Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2012 Apr;142(4):886-896.e9.
doi: 10.1053/j.gastro.2011.12.047. Epub 2012 Jan 10.

Strand-specific miR-28-5p and miR-28-3p Have Distinct Effects in Colorectal Cancer Cells

Affiliations
Free PMC article

Strand-specific miR-28-5p and miR-28-3p Have Distinct Effects in Colorectal Cancer Cells

Maria I Almeida et al. Gastroenterology. .
Free PMC article

Abstract

Background & aims: MicroRNAs (miRNAs) can promote or inhibit tumor growth and are therefore being developed as targets for cancer therapies. They are diverse not only in the messenger RNAs (mRNA) they target, but in their production; the same hairpin RNA structure can generate mature products from each strand, termed 5p and 3p, that can bind different mRNAs. We analyzed the expression, functions, and mechanisms of miR-28-5p and miR-28-3p in colorectal cancer (CRC) cells.

Methods: We measured levels of miR-28-5p and miR-28-3p expression in 108 CRC and 49 normal colorectal samples (47 paired) by reverse transcription, quantitative real-time polymerase chain reaction. The roles of miR-28 in CRC development were studied using cultured HCT116, RKO, and SW480 cells and tumor xenograft analyses in immunodeficient mice; their mRNA targets were also investigated.

Results: miR-28-5p and miR-28-3p were down-regulated in CRC samples compared with normal colon samples. Overexpression of miRNAs in CRC cells had different effects and the miRNAs interacted with different mRNAs: miR-28-5p altered expression of CCND1 and HOXB3, whereas miR-28-3p bound NM23-H1. Overexpression of miR-28-5p reduced CRC cell proliferation, migration, and invasion in vitro, whereas miR-28-3p increased CRC cell migration and invasion in vitro. CRC cells overexpressing miR-28 developed tumors more slowly in mice compared with control cells, but miR-28 promoted tumor metastasis in mice.

Conclusion: miR-28-5p and miR-28-3p are transcribed from the same RNA hairpin and are down-regulated in CRC cells. Overexpression of each has different effects on CRC cell proliferation and migration. Such information has a direct application for the design of miRNA gene therapy trials.

Conflict of interest statement

Conflicts of interest:

The authors declare no competing financial interests in relation to the work described.

Figures

Figure 1
Figure 1
Expression of miR-28-5p and miR-28-3p in colon tissue samples. (A) Quantitative real-time polymerase chain reaction analysis shows that miR-28-5p and miR-28-3p are downregulated in colon cancer samples compared with normal colorectal tissue samples. (B) Both microsatellite stable (MSS) and microsatellite unstable (MSI) tumors express significantly less miR-28-5p and miR-28-3p levels when compared with normal colon tissue. No differences were found when comparing miR-28-5p and miR-28-3p levels of MSS and MSI tumors. (C) miRNAs downregulation in CRC tumors paired with normal tissue from the second set of patients. All values of miRNA expression levels were normalized to small nuclear RNA U6. Mean +/- standard error of the mean (SEM) are represented on the images (***P < 0.005, Mann-Whitney-Wilcoxon test, and paired t test for paired normal vs tumor groups).
Figure 2
Figure 2
Biological effects of miR-28-5p in proliferation, apoptosis, and cell cycle in vitro. (A and B) Representative experiment of the proliferation effect of miR-28-5p and miR-28-3p in HCT116 and RKO colon cell lines. Cell numbers were counted every 24 h for 4 days posttransfection with scrambled negative control (SCR), miR-28-5p, or miR-28-3p. miR-28-5p, but not miR-28-3p, inhibited growth in both HCT116 and RKO cell lines. Values represent the average of 3 replicates +/- standard deviation (SD) (***P < 0.005, Student t test). Two independent experiments were performed. (C) Immunoblotting with anti-poly(adenosine diphosphate-ribose) polymerase 1 (PARP1) 48 h after transfection of HCT116 and RKO cell lines with SCR, miR-28-5p, or miR-28-3p. Graphic represents the ratio between cleavage and total PARP1 form. miR-28-5p, but not miR-28-3p, increased PARP1 cleavage form. (D) Fluorescent-activated cell sorting (FACS) analysis 48 h posttransfection with SCR, miR-28-5p, or miR-28-3p. Representative experiment was performed in duplicate; average +/- SD (*P < 0.05, Student t test). Two independent experiments were performed.
Figure 3
Figure 3
miR-28 decreases tumor volume in mice xenografts. (A and B) HCT116-pBABE-empty (control) and HCT116-pBABE-miR-28 (stably expressing miR-28) were subcutaneously injected in the left and right flanks of 9 mice, and tumor volume was measured during the (A) course of the experiment and (B) at the end of the experiment (21 days postinoculation). The tumor volumes in the HCT116-pBABE-miR-28 group were lower than those in the HCT116-pBABE-empty group (**P < 0.01, Student t test). (C) Photographs show tumors excised from 5 mice in each group. (D) Quantitative real-time polymerase chain reaction analysis shows miR-28-5p and miR-28-3p expression in the tumors extracted from the mice (mean +/- standard deviation) (**P < 0.01, Student t test).
Figure 4
Figure 4
Effect of mir-28-5p and miR-28-3p in migration and invasion in vitro. Absorbance was measured for cells on the bottom of noncoated and matrigel-coated transwell chambers at 24 h (for migration) and 48 h (for invasion) after HCT116 cells expressing miR-28-5p or miR-28-3p were plated. Results are shown relative to scrambled negative control (SCR). A representative experiment is shown. Average (of triplicates) +/- standard deviation is shown (*P < 0.05, **P < 0.01, Student t test). (B) Microscopy images (x50) show the migratory and invasive cells on transwell assays.
Figure 5
Figure 5
miR-28 increases metastasis in vivo. (A and B) HCT116-pBABE-empty (control) and HCT116-pBABE-miR-28 (stably expressing miR-28) were injected in the vein tail of mice. (A) Thirty-five days postinjection metastases were detected in the liver, kidney, lung, and spinal cord. The percentage of mice with metastases in these organs was consistently higher in miR-28-expressing tumors than in the control. (B) Microscopy images (x100) show hematoxylin and eosin (HE) and anti-green fluorescent protein (GFP) immunohistochemical staining for liver, kidney, and lung metastatic tumors. (C) Number of tumors observed within the liver and kidneys. (D) Photographs of HCT116-pBABE-empty (left panel) and HCT116-pBABE-miR-28 (right panel) mice show the sites with metastasis (white arrows) found in more than 30% of each group of mice.
Figure 6
Figure 6
miR-28-5p targets Cyclin D1 and HoxB3, and miR-28-3p targets Nm23-H1. Western blot analysis shows (A) Cyclin D1, (B) HoxB3, and (C) Nm23-H1 expression in scrambled, miR-28-5p, and miR-28-3p transfected HCT116 cells. Expression levels were normalized for vinculin or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) protein levels and were compared to the scrambled negative control transfection (=1). (D and E) The predicted miRNA∷mRNA interaction sites are shown in the top panels. The bottom panels show luciferase activity for the predicted interaction sites (D) PGL3-HOXB3-WT constructs cotransfected with scrambled negative control (n=1) or miR-28-5p and (E) PGL3-NM23-H1-WT construct cotransfected with scrambled negative control (n=1) or miR-28-3p. The same experiment was also performed using constructs with a mutated interaction site—PGL3-HOXB3-Mut and PGL3-NM23-H1-Mut. Values represent the average +/− standard deviation of 2 independent experiments performed in 4 replicates (**P < 0.01, Student t test). (F) The proposed mechanism for miR-28-5p and -3p function in CRC is shown.

Similar articles

See all similar articles

Cited by 72 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback