miR-125a-5p: a novel regulator of SLC26A6 expression in intestinal epithelial cells

Abstract

Putative anion transporter 1 (PAT1, SLC26A6), an intestinal epithelial Cl⁻/${\text{HCO}}_3^{-}$ exchanger, also plays a key role in oxalate homeostasis via mediating intestinal oxalate secretion. Indeed, Slc26a6-null mice showed defect in intestinal oxalate secretion and high incidence of kidney stones. Recent emergence of PAT-1 as a novel therapeutic target for nephrolithiasis warrants detailed understanding of the mechanisms of PAT-1 regulation in health and disease. Therefore, we investigated the regulation of PAT-1 expression by microRNAs (miRNA), as they have been shown to play key role in modulating expression of other ion transporters. In silico analysis of PAT-1 3′-untranslated region (UTR) revealed potential binding sites for several miRNAs, suggesting the role of miRNAs in modulating PAT1 expression. miRNAs showing highest context scores (125a-5p, 339-5p, 423-5p, 485-5p, and 501-3p) were selected as candidates for their effects on the activity of a 263-bp PAT-1 3′-untranslated region (UTR) fragment cloned into pmirGLO vector upstream of luciferase. The 3′-UTR activity was measured by dual luciferase reporter assay in Caco-2, T-84, HT-29, and SK-CO15 cells. Transient transfection of PAT-1 3′-UTR significantly decreased the relative luciferase activity compared with the empty vector suggesting binding of potential miRNA(s) to the PAT-1 3′-UTR. Among all the selected candidates, cotransfection with miRNA mimics 125a-5p and 423-5p further decreased PAT-1 3′-UTR activity. Furthermore, increasing miR-125a-5p abundance via mimic transfection in Caco-2 cells decreased both mRNA and protein levels of PAT-1. Our results demonstrate a novel regulatory mechanism of intestinal PAT-1 expression via miR-125a-5p that could be of therapeutic importance in disorders associated with decreased PAT-1 expression and function.

INTRODUCTION

Calcium oxalate urolithiasis is the most common renal disease in the developed countries (10, 16, 21) and is a critical risk factor for chronic kidney disease and progression to end-stage renal disease (1). Both genetic as well as environmental factors are thought to contribute to pathogenesis of nephrolithiasis (29, 47). Also, there is an increasing evidence for association between nephrolithiasis and a number of pathologies such as inflammatory bowel diseases (8, 18, 36), obesity (60), and diabetes (13, 30) and procedures including bariatric surgery (44, 57, 60). Excessive urinary excretion of oxalate (hyperoxaluria) is identified as the primary risk factor for kidney stone formation. Although kidneys serve as the primary route for oxalate excretion, intestinal oxalate transport has also been shown to have a significant impact on oxalate homeostasis (22). Enhanced intestinal absorption of dietary oxalate arising from impaired enteric oxalate secretion can lead to elevated renal oxalate excretion and lithogenicity (22, 65).

In the intestine, SLC26A6 or putative anion transporter-1 (PAT-1) has been shown to play an important role in transcellular oxalate transport. PAT-1 is a multifunctional anion exchanger present on the apical membrane of intestinal villi in duodenum, jejunum, and ileum (64). PAT-1 is capable of operating in Cl⁻/oxalate, Cl⁻/${\text{HCO}}_3^{-}$, $\text{S}{\text{O}}_4^{2-}$/oxalate, $\text{S}{\text{O}}_4^{2-}$/Cl⁻, Cl⁻/formate, and Cl⁻/OH⁻ exchange modes (3, 27, 32, 64, 67). In this regard, previous studies utilizing siRNA-mediated knockdown of PAT-1 in Caco-2-BBe1 cells suggested that 50% of apical oxalate exchange was mediated by PAT-1 (16). The physiological role of PAT-1 as a major Cl⁻/oxalate exchanger was evident from studies in PAT-1 knockout mice. These mice exhibited a significant reduction in oxalate secretion in the small intestine, enhanced net oxalate absorption with a subsequent increase in serum and urinary oxalate concentration, and excessive calcium oxalate stone formation in the kidneys (16, 26). These observations imply that hyperoxaluria associated with deletion of SLC26A6 poses a major risk for nephrolithiasis. Mutations in SLC26A6 have been shown in patients with kidney stones (41, 45). Therefore, it is critical to understand the regulation of PAT-1 at the molecular level in the intestine. Very few reports are available with respect to PAT-1 regulation in the intestine. In this regard, the role of PAT-1 is still underappreciated, and little emphasis has been paid on the analysis of its regulation. Neural regulation of PAT-1 expression and function has been shown in vitro where cholinergic stimulation of T-84 cells reduced cell surface expression with concomitant reduction in Cl⁻/oxalate exchange (20). Similar regulation of PAT-1 has been shown by ATP and UTP treatment in Caco2.BBe cells (4). Transcriptionally, interferon-γ treatment has been shown to decrease PAT-1 mRNA and promoter activity (54). Furthermore, a recent study identified that conditioned medium from Oxalobacter formigenes significantly stimulated PAT-1-mediated oxalate uptake in Caco-2.BBe cells via a PKA-dependent pathway (6) and significantly decreased urinary oxalate excretion in mouse model of primary hyperoxaluria type 1 (6). All these results demonstrate that PAT-1-mediated oxalate transport can be modulated by various agents.

In recent years, microRNAs (miRNA), a group of small noncoding RNAs (21–23 nucleotides), have been shown to play major role as regulators of gene expression, via binding to the 3′-untranslated regions (3′-UTRs) of their target mRNAs leading to mRNA turnover (7, 38) and/or translational repression (46, 56). Recent reports demonstrated that specific miRs regulate the expression of tight junction proteins such as occludin (69) and intestinal transporters such as cystic fibrosis transmembrane conductance regulator (CFTR; a chloride channel) (17), downregulated in adenoma (DRA; Cl⁻/${\text{HCO}}_3^{-}$ exchanger) (5), Na⁺/K⁺/Cl⁻ cotransporter (17), and peptide transporter 1 (pepT1; an oligopeptide transporter) (11). However, role of miRNAs in the regulation of PAT-1 has not been investigated. Our current studies showed that PAT-1 3′-UTR harbors potential binding sites for several miRNAs including miR125a-5p, 339-5p, 423-5p, 485-5p, and 501-3p. Among these, miR125a-5p was found to repress mRNA and protein expression of PAT-1.

MATERIALS AND METHODS

Cell culture.

Human intestinal epithelial cell lines Caco-2 (colorectal adenocarcinoma), HT-29 (colorectal adenocarcinoma), and T-84 (carcinoma) cells were obtained from American Type Culture Collection (ATCC, Manassas, VA), and SK-CO15 (colorectal adenocarcinoma) cells were a kind gift from Dr. Jun Sun (Univ. of Illinois at Chicago). Caco-2, T-84, and SK-CO15 cells were cultured in Eagle’s minimum essential medium, Dulbecco’s modified minimum Eagle’s medium F-12 (1:1), and Dulbecco’s modified minimum Eagle’s medium (high glucose), respectively. The culture medium for all cell types was supplemented with 10% fetal bovine serum, 100 U/ml penicillin, 100 μg/ml streptomycin, and 50 μg/ml gentamicin. The cells were maintained at 37°C in a 5% CO₂-95% air environment in T-75 (75 cm²) plastic flasks. The cells between passage 25 and 45 were utilized for the present study.

3′-UTR of PAT-1 and miR-125a-5p binding region mutated constructs.

A 263-bp fragment of the PAT-1 mRNA-3′-UTR was custom synthesized (GenScript, Piscataway, NJ). Briefly, synthesized DNA fragments were verified for their sequences and then ligated in vector pmirGLO (Promega, Madison, WI), a special kind of vector based on dual-luciferase technology, where firefly luciferase (luc2) is used as the primary reporter to monitor mRNA regulation and Renilla luciferase (hRluc-neo) as a control reporter for normalization. We utilized NheI-XhoI restriction enzymes for digesting the vector. The sequence of the product was verified by using the sequencing primers: M13F CGCCAGGGTTTTCCCAGTCACGAC, T7ter TGCTAGTTATTGCTCAGCGG. The miR-125a-5p binding seed sequence of PAT-1–3′UTR with desired mutations was custom synthesized by GenScript (see Fig. 6), and mutations were confirmed by sequencing.

