The Study Of Epigenetic Mechanism Of Estrogen Inducing CRC Cells Apoptosis

BackgroundColorectal cancer (CRC) is the fourth most common cause of death from cancer,affects around1.2million people world wide, with about608,700fatalities per year. Boththe prevalence and the lifetime risk of CRC are significantly lower in females than in males.This is particularly true when comparing pre-menopausal women to age-matched men. Thesex-specific difference in CRC incidence has been attributed to estrogen, but the exactmechanism is not clear. Since expression alterations of some miRNAs are associated with aseries of genetic events, their expression patterns may either have pro-or anti-cancereffects. Several miRNAs expression alterations have been reported in CRC, includingmiR-135b, miR-96, miR-183, miR-133a, miR-133b, miR-21, miR-31, miR-145, miR-203,miR-223, miR-155. However, the mechanism for miRNA change in CRC is not wellunderstood. Recent studies have indicated that estrogens (especially estradiol E2) canregulate miRNAs expression in human breast cancer cells, endometrial stromal cells, andmyometrial smooth muscle cells. Estrogen receptor (ER) expression is associated withmicroRNA (miRNA) expression in ER-positive tumors. Thus, estrogen’s protect effect maybe related to the regulation of some miRNAs. DNA mismatch repair (MMR) system playsan important role in maintaining genomic stability. The MMR gene alteration includinggenetic and sporadic mutations, commonly occur in the early stages of the neoplasticprocess. MMR dysfunction can cause microsatellite instability (MSI). It has been reportedthat estrogen is associated with MSI. In previous study, estradiol (E2) has shown to havethe ability to induce CRC cells to apoptosis, and to up-regulate major MMR gene (hMLH1)activity in colonic epithelial cells. These findings suggest that estrogen’s anti-colorectalcancer effect may be mediated by MMR function in colonic cells. However, the pathwaythrough which estrogen regulates MMR function is still not well understood.From our knowledge, studies the direct link between estrogen, miRNA and MMRexpression in cultured cells and patient tissue/serum samples, the relationship of miRNA-target gene in CRC cell apoptosis, have not been done. In this study, we firstevaluated the effects of estrogen and its antagonist ICI182,780on the expression ofmiRNAs (miR-31, miR-135b, miR-155, miR-203and miR-223) and MMR using COLO205,SW480and MCF-7cell lines, followed by examining the association of tissue miRNAexpression, MMR expression and serum E2levels using samples collected from18colorectal cancer patients. Then the miRNA target gene will be confirmed by luciferaseassay.Objective1. Studies of direct link between estrogen and miRNA expression in CRC cells.2. Studies of direct link between estrogen and MMR expression in CRC cells.3. Studies the association of tissue miRNA expression, MMR expression and serum E2levels using samples collected from18colorectal cancer patients.4. To check whether miRNAs participates in estrogen-induced apoptosis in culturedcells. Then the target gene will be confirmed.Materials and methods1. Studies of direct link between estrogen and miRNA expression in CRC.COLO205and SW480cells were incubated in different concentrations of estradiol E2for48h;100nM of estrogen receptor antagonist ICI182,780was then added followed byAnnexin-V apoptosis assay. COLO205cells were treated with10nM E2with a finalvolume of0.01%ethanol for0h,6h,12h and24h to determine miRNA expression.SW480and MCF-7cells were treated with EtOH,1nM or10nM E2, for12h alone. Forthe indicated experiments, cells were pretreated with100nM ICI182,780for6h prior toEtOH (control) or E2treatment. SW480cells were transiently transfected with an emptypRST7vector and pRST7-ER-β. After24h of transfection, Western blotting was used toconfirm the transfection efficiency after24h or48h of transfection. After24h oftransfection, SW480(pERβ) and SW480(vector) were treated with EtOH or10nM E2for12h. After incubation, cells were harvested and used for quantitative reversetranscription-PCR (RT-qPCR).2. Studies of direct link between estrogen and MMR in CRC.COLO205cells were treated with10nM E2with a final volume of0.01%ethanol for0h,6h,12h and24h to determine mRNA expression, or for0h,24h and48h to determine protein expression. SW480and MCF-7cells were treated with EtOH,1nM or10nM E2for12h to determine mRNA expression, or for0h,24h and48h to determineprotein expression. After24h of transfection, SW480(pERβ) and SW480(vector) weretreated with EtOH or10nM E2for12h. After incubation, cells were harvested and usedfor quantitative reverse transcription-PCR (RT-qPCR) and Western blotting.3. Studies the association of tissue miRNA expression, MMR expression and serum E2levels using samples collected from18colorectal cancer patients.A total of18patients9males and9females diagnosed with CRC were enrolled. Thesepatients were randomly selected from patient pool of the hospital’s gastrointestinal clinic;none of the subjects had undergone estrogen replacement therapy. Blood was taken half anhour after colonoscopy for determining the serum E2level. During colonoscopy, four orfive samples from cancer tissues and adjacent non-cancer tissues, respectively, wereobtained and stored in liquid nitrogen. Correlation analyses between serum E2level andmiRNA expression, serum E2level and mRNA expression, and miRNA and mRNAexpression were performed.4. To check whether miRNAs participates in estrogen-induced apoptosis in culturedcells. Then the target gene will be confirmed.COLO205were transfected with miR-135b inhibitor, and incubated in differentconcentrations of estradiol E2for0h,6h,12h and24h to determine miRNA expressionand apoptosis. According to the UCSC (www.geome.ucsc.edu), LATS23′UTR wasamplified from the genomic DNA of human by PCR and cloned into luciferase reportergene pGL3-vector.The recombinant pGL3-LATS23’UTR Wt and pGL3-LATS23’UTRMut were respectively transient co-transfected with miR-135b mimic or miRNA NC intoHEK293cells using Lipofectamine. Luciferase activities were measured consecutively.Rusults1. After treatment with different concentrations of E2for48h, the apoptotic cells weresignificantly increased in COLO205, and the effect was the strongest at1×10-9M of E2.Estrogen receptor antagonist, ICI182,780, inhibited E2-induced apoptosis in COLO205cells. However, the effect of E2on apoptotic induction was not seen in SW480cells.RT-qPCR results indicated that E2decreased the levels of miR-31, miR-155and miR-135bin a time-dependent manner. ICI182,780reversed E2-induced repression of miR-31, miR-155and miR-135b in COLO205cells. E2did not however affect miRNA expressionin either SW480cells, or MCF-7cells. ER-β protein expression was not up-regulated invector transfected SW480cells, whereas it was up-regulated in pRST7-ER-β transfectedSW480cells after24and48h of transfection by western blot. RT-qPCR results indicatedthat E2decreased the levels of miR-31, miR-155and miR-135b in pRST7-ER-β transfectedSW480cells, but not in vector cells.2. RT-qPCR and Western blotting showed that E2increased the expression of ER-βand hMLH1, but at less degree of hMSH2, at both mRNA and protein levels in COLO205cells, in a time and dose-dependent manners. These effects were again inhibited byICI182,780. No changes in the expression of hMLH1, hMSH2and ER-β mRNA were seenin MCF-7and SW480cells. RT-qPCR and Western blotting showed that E2increased theexpression of hMLH1, but at less degree of hMSH2, at both mRNA and protein levels inpRST7-ER-β transfected SW480cells.3. The results indicated that