Previous Article |
Table of Contents
| Next Article 
BIOLOGICAL SCIENCES / MEDICAL SCIENCES
Tumor-suppressive miR-34a induces senescence-like growth arrest through modulation of the E2F pathway in human colon cancer cells
Biochemistry Division, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan
Communicated by Takashi Sugimura, National Cancer Center, Tokyo, Japan, August 4, 2007 (received for review June 6, 2007)
 |
Abstract |
|---|
 Top Abstract ResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
microRNA | p53 | adriamycin | atelocollagen
miRNA consists of 
22 nucleotides and regulates gene expression in a posttranscriptional manner by pairing with complementary nucleotide sequences in 3' untranslated regions (UTRs) of target mRNAs (5). Precise chronological and topological regulation of posttranscriptional gene silencing by miRNA is essential for animal development and tissue differentiation (6), and abnormal expression is suggested to be associated with various human disorders, including cancer (7–9). Recently, mutations of miR-16-1 and miR-15a genes have been reported in chronic lymphocytic leukemia patients (10), and the available results suggest a crucial involvement of aberrant miRNA expression in human carcinogenesis. However, the precise, critical roles of individual miRNAs largely remain to be elucidated.
We recently identified SND1/Tudor-SN as a C-rich DNA/RNA-binding protein (11), and Caudy et al. (12) reported it to be a component of RNA-induced silencing complex (RISC). We also demonstrated its frequent up-regulation in human colon cancers (13). Furthermore, it was also overexpressed in precancerous lesions induced by chemical carcinogens in rats (13). Although the detailed molecular mechanisms underlying the induction of SND1 in colon epithelial cells are not yet clear, alteration of its expression could be accompanied by the changes in the expression of miRNA species caused by some environmental insults. Therefore, we hypothesize that expression of a subset of miRNA species and components of miRNA effector complexes, including SND1, is affected by cytotoxic stresses and could play an important role in the onset and progression of colon carcinogenesis.
Recently, aberrant up- and down-regulation of miRNA species in human colon cancers has been reported (7–9). However, which miRNA species are actually implicated in human colon cancer development remains to be elucidated. Therefore, we have attempted to isolate miRNA species associated with cell proliferation control in colon epithelial cells. Aiming to this goal, we here used human colon cancer HCT 116 cells harboring wild-type p53 to identify miRNA species induced after cell proliferation arrest when treated with a low concentration of ADR. By comparing the miRNA responses after ADR treatment between HCT 116 and HCT 116 p53 knockout (HCT 116 p53–/–) cell lines (14), we identified the miR-34 family as an ADR-responsive miRNA group in a p53-dependent manner. Focusing on miR-34a that showed relatively high-expression levels among the miR-34 family members in HCT 116 cells, we further investigated the biological effects of miR-34a on cell proliferation both in vitro and in vivo settings. Expression levels of miR-34a in human colon cancers were also determined, and their possible role in human colon cancer development is discussed below.
|
Results |
|---|
|
TopAbstractResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
miR-34a Induction Depends on p53 Activation. miR-34a expression was increased in a time-dependent manner after ADR treatment, rising 3.2-fold at 8 h and >10-fold at 48 h (Fig. 1A). We also observed that p53 and p21 started to accumulate at 2 and 4 h, respectively, and the accumulation continued until 48 h after treatment (Fig. 1A). HCT 116 p53–/– cells showed no change in expression of miR-34a after ADR treatment (Fig. 1B). To confirm that the induction of miR-34a depends on p53, other human colon cancer cell lines, either with wild-type p53 genes (LoVo and RKO) or mutated p53 genes (DLD1 and HT29), were analyzed (Fig. 1B). As expected, LoVo and RKO cells exhibited increased expression of miR-34a similar to HCT 116 cells, but DLD1 and HT29 cells showed no change, like the HCT 116 p53–/– cells. Accumulation of p53 and p21 was observed in HCT 116, LoVo, and RKO cells, whereas HCT 116 p53–/– cells showed no accumulation, and DLD1 and HT29 cells expressing mutant p53 showed consistent levels of p53 and no accumulation of p21. These results indicate that miR-34a is induced in a p53-dependent manner after ADR treatment.