Fig. 6.

Mutation of miR-125a-5p binding region in putative anion transporter-1 (PAT-1) 3′-untranslated region (UTR) abrogated the effects of mimic-125a-5p transfection on relative luciferase activity. A: mutated miR-125a-5p seed sequence in 3′-UTR of PAT-1 Caco-2 cells (grown in T-75 flask for 3–5 days to achieve ~80% confluency) were cotransfected with miR-125a-5p and pmirGLO-PAT-1 or pmirGLO-PAT-1 mut125a-5p. B: firefly luciferase activities were measured and normalized with respective Renilla luciferase activities. miR-125a-5p binding region in 3′-UTR of PAT-1 is denoted in red color and respective mutated sequence is in blue color. Results are means ± SE of 3 independent experiments. Statistical analysis using one-way ANOVA followed by Tukey’s multiple comparison test showed a significant difference between pmirGLO-3′-UTR-PAT-1 + -ve control and pmirGLO-3′-UTR-PAT-1 + mimic-125a-5p (***P < 0.0001).

In silico analysis.

Potential miRNAs targeting PAT-1 3′-UTR were predicted using common prediction algorithm TargetScan (http://www.targetscan.org/) (19). The top five microRNAs with high context scores and context score percentile were selected for the experiments.

Transient transfection and 3′-UTR activity.

Caco-2, HT-29, T-84, and SK-CO15 cells were transiently transfected with 1.5 μg of pmirGLO 3′-UTR of PAT-1 alone or in combination with 20 nM different microRNA mimics/negative control miRNA (Sigma, St. Louis, MO) as previously described us (5). Breifly, cells grown in T-150 flask to 80% confluency were trypsinized and suspended in complete growth medium without antibiotic. Transfection mixture (100 μl optiMEM, 1 μl of RNAiMax reagent, and 20 nM mimic/negative control) prepared according to the manufacturer’s instructions was added to the cell suspension and plated into 24-well plates. Lipofectamine 2000 (Invitrogen) was used for all the cell lines except Caco-2 cells, where we utilized Lipofectamine RNAiMax (Invitrogen Life Technologies, Carlsbad, CA). Cells were harvested 48 h after transfection and lysed in passive lysis buffer (Promega). The luciferase activity was determined using the Dual Luciferase Assay Kit (Promega) and a GLOMAX 20/20 Luminometer (Promega) equipped with double injectors. The effect of 3′-UTR activity on the reporter gene was calculated as a ratio of firefly luciferase to Renilla luciferase and is expressed as percentage of control.

RNA extraction and real-time PCR.

To quantitate the PAT-1 mRNA, total RNA from Caco-2 cells was extracted with Qiazol using miRNeasy mini kit (Qiagen, Frederick, MD) according to the manufacturer’s instructions as previously described (5). Briefly, extracted RNA was amplified by Brilliant SYBR Green qRT-PCR Master Mix kit (Agilent Technologies, Santa Clara; CA) utilizing gene-specific primers for PAT-1 and GAPDH (Table 1). The relative mRNA levels of PAT-1 were expressed as percentage of control normalized to GAPDH used as internal control gene.

Table 1.

Primers used in the study

	Sequence
Real-time PCR primers
SLC26A6
Forward	5′-AGATGCCCCACTACTCTGTCCT-3′
Reverse	5′-ATCCACACCACACCTCTGCTT-3′
GAPDH
Forward	5′-GAAATCCCATCACCATCTTCC-3′
Reverse	5′-AAATGAGCCCCAGCCTTCT-3′
Primers used for quantifying microRNA expression
Forward
hsa-miR-125a-5p	5′-TCCCTGAGACCCTTTAACCTGTGA-3′
hsa-miR-501-3p	5′-AATGCACCCGGGCAAGGATTCT-3′
hsa-miR-423-5p:	5′-TGAGGGGCAGAGAGCGAGACTTT-3′
hsa-miR-485-5p	5′-AGAGGCTGGCCGTGATGAATTC-3′
hsa-miR-339-5p	5′-TCCCTGTCCTCCAGGAGCTCACG-3′
U6B	5′-CGCAAGGATGACACGCAAATTCG-3′

A common universal reverse primer provided in the kit was used for quantifying microRNA expression.

Quantification of mature miR expression.

Quantification of mature microRNA expression was done as described previously (5). Briefly, 1 μg of total RNA was reverse transcribed to produce cDNA using the NCode miRNA first-strand cDNA synthesis kit (Invitrogen) in a 10-μl reaction according to the manufacturer’s instructions. Mature miRNA levels were quantified by Express SYBR Green qPCR supermix and universal reverse primer provided in the NCode miRNA first-strand cDNA synthesis kit (Invitrogen), and the specific microRNA forward primers for miR-125a-5p, miR-339-5p, miR-423-5p, miR-485-5p, miR-501-3p, and small RNA U6 (used as housekeeping gene) were obtained from MWG Operon Eurofins (Huntsville, AL) the primer sequences listed in Table 1. Real-time PCR amplification and data capture were performed using the Stratagene Mx3005P (Agilent Technologies, Santa Clara, CA).

Western blotting.

Cell lysates were prepared after 48 h of transient transfection with mimics as previously described (5). Briefly, media were aspirated and cells were washed with ice-cold 1× PBS and lysed in 1× cell lysis buffer (Cell Signaling, Danvers, MA) and 1× protease cocktail inhibitor mixture (Roche, Indianapolis, IN). Lysed cells were sonicated, and the lysate was centrifuged at 7,000 rpm for 7 min at 4°C. The concentration of the protein was determined by the Bradford assay. Cell lysates containing 75–100 μg protein were loaded on 7.5% SDS-polyacrylamide gels and transblotted to nitrocellulose membranes to examine the protein levels of PAT-1 and GAPDH as a loading control. Nonfat dry milk (5%) was used as a blocking buffer for 1 h. Membranes were then probed with human anti-PAT-1 antibody (1:3,000 dilution), which was a generous gift from Dr. Peter S. Aronson (Yale University, New Haven, CT), which is an affinity-purified rabbit polyclonal antibody (R29; Pocono Rabbit Farm and Laboratory, Canadensis, PA) directed against a 22 amino acid peptide (CDLRRRDYHMERPLLNQEHLEE) from the NH₂-terminal region of human SLC26A6 (31, 61) or GAPDH antibodies (1:3,000 dilution) in 1% nonfat dry milk overnight at 4°C. The membranes were washed four times with the wash buffer containing 1× PBS and 0.1% Tween-20 for 5 min. Later, the membranes were probed with horseradish peroxidase-conjugated goat anti-rabbit IgG antibody (1:2000 dilution) for 1 h, and the bands were visualized using Enhanced Chemiluminescence detection reagents (Bio-Rad, Hercules, CA). We utilized ImageJ software for densitometric analysis of the relative band intensities.

Statistical analyses.

All data were analyzed by Prism (Prism Graph Pad Software). Results are expressed as means ± SE and represent the data from three to six independent experiments. One-way ANOVA with Tukey’s multiple comparison test and unpaired t-test was used for statistical analysis. P < 0.05 was considered as statistically significant.

RESULTS

Identification of potential miRNAs targeting the PAT-1 3′-UTR.

First, Targetscan algorithm was utilized to predict the potential miRNAs that bind to PAT-1 3′-UTR (Fig. 1). Among the five microRNAs, miR-125a-5p, miR-339-5p, miR-423-5p, miR-485-5p, and miR-501-3p predicted by Targetscan (Table 2) were found to be potentially targeting PAT-1 3′-UTR, and miR-125a-5p demonstrated the highest context score of −0.61. Figure 1 illustrates the potential binding sites of several miRNAs along the length of PAT-1 3′-UTR (263 bp) and conserved sites for miRNA families among vertebrates and mammals. In silico results from Targetscan indicated that miR-125a-5p is the only microRNA broadly conserved among the vertebrates that bind to the 3′-UTR of PAT-1 with the highest context score.

Fig. 1.

Map of potential microRNA binding sites in 3′-untranslated region (UTR) of putative anion transporter-1 (PAT-1): in silico analysis. Figure shows the Targetscan analysis of microRNAs that target 3′-UTR of PAT-1 and their respective location to where they bind and several microRNAs whose binding region in 3′-UTR of PAT-1 is broadly or poorly conserved among mammals and vertebrates. [Figure is based on the output from http://genome.ucsc.edu (hg19 assembly) and http://www.targetscan.org Release 7.1.]

Table 2.

miRNAs predicted to target SLC26A6

	Context Score
miR125a-5p	−0.61
miR-501-3p	−0.55
miR-423-5P	−0.51
miR-485-5p	−0.30
miR-339-5p	−0.28

Some of the micro (mi)RNAs predicted to target SLC26A6 3′-untranslated region according to context score (targetscan 7.1).

Transient transfection of PAT-1 3′-UTR in intestinal epithelial cells decreased luciferase reporter activity.

Expression regulation by UTRs engage binding of various interacting partners like RNA binding proteins or noncoding RNAs (43). Targetscan analysis revealed that PAT-1 3′-UTR could potentially interact with a number of miRNAs. To verify a possible gene-regulatory role for PAT-1 3′-UTR, a 263-bp fragment of the region was commercially synthesized by Genescript and cloned into a pmiRGLO Dual Luciferase miRNA target expression vector. Caco-2, HT-29, T-84, and SK-CO15 cells were transiently transfected with pmirGLO empty vector or pmirGLO-3′-UTR-PAT-1. Here, a decline in reporter luciferase activity would signify a likely interaction of PAT-1 3′-UTR with binding partners e.g., miRNAs. As depicted in Fig. 2, a significant reduction (50–60%) in luciferase activity in all the cell lines was associated with the transfection with pmirGLO-3′-UTR-PAT-1. This repression of the reporter luciferase activity thus confirmed that certain binding partners like miRNAs or RNA binding proteins could interact with the PAT-1 3′-UTR and modulate the expression of this transporter.

Fig. 2.

Transient transfection of putative anion transporter-1 (PAT-1) 3′-untranslated region (UTR) decreased relative luciferase activity in intestinal epithelial cells. A–D: relative luciferase activity after 48-h transient transfection of pmiRGLO and 3′-UTR PAT-1 in Caco-2 (A), T-84 cells (B), HT-29 (C), and SK-CO15 cells (D) (grown in T-75 flask for 3–5 days ~80% confluent). Results are shown as % control in response to pmirGLO-3′-UTR-PAT-1 transfection compared with pmirGLO empty vector transfection. All the results are means ± SE of 4 independent experiments. ***P < 0.0001, ****P < 0.00001 vs. pmiRGLO was considered as statistically significant using unpaired t-test.

miRNAs 125a-5p and 423-5p target PAT-1 3′-UTR.

A similar approach involving co-delivery of pmirGLO-3′-UTR-PAT-1 and predicted miRNA mimics was next utilized to specifically identify miRNAs interacting with the 3′-UTR of PAT-1. Mimics of top five Targetscan miRNA hits, 125a-5p, -339-5p, -423-5p, -485-5p, and -501-3p, were cotransfected with pmirGLO-3′-UTR-PAT-1 in Caco-2 cells. Overexpression of miRNAs was confirmed by PCR as outlined in materials and methods (data not shown). Only mimic-125a-5p and mimic-423-5p were able to further significantly reduce the relative luciferase activity compared with negative control (Fig. 3). These data indicate a functional role of 125a-5p, and -423-5p microRNAs in regulating PAT-1 expression via the PAT-1 3′-UTR. Whereas the other three mimics, mimic-339-5p, -485-5p, and -501-3p, showed no significant changes in the relative luciferase activity, suggesting that these three microRNAs do not bind directly to the 3′-UTR of PAT-1.

Fig. 3.

