN and other enterocyte markers (e.g., Na,K-ATPase and CDX2) was demonstrated by western blotting (Figure 1D) and IF staining (Figure 1E). These results show that repression of miR-27a-3p promotes intestinal enterocyte differentiation, which suggests an important role of miR-27a-3p in the maintenance of intestinal homeostasis.

Figure S1 Expression pattern of miR-27a-3p in mouse small intestinal mucosa. Mouse small intestinal crypts were isolated and total miRNA was extracted. Expression of miR-27a-3p was determined by qPCR. Expression was normalised to U6 expression. n = 4 mice.

3.2 miR-27a-3p repression decreases stem cell proliferation in mouse intestinal organoids

To determine the effects of miR-27a-3p on intestinal stem cell proliferation, mouse SI organoids were incubated with miR-27a-3p inhibitor. As shown in Figure 2, miR-27a-3p inhibitor repressed organoid growth as noted by decreased organoid formation (Figure 2A,B). Moreover, miR-27a-3p repression decreased the expression of stem cell markers LGR5, ASCL2 and OLFM4 mRNA (Figure 2C). Decreased expression of OLFM4 and proliferating marker PCNA was demonstrated by western blot (Figure 2D). These results demonstrate that miR-27a-3p is required for the maintenance of stem cell proliferation.

3.3 Repression of miR-27a-3p promotes enterocyte differentiation and inhibits stem cell proliferation in human duodenum organoids

We have shown an important role of miR-27a-3p in mouse IEC proliferation and differentiation. To determine whether this occurs in human IECs, we incubated human duodenum organoids with the miR-27a-3p inhibitor. Treatment with miR-27a-3p inhibitor significantly repressed endogenous miR-27a-3p expression in human IECs as expected (Figure 3A). Concomitantly, miR-27a-3p inhibitor increased enterocyte maturation as noted by the increased expression of enterocyte markers FABP1, IAP and KRT20 mRNA (Figure 3B). Increased KRT20 and FABP1 protein expression was also demonstrated by IF analysis (Figure 3C). Similar to mouse IECs, miR-27a-3p inhibitor repressed organoid growth as noted by decreased organoid formation (Figure 3D,E). Moreover, miR-27a-3p repression resulted in decreased expression of intestinal stem cell markers LGR5 and ASCL2 detected by qPCR (Figure 3F) and decreased protein expression of proliferating marker PCNA determined by western blotting (Figure 3G). These data demonstrate a general role for miR-27a-3p in the regulation of proliferation and differentiation in human and mouse IECs.

3.4 miR-27a-3p mimic inhibits enterocytes differentiation in mouse organoids

We have shown the effects of miR-27a-3p inhibitor on IECs. To further demonstrate the role of miR-27a-3p, we used miR-27a-3p mimics. Mouse SI organoids were infected with lentivirus vector expressing miR-27a-3p mimic or mimic NC oligo and selected with puromycin. Lentivirus vector-mediated miR-27a-3p mimic overexpression was confirmed (Figure 4A). In contrast to miR-27a-3p repression, overexpression of miR-27a-3p increased numbers of organoids with round morphology (Figure 4B), indicating an increased stemness.²⁸ In addition, overexpression of miR-27a-3p repressed enterocyte differentiation as noted by decreased IAP activity (Figure 4C) and decreased expression of enterocyte markers as determined by qPCR (Figure 4D), western blotting (Figure 4E) and IF staining (Figure 4F). Together, our findings further confirm an important role of miR-27a-3p in the regulation of IEC proliferation and differentiation.

3.5 miR-27a-3p regulates RXRα/β-catenin signalling

miR-27a-3p promotes Wnt/β-catenin signalling pathway in mouse astrocytes.¹¹ To determine whether miR-27a-3p regulates intestinal cell proliferation and differentiation through the Wnt/β-catenin pathway, we first determined the effects of miR-27a-3p on Wnt signalling in normal IECs. Treatment with the miR-27a-3p inhibitor repressed Wnt signalling as shown by the decreased expression of Wnt target genes AXIN2 and cyclin D1 (Figure 5A) and decreased active (non-phosphorylated) β-catenin²⁹ (Figure 5B). miR-27a-3p repression also decreased cyclin D1 positive cells as demonstrated by IHC staining followed by HALO analysis (Figure 5C). miR-27a-3p repression in human IECs decreased the expression of AXIN2 and cyclin D1 mRNA (Figure 5D), decreased the expression of cyclin D1 and β-catenin protein (Figure 5E) and decreased cyclin D1 positive cells (Figure 5F). Consistently, overexpression of the miR-27a-3p mimic increased cyclin D1 mRNA expression (Figure 5G) accompanied by the increased total and active β-catenin and cyclin D1 protein expression (Figure 5H) and cyclin D1 positive cell numbers (Figure 5I) in mouse SI organoids. Together, these data demonstrate that miR-27a-3p activates Wnt/β-catenin in both mouse and human IECs.

miR-27a-3p has been shown to repress the expression of GSK3β,³⁰ adenomatous polyposis coli (APC),³¹ and RXRα¹² in certain cell types, which inactivate Wnt/β-catenin signalling.^{12, 32, 33} To determine how miR-27a-3p impacts Wnt/β-catenin signalling, the effects of miR-27a-3p on the expression of GSK3β, APC and RXRα were determined. Treatment with miR-27a-3p inhibitor increased RXRα expression without affecting the expression of GSK3β and APC in mouse SI organoids (Figure 5J). In contrast, overexpression of the miR-27a-3p mimic significantly decreased RXRα expression in mouse IECs (Figure 5K) suggesting that miR-27a-3p activates Wnt/β-catenin signalling through the regulation of RXRα expression in IECs.

3.6 miR-27a-3p regulates intestinal enterocyte differentiation through the Wnt/β-catenin pathway

Wnt/β-catenin pathway plays a crucial role in IEC proliferation and differentiation; therefore, we next determined whether miR-27a-3p regulated IEC proliferation and differentiation through the regulation of Wnt/β-catenin signalling. Mouse SI organoids infected with lentivirus expressing miR-27a-3p mimic or miRNA mimic NC were incubated with or without XAV939, which inhibits Wnt signalling.³⁴ As shown in Figure 6, overexpression of the miR-27a-3p mimic decreased enterocyte differentiation, as shown by decreased IAP activity (Figure 6A) and expression of IAP, FABP1, Na,K-ATPase and CDX2 (Figure 6B). These decreases were attenuated or blocked by combination treatment with XAV939. In contrast, treatment with the miR-27a-3p mimic increased the expression of cyclin D1 mRNA (Figure 6C) and protein (Figure 6D) and decreased VILLIN protein and increased active β-catenin expression (Figure 6D). These alterations were blocked by treatment with XAV939. Together, our results demonstrate that miR-27a-3p regulates IEC proliferation and differentiation through the regulation of Wnt/β-catenin signalling (Figure 6E).