|
|
We then examined the protein levels of E2F-family proteins, E2F-1, -2, and -3, in HCT 116 and RKO cells. E2F-3 is a predicted target for miR-34a in the databases (PicTar web site, http://pictar.bio.nyu.edu), but no further information on E2F-3 with regard to its biological relationship with miR-34a is available. As shown in Fig. 2C, introduction of miR-34a decreased the accumulation of E2F-1 and -3. E2F-2 was not detectable in either case (data not shown). Accumulation of the p53 and p21 proteins was also observed by introduction of miR-34a in both cell lines, reflecting the results of global gene-expression analysis by microarray system. These observations indicate that introduction of miR-34a causes the down-regulation of the E2F pathway, leading to the up-regulation of the p53/p21 pathway. A possible mechanism for the up-regulation of p53 and its downstream target is discussed below.
miR-34a Induces Senescence-Like Phenotypes.
We observed that introduction of miR-34a in HCT 116 cells caused senescence-like phenotypes with positive staining for senescence-associated 
-galactosidase (SA--gal) and enlarged cellular size (Fig. 2D). RKO cells also showed similar morphological changes with enlarged cellular size by miR-34a introduction, although few SA--gal-positive cells were observed (Fig. 2D). We also observed that p53-mutated human colon cancer cells exhibited morphological changes and SA--gal positive staining characteristic of cellular senescence (SI Fig. 5). These results suggest that suppression of cell proliferation by miR-34a is mainly associated with the induction of senescence-like phenotypes.
Administration of miR-34a with Atelocollagen Suppresses Tumor Growth in Vivo. Atelocollagen has been recently shown to be a very useful system to efficiently deliver small interfering RNA molecules into tumors in vivo (15–17). Subcutaneous administration of miR-34a/atelocollagen complexes caused significant suppression of growth of both HCT 116 and RKO cells by day 4 compared with the administration of the control miRNA/atelocollagen complex (Fig. 3). Significant reduction of tumor volume was also observed for HCT 116 cells until day 6 after the miR-34a administration. The averaged tumor volumes were decreased significantly with administration of miR-34a compared with those that were administered control miRNA throughout the entire experimental period of 14 days. Tumor tissues treated with miR-34a showed a considerable amount of necrotic tissue but showed no significant differences in Ki67 and p21 immunostaining compared with those treated with control miRNA (SI Fig. 6). These results indicate that introduction of miR-34a suppresses the growth of human colon cancer cells to tumors in an in vivo setting as well.
|
|
|
Discussion |
|---|
|
TopAbstractResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
The introduction of miR-34a into two colon cancer cell lines, HCT 116 and RKO, showed a profound inhibition of cell proliferation accompanying the down-regulation of the E2F family. The E2F family is a well characterized group of transcription factors comprising eight subclasses, E2F1–E2F8 (21). E2F1–E2F3 are positive regulators for the cell cycle, whereas E2F4 and -5 negatively regulate, and the functions of E2F6–8 are unknown. Embryonic fibroblasts derived from knockout mice lacking E2f1, -2, or -3 demonstrate slow cell-cycle progression, especially at entry to the S phase (22, 23), but there is not a complete block of the cell cycle and of proliferation. In contrast, fibroblasts with triple knockout of E2f1–E2f3 demonstrate strong inhibition of cell proliferation throughout the cell cycle. Interestingly, induction of p53 and its downstream target p21 was observed in the triple knockout cells (23–25). This phenotype resembles the results obtained in our current study with miR-34a introduction in colon cancer cells.
Among the E2F family, E2F-3 seems to be a strong candidate for the miR-34a targets. Because E2F-3 is reported to be a regulator for E2F-1 gene expression at the transcriptional level (26), repression of other E2F family proteins might thus occur as a subsequent event by miR-34a overexpression. Activation of the p53 pathway could be caused as a consequence of the repression of the E2F family members as in the case of fibroblast with triple knockout of E2f1–E2f3 (23–25) as described above. A positive feedback loop through the p53 pathway for miR-34a could be another important molecular basis for its transcriptional control, explaining the time-dependent increment in miR-34a induction in p53 wild-type colon cancer cells upon continuous exposure to ADR (Fig. 1A).
It is also important to note that introduction of miR-34a represses cell proliferation of p53-knockout HCT 116 and other colon cancer cell lines with mutant p53 (SI Fig. 5). This observation indicates that enforced induction of exogenous miR-34a causes cell growth arrest in a p53-independent manner. The down-regulation of E2F family members, possibly through down-regulation of E2F-3 by overexpression of miR-34a, appears to be a key event for the induction of cellular senescence. Maehara et al. (27), indeed, recently revealed that reduction of E2F/DP activity induces senescence-like phenotypes in human cancer cells. Introduction of miR-34a also caused the up-regulation of the HBP1 gene (SI Table 4), which is associated with oncogenic RAS-induced premature senescence (28). Thus, miR-34a-mediated modulation of several signaling pathways, including E2F- and RAS-related pathways, would collectively and effectively contribute to the induction of senescence-like phenotypes in human colon cancer cells.
Tumor cell growth in nude mice was significantly inhibited for 4–6 days after administration of miR-34a/atelocollagen complexes, but the suppressive effect was less prominent at later time points. Because complexed miRNA with atelocollagen is expected to stay stable for 1 week in this system (17), repeated injections of miR-34a/atelocollagen complexes may be required to exert tumor suppressive effects more efficiently and continuously.
Finally, it is of great interest to note that approximately one-third of human colon cancer specimens revealed down-regulation of the miR-34a expression. Similarly, quite recently, miR-34a was reported to be down-regulated in human primary neuroblastomas with heterozygous deletion of chromosome 1p36, in which miR-34a resides (29). Because the same genomic region is also frequently deleted in colon cancers (30), down-regulation of miR-34a could be associated with the deletion of chromosome 1p36. Furthermore, deletions or mutations of the p53 gene are also suggested to be causative genetic events underlying the down-regulation of miR-34a because its induction is tightly associated with p53 status (Fig. 1B). We indeed observed p53 mutations and chromosomal losses of 17p, including the p53 locus, in some of the colon cancer specimens that showed down-regulation of miR-34a (unpublished observations). Epigenetic inactivation of miR-34a should also be considered as an underlying mechanism. Unexpectedly, we also observed that one-third of colon cancer cases showed up-regulation of miR-34a. There may be alternative mechanisms that remain to be identified in the regulation of miR-34a.
To conclude, the present demonstration of previously uncharacterized biological functions of miR-34a, with the ability to control cell proliferation and induce senescence-like changes in cancer cells, points to its significance as a unique type of tumor suppressor in colon cancer development.
|
Materials and Methods |
|---|
|
TopAbstractResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
ADR was purchased from Sigma–Aldrich (St. Louis, MO). For isolation of total RNA and protein samples from various colon cancer cell lines, cells were seeded at 1.5 x 106 cells per 100-mm dish and treated with ADR at a concentration of 100 ng/ml for 16 h.
For the time course experiment of ADR treatment, HCT 116 cells were seeded at 6 x 105 cells per 60-mm dish and then incubated afterward with or without ADR in the culture medium. Cells were collected and subjected to extraction of total RNA and protein at 0, 1, 2, 4, 8, 12, 24, and 48 h thereafter.
Isolation of miRNA and Quantitative Real-Time RT-PCR Analysis.
Total RNA, including miRNA, was extracted from the cells by using a mirVana miRNA Isolation Kit (Ambion, Austin, TX) according to the manufacturer's instructions and quantified with a NanoDrop spectrophotometer (NanoDrop Technologies, Wilmington, DE). cDNA was synthesized from 10 ng of total RNA by using a High Capacity cDNA Archive kit (Applied Biosystems, Foster City, CA), and the expression levels of 157 mature miRNA species were quantified by using a TaqMan MicroRNA Assays Human Panel (Applied Biosystems). Real-time RT-PCR was performed by using the Applied Biosystems 7300 Sequence Detection system (Applied Biosystems), the expression of each miRNA was defined from the threshold cycle (Ct), and relative expression levels were calculated by using the 2–
Ct method (32) after normalization with reference to expression of U6 small nuclear RNA.