Cotransfection of microRNA mimics that target putative anion transporter-1 (PAT-1) 3′-untranslated region (UTR) decreased luciferase reporter activity. Caco-2 cells (grown to 80% confluency, normally achieved in 3–5 days depending on seeding density) were co-transfected with microRNA mimics (that target 3′-UTR of PAT-1) or negative control and PAT-1 3′UTR or pmiR-GLO. Forty-eight hours posttransfection, firefly luciferase activities were measured and normalized with respective Renilla luciferase activities. Results are means ± SE of 4 independent experiments. Difference between 3′-UTR-PAT-1 -ve control vs. pmirGLO-3′-UTR-PAT-1 + mimic-125a-5p (****P < 0.00001) and vs. pmirGLO-3′UTR-PAT-1 + mimic-423-5p (***P < 0.0001) was considered statistically significant using one-way ANOVA followed by Tukey’s multiple comparison test.

miR-125a-5p decreased both mRNA and protein levels of PAT-1.

miRNAs can exert regulatory control on the genes both posttranscriptionally and posttranslationally. To identify which regulatory mechanism is active in miR-125a-5p regulation of PAT-1, we investigated the effects of mimics on PAT-1 mRNA and protein levels by transiently transfecting Caco-2 cells with mimics and measuring PAT-1 mRNA and protein levels. As expected, among all the mimics transfected, only mimic-125a-5p caused a significant decline in PAT-1 mRNA levels (~49%; Fig. 4). This shows that overexpression of miR-125a-5p downregulates PAT-1 by promoting mRNA degradation. Next, we examined the effect of overexpression of miR-125a-5p on PAT-1 protein levels. Parallel to the decrease in PAT-1 mRNA, mimic-125a-5p also caused a significant decrease in PAT-1 protein expression (~44%) compared with negative control (Fig. 5). These data clearly indicate that binding of miR-125a-5p to 3′-UTR of PAT-1 destabilizes PAT-1 transcript by degrading the mRNA subsequently leading to decreased PAT-1 protein expression. Caco-2 cells transiently transfected with mimic of another miR, miR-423-5p, did not change PAT-1 mRNA (Fig. 4) and protein levels (data not shown).

Fig. 4.

miR-125a-5p mimic transfection decreased putative anion transporter-1 (PAT-1) mRNA expression. RNA was extracted from Caco-2 cells (grown in T-75 flask for 3–5 days to achieve ~80% confluency) 48 h posttransfection with mimics of microRNAs that target PAT-1. PAT-1 mRNA levels were measured by quantitative real-time PCR as described in materials and methods. Values of mRNA levels for PAT-1 were normalized against GAPDH mRNA levels. Results are means ± SE of 6 independent experiments (**P < 0.01).

Fig. 5.

Mimic-125a-5p transfection decreased putative anion transporter-1 (PAT-1) protein expression. A: cell lysates of Caco-2 cells (grown in T-75 flask for 3–5 days to achieve ~80% confluency) transiently transfected with negative control and mimic-125a-5p were subjected to 7.5% SDS-PAGE followed by transfer to nitrocellulose membrane. The blots were probed with anti-PAT-1 or anti-GAPDH antibody. B: densitometric analysis of the relative band intensities was performed using ImageJ software. Results represent means ± SE of 5 different experiments. Differences between negative control versus mimic-125a-5p transfected groups (**P < 0.001) were found to be statistically significant using unpaired t-test.

Mutating miR-125a-5p binding region of 3′-UTR of PAT-1 abrogated the effects on PAT-1.

From the results above it is evident that miR-125a-5p affects PAT-1 expression. To experimentally substantiate that all of the effects noted above were due to miR-125a-5p’s direct binding to 3′-UTR of PAT-1, the binding region of the miR-125a-5p on 3′-UTR of PAT-1 was mutated (sequences are depicted in Fig. 6A, miR-125a-5p binding region in 3′-UTR of PAT-1 is denoted in red color and respective mutated sequence in blue color). Cells transiently cotransfected with mimic-125a-5p with either native 3′-UTR PAT-1 or mutated 3′-UTR PAT-1 were harvested after 48 h, and relative luciferase activity was measured. Mutation in miR-125a-5p binding region in the 3′-UTR of PAT-1 prevented a decline in luciferase activity by mimic-125a-5p while native (unmutated) 3′-UTR of PAT-1 showed a repression of relative luciferase activity (Fig. 6B).

DISCUSSION

Nephrolithiasis is a common health problem affecting 1 in 11 people in United States (55). Elevation in urinary oxalate concentration is the major risk factor for kidney oxalate stone formation. Recent studies have shown that gastrointestinal oxalate transport can influence renal oxalate excretion (16, 21, 26, 65). In this regard, PAT-1 is the key anion transporter in the intestine involved in transcellular oxalate-secreting pathway. Therefore, it is important to understand the mechanisms by which PAT-1 expression and function can be modulated.

