miR-346 and miR-138 competitively regulate hTERT in GRSF1- and AGO2-dependent manners, respectively

miRNAs typically downregulate the expression of target genes by binding to their 3′UTR, and dysregulation of miRNAs may contribute to tumorigenesis. Here, we found that miR-346 and miR-138 competitively bind to a common region in the 3′UTR of hTERT mRNA and have opposite effects on the expression and function of hTERT in human cervical cancer cells. Furthermore, G-rich RNA sequence binding factor 1 (GRSF1) mediates the miR-346-dependent upregulation of hTERT by binding to the miR-346 middle sequence motif (CCGCAU) which forms a “bulge loop” when miR-346 is bound to the hTERT 3′UTR, facilitating the recruitment of hTERT mRNA to ribosomes to promote translation in an AGO2-independent manner. Conversely, miR-138 suppresses hTERT expression in an AGO2-dependent manner. Interestingly, replacement of the miR-138 middle sequence with that of miR-346 results in an upregulation of hTERT expression in a GRSF1-dependent manner. Moreover, miR-346 depends on GRSF1 to upregulate another target gene, activin A receptor, type IIB (ACVR2B), in which miR-346 “CCGCAU” motif is essential. These findings reveal novel mechanisms of miRNA-mediated upregulation of target gene expression and describe the coordinated action of multiple miRNAs to control the fate of a single target mRNA through binding to its 3′UTR.

Figures S1-S10 Figure S1. Validation of efficiency of plasmids and oligomers in transfected HeLa cells.
(a-e) Small RNA was extracted from HeLa cells transfected with ASO-miR-346 or ASO-miR-138 and control oligomers; pri-miR-346, pri-miR-346 loop mut (Note: used different forward primers in Table S1, 346 Frd and 346 mut Frd to amplify miR-346, miR-346 loop mut, respectively) or pri-miR-138 and pcDNA3 as the control vector, and the expression of miR-346 and miR-138 was measured by qRT-PCR. U6 snRNA was regarded as the endogenous normalizer and the relative miR-346 and miR-138 expression levels (mean ± SD) were shown (* p < 0.05).         Tables S1-S2 Table S1. Primer sequences and oligomers used in this study

Oligonucleotides and plasmids
The primers used to construct the pri-miRNA, hTERT, ACVR2B and SMAD3 mRNA  Table S1.

EGFP reporter assay
The EGFP reporter assay was performed as previously described 1 . In brief, the 3'-untranslated region of hTERT mRNA or a series of hTERT mRNA 3'UTR mutants were amplified by PCR using the primers shown in Table S1. The PCR products were then cloned into the pcDNA3/EGFP vector at the BamHI and EcoRI sites downstream of the EGFP coding region.
The ACVR2B and SMAD3 mRNA 3'UTRs were annealed using the primers listed in Table   S1 and cloned into the same vector mentioned above. HeLa cells were co-transfected with a reporter plasmid and either pri-miR-346, pri-miR-138 or the pcDNA3 control vector in 48-well plates. The pDsRed2-N1 vector (Clontech) expressing RFP was spiked in to normalize the transfection. After 48 hours, the cells were lysed and the fluorescence 13 intensities of EGFP and RFP were detected with a F-4500 fluorescence spectrophotometer (HITACHI).

Telomerase activity detection
The activity and quantification of telomerase in the parental and transfected cervical cancer cells was detected using the TeloTAGGG Telomerase PCR ELISA PLUS kit (Roche) following the manufacturer's instructions.

Northern blot assay
Northern blot assay was performed as previously described 4 . Small RNAs were separated by electrophoresis on 20% PAGE-urea gels. The RNA was blotted and cross-linked onto nylon 14 membranes (Ambion) and hybridized with γ-32 P end-labeled (Furui Company, Beijing, China) ASO-miR-346, 138 and U6 snRNA, respectively (sequences can be found in Table S1). All of the radioisotopes were imaged at −80°C.

RNA immunoprecipitation assay
RIP assay was carried out following the method described by Christoph Ufer 5  to the non-specific control sample. After a rotation at 4°C overnight, 50 μl of protein G slurry was added to the supernatants and the samples were rotated at 4°C for 2 more hours.
Following centrifugation at 5,000 g for 5 minutes at 4°C, the supernatants were discarded and the samples were resuspended in 1 ml lysis buffer. The samples were centrifuged at 5,000 g for 5 minutes at 4°C and then the supernatants were discussed. This step was then repeated.
The pellet was then washed twice with high salt lysis buffer (the same as the lysis buffer 15 except that NaCl is at 700 mM). The RNAs in the pellets were released from the samples after proteinase K treatment for 30 minutes at 50°C, extracted with phenol: chloroform: isoamyl alcohol (25: 24: 1), and then precipitated in alcohol overnight. The RNAs were then quantified by qRT-PCR.

Dot blot hybridization assay
RNA dot hybridization was performed as described previously 6  probe were added to the reactions.

RNA electrophoretic mobility shift assay
Single