Transient Transfection of Cells with miR-34a and Cell Proliferation Assay. HCT 116 and RKO cells were seeded at a density of 1.0 x 104 cells in 24-well tissue culture plates 24 h before transfection with 5 nM of either miR-34a (Ambion) or nontargeting control miRNA (control miRNA; Ambion) by using HiPerfect transfection reagents (Qiagen, Valencia, CA). The transfection efficiency under the conditions we adopted in this study was estimated to be close to 100% according to our observations made by using a fluorescence-labeled double-stranded short interference RNA. After transfection, cells were harvested and counted every day for 4 days. The average number of cells was determined at each time point in triplicate.
Microarray Analysis. Comprehensive gene-expression analysis of HCT 116 and RKO cells transfected with miR-34a or control miRNA was carried out by using the Whole Human Genome (4 x 44K) Oligo Microarray system (Agilent Technologies). Cells, seeded at 5 x 104 cells per ml in six-well tissue culture plates, were transfected with miR-34a or control miRNA as described above and propagated for 3 days. Total RNAs were isolated with an RNeasy column (Qiagen), and Cy3-labeled cRNA was prepared with an Agilent One Color Spike Mix Kit (Agilent Technologies) and then hybridized by using a Gene Expression Hybridization Kit (Agilent Technologies) at 65°C for 17 h. Signal intensity was calculated from digitized images captured by Laser Scanner (Agilent Technologies), and data analysis was performed by using GeneSpring GX software (Agilent Technologies).
Immunoblot Analysis.
At day 3 after transfection with miR-34a or control miRNA, whole cell lysate was prepared in a lysis buffer (50 mM Tris·HCl, pH 7.4/150 mM NaCl/1% Triton X-100) with a protease inhibitor mixture (Complete Mini; Roche, Indianapolis, IN). Proteins were electrophoresed on a 5–20% linear gradient Tris·HCl-ready gels (Bio-Rad, Hercules, CA) and transferred to polyvinylidene fluoride membranes (Immobiron-P; Millipore, Billerica, MA). Blots were blocked with 3% nonfat dry milk in TBS-T (Tris-buffered saline/0.1% Tween-20, pH 7.4) at room temperature for 30 min. For primary antibodies, we used mouse anti-p53 mAb (1:200, DO-1; Santa Cruz Biotechnology, Santa Cruz, CA), mouse anti-p21 mAb (1:2,000, DCS60; Cell Signaling Technology, Beverly, MA), mouse anti-
-tubulin mAb (1:5,000, DM1A, ICN Biomedicals, Irvine, CA), mouse PARP mAb (1:1,000, C-2–10, Oncogene Research Products, San Diego, CA), mouse anti-caspase-3 mAb (1:1,000, 3G2; Cell Signaling Technology), rabbit anti-E2F-1 polyclonal antibody (pAb) (1:500, C-20; Santa Cruz Biotechnology), rabbit anti-E2F-2 pAb (1:1,000, C-20; Santa Cruz Biotechnology), and rabbit anti-E2F-3 pAb (1:1,000, C-18; Santa Cruz Biotechnology). Horseradish peroxidase-conjugated antibodies against mouse IgG or rabbit IgG (1:5,000–10,000, NA9310V; Amersham Biosciences, Piscataway, NJ) were used as secondary antibodies. Immunoreactive bands on the blots were visualized with enhanced chemiluminescence substrates (Immobilon Western; Millipore).
SA--Gal Staining.
At day 4 after transfection with miR-34a or control miRNA, HCT 116 and RKO cells were fixed and stained by using an SA--gal kit (Cell Signaling Technology) according to the manufacturer's instructions.
Human Tumor Xenograft Model and Administration of miR-34a/Atelocollagen Complexes. Animal experimental protocols were approved by our institute's Committee for Ethics in Animal Experimentation, and the experiments were conducted in accordance with the guidelines for Animal Experiments of the National Cancer Center. HCT 116 or RKO cells were inoculated with 5 x 106 cells per site bilaterally on the backs of female athymic nude mice aged 6 weeks (Charles River Laboratories, Wilmington, MA). Tumor size was monitored by measuring the length and width with calipers, and volumes were calculated with the formula: (L x W2) x 0.5, where L is length and W is width of each tumor.
Atelocollagen was pretreated with pepsin to remove the telopeptides, which confer most of the collagen's antigenicity (33), and complexes of miR-34a (Ambion) and control miRNA (Ambion) with atelocollagen were prepared as described in refs. 15–17. The miR-34a and control miRNA (15 µg) with 0.5% atelocollagen in a 200-µl volume were administered into the s.c. spaces around the tumors when they had reached a volume of 50–100 mm3 at day 6 after inoculation. The volume of each tumor at the point of miR-34a or control miRNA administration was set at 100%, and the relative tumor volumes at each time point was evaluated every other day from days 2 to 14.
Immunohistochemistry. Tumors were fixed in 10% neutralized formalin and embedded in paraffin blocks. Sections (4 µm) were prepared for hematoxylin/eosin staining and also for immunohistochemical examination. After deparaffinization and rehydration, antigen retrieval was performed by microwave irradiation at 600 W for 10 min in 10 mM citrate buffer (pH 6.0). Tissue sections were incubated with rabbit anti-Ki67 polyclonal antibody (pAb) (1:1,000; Novocastra Laboratories, Newcastle, U.K.), mouse anti-p21 mAb (1:100) and rabbit anti-cleaved caspase-3 mAb (1:200, 5A1; Cell Signaling Technology). The sections were then incubated with goat biotinylated anti-rabbit IgG antibody and goat biotinylated anti-mouse IgG antibody (1:200; Vector Laboratories, Burlingame, CA). Immunoreactive signals were visualized by using 3,3'-diaminobenzidine tetrahydrochloride solution, and the nuclei were counterstained with hematoxylin.
Human Colon Cancer Samples.
Primary human colon cancers and paired noncancerous normal colon samples were obtained from 25 patients treated at the National Cancer Center Hospital, Tokyo, with documented informed consent in each case. Samples were frozen in liquid nitrogen and stored at –80°C until use. Total RNA was isolated from frozen samples by using TRIzol reagent (Invitrogen), and then quantitative real-time RT-PCR for miR-34a and U6 small nuclear RNA was performed as described above. The relative expression level of miR-34a in each sample was calculated and quantified by using the 2–Ct method after normalization for expression of U6 small nuclear RNA as described above.
Statistical Analysis. The Mann–Whitney U test was used to test for statistical significance of differences in tumor volumes at each time point after administration of miR-34a and control miRNA. We also used linear mixed effect models (34) to evaluate the differences in averaged volumes of tumors throughout the entire experimental period between the two experimental groups, namely miR-34a- and control miRNA-administered groups. P < 0.05 was considered significant.
During the process of submitting our manuscript, three related papers by Raver-Shapira et al. (35), Chang et al. (36), and He et al. (37) were published.
|
Acknowledgements |
|---|
|
TopAbstractResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
|
Footnotes |
|---|
Abbreviations: miRNA, microRNA; ADR, adriamycin; SA--gal, senescence-associated -galactosidase.
*To whom correspondence should be addressed. E-mail: hnakagam{at}gan2.res.ncc.go.jp
Author contributions: H.T., N.T, and H.N. designed research; H.T., N.T., and M.I. performed research; H.T., N.T., M.I., and H.N. analyzed data; and H.T., N.T., and H.N wrote the paper.
The authors declare no conflict of interest.
This article contains supporting information online at www.pnas.org/cgi/content/full/0707351104/DC1.
© 2007 by The National Academy of Sciences of the USA
|
References |
|---|
|
TopAbstractResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg What's this?
This article has been cited by other articles in HighWire Press-hosted journals:
|
|
 |
|
  I. Ivanovska, A. S. Ball, R. L. Diaz, J. F. Magnus, M. Kibukawa, J. M. Schelter, S. V. Kobayashi, L. Lim, J. Burchard, A. L. Jackson, et al. MicroRNAs in the miR-106b Family Regulate p21/CDKN1A and Promote Cell Cycle Progression Mol. Cell. Biol., April 1, 2008; 28(7): 2167 - 2174. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
, P < 0.05). Significant differences (
, P < 0.05) in the averaged tumor volumes were also observed in HCT 116 and RKO tumors between mir-34a- and control miRNA-administered tumors throughout the entire experimental period by adopting the linear mixed effect model (