Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 70 (16), 6609-18

Epigenetic Silencing of miR-137 Is an Early Event in Colorectal Carcinogenesis


Epigenetic Silencing of miR-137 Is an Early Event in Colorectal Carcinogenesis

Francesc Balaguer et al. Cancer Res.


Global downregulation of microRNAs (miRNA) is a common feature in colorectal cancer (CRC). Whereas CpG island hypermethylation constitutes a mechanism for miRNA silencing, this field largely remains unexplored. Herein, we describe the epigenetic regulation of miR-137 and its contribution to colorectal carcinogenesis. We determined the methylation status of miR-137 CpG island in a panel of six CRC cell lines and 409 colorectal tissues [21 normal colonic mucosa from healthy individuals (N-N), 160 primary CRC tissues and their corresponding normal mucosa (N-C), and 68 adenomas]. TaqMan reverse transcription-PCR and in situ hybridization were used to analyze miR-137 expression. In vitro functional analysis of miR-137 was performed. Gene targets of miR-137 were identified using a combination of bioinformatic and transcriptomic approaches. We experimentally validated the miRNA:mRNA interactions. Methylation of the miR-137 CpG island was a cancer-specific event and was frequently observed in CRC cell lines (100%), adenomas (82.3%), and CRC (81.4%), but not in N-C (14.4%; P < 0.0001 for CRC) and N-N (4.7%; P < 0.0001 for CRC). Expression of miR-137 was restricted to the colonocytes in normal mucosa and inversely correlated with the level of methylation. Transfection of miR-137 precursor in CRC cells significantly inhibited cell proliferation. Gene expression profiling after miR-137 transfection discovered novel potential mRNA targets. We validated the interaction between miR-137 and LSD-1. Our data indicate that miR-137 acts as a tumor suppressor in the colon and is frequently silenced by promoter hypermethylation. Methylation silencing of miR-137 in colorectal adenomas suggests it to be an early event, which has prognostic and therapeutic implications.

Conflict of interest statement

Conflict of interest: There is no conflict of interest to disclose for all authors.


Figure 1
Figure 1. Methylation analysis of miR-137 CpG island
(A) Map of miR-137 CpG island, position of mature miR-137 and PCR products used for methylation analysis. Orange box, miR-137 CpG island; vertical tick marks, CpG sites; TSS, putative transcription start site. (B) MSP analyses for miR-137 methylation in CRC cell lines. U, unmethylated state; M, methylated state; U-DNA, normal lymphocytes; M-DNA, in vitro methylated DNA. Top, parental cell lines. Bottom, analysis after 5-AZA treatment. (C) Electropherogram corresponding to the bisulfite sequencing of 4 CpG sites in HCT116 cell line. Top panel, parental cell lines. Bottom panel, methylation analysis after 5-AZA treatment. Horizontal lines, CpG sites. (D) Results of bisulfite pyrosequencing of miR-137 in HCT116. Methylation percentages of 4 CpG sites (marked within gray vertical boxes) are indicated in the pyrogram. Top panel, parental cell lines with high levels of miR-137 methylation; Bottom panel, miR-137 demethylation after 5-AZA treatment.
Figure 2
Figure 2. Methylation and expression analyses of miR-137 CpG island in CRC cell lines and in CRC tissues
(A) Bisulfite pyrosequencing results of miR-137 in 6 CRC cell lines. (B) TaqMan RT-PCR analysis of miR-137 expression in CRC cell lines. Results are expressed as 2−ΔΔCt (log10) and normalized to RNU6b. Error bars represent the standard deviation. (C) Bisulfite pyrosequencing results of miR-137 in colorectal tissues. N-N (n=21); N-C (n=111); CRC (n=113); adenomas (n=68). The black horizontal bar indicates mean methylation level. (D) TaqMan RT-PCR analysis of miR-137 expression in 15 paired CRC tissues and N-C. Results are expressed as 2−ΔCt and normalized to RNU6b.
Figure 3
Figure 3. In situ hybridization (ISH) analysis of miR-137 in normal colorectal mucosa and CRC
miR-137, positive control (U6) and negative control (no probe) ISH analysis were performed in normal colorectal mucosa (A) and a group of adenomas (B) and CRC (C). Staining for miR-137 was observed in the epithelium throughout the colonic crypt, with no staining of the stromal cells. However, miR-137 was not expressed in any of the neoplastic tissues evaluated. Hematoxylin-eosin (H&E) staining of the corresponding tissues is shown.
Figure 4
Figure 4. Identification of potential targets of miR-137 in CRC
(A) The Venn diagram represents the downregulated genes (≥2 fold-change) observed in the gene expression microarray analysis after transfection of miR-137 precursor (in blue), and the predicted targets generated by the in silico prediction tool miRecords (in yellow). (B) The list of genes identified using the mentioned strategy. The expression intensity of each mRNA varies from red (above the average) to green (below the average). ID, gene name; FC, fold-change. (C) Comparison between the qRT-PCR and microarray results. For each gene, the variation in expression compared to control is represented as average fold-change for both the microarray and the qRT-PCR analysis (error bars represent the standard deviation). (D) Western-blot analysis of potential miR-137 targets. Densitometric analysis of protein expression is shown below each blot, in relation to NC and normalized to β-actin expression.
Figure 5
Figure 5. Interaction between miR-137 and LSD1 and tumor suppressor features of miR-137
(A) The predicted hybridization of miR-137 (green) with the 3′UTR region of LSD1 mRNA (red) using RNAhybrid software. The minimum free energy (mfe) required for RNA hybridization is shown. The conserved predicted binding site of miR-137 with the 3′UTR region of LSD1 mRNA is also represented. The mutated binding site used for the luciferase assay is shown in red. (B) Correlation between miR-137 expression and LSD1 mRNA levels across a panel of 6 CRC cell lines. Results are expressed as 2−ΔCt. RNU6b and β-actin were used for normalization, respectively. Error bars represent the standard deviation. (C) Direct recognition of LSD1 mRNA 3′UTR by miR-137. Luciferase assay of HCT116 cells transfected with firefly luciferase constructs containing LSD1-mut or LSD1-wt. The parental luciferase plasmid (empty) was also transfected as a control. β-galactosidase activity was calculated for normalization. Error bars represent standard deviation. (D) Effect of miR-137 on cell proliferation (BrdU assay). Error bars represent standard deviation.

Similar articles

See all similar articles

Cited by 103 PubMed Central articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources