- Email: [email protected]
P53-dependent miRNAs mediate nitric oxide-induced apoptosis in colonic carcinogenesis
Author's Accepted Manuscript
P53-dependent miRNAs mediate nitric oxideinduced apoptosis in colonic carcinogenesis Weiwei Li, Wenxiao Han, Yiming Ma, Liang Cui, Yantao Tian, Zhixiang Zhou, Hongying Wang
www.elsevier.com/locate/freeradbiomed
PII: DOI: Reference:
S0891-5849(15)00179-3 http://dx.doi.org/10.1016/j.freeradbiomed.2015.04.016 FRB12395
To appear in:
Free Radical Biology and Medicine
Received date: 17 November 2014 Revised date: 18 March 2015 Accepted date: 13 April 2015 Cite this article as: Weiwei Li, Wenxiao Han, Yiming Ma, Liang Cui, Yantao Tian, Zhixiang Zhou, Hongying Wang, P53-dependent miRNAs mediate nitric oxide-induced apoptosis in colonic carcinogenesis, Free Radical Biology and Medicine, http://dx.doi.org/10.1016/j.freeradbiomed.2015.04.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
P53-dependent miRNAs mediate Nitric Oxide-induced apoptosis in colonic carcinogenesis
Weiwei Li1, Wenxiao Han1, Yiming Ma1, Liang Cui2, Yantao Tian2, Zhixiang Zhou2 and Hongying Wang1
1
State Key Laboratory of Molecular Oncology, 2Department of gastrointestinal cancer
surgery, Cancer Institute/Cancer Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, China.
*Correspondence to: Hongying Wang, 17 Panjiayuan Nanli, Chaoyang District, Beijing,
China
100021.
Phone:
[email protected].
86-10-87787389;
email:
Abstract Both miRNAs and nitric oxide (NO) play important roles in colonic inflammation and tumorigenesis. Resistance of colonic epithelial cells to apoptosis may contribute to tumor development. We hypothesized that some miRNAs could increase the resistance of colonic cancer cells to nitric oxide-induced apoptotic cell death. Here we show that NO induced apoptosis and stimulated expression of some miRNAs. Loss of p53 not only blocked NO-induced apoptosis but also dramatically inhibited the expression of NO-related miRNAs, such as miR-34, -203 and -1301. In addition, blockage of p53-dependent miRNAs significantly reduced NO-induced apoptosis. Furthermore, forced expression of these miRNAs rendered HT-29 cells, which are resistant to apoptosis with mutant p53, more sensitive to NO-induced apoptotic cell death. Most interestingly, in a colitis-associated colon cancer mouse model, the level of miRNAs dropped significantly, accompanied by downregulation of p21 which is a key target gene of p53. In human colorectal cancer samples, the expression of miR-34 significantly correlated with the level of inducible nitric oxide synthase (iNOS). We contend that increased NO production may select cells with low level of p53-dependent miRNAs which contributes to human colonic carcinogenesis and tumor progression.
z z
z
Key words: Nitric Oxide, apoptosis, inflammation, colon cancer, miRNA, p53
Introduction Tumor progression is the culmination of the collective behavior of cells involving multiple cellular processes: survival and proliferation, migration and invasiveness, and cellular ability to induce angiogenesis. Apoptosis is the intrinsic cellular mechanism to eliminate cells that are damaged or transformed. Deregulation of apoptosis is a critical step for cancer, as it allows the genetically unstable cells to survive and accumulate further mutations that eventually lead to tumorigenesis. Critical genetic changes of colorectal carcinoma (CRC) such as ectopic activation of beta-catenin and mutation of p53 can both promote tumorigenesis through evasion of apoptosis [1]. Chronic inflammation plays a critical role during gastrointestinal carcinogenesis. INOS is highly expressed during gastrointestinal inflammation and can generate mutagenic concentrations of nitric oxide (NO), which may be sustained in the micromole range for long periods of time [2]. Prolonged exposure to NO in conditions like chronic inflammation may predispose cells to tumorigenesis. Inhibition or genetic ablation of iNOS can attenuate tumorigenesis in a mouse model of CRC induced by either APC loss of function or colitis [3, 4]. Increasing evidence suggests that iNOS/NO not only contributes to genomic instability and genetic changes [5, 6] but also involves clonal selection through the modulation of apoptosis [7], both of which are the primary mechanisms responsible for tumor initiation. MiRNAs are small non-coding RNAs regulating gene expression at the post-transcriptional and/or translational levels. Accumulating evidence suggests that miRNAs play vital roles in both physiological and pathological processes including carcinogenesis [8]. In general, miRNAs can function either as tumor suppressors or as oncogenes, initiating tumor growth, invasion, metastases, as well as regulating the overall stemness of cancer cells [9]. As the critical player of tumorigenesis, miRNAs perform various functions in the regulation of apoptosis. It has been reported that miR-21, miR-24 and miR-200c function as anti-apoptotic regulators and inhibit FAS-mediated cell death through negative regulation of FasL, FAF1 (Fas-associated factor-1) or FAP-1 (Fas-associated phosphotase-1) respectively [10-12]. On the
contrary, anti-apoptotic genes and proteins such as Bcl-2, Bcl-xL can be targeted by pro-apoptotic miRNAs including miR-491, Let-7, miR-34, miR-143 and miR-195 [13-17]. Although there is some evidence for the coincidence of deregulation of miRNAs and high expression of inflammatory mediators, such as NO, little is known on how miRNAs interact with iNOS/NO to play a role in the transition from inflammation to cancer. It has been demonstrated that inflammation significantly increased miR-21, miR-29b and miR-34a/b/c and decreased miR-29c and miR-181a/c expression in mice, some of which were mediated by iNOS and p53 [18]. These indicated that inflammation modulated miRNA expression in vivo. However,
whether
these
miRNAs
contribute
to
inflammation-related
carcinogenesis is not known. In the current study, we demonstrated that nitric oxide induced apoptosis in colon cancer cells in a p53-dependent manner. MicroRNAs, including miR-34 and miR-1301, mediated NO-induced and p53-mediated apoptosis through targeting Survivin and MET. The perturbation of p53/miRNA mediated apoptosis may contribute to inflammation-related carcinogenesis in colon.
Methods and Materials Cell lines and reagents Human colon carcinoma cell lines HT-29, RKO and RKO-E6 were purchased from American Type Culture Collection. Human colon cancer cell lines HCT116 p53+/+ and HCT116 p53-/- were obtained from Institute of Basic medical Sciences, Chinese Academy of Medical Sciences, Beijing, China. RKO, HT-29, HCT116 p53+/+ and HCT116 p53-/- were grown in complete medium consisting of DMEM/F12 medium (HyClone) supplemented with 10% fetal bovine serum (Gibco) and 1% penicillin-streptomycin solution (HyClone). RKO-E6 cells were maintained in complete medium containing 500µg/mL Geneticin (Life Technologies). Cells were incubated in a humidified 5% CO2 environment at 37 . Mouse monoclonal antibody of p53 was from Santa Cruz Biotech; Rabbit polyclonal antibody of PARP, rabbit monoclonal antibody of MET, anti-mouse and anti-rabbit HRP-linked secondary antibodies were from Cell Signaling Technology. !
Rabbit monoclonal antibody of Survivin was from Abcam Ltd.. The nitric oxide donor diethylenetriamine-NONOate (DETA NONOate) was purchased from Cayman Chemical. Mouse monoclonal antibody of ȕ-actin, NO scavenger Carboxy-PTIO (2-(4-Carboxyphenyl)-4,4,5,5- tetramethylimidazoline-1-oxyl-3-oxide potassium salt) and iNOS inhibitor 1400W were purchased from Sigma Aldrich.
Quantitative RT-PCR Total RNA, extracted using Trizol (Invitrogen) according to the manufacturer’s instructions, was treated with DNase
(Thermo Scientific) and reversely transcribed
with a transcription kit (Thermo Scientific) for mRNA detection, or transcribed with TaKaRaTM microRNA transcription kit for microRNA detection. Real time PCR was performed with a BioRad PCR System. Primers of microRNAs used for real time PCR were purchased from GeneCopoeia (Guangzhou, China). The latest names of human miR-34a, miR-34c, miR-122, miR-149, miR-203 and miR-1301 in miRBase 21.0 are has-miR-34a-5p, has-miR-34c-5p, has-miR-122-5p, has-miR-149-3p, has-miR-203a-3p and has-miR-1301-5p. Mouse miR-34c and miR-203 are named as mmu-miR-34c-5p and mmu-miR-203-3p. Other primers included: human GAPDH, forward: CTCTGCTCCTCCTGTTCGAC, reverse: GCGCCCAATACGACCAAATC; human
MET,
forward:
GAGCGCTTTGTGAGCAGATG,
AACCAGTGGAGAAGTCAGCG;
human
Survivin,
reverse: forward:
GACGACCCCATAGAGGAACA, reverse: CGCACTTTCTCCGCAGTTTC; human iNOS,
forward:
TGGCCATGGAACATCCCAAA,
reverse:
AGGATGTTGTAGCGCTGGAC; human p21, CACCAGACTTCTCTGAGCCC, reverse:
ATGGCGCCTGAACAGAAGAA;
mouse
GAPDH,
forward:
AAGAGGGATGCTGCCCTTAC, reverse: TACGGCCAAATCCGTTCACA; mouse iNOS,
forward:
TCACGCTTGGGTCTTGTTCA,
GGGGAGCCATTTTGGTGACT;
mouse
p21,
reverse: forward:
CTAGGGGAATTGGAGTCAGGC, reverse: CATGAGCGCATCGCAATCAC. The relative levels of microRNA and mRNA transcripts were normalized to U6 or GAPDH (glyceraldehyde-3-phosphate dehydrogenase) expression for microRNA and
mRNA respectively.