In the present study, we report that microRNA-125a-5p can modulate PAT-1 expression posttranscriptionally via mRNA degradation. miRNAs (21–23 nucleotides long) are nonprotein-coding RNA molecules that can modulate gene expression either via translation repression or by destabilization of target transcripts (14). miRNAs have been shown to play crucial roles in numerous cellular processes (42) including cell differentiation (68), proliferation (28), apoptosis (9), and metabolism (53), and their aberrant expression has been shown to be associated with various pathologies such as ulcerative colitis (48, 66), Crohn’s disease (48, 66), and cancer (39). The miR-125a group of miRNAs are known to have ubiquitous expression and functional targets in a wide range of tissues (51) including intestine. Expression of miR-125a-5p generally increases with cell differentiation whereas it is downregulated in several types of tumors, including breast, gastric, cervical, lung, ovarian, and colon cancers, retinoblastoma, medulloblastoma, glioblastoma, neuroblastoma, and hepatocellular carcinoma (50). miRNAs are known to have differential role in modulating differentiation and proliferation in different tissues. In this regard, an earlier study (12) reported ~50-fold higher expression of miR125a-5p in colonocyte-like HT-29cl.19 cells compared with enterocyte-like Caco-2-BBe cells, independent of differentiation status. In both cell types, however, expression levels were higher in differentiated stage.

The expression pattern of various miRNAs and their functions in nephrolithiasis pathogenesis are slowly surfacing. In this regard, Hu et al. (23) reported the possible involvement of miR-155 in pathogenesis of nephrolithiasis in a study conducted in 60 nephrolithiasis patients. This study (23) reported significantly higher expression of miR-155 in the serum and urine of kidney stone patients than those in the healthy subjects and showed that miR-155 also regulated the expression of inflammatory cytokines. Another recent study identified miRNA expression profiles in the kidney tissues of hyperoxaluric rats highlighting the role of miRNAs in the formation of calcium oxalate stone (40). Taken together, these studies emphasize the potential role of miRNAs in nephrolithiasis. As PAT-1 knockout mice exhibit calcium oxalate nephrolithiasis (16, 26), we hypothesized that miRNAs may be involved in modulation of PAT-1 expression.

The key feature of miRNA-mediated regulation involves base pairing between the seed sequence of the miRNA and complementary seed match in the 3′-UTR region of the mRNA sequence of the target gene (52). Our in silico analysis by Targetscan identified miR-125a-5p, miR-339-5p, miR-423-5p, miR-485-5p, and miR-501-3p as potential miRNA candidates for PAT-1 3′-UTR. The Targetscan algorithm relies not only on the complementarity of pairing for miRNA-mRNA interactions but also provides with a “context score,” which is a sum total of 14 additional parameters (such as stability of seed pairing, 3′-UTR length of the target mRNA, conservation in orthologous UTRs, target site abundance, free energy calculation, etc.) (2, 15). Therefore, predictions by this algorithm were considered for experimental validation in the present study. In fact, we experimentally substantiated that PAT-1 expression is susceptible to miRNA-mediated regulation and that it is probably not cell type specific. All four intestinal epithelial (T-84, HT-29, Caco-2, and SK-CO15) cell lines exhibited significant reduction in reporter luciferase activity upon transfection with PAT-1 3′-UTR. The direct target relationship between various miRNA candidates and 3′-UTR of PAT-1 was confirmed by luciferase reporter assay after cotransfection of PAT-1 3′-UTR with various miRNA mimics. Notable suppression of the firefly luciferase signal was observed only for miR-125a-5p and miR-423-5p when compared with negative control and pmirGLO-3′-UTR-PAT-1 cotransfected cells. These data indicated that PAT-1 is a direct novel target of both miR-125a-5p and miR-423-5p. The other miRNAs (miR-339-5p, miR-485-5p, and miR-501-3p) were considered false positive computational predictions and were not tested further. It is interesting to note that miR-125a-5p was predicted to 1) have 8mer seed match in PAT-1 3′-UTR (defined as a perfect match to positions 2-8 of the mature miRNA followed by an adenosine residue), and 2) be conserved among various mammalian species, a higher context score (−0.61) in comparison to miR-423-5p (context score; −0.51) suggestive of higher probability of its in vivo functionality.

Mechanistically, target mRNA decay/destabilization underlies miRNA-mediated regulation of target expression (33). Between miR-125a-5p and miR-423-5p, only the former repressed PAT-1 mRNA levels. Interestingly, 49% (P < 0.05) repression in PAT-1 mRNA by miR-125a-5p was complemented by significant decrease in PAT-1 protein. This indicates that miR-125a-5p is likely to repress the PAT-1 expression through mRNA degradation rather than through translational repression. During the process of mRNA decay, mature miRNAs are incorporated into the RNA-induced silencing complex, the functional protein complex involved in association between the miRNA and its target mRNA (25), and guide cleavage of mRNA targets with perfect or nearly perfect complementarity (15, 34, 63). Similarly, miR-125a-5p mediated decrease in PAT-1 transcripts can be explained based on the perfect complementary of the miR-125a-5p seed sequence at the predicted binding positions of the PAT-1 3′-UTR. The reason why miR-423-5p failed to exhibit functional interaction with PAT-1 mRNA could be due to nonphysiological interaction observed in the reporter system (35). Further validation of the specificity of miR-125a-5p and PAT-1 3′-UTR interaction is exhibited by reversal of the inhibitory effect of miR-125a-5p on PAT-1 mRNA expression induced by mutation of its specific binding site.

Recent studies revealed that miRNAs play pivotal roles in physiology and disease, and therapeutic targeting has started to be investigated. miR-125a-5p has been shown to act both as a tumor suppressor in several cancers such as lung cancer (70), hepatocellular carcinoma (37, 49), colon cancer (59, 62), as well as tumor promoter (58). Our current studies on this miR have several physiological implications. It is the first study to report miR-dependent mechanism of regulation of expression of PAT-1 in intestinal epithelial cells. We show that miR-125a-5p inhibits PAT-1 mRNA and protein expression through its direct interactions with the PAT-1 mRNA 3′-UTR. Additionally, our studies also implicate potential role of this miRNA in the pathophysiology of hyperoxaluria and kidney stone disease and potential therapeutic modality based on manipulating the activity of this miRNA. Generally, the upregulation of miRNAs is achieved through administration of synthetic miRNAs or administration of miRNA-expressing vectors. The downregulation of miRNAs is achieved through administration of antisense nucleotides, often chemically modified to ensure stability and specificity (24). There are multiple potential limitations associated with the development and testing of miRNA-based therapeutics. However, given the relatively short period of time since discovery of miRNAs, the progress appears sufficient to justify optimism regarding developing novel therapeutics based on miRNAs (24). Therefore, results of our studies define avenues to develop miRNA-based treatment strategies aimed at downregulation of miR-125a-5p for the management of diseases associated with impairment of PAT-1-mediated intestinal oxalate transport.

GRANTS

These studies were supported by the Department of Veterans Affairs, Veterans Heath Administration, Office of Research and Development, Biomedical Laboratory Research and Development: Merit Review Awards: BX002011 (P. K. Dudeja), BX000152 (W. A. Alrefai), BX002867 (S. Saksena); VA Research Career Scientist Award (P. K. Dudeja and W. A. Alrefai); NIH (National Institute of Diabetes and Digestive and Kidney Diseases/National Institute of Allergy and Infectious Diseases) Grants DK-54016, DK-81858, DK-92441 (P. K. Dudeja), DK-109709 (W. A. Alrefai), DK-98170 (R. Gill), and AI-130790 (A. Borthakur).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

P.K.D. conceived and designed research; A.N.A. and S.P. performed experiments; A.N.A. and S.P. analyzed data; A.N.A., A.B., S.S., R.K.G., and W.A.A. interpreted results of experiments; A.N.A. prepared figures; A.N.A. and A.B. drafted manuscript; A.B., S.S., R.K.G., W.A.A., and P.K.D. edited and revised manuscript; P.K.D. approved final version of manuscript.