Western blot Total protein was extracted with cell extraction buffer (RIPA Lysis Buffer with proteinase inhibitor cocktail and protein phosphatase inhibitor). Protein concentration was quantified by a protein assay kit (Bio-Rad). The equivalent aliquots of protein were separated on 10% or 15% SDS-PAGE gel and transferred to PVDF membranes. The membranes were blocked with 5% nonfat milk for 1 hour before incubation with primary antibodies overnight at 4 . Following incubated with HRP-linked secondary antibodies the immune complexes on the membrane were detected by ECL (Thermo scientific). The results were recorded with LAS-4000 fluorescent chemiluminescence imager (Fujifilm).
Measurement of apoptosis The percentages of apoptotic cells were determined by flow cytometry using sub-G1 and Annexin V/ PI staining assays. For the sub-G1 analysis, both floating and adherent cells were harvested and washed with PBS twice. Cells were fixed in 70% cold ethanol at -20
overnight and
then stained with propidium iodide (1mg/mL) containing RNase A (1mg/mL) at room temperature for 30 min in the dark. Cells were analyzed using a BD LSR
! flow
cytometer. Sub-G1 phase represented the apoptotic population. The Annexin V/ PI staining assay was performed with a detection kit (BD) according to the manufacturer’s instructions. Briefly, floating and adherent cells were harvested and resuspended in binding buffer. Cells (106/mL) were incubated with 5 µl Annexin V-FITC and PI at room temperature for 20 min in the dark. After incubation, the cells were analyzed using flow cytometry. Every experiment was performed in triplicate.
Transfection with mimic or inhibitor of miRNA Transfection was performed using Lipofectamine™ 2000 (Invitrogen) according "
to the manufacturer’s instructions. MicroRNA mimics and inhibitors (Guangzhou RiboBio Co. Ltd., China) were delivered at a final concentration of 50 nM and 100 nM respectively. Twenty four hours after transfection, cells were treated with NO donor. For negative control, cells were transfected with microRNA mimic/inhibitor negative control (Guangzhou RiboBio Co. Ltd.) or Lipofectamine™ 2000 alone.
Human colon cancer samples Human colon cancer tissues and paired normal tissues (5cm adjacent to tumor) were obtained from 57 patients in Cancer Hospital, Chinese Academy of Medical Sciences, Beijing, China. Cancer tissues and matched normal tissues were dissected and submerged in RNAlaterTM (Ambion) for 24 hours and then stored at -80
until
use. All the patients received no chemotherapy before surgical resection for colon cancer and signed informed consent forms for sample collection. This study was approved by the Institutional Review Board of the Chinese Academy of Medical Sciences Cancer Institute.
Mouse models AOM-DSS mouse model is a well-accepted model of colitis-associated colon cancer [19]. Briefly, 6-8-week-old male C57BL/6 mice were injected intraperitoneally (i.p.) with 12.5 mg/kg azoxymethane (AOM, sigma). A week later 2.5% dextran sulfate sodium (DSS, MP Biomedicals) was given in the drinking water for five days, followed by two weeks of regular drinking water. The cycle was repeated three times. Mice were sacrificed after the third or fourth cycle and named as AD3 or AD4 respectively (Figure 4A). Mouse colonic mucosa was collected and saved with RNAlaterTM. To study the effect of iNOS inhibitor on NO-induced miRNA expression in vivo, 6-8-week-old male C57BL/6 mice received 2.5% DSS for five continuous days and were administrated with or without iNOS inhibitor 1400W at the dosage of 10mg/kg/day (i.p.) for another 5 days. And then all animals were sacrificed. Mouse colonic mucosa was collected and saved with RNAlaterTM. Mouse serum was #
collected for the measurement of nitrate/nitrite with Greiss reaction kit (Beyotime Biotechnology China)[20].
Statistical analysis Statistical analysis was carried out with a statistical program GraphPad Prism (GraphPad Software version 5.0). All the data were presented as mean ± SD. Differences were determined by Student’s t test. p
! 0.05
was considered as
statistically significant.
Results 1. NO induced apoptosis followed by the change of miRNAs in colon cancer cells Consistent with previous reports, NO donor significantly induced apoptosis at 48 hours which was measured with sub-G1 analysis (Figure 1A) and Annexin V/PI staining assay in RKO cells (Figure 1B). In addition, western blot showed cleaved PARP after the treatment with 0.25 mM NO donor for 24 and 48 hours (Figure 1C), which confirmed the apoptotic cell death. To test whether miRNAs play roles in NO-induced apoptosis, miRNAs involved in apoptosis were screened according to literature reviews and their expression levels were measured by real time PCR. Six miRNAs (miR-34a, miR-34c, miR-122, miR-149, miR-203 and miR-1301) were found to be remarkably upregulated by NO donor (Figure 1D). To confirm that free radical NO stimulate the expression of miRNAs, decomposed NO donor was used to treat RKO cells and did not show any effect on the expression of miR-34c (Figure 1E). Furthermore, pretreatment with NO scavenger carboxy-PTIO significantly reduced the level of miR-34c induced by NO donor (Figure 1E). Thus, the data strongly suggest that NO induces the expression of miRNAs and it is possible that miRNAs involved in NO-induced apoptosis.
2. Deficiency of p53 blocked NO-induced apoptosis and NO-related change of miRNAs $
Since p53 was involved in NO-induced apoptosis and transcriptional regulation of miRNAs [21, 22], we tested whether p53 mediated NO-induced expression of miRNAs. Firstly, loss of function of p53 in RKO-E6 or HCT116 p53-/- cells not only blocked NO-induced accumulation of p53 (Figure 2A) but also reduced NO-induced apoptosis compared with RKO or HCT116 p53+/+ cells (Figure 2B). Secondly, the time course study showed that the accumulation of p53 preceded the elevation of miRNAs (miR-34c, miR-203 and miR-1301) expression induced by NO donor (Figure 2C). Most importantly, deficiency of p53 abolished the upregulation of miR-34c, miR-203 and miR-1301 induced by NO donor (Figure 2 D-F), but not miR-122 and miR-149 (data not show). All these strongly indicated that p53 mediated NO-induced expression of miRNAs in colon cancer cells.
3. MiRNAs mediated NO-induced apoptosis In order to test whether miR-34c, miR-203 and miR-1301 are required by NO-induced apoptosis, specific inhibitors were used to block the increase of respective miRNAs. It showed that apoptosis induced by NO donor was significantly reduced to a certain degree by the inhibitors of miR-34c, miR-203 or miR-1301 in RKO cells which have wild type p53 (Figure 3A). To test whether p53-dependent miRNAs mediate the resistance to NO-induced apoptosis in colon cancer cells, HT-29 cells were selected for the further study because HT-29 cells had mutated p53 (R273H) and were resistant to NO-induced apoptosis compared with RKO cells (Figure 3B). As expected, miR-34c, miR-203 and miR-1301 were not elevated in HT-29 cells after NO treatment (Figure 3C). Interestingly, overexpression of miR-34c, miR-203 or miR-1301 by transient transfection of their specific mimics significantly sensitized HT-29 cells to NO-induced apoptosis (Figure 3D). This suggested that p53-dependent miRNAs mediated NO-induced apoptosis in colon cancer cells.
4. MiRNAs were negatively correlated with iNOS in mouse model To test whether p53/miRNAs mediates colonic carcinogenesis in vivo, %
colitis-associated colon cancer was generated in the AOM-DSS mouse model. Mice received one intraperitoneal injection of AOM and four cycles of DSS treatment (Figure 4A). Consistent with previous reports, adenoma or hyperplasia was observed in the colon of most mice after three-cycles of treatment of DSS (AD3). After the fourth cycle of DSS, HE staining showed adenocarcinoma in mouse colon (AD4) characterized by the invasion of cancer cells to the submucosa. Therefore, the fourth cycle of DSS treatment induced the transformation of adenoma to adenocarcinoma in mouse colon. We found that the expression of iNOS at mRNA level was significantly higher in both AD3 and AD4 groups compared with control group (Figure 4B). Strikingly, the levels of miR-34c and miR-203 remarkably increased in the AD3 group, but dropped notably in AD4 group compared with AD3 (Figure 4C). Since we have shown that the expression of miR-34c and miR-203 was dependent on the activity of p53, we tested whether p53 became inactivated during the transformation from adenoma to adenocarcinoma. In accordance with expectation, the mRNA expression of p21, an important downstream target gene of p53, reduced significantly in AD4 group compared with AD3 and control group (Figure 4D). To address the causative relationship between iNOS and miRNA expression in vivo, the experiment with iNOS inhibitor 1400W was conducted in DSS-induced colitis mouse model. Firstly, consistent with previous report, we found that iNOS gene expression is strongly induced after the induction of colitis with 2.5% DSS (data not shown). As expected, the level of miR-34c was also significantly higher in colitis (DSS) group compared with control group (Figure 4E). Importantly, treatment with the inhibitor of iNOS activity 1400W dramatically blocked colitis-related upregulation of miR-34c in mouse colon (Figure 4E). The efficiency of inhibitory effect of 1400W on iNOS activity in vivo was confirmed with nitrate/nitrite level in mouse serum (Figure 4F). All these suggested that iNOS regulated the expression of miRNA in vivo and p53/miRNA may contribute to the development of CRC in mouse model.
&
5. MiRNAs were negatively correlated with iNOS in human colon cancer samples To validate whether p53-dependent miRNAs contribute to colonic carcinogenesis in human, the expression of iNOS and miRNAs were tested in human colorectal cancer tissues. Firstly, iNOS expression was measured with real time PCR in 57 pairs CRC specimens. Forty seven percent (27/57) of CRC samples showed higher level of iNOS in tumor site compared with paired normal tissue. We then measured the expression of miR-34c, miR-203 and miR-1301 in these 27 cases. It showed that the level of miR-34c negatively correlated with that of iNOS in human CRC samples (p=0.031) (Figure 5A). The same trend was observed for miR-1301 and iNOS but had no significance (p=0.088) (Figure 5B). In addition, consistent with the mouse model, the mRNA level of p21 was significantly downregulated in tumor tissues compared with matched normal tissues (Figure 5C).
6. The targets of miRNAs in NO-induced apoptosis To identify the downstream targets of miRNAs which mediate apoptosis induced by NO, MET and Survivin were measured in colon cancer cells. Although miRNA mimics successfully elevated the level of the miRNAs respectively (Figure 6A), they had no any effect on the mRNA levels of MET and Survivin in RKO cells (Figure 6B). In contrast, the protein level of Survivin was dramatically inhibited by each miRNA mimic while MET protein was only suppressed by miR-34c mimic (Figure 6C). Furthermore, the inhibitor of miR-34c blocked repression of MET induced by NO donor (Figure 6D). Except for miR-203, inhibition of miR-34c and miR-1301 also abolished downregulation of Survivin induced by NO donor (Figure 6E). All these strongly indicated that miRNAs mediated NO-induced apoptosis likely through the reduction of anti-apoptotic protein MET and Survivin in colon cancer cells.
Discussion Chronic inflammation accelerates oncogenic evolution through various mechanisms. On the one hand, great amounts of reactive oxygen/nitrogen species are
produced and create a high selection pressure during inflammatory process; on the other hand, they induce DNA damage and instability of cell genome. Here we show that NO stress induced apoptosis through p53-dependent overexpression of miR-34, miR-203 and miR-1301 all of which can negatively regulate anti-apoptotic protein Survivin and MET. When p53 mutated and became inactivated, interference to NO-induced miRNA expression mitigated the downregulation of Survivin and MET and then reduced apoptosis, which may contribute to malignant transformation in vivo. Our study reveals a new mechanism by which miRNAs contribute to clonal selection under NO stress during inflammation-associated tumorigenesis. Tumor suppressor p53 plays a central role in tumorigenesis. Mutation of p53 even occurs as an early event in inflammation-related colonic carcinogenesis in human [23]. Consistent with previous report, we showed in the current study that p53 became inactivated during the transformation from adenoma to adenocarcinoma in AOM-DSS model, which was supported by the abolishment of p21 mRNA expression. Notably, it has been demonstrated that p53 is an important transcription repressor of iNOS expression in vivo [24]. That is to say inactivated mutation of p53 would enhance the excessive NO production. Therefore, inactivation of p53 not only influenced miRNA expression and apoptosis but also worsened the reactive nitrogen stress. All these accelerated the carcinogenesis under inflammation. As a new class of p53 target genes, miRNAs are important components in the p53 network and crosstalk with p53 at multiple levels. First of all, p53 regulates the transcription and the maturation of a group of miRNAs [25]. In the current study, we showed that miR-34, miR-203 and miR-1301 were regulated by NO in a p53 dependent manner in colon cancer cells. Actually, p53 regulates the expression of each miRNA at different levels. For example, it can directly bind to the promoter and activates transcription of miR-34, which includes miR-34a, miR-34b and miR-34c [26]. In contrast, p53 has been shown to promote the maturation of miR-203 through augmentation of acetylation-dependent Drosha-mediated processing [27]. In current study we show by the first time that NO induced miR-1301 expression in a p53-dependent manner in colon cancer cells. However, the molecular mechanism
behind it needs further study in the future. On the other hand, miRNAs can regulate the activity and function of p53 through direct repression of p53 or its regulators in cells. It has been demonstrated that miR-34a positively regulated p53 function in apoptosis through direct negative regulation of SIRT1, which is a negative regulator of p53[26]. Overexpression of miR-34a decreased SIRT1 expression, leading to the increase in acetylated p53 levels and p53 activity, which in turn induced apoptosis in a p53-dependent manner. In addition, miR-1301 mimic could increase the expression of p53 at mRNA and protein levels through an unknown mechanism in HpG2 cells [28]. MiRNAs have emerged as a class of abundant and critical regulatory genes in human tumorigenesis. Over 2000 miRNAs have been identified in human. Aberrant miRNA expression, and amplification or deletion of miRNAs has been frequently observed in various human tumors. Here, we demonstrated that inactivation of the tumor suppressor p53 might also participate in tumorigenesis through the regulation of miRNAs and apoptosis. On the contrary, dysregulation of miRNAs has been suggested to be an important mechanism for the aberrant expression of a list of oncogenes and tumor suppressor genes which are not affected by genetic mutations or transcriptional regulation in tumorigenesis. More importantly, each miRNA is predicted to have numerous targets that have disparate functions. The miRNAs which were found to involve in inflammation-associated carcinogenesis in the current study, participated in the multiple aspect of carcinogenesis besides apoptosis. Low levels of the miR-34 family (which includes miR-34a, miR-34b) and miR-34c has been frequently observed in various tumors, including CRC [29]. Inactivation of endogenous miR-34a stimulates cell proliferation and inhibits p53-dependent apoptosis through targeting cyclin E2, cyclin-dependent kinases 4/6 (CDK4/6), and Bcl-2[29]. In addition, miR-34 plays an important role in cell fate determination. It was shown that miR-34 involved in self-renewal of cancer stem cell potentially via the downstream targets Bcl-2 and Notch [30]. Since miR-34 can partly restore the function of p53, it may hold significant promise as a novel molecular therapy for human cancer with loss of p53-miR-34 [30].
Different lines of evidence support the notion that miR-203 functions as a tumor suppressor. In some types of cancer, miR-203 is epigenetically silenced, and the silencing promotes tumor cell growth and invasion through different targets [31]. MiR-203 has been shown to promote apoptosis by targeting anti-apoptotic Survivin [32] and Bcl-w in a p53-dependent manner [27]. In addition, miR-203 suppressed proliferation and migration through the inhibition of SRC translation in lung cancer cells [33]. It was found recently that miR-203 directly targets Lef1, which is one of the co-transcriptional factors and mediates Wnt/beta-catenin-induced transcription [34]. Moreover, miR-203 mediates the modulation of stemness and EMT. By coordinating with miR-200c, miR-203 can suppress the expression of Bmi1 and other stem cell factors [35] [36]. Importantly, overexpression of SNAI1 induced EMT and suppressed miR-203, which repressed endogenous SNAI1, forming a double negative miR-203/SNAI1 feedback loop. This miR-203/SNAI1 with the known miR-200/ZEB feedback loops construct an a priori EMT core network [37]. The information on miR-1301 is limited. Consistent with our findings, it has been indicated that miR-1301 may be an inhibitor of tumorigenesis in HepG2 cells. MiR-1301 mimics promoted hepatocellular apoptosis through down-regulating Bcl-2 and Bcl-xL mRNA and protein expression [38]. Notably, miR-1301 mimics induced the upregulation of p53 [38]. It implied that miR-1301 might crosstalk with p53 network. However, further studies are required to elucidate the underlying mechanisms. Taken together, we demonstrated here that p53-dependent miRNAs mediated nitric oxide-induced apoptosis through targeting Survivin and MET in colon cancer cells. The perturbation of p53/miRNA mediated apoptosis may contribute to inflammation-related carcinogenesis in colon. MiRNAs may be good candidates as biomarkers to monitor the transformation from chronic inflammation to cancer in colon.
!
Acknowledgement This work was supported by funding from the 973 National Key Fundamental Research Program of China (2012CB967003) and the National Nature Science Foundation of China (91129717 and 81172034).
Figure legends Figure 1. NO induced apoptosis and changed the expression of miRNAs in colon cancer cells. RKO cells were treated with NO donor, DETA NONOate, at different concentrations for 48 hours and then were collected to measure apoptosis by flow cytometry using (A) sub-G1 analysis and (B) Annexin V/ PI staining assay. (C) Cleaved PARP was detected by western blot in RKO cells after treatment with DETA NONOate at 0.25 mM for 0, 6, 12, 24 or 48 hours. (D) The levels of miRNAs were measured by real time PCR in RKO cells after treatment with 0.25 mM DETA NONOate for 24 hours. (E) DETA NONOate (0.25mM) was left at 37
!"for 72
hours and then used as “composed NO donor”. Carboxy-PTIO at concentration of 0.5mM was used to scavenge NO in the cell culture medium. RKO cells were exposed to decomposed NO donor, DETA NONOate (0.25 mM), Carboxy-PTIO alone or DETA NONOate combination with Carboxy-PTIO for 24 hours and then the
"
level of miR-34c was measured by real time PCR. The data showed the means ± SD from three independent experiments.
##"p<0.01 and ###"p<0.001.
Figure 2. P53 mediated NO-induced apoptosis and the change of miRNAs. RKO and HCT116 cells were treated with or without DETA NONOate for 24 hours at 0.25 mM or 0.5 mM respectively. Then the cells were collected to analyze (A) p53 by western blot, and (B) apoptosis by flow cytometry. (C) RKO cells were treated with DETA NONOate for different time and then were collected to analyze p53 by western blot
(insert) and the expression of miRNAs by real time PCR. (D-F) RKO and RKO-E6 cells were treated with DETA NONOate at 0.25 mM for 24 hours and then were collected to measure the levels of miRNAs by real time PCR. The data showed as the means ± SD of three independent experiments. ! p<0.05), !! p <0.01 and !!! p <0.001.
Figure 3. MiRNAs mediated NO-induced apoptosis. (A) RKO cells were transfected with the specific inhibitors of miR-34c, miR-203 or miR-1301 at 100 nM. Twenty four hours later, cells were treated with DETA NONOate at 0.25 mM for 48 hours and then were collected to analyze apoptosis with Annexin V/ PI staining assay. (B) Apoptosis was analyzed with Annexin V/ PI staining assay in HT-29 and RKO cells after treatment with DETA NONOate for 48 hours. (C) MiRNA expression was measured by real time PCR in HT-29 cells 24 hours after treatment with DETA NONOate. (D) After transfection with the specific mimics of miR-34c, miR-203 or miR-1301 respectively, HT-29 cells were treated with DETA NONOate at 0.5 mM for 48 hours and then were collected to analyze apoptosis with Annexin V/ PI staining assay. ! p <0.05, !! p <0.01 and !!! p <0.001, n.s. no significance.
Figure 4. MiRNAs were negatively correlated with iNOS in mouse model. (A) Schematic overview of AOM-DSS mouse model of colitis-associated colon cancer. The expression levels of (B) iNOS mRNA and (C) miRNAs in mouse colon were measured by real time PCR. (D) The mRNA expression of p21, an important downstream target gene of p53, was measured by regular PCR in mouse colon. GAPDH was used as a reference. (E-F) Mice received 2.5% DSS for five continuous days and were administrated with or without iNOS inhibitor 1400W at the dosage of 10mg/kg/day (i.p.) for another 5 days.
Then all animals were sacrificed. (E) The
expression of miR-34c in mouse colon was measured with real time PCR and (F) the level of nitrate/nitrite in mouse serum was measured with Greiss reaction kit. N>7 and ! p <0.05, !! p <0.01 and !!! p <0.001.
"
Figure 5. MiRNAs were negatively correlated with iNOS in human colon cancer samples. The expression of miR-34c, miR-203 and miR-1301 were measured by real time PCR in 27 paired human colorectal cancer tissues. The correlation between (A) iNOS and miR-34c and (B) iNOS and miR-1301 were analyzed. (C) The mRNA expression of p21 was measured by real time PCR. !! p <0.01.
Figure 6. The targets of miRNAs in NO-induced apoptosis. (A) Fourty eight hours after transfection with the specific mimics of miR-34c, miR-203 or miR-1301 respectively, RKO cells were collected to measure the expression of miRNAs. The levels of MET and Survivin were measured at (B) mRNA and (C) protein levels. !! p <0.01 and !!! p <0.001. (D and E) Twenty four hours after transfection with the specific inhibitors of miR-34c, miR-203 or miR-1301 respectively, RKO cells were treated with or without DETA NONOate at 0.25 mM for another 24 hours and then were collected to analyze the expression of MET and Survivin by western blot.
'(
)* + , + - ./
0 1ϲϱ2$"&!#1&&
)* ( - , 30 3 '
1Ϯ2!% "1&&
) * - , 40 , 5 5 , 6 7 0 , 6 , -0 5 ' , ..#! ' 8 9 :;< 1ϭϮϭ2 &" 1&&#
)!* 6 = , ( > , 5 ? 5 3 /
' 8 ' / 1ϰϲ2 ! 1&&#
) * 68 4 , . / = '
1ϭϰ2%&&1&&"
)"* 60 ,' ( > ,3 / ,3 / ,+> ,+5 3 @0A / 3
.6 1ϭϬϵ2!&&
#
)#* 6 + ,- , 7 , 0 > + ,0 - ,+ B ' /
1ϵ2!%$%1&&#
)$* 6 ,68
A ,4' ,@C ,- ,D8 0 + 1ϰϱϱ2 $" 1&&$
)%* 5 , ( + , + ' / ' 1&
)&* 60 , 06 , E , E && ' 3=% @ 1ϯϴ2%&$ 1&&
)* F+ ,65 ,CA ,53 ,>( , > ! '9.@<'@@
1ϱ2%!%1&&
)* + ,C ( ,C> ,@> ,( > ,3 4 / / ' / ' / ,
' @ ( 1ϳ2
!! 1&
) * ( ( , 3 ( , B 5 , ( , = B %
'
!
1ϰϬϬ2 "!&1&&
)!* A ,6 E ,? 6 E ,( 4 ,3/4 ,@= ! /
8 ' ' 8 ' ''
1ϲ2 #$#1&
) * > F , B , C , 4 + , 5 , ( ( !
1ϰ2"$"1&&%
)"* 67 6 , 40 4 ,0 ,-4 , 4 , 4 # ' ' A(
' " 1ϱϮ2"%$#&!1&&
)#* ? ,+5 ,6+ 5 ,( A , 6 F ,( ( !% 4 AB( 6+%%& 8 1ϭϳ2!#
!#1&
)$* E , ? , @ 7 , , 0 , 3 3 G'
8 1ϭϯϭ2#"& 1&
)%* ? @ , E0 ( , ? 4 @ , 0 > , ( C + , E ( > G-- 0 ' '
1ϭϭϴ2$ %"1&&!
)&* 6 F ,5> 6 ,C5 5 ,6 3 3 ,( = ,+ 5 6 '
$
0
/
'
1Ϯϴϳ2 "!##1&
)* 304 , +C , 6 ,( - ,+ 0 6 3 ,H/ 75 7 8 ' / 1ϲϰ2$& 1&&!
)* 0 802 '
#1ϭϮ2" "1&
) * G 708 7 6 ' ' 8 $ 1ϯϱ2 #1&&"
)!* 6 , . ' 0 . , + ? , A + , A 4 , 3 3 I '
00 1ϵϱ2$$ $1%%$
) * @ C , C 3 , + , + 4
1ϯ2!! &1&
)"* 50 ,@ ,(8 3 > !
'6G4 1ϭϬϱ2 !"1&&$
)#* 3> ,=/ = A ,- ,>B , ,6
'
//'8
% 1ϯϮ2 %& 1&
)$* @ ( , 5 , > , @ 5 , 5 ? 6 & ' 1Ϯϳ2%% !1&
)%* 478 , 4 , G7 , 0 4
/ !
0 8 ' E@ 8 1ϭϬϰ2 !##1&&#
) &* > F , B , C , 4 + , 5 , ( ( !
1ϰ2"$"1&&%
) * C C , C A , ( + , @ ( , @ ( , C C E 6 ' & I 6G 3 G/ /
' A 3 3 $ 1Ϯ2#$%1&
) * B = , + F , 5 , B ( & ' ,
' 6 // 1ϴ2 #%$!1&
)
* + , ( , C 5 , + 3 , C 6 , 5 &
' ' 63
1ϵ2& #&1&!
%
) !* 4 E > , G , - - , > ? '7 ' ' 1ϭϬϱ2$ $!%1&&$
) * + I , 6 > , A 0 I 3 , 6' . , C @ , 6 4 E 4 / CEA
1ϭϭ2!$#% 1&&%
) "* 5 > , C ? , > B , C C , C + , 6 5 A
0 '
1ϴ2# "$1&
) #* H , 3 , ( A > , A > ( , 3 - , 378 6 ! '' 8
'
1ϳ2!$1&
) $* @ ( , 5 , > , @ 5 , 5 ? 6 & ' 1Ϯϳ2%% !1&
&
Graphical Abstract (for review)
Survivin and MET
Survive
Apoptosis
X
miRNAs
X
p53
Survivin and MET
miRNAs
p53
Nitric Oxide
Figure
D
C
A
miR-34a
miR-34c
miR-122
β-actin
Time (h) PARP˄116KDa˅ Cleaved PARP˄89KDa˅
miR-149
0
B
miR-1301
E
0.25mM NO donor 12 24
Figure 1
miR-203
6
control
48
miR-34c
0.25 mM NO donor
β-actin
p53
NO donor
C
B
A RKO + ˉ
RKO-E6 + ˉ
β-actin
Time (h) p53
ˉ
6
12
24
0.25mM NO donor
ˉ
48
+
HCT-116 p53 -/-
Figure 2
0
+
HCT-116 p53 +/+
F
E
D
miR-1301
miR-203
miR-34c
D
C
Figure 3
B
A
B
A
D
GAPDH
p21
1
2
DSS
control
Time˄weeks˅
AOM 3
4
5
6
C
7
AD3
DSS 8
DSS 9
DSS
AD3
11
12
14
miR-203
AD4
AD4
13
Figure 4
miR-34c
10
F
E miR-34c
C
A
P= 0.031 r= -0.415
Figure 5
B
P= 0.088 r= -0.334
D
A
ˇ
Figure 6
ˇ
—
ˇ
miR-34c
β-actin
Survivin
MET
mimic
negative control
C
NO donor —
inhibitor
β-actin
—
miR-34c inhibitor
E
β-actin
ˇ
miR-1301
Survivin
—
negative control
miR-203
MET
NO donor
miR-34c
B
—
ˇ
miR-203
—
ˇ
miR-1301
P53-dependent miRNAs mediate nitric oxideinduced apoptosis in colonic carcinogenesis Weiwei Li, Wenxiao Han, Yiming Ma, Liang Cui, Yantao Tian, Zhixiang Zhou, Hongying Wang
www.elsevier.com/locate/freeradbiomed
PII: DOI: Reference:
S0891-5849(15)00179-3 http://dx.doi.org/10.1016/j.freeradbiomed.2015.04.016 FRB12395
To appear in:
Free Radical Biology and Medicine
Received date: 17 November 2014 Revised date: 18 March 2015 Accepted date: 13 April 2015 Cite this article as: Weiwei Li, Wenxiao Han, Yiming Ma, Liang Cui, Yantao Tian, Zhixiang Zhou, Hongying Wang, P53-dependent miRNAs mediate nitric oxide-induced apoptosis in colonic carcinogenesis, Free Radical Biology and Medicine, http://dx.doi.org/10.1016/j.freeradbiomed.2015.04.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
P53-dependent miRNAs mediate Nitric Oxide-induced apoptosis in colonic carcinogenesis
Weiwei Li1, Wenxiao Han1, Yiming Ma1, Liang Cui2, Yantao Tian2, Zhixiang Zhou2 and Hongying Wang1
1
State Key Laboratory of Molecular Oncology, 2Department of gastrointestinal cancer
surgery, Cancer Institute/Cancer Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, China.
*Correspondence to: Hongying Wang, 17 Panjiayuan Nanli, Chaoyang District, Beijing,
China
100021.
Phone:
[email protected].
86-10-87787389;
email:
Abstract Both miRNAs and nitric oxide (NO) play important roles in colonic inflammation and tumorigenesis. Resistance of colonic epithelial cells to apoptosis may contribute to tumor development. We hypothesized that some miRNAs could increase the resistance of colonic cancer cells to nitric oxide-induced apoptotic cell death. Here we show that NO induced apoptosis and stimulated expression of some miRNAs. Loss of p53 not only blocked NO-induced apoptosis but also dramatically inhibited the expression of NO-related miRNAs, such as miR-34, -203 and -1301. In addition, blockage of p53-dependent miRNAs significantly reduced NO-induced apoptosis. Furthermore, forced expression of these miRNAs rendered HT-29 cells, which are resistant to apoptosis with mutant p53, more sensitive to NO-induced apoptotic cell death. Most interestingly, in a colitis-associated colon cancer mouse model, the level of miRNAs dropped significantly, accompanied by downregulation of p21 which is a key target gene of p53. In human colorectal cancer samples, the expression of miR-34 significantly correlated with the level of inducible nitric oxide synthase (iNOS). We contend that increased NO production may select cells with low level of p53-dependent miRNAs which contributes to human colonic carcinogenesis and tumor progression.
z z
z
Key words: Nitric Oxide, apoptosis, inflammation, colon cancer, miRNA, p53
Introduction Tumor progression is the culmination of the collective behavior of cells involving multiple cellular processes: survival and proliferation, migration and invasiveness, and cellular ability to induce angiogenesis. Apoptosis is the intrinsic cellular mechanism to eliminate cells that are damaged or transformed. Deregulation of apoptosis is a critical step for cancer, as it allows the genetically unstable cells to survive and accumulate further mutations that eventually lead to tumorigenesis. Critical genetic changes of colorectal carcinoma (CRC) such as ectopic activation of beta-catenin and mutation of p53 can both promote tumorigenesis through evasion of apoptosis [1]. Chronic inflammation plays a critical role during gastrointestinal carcinogenesis. INOS is highly expressed during gastrointestinal inflammation and can generate mutagenic concentrations of nitric oxide (NO), which may be sustained in the micromole range for long periods of time [2]. Prolonged exposure to NO in conditions like chronic inflammation may predispose cells to tumorigenesis. Inhibition or genetic ablation of iNOS can attenuate tumorigenesis in a mouse model of CRC induced by either APC loss of function or colitis [3, 4]. Increasing evidence suggests that iNOS/NO not only contributes to genomic instability and genetic changes [5, 6] but also involves clonal selection through the modulation of apoptosis [7], both of which are the primary mechanisms responsible for tumor initiation. MiRNAs are small non-coding RNAs regulating gene expression at the post-transcriptional and/or translational levels. Accumulating evidence suggests that miRNAs play vital roles in both physiological and pathological processes including carcinogenesis [8]. In general, miRNAs can function either as tumor suppressors or as oncogenes, initiating tumor growth, invasion, metastases, as well as regulating the overall stemness of cancer cells [9]. As the critical player of tumorigenesis, miRNAs perform various functions in the regulation of apoptosis. It has been reported that miR-21, miR-24 and miR-200c function as anti-apoptotic regulators and inhibit FAS-mediated cell death through negative regulation of FasL, FAF1 (Fas-associated factor-1) or FAP-1 (Fas-associated phosphotase-1) respectively [10-12]. On the
contrary, anti-apoptotic genes and proteins such as Bcl-2, Bcl-xL can be targeted by pro-apoptotic miRNAs including miR-491, Let-7, miR-34, miR-143 and miR-195 [13-17]. Although there is some evidence for the coincidence of deregulation of miRNAs and high expression of inflammatory mediators, such as NO, little is known on how miRNAs interact with iNOS/NO to play a role in the transition from inflammation to cancer. It has been demonstrated that inflammation significantly increased miR-21, miR-29b and miR-34a/b/c and decreased miR-29c and miR-181a/c expression in mice, some of which were mediated by iNOS and p53 [18]. These indicated that inflammation modulated miRNA expression in vivo. However,
whether
these
miRNAs
contribute
to
inflammation-related
carcinogenesis is not known. In the current study, we demonstrated that nitric oxide induced apoptosis in colon cancer cells in a p53-dependent manner. MicroRNAs, including miR-34 and miR-1301, mediated NO-induced and p53-mediated apoptosis through targeting Survivin and MET. The perturbation of p53/miRNA mediated apoptosis may contribute to inflammation-related carcinogenesis in colon.
Methods and Materials Cell lines and reagents Human colon carcinoma cell lines HT-29, RKO and RKO-E6 were purchased from American Type Culture Collection. Human colon cancer cell lines HCT116 p53+/+ and HCT116 p53-/- were obtained from Institute of Basic medical Sciences, Chinese Academy of Medical Sciences, Beijing, China. RKO, HT-29, HCT116 p53+/+ and HCT116 p53-/- were grown in complete medium consisting of DMEM/F12 medium (HyClone) supplemented with 10% fetal bovine serum (Gibco) and 1% penicillin-streptomycin solution (HyClone). RKO-E6 cells were maintained in complete medium containing 500µg/mL Geneticin (Life Technologies). Cells were incubated in a humidified 5% CO2 environment at 37 . Mouse monoclonal antibody of p53 was from Santa Cruz Biotech; Rabbit polyclonal antibody of PARP, rabbit monoclonal antibody of MET, anti-mouse and anti-rabbit HRP-linked secondary antibodies were from Cell Signaling Technology. !
Rabbit monoclonal antibody of Survivin was from Abcam Ltd.. The nitric oxide donor diethylenetriamine-NONOate (DETA NONOate) was purchased from Cayman Chemical. Mouse monoclonal antibody of ȕ-actin, NO scavenger Carboxy-PTIO (2-(4-Carboxyphenyl)-4,4,5,5- tetramethylimidazoline-1-oxyl-3-oxide potassium salt) and iNOS inhibitor 1400W were purchased from Sigma Aldrich.
Quantitative RT-PCR Total RNA, extracted using Trizol (Invitrogen) according to the manufacturer’s instructions, was treated with DNase
(Thermo Scientific) and reversely transcribed
with a transcription kit (Thermo Scientific) for mRNA detection, or transcribed with TaKaRaTM microRNA transcription kit for microRNA detection. Real time PCR was performed with a BioRad PCR System. Primers of microRNAs used for real time PCR were purchased from GeneCopoeia (Guangzhou, China). The latest names of human miR-34a, miR-34c, miR-122, miR-149, miR-203 and miR-1301 in miRBase 21.0 are has-miR-34a-5p, has-miR-34c-5p, has-miR-122-5p, has-miR-149-3p, has-miR-203a-3p and has-miR-1301-5p. Mouse miR-34c and miR-203 are named as mmu-miR-34c-5p and mmu-miR-203-3p. Other primers included: human GAPDH, forward: CTCTGCTCCTCCTGTTCGAC, reverse: GCGCCCAATACGACCAAATC; human
MET,
forward:
GAGCGCTTTGTGAGCAGATG,
AACCAGTGGAGAAGTCAGCG;
human
Survivin,
reverse: forward:
GACGACCCCATAGAGGAACA, reverse: CGCACTTTCTCCGCAGTTTC; human iNOS,
forward:
TGGCCATGGAACATCCCAAA,
reverse:
AGGATGTTGTAGCGCTGGAC; human p21, CACCAGACTTCTCTGAGCCC, reverse:
ATGGCGCCTGAACAGAAGAA;
mouse
GAPDH,
forward:
AAGAGGGATGCTGCCCTTAC, reverse: TACGGCCAAATCCGTTCACA; mouse iNOS,
forward:
TCACGCTTGGGTCTTGTTCA,
GGGGAGCCATTTTGGTGACT;
mouse
p21,
reverse: forward:
CTAGGGGAATTGGAGTCAGGC, reverse: CATGAGCGCATCGCAATCAC. The relative levels of microRNA and mRNA transcripts were normalized to U6 or GAPDH (glyceraldehyde-3-phosphate dehydrogenase) expression for microRNA and
mRNA respectively.
Western blot Total protein was extracted with cell extraction buffer (RIPA Lysis Buffer with proteinase inhibitor cocktail and protein phosphatase inhibitor). Protein concentration was quantified by a protein assay kit (Bio-Rad). The equivalent aliquots of protein were separated on 10% or 15% SDS-PAGE gel and transferred to PVDF membranes. The membranes were blocked with 5% nonfat milk for 1 hour before incubation with primary antibodies overnight at 4 . Following incubated with HRP-linked secondary antibodies the immune complexes on the membrane were detected by ECL (Thermo scientific). The results were recorded with LAS-4000 fluorescent chemiluminescence imager (Fujifilm).
Measurement of apoptosis The percentages of apoptotic cells were determined by flow cytometry using sub-G1 and Annexin V/ PI staining assays. For the sub-G1 analysis, both floating and adherent cells were harvested and washed with PBS twice. Cells were fixed in 70% cold ethanol at -20
overnight and
then stained with propidium iodide (1mg/mL) containing RNase A (1mg/mL) at room temperature for 30 min in the dark. Cells were analyzed using a BD LSR
! flow
cytometer. Sub-G1 phase represented the apoptotic population. The Annexin V/ PI staining assay was performed with a detection kit (BD) according to the manufacturer’s instructions. Briefly, floating and adherent cells were harvested and resuspended in binding buffer. Cells (106/mL) were incubated with 5 µl Annexin V-FITC and PI at room temperature for 20 min in the dark. After incubation, the cells were analyzed using flow cytometry. Every experiment was performed in triplicate.
Transfection with mimic or inhibitor of miRNA Transfection was performed using Lipofectamine™ 2000 (Invitrogen) according "
to the manufacturer’s instructions. MicroRNA mimics and inhibitors (Guangzhou RiboBio Co. Ltd., China) were delivered at a final concentration of 50 nM and 100 nM respectively. Twenty four hours after transfection, cells were treated with NO donor. For negative control, cells were transfected with microRNA mimic/inhibitor negative control (Guangzhou RiboBio Co. Ltd.) or Lipofectamine™ 2000 alone.
Human colon cancer samples Human colon cancer tissues and paired normal tissues (5cm adjacent to tumor) were obtained from 57 patients in Cancer Hospital, Chinese Academy of Medical Sciences, Beijing, China. Cancer tissues and matched normal tissues were dissected and submerged in RNAlaterTM (Ambion) for 24 hours and then stored at -80
until
use. All the patients received no chemotherapy before surgical resection for colon cancer and signed informed consent forms for sample collection. This study was approved by the Institutional Review Board of the Chinese Academy of Medical Sciences Cancer Institute.
Mouse models AOM-DSS mouse model is a well-accepted model of colitis-associated colon cancer [19]. Briefly, 6-8-week-old male C57BL/6 mice were injected intraperitoneally (i.p.) with 12.5 mg/kg azoxymethane (AOM, sigma). A week later 2.5% dextran sulfate sodium (DSS, MP Biomedicals) was given in the drinking water for five days, followed by two weeks of regular drinking water. The cycle was repeated three times. Mice were sacrificed after the third or fourth cycle and named as AD3 or AD4 respectively (Figure 4A). Mouse colonic mucosa was collected and saved with RNAlaterTM. To study the effect of iNOS inhibitor on NO-induced miRNA expression in vivo, 6-8-week-old male C57BL/6 mice received 2.5% DSS for five continuous days and were administrated with or without iNOS inhibitor 1400W at the dosage of 10mg/kg/day (i.p.) for another 5 days. And then all animals were sacrificed. Mouse colonic mucosa was collected and saved with RNAlaterTM. Mouse serum was #
collected for the measurement of nitrate/nitrite with Greiss reaction kit (Beyotime Biotechnology China)[20].
Statistical analysis Statistical analysis was carried out with a statistical program GraphPad Prism (GraphPad Software version 5.0). All the data were presented as mean ± SD. Differences were determined by Student’s t test. p
! 0.05
was considered as
statistically significant.
Results 1. NO induced apoptosis followed by the change of miRNAs in colon cancer cells Consistent with previous reports, NO donor significantly induced apoptosis at 48 hours which was measured with sub-G1 analysis (Figure 1A) and Annexin V/PI staining assay in RKO cells (Figure 1B). In addition, western blot showed cleaved PARP after the treatment with 0.25 mM NO donor for 24 and 48 hours (Figure 1C), which confirmed the apoptotic cell death. To test whether miRNAs play roles in NO-induced apoptosis, miRNAs involved in apoptosis were screened according to literature reviews and their expression levels were measured by real time PCR. Six miRNAs (miR-34a, miR-34c, miR-122, miR-149, miR-203 and miR-1301) were found to be remarkably upregulated by NO donor (Figure 1D). To confirm that free radical NO stimulate the expression of miRNAs, decomposed NO donor was used to treat RKO cells and did not show any effect on the expression of miR-34c (Figure 1E). Furthermore, pretreatment with NO scavenger carboxy-PTIO significantly reduced the level of miR-34c induced by NO donor (Figure 1E). Thus, the data strongly suggest that NO induces the expression of miRNAs and it is possible that miRNAs involved in NO-induced apoptosis.
2. Deficiency of p53 blocked NO-induced apoptosis and NO-related change of miRNAs $
Since p53 was involved in NO-induced apoptosis and transcriptional regulation of miRNAs [21, 22], we tested whether p53 mediated NO-induced expression of miRNAs. Firstly, loss of function of p53 in RKO-E6 or HCT116 p53-/- cells not only blocked NO-induced accumulation of p53 (Figure 2A) but also reduced NO-induced apoptosis compared with RKO or HCT116 p53+/+ cells (Figure 2B). Secondly, the time course study showed that the accumulation of p53 preceded the elevation of miRNAs (miR-34c, miR-203 and miR-1301) expression induced by NO donor (Figure 2C). Most importantly, deficiency of p53 abolished the upregulation of miR-34c, miR-203 and miR-1301 induced by NO donor (Figure 2 D-F), but not miR-122 and miR-149 (data not show). All these strongly indicated that p53 mediated NO-induced expression of miRNAs in colon cancer cells.
3. MiRNAs mediated NO-induced apoptosis In order to test whether miR-34c, miR-203 and miR-1301 are required by NO-induced apoptosis, specific inhibitors were used to block the increase of respective miRNAs. It showed that apoptosis induced by NO donor was significantly reduced to a certain degree by the inhibitors of miR-34c, miR-203 or miR-1301 in RKO cells which have wild type p53 (Figure 3A). To test whether p53-dependent miRNAs mediate the resistance to NO-induced apoptosis in colon cancer cells, HT-29 cells were selected for the further study because HT-29 cells had mutated p53 (R273H) and were resistant to NO-induced apoptosis compared with RKO cells (Figure 3B). As expected, miR-34c, miR-203 and miR-1301 were not elevated in HT-29 cells after NO treatment (Figure 3C). Interestingly, overexpression of miR-34c, miR-203 or miR-1301 by transient transfection of their specific mimics significantly sensitized HT-29 cells to NO-induced apoptosis (Figure 3D). This suggested that p53-dependent miRNAs mediated NO-induced apoptosis in colon cancer cells.
4. MiRNAs were negatively correlated with iNOS in mouse model To test whether p53/miRNAs mediates colonic carcinogenesis in vivo, %
colitis-associated colon cancer was generated in the AOM-DSS mouse model. Mice received one intraperitoneal injection of AOM and four cycles of DSS treatment (Figure 4A). Consistent with previous reports, adenoma or hyperplasia was observed in the colon of most mice after three-cycles of treatment of DSS (AD3). After the fourth cycle of DSS, HE staining showed adenocarcinoma in mouse colon (AD4) characterized by the invasion of cancer cells to the submucosa. Therefore, the fourth cycle of DSS treatment induced the transformation of adenoma to adenocarcinoma in mouse colon. We found that the expression of iNOS at mRNA level was significantly higher in both AD3 and AD4 groups compared with control group (Figure 4B). Strikingly, the levels of miR-34c and miR-203 remarkably increased in the AD3 group, but dropped notably in AD4 group compared with AD3 (Figure 4C). Since we have shown that the expression of miR-34c and miR-203 was dependent on the activity of p53, we tested whether p53 became inactivated during the transformation from adenoma to adenocarcinoma. In accordance with expectation, the mRNA expression of p21, an important downstream target gene of p53, reduced significantly in AD4 group compared with AD3 and control group (Figure 4D). To address the causative relationship between iNOS and miRNA expression in vivo, the experiment with iNOS inhibitor 1400W was conducted in DSS-induced colitis mouse model. Firstly, consistent with previous report, we found that iNOS gene expression is strongly induced after the induction of colitis with 2.5% DSS (data not shown). As expected, the level of miR-34c was also significantly higher in colitis (DSS) group compared with control group (Figure 4E). Importantly, treatment with the inhibitor of iNOS activity 1400W dramatically blocked colitis-related upregulation of miR-34c in mouse colon (Figure 4E). The efficiency of inhibitory effect of 1400W on iNOS activity in vivo was confirmed with nitrate/nitrite level in mouse serum (Figure 4F). All these suggested that iNOS regulated the expression of miRNA in vivo and p53/miRNA may contribute to the development of CRC in mouse model.
&
5. MiRNAs were negatively correlated with iNOS in human colon cancer samples To validate whether p53-dependent miRNAs contribute to colonic carcinogenesis in human, the expression of iNOS and miRNAs were tested in human colorectal cancer tissues. Firstly, iNOS expression was measured with real time PCR in 57 pairs CRC specimens. Forty seven percent (27/57) of CRC samples showed higher level of iNOS in tumor site compared with paired normal tissue. We then measured the expression of miR-34c, miR-203 and miR-1301 in these 27 cases. It showed that the level of miR-34c negatively correlated with that of iNOS in human CRC samples (p=0.031) (Figure 5A). The same trend was observed for miR-1301 and iNOS but had no significance (p=0.088) (Figure 5B). In addition, consistent with the mouse model, the mRNA level of p21 was significantly downregulated in tumor tissues compared with matched normal tissues (Figure 5C).
6. The targets of miRNAs in NO-induced apoptosis To identify the downstream targets of miRNAs which mediate apoptosis induced by NO, MET and Survivin were measured in colon cancer cells. Although miRNA mimics successfully elevated the level of the miRNAs respectively (Figure 6A), they had no any effect on the mRNA levels of MET and Survivin in RKO cells (Figure 6B). In contrast, the protein level of Survivin was dramatically inhibited by each miRNA mimic while MET protein was only suppressed by miR-34c mimic (Figure 6C). Furthermore, the inhibitor of miR-34c blocked repression of MET induced by NO donor (Figure 6D). Except for miR-203, inhibition of miR-34c and miR-1301 also abolished downregulation of Survivin induced by NO donor (Figure 6E). All these strongly indicated that miRNAs mediated NO-induced apoptosis likely through the reduction of anti-apoptotic protein MET and Survivin in colon cancer cells.
Discussion Chronic inflammation accelerates oncogenic evolution through various mechanisms. On the one hand, great amounts of reactive oxygen/nitrogen species are
produced and create a high selection pressure during inflammatory process; on the other hand, they induce DNA damage and instability of cell genome. Here we show that NO stress induced apoptosis through p53-dependent overexpression of miR-34, miR-203 and miR-1301 all of which can negatively regulate anti-apoptotic protein Survivin and MET. When p53 mutated and became inactivated, interference to NO-induced miRNA expression mitigated the downregulation of Survivin and MET and then reduced apoptosis, which may contribute to malignant transformation in vivo. Our study reveals a new mechanism by which miRNAs contribute to clonal selection under NO stress during inflammation-associated tumorigenesis. Tumor suppressor p53 plays a central role in tumorigenesis. Mutation of p53 even occurs as an early event in inflammation-related colonic carcinogenesis in human [23]. Consistent with previous report, we showed in the current study that p53 became inactivated during the transformation from adenoma to adenocarcinoma in AOM-DSS model, which was supported by the abolishment of p21 mRNA expression. Notably, it has been demonstrated that p53 is an important transcription repressor of iNOS expression in vivo [24]. That is to say inactivated mutation of p53 would enhance the excessive NO production. Therefore, inactivation of p53 not only influenced miRNA expression and apoptosis but also worsened the reactive nitrogen stress. All these accelerated the carcinogenesis under inflammation. As a new class of p53 target genes, miRNAs are important components in the p53 network and crosstalk with p53 at multiple levels. First of all, p53 regulates the transcription and the maturation of a group of miRNAs [25]. In the current study, we showed that miR-34, miR-203 and miR-1301 were regulated by NO in a p53 dependent manner in colon cancer cells. Actually, p53 regulates the expression of each miRNA at different levels. For example, it can directly bind to the promoter and activates transcription of miR-34, which includes miR-34a, miR-34b and miR-34c [26]. In contrast, p53 has been shown to promote the maturation of miR-203 through augmentation of acetylation-dependent Drosha-mediated processing [27]. In current study we show by the first time that NO induced miR-1301 expression in a p53-dependent manner in colon cancer cells. However, the molecular mechanism
behind it needs further study in the future. On the other hand, miRNAs can regulate the activity and function of p53 through direct repression of p53 or its regulators in cells. It has been demonstrated that miR-34a positively regulated p53 function in apoptosis through direct negative regulation of SIRT1, which is a negative regulator of p53[26]. Overexpression of miR-34a decreased SIRT1 expression, leading to the increase in acetylated p53 levels and p53 activity, which in turn induced apoptosis in a p53-dependent manner. In addition, miR-1301 mimic could increase the expression of p53 at mRNA and protein levels through an unknown mechanism in HpG2 cells [28]. MiRNAs have emerged as a class of abundant and critical regulatory genes in human tumorigenesis. Over 2000 miRNAs have been identified in human. Aberrant miRNA expression, and amplification or deletion of miRNAs has been frequently observed in various human tumors. Here, we demonstrated that inactivation of the tumor suppressor p53 might also participate in tumorigenesis through the regulation of miRNAs and apoptosis. On the contrary, dysregulation of miRNAs has been suggested to be an important mechanism for the aberrant expression of a list of oncogenes and tumor suppressor genes which are not affected by genetic mutations or transcriptional regulation in tumorigenesis. More importantly, each miRNA is predicted to have numerous targets that have disparate functions. The miRNAs which were found to involve in inflammation-associated carcinogenesis in the current study, participated in the multiple aspect of carcinogenesis besides apoptosis. Low levels of the miR-34 family (which includes miR-34a, miR-34b) and miR-34c has been frequently observed in various tumors, including CRC [29]. Inactivation of endogenous miR-34a stimulates cell proliferation and inhibits p53-dependent apoptosis through targeting cyclin E2, cyclin-dependent kinases 4/6 (CDK4/6), and Bcl-2[29]. In addition, miR-34 plays an important role in cell fate determination. It was shown that miR-34 involved in self-renewal of cancer stem cell potentially via the downstream targets Bcl-2 and Notch [30]. Since miR-34 can partly restore the function of p53, it may hold significant promise as a novel molecular therapy for human cancer with loss of p53-miR-34 [30].
Different lines of evidence support the notion that miR-203 functions as a tumor suppressor. In some types of cancer, miR-203 is epigenetically silenced, and the silencing promotes tumor cell growth and invasion through different targets [31]. MiR-203 has been shown to promote apoptosis by targeting anti-apoptotic Survivin [32] and Bcl-w in a p53-dependent manner [27]. In addition, miR-203 suppressed proliferation and migration through the inhibition of SRC translation in lung cancer cells [33]. It was found recently that miR-203 directly targets Lef1, which is one of the co-transcriptional factors and mediates Wnt/beta-catenin-induced transcription [34]. Moreover, miR-203 mediates the modulation of stemness and EMT. By coordinating with miR-200c, miR-203 can suppress the expression of Bmi1 and other stem cell factors [35] [36]. Importantly, overexpression of SNAI1 induced EMT and suppressed miR-203, which repressed endogenous SNAI1, forming a double negative miR-203/SNAI1 feedback loop. This miR-203/SNAI1 with the known miR-200/ZEB feedback loops construct an a priori EMT core network [37]. The information on miR-1301 is limited. Consistent with our findings, it has been indicated that miR-1301 may be an inhibitor of tumorigenesis in HepG2 cells. MiR-1301 mimics promoted hepatocellular apoptosis through down-regulating Bcl-2 and Bcl-xL mRNA and protein expression [38]. Notably, miR-1301 mimics induced the upregulation of p53 [38]. It implied that miR-1301 might crosstalk with p53 network. However, further studies are required to elucidate the underlying mechanisms. Taken together, we demonstrated here that p53-dependent miRNAs mediated nitric oxide-induced apoptosis through targeting Survivin and MET in colon cancer cells. The perturbation of p53/miRNA mediated apoptosis may contribute to inflammation-related carcinogenesis in colon. MiRNAs may be good candidates as biomarkers to monitor the transformation from chronic inflammation to cancer in colon.
!
Acknowledgement This work was supported by funding from the 973 National Key Fundamental Research Program of China (2012CB967003) and the National Nature Science Foundation of China (91129717 and 81172034).
Figure legends Figure 1. NO induced apoptosis and changed the expression of miRNAs in colon cancer cells. RKO cells were treated with NO donor, DETA NONOate, at different concentrations for 48 hours and then were collected to measure apoptosis by flow cytometry using (A) sub-G1 analysis and (B) Annexin V/ PI staining assay. (C) Cleaved PARP was detected by western blot in RKO cells after treatment with DETA NONOate at 0.25 mM for 0, 6, 12, 24 or 48 hours. (D) The levels of miRNAs were measured by real time PCR in RKO cells after treatment with 0.25 mM DETA NONOate for 24 hours. (E) DETA NONOate (0.25mM) was left at 37
!"for 72
hours and then used as “composed NO donor”. Carboxy-PTIO at concentration of 0.5mM was used to scavenge NO in the cell culture medium. RKO cells were exposed to decomposed NO donor, DETA NONOate (0.25 mM), Carboxy-PTIO alone or DETA NONOate combination with Carboxy-PTIO for 24 hours and then the
"
level of miR-34c was measured by real time PCR. The data showed the means ± SD from three independent experiments.
##"p<0.01 and ###"p<0.001.
Figure 2. P53 mediated NO-induced apoptosis and the change of miRNAs. RKO and HCT116 cells were treated with or without DETA NONOate for 24 hours at 0.25 mM or 0.5 mM respectively. Then the cells were collected to analyze (A) p53 by western blot, and (B) apoptosis by flow cytometry. (C) RKO cells were treated with DETA NONOate for different time and then were collected to analyze p53 by western blot
(insert) and the expression of miRNAs by real time PCR. (D-F) RKO and RKO-E6 cells were treated with DETA NONOate at 0.25 mM for 24 hours and then were collected to measure the levels of miRNAs by real time PCR. The data showed as the means ± SD of three independent experiments. ! p<0.05), !! p <0.01 and !!! p <0.001.
Figure 3. MiRNAs mediated NO-induced apoptosis. (A) RKO cells were transfected with the specific inhibitors of miR-34c, miR-203 or miR-1301 at 100 nM. Twenty four hours later, cells were treated with DETA NONOate at 0.25 mM for 48 hours and then were collected to analyze apoptosis with Annexin V/ PI staining assay. (B) Apoptosis was analyzed with Annexin V/ PI staining assay in HT-29 and RKO cells after treatment with DETA NONOate for 48 hours. (C) MiRNA expression was measured by real time PCR in HT-29 cells 24 hours after treatment with DETA NONOate. (D) After transfection with the specific mimics of miR-34c, miR-203 or miR-1301 respectively, HT-29 cells were treated with DETA NONOate at 0.5 mM for 48 hours and then were collected to analyze apoptosis with Annexin V/ PI staining assay. ! p <0.05, !! p <0.01 and !!! p <0.001, n.s. no significance.
Figure 4. MiRNAs were negatively correlated with iNOS in mouse model. (A) Schematic overview of AOM-DSS mouse model of colitis-associated colon cancer. The expression levels of (B) iNOS mRNA and (C) miRNAs in mouse colon were measured by real time PCR. (D) The mRNA expression of p21, an important downstream target gene of p53, was measured by regular PCR in mouse colon. GAPDH was used as a reference. (E-F) Mice received 2.5% DSS for five continuous days and were administrated with or without iNOS inhibitor 1400W at the dosage of 10mg/kg/day (i.p.) for another 5 days.
Then all animals were sacrificed. (E) The
expression of miR-34c in mouse colon was measured with real time PCR and (F) the level of nitrate/nitrite in mouse serum was measured with Greiss reaction kit. N>7 and ! p <0.05, !! p <0.01 and !!! p <0.001.
"
Figure 5. MiRNAs were negatively correlated with iNOS in human colon cancer samples. The expression of miR-34c, miR-203 and miR-1301 were measured by real time PCR in 27 paired human colorectal cancer tissues. The correlation between (A) iNOS and miR-34c and (B) iNOS and miR-1301 were analyzed. (C) The mRNA expression of p21 was measured by real time PCR. !! p <0.01.
Figure 6. The targets of miRNAs in NO-induced apoptosis. (A) Fourty eight hours after transfection with the specific mimics of miR-34c, miR-203 or miR-1301 respectively, RKO cells were collected to measure the expression of miRNAs. The levels of MET and Survivin were measured at (B) mRNA and (C) protein levels. !! p <0.01 and !!! p <0.001. (D and E) Twenty four hours after transfection with the specific inhibitors of miR-34c, miR-203 or miR-1301 respectively, RKO cells were treated with or without DETA NONOate at 0.25 mM for another 24 hours and then were collected to analyze the expression of MET and Survivin by western blot.
'(
)* + , + - ./
0 1ϲϱ2$"&!#1&&
)* ( - , 30 3 '
1Ϯ2!% "1&&
) * - , 40 , 5 5 , 6 7 0 , 6 , -0 5 ' , ..#! ' 8 9 :;< 1ϭϮϭ2 &" 1&&#
)!* 6 = , ( > , 5 ? 5 3 /
' 8 ' / 1ϰϲ2 ! 1&&#
) * 68 4 , . / = '
1ϭϰ2%&&1&&"
)"* 60 ,' ( > ,3 / ,3 / ,+> ,+5 3 @0A / 3
.6 1ϭϬϵ2!&&
#
)#* 6 + ,- , 7 , 0 > + ,0 - ,+ B ' /
1ϵ2!%$%1&&#
)$* 6 ,68
A ,4' ,@C ,- ,D8 0 + 1ϰϱϱ2 $" 1&&$
)%* 5 , ( + , + ' / ' 1&
)&* 60 , 06 , E , E && ' 3=% @ 1ϯϴ2%&$ 1&&
)* F+ ,65 ,CA ,53 ,>( , > ! '9.@<'@@
1ϱ2%!%1&&
)* + ,C ( ,C> ,@> ,( > ,3 4 / / ' / ' / ,
' @ ( 1ϳ2
!! 1&
) * ( ( , 3 ( , B 5 , ( , = B %
'
!
1ϰϬϬ2 "!&1&&
)!* A ,6 E ,? 6 E ,( 4 ,3/4 ,@= ! /
8 ' ' 8 ' ''
1ϲ2 #$#1&
) * > F , B , C , 4 + , 5 , ( ( !
1ϰ2"$"1&&%
)"* 67 6 , 40 4 ,0 ,-4 , 4 , 4 # ' ' A(
' " 1ϱϮ2"%$#&!1&&
)#* ? ,+5 ,6+ 5 ,( A , 6 F ,( ( !% 4 AB( 6+%%& 8 1ϭϳ2!#
!#1&
)$* E , ? , @ 7 , , 0 , 3 3 G'
8 1ϭϯϭ2#"& 1&
)%* ? @ , E0 ( , ? 4 @ , 0 > , ( C + , E ( > G-- 0 ' '
1ϭϭϴ2$ %"1&&!
)&* 6 F ,5> 6 ,C5 5 ,6 3 3 ,( = ,+ 5 6 '
$
0
/
'
1Ϯϴϳ2 "!##1&
)* 304 , +C , 6 ,( - ,+ 0 6 3 ,H/ 75 7 8 ' / 1ϲϰ2$& 1&&!
)* 0 802 '
#1ϭϮ2" "1&
) * G 708 7 6 ' ' 8 $ 1ϯϱ2 #1&&"
)!* 6 , . ' 0 . , + ? , A + , A 4 , 3 3 I '
00 1ϵϱ2$$ $1%%$
) * @ C , C 3 , + , + 4
1ϯ2!! &1&
)"* 50 ,@ ,(8 3 > !
'6G4 1ϭϬϱ2 !"1&&$
)#* 3> ,=/ = A ,- ,>B , ,6
'
//'8
% 1ϯϮ2 %& 1&
)$* @ ( , 5 , > , @ 5 , 5 ? 6 & ' 1Ϯϳ2%% !1&
)%* 478 , 4 , G7 , 0 4
/ !
0 8 ' E@ 8 1ϭϬϰ2 !##1&&#
) &* > F , B , C , 4 + , 5 , ( ( !
1ϰ2"$"1&&%
) * C C , C A , ( + , @ ( , @ ( , C C E 6 ' & I 6G 3 G/ /
' A 3 3 $ 1Ϯ2#$%1&
) * B = , + F , 5 , B ( & ' ,
' 6 // 1ϴ2 #%$!1&
)
* + , ( , C 5 , + 3 , C 6 , 5 &
' ' 63
1ϵ2& #&1&!
%
) !* 4 E > , G , - - , > ? '7 ' ' 1ϭϬϱ2$ $!%1&&$
) * + I , 6 > , A 0 I 3 , 6' . , C @ , 6 4 E 4 / CEA
1ϭϭ2!$#% 1&&%
) "* 5 > , C ? , > B , C C , C + , 6 5 A
0 '
1ϴ2# "$1&
) #* H , 3 , ( A > , A > ( , 3 - , 378 6 ! '' 8
'
1ϳ2!$1&
) $* @ ( , 5 , > , @ 5 , 5 ? 6 & ' 1Ϯϳ2%% !1&
&
Graphical Abstract (for review)
Survivin and MET
Survive
Apoptosis
X
miRNAs
X
p53
Survivin and MET
miRNAs
p53
Nitric Oxide
Figure
D
C
A
miR-34a
miR-34c
miR-122
β-actin
Time (h) PARP˄116KDa˅ Cleaved PARP˄89KDa˅
miR-149
0
B
miR-1301
E
0.25mM NO donor 12 24
Figure 1
miR-203
6
control
48
miR-34c
0.25 mM NO donor
β-actin
p53
NO donor
C
B
A RKO + ˉ
RKO-E6 + ˉ
β-actin
Time (h) p53
ˉ
6
12
24
0.25mM NO donor
ˉ
48
+
HCT-116 p53 -/-
Figure 2
0
+
HCT-116 p53 +/+
F
E
D
miR-1301
miR-203
miR-34c
D
C
Figure 3
B
A
B
A
D
GAPDH
p21
1
2
DSS
control
Time˄weeks˅
AOM 3
4
5
6
C
7
AD3
DSS 8
DSS 9
DSS
AD3
11
12
14
miR-203
AD4
AD4
13
Figure 4
miR-34c
10
F
E miR-34c
C
A
P= 0.031 r= -0.415
Figure 5
B
P= 0.088 r= -0.334
D
A
ˇ
Figure 6
ˇ
—
ˇ
miR-34c
β-actin
Survivin
MET
mimic
negative control
C
NO donor —
inhibitor
β-actin
—
miR-34c inhibitor
E
β-actin
ˇ
miR-1301
Survivin
—
negative control
miR-203
MET
NO donor
miR-34c
B
—
ˇ
miR-203
—
ˇ
miR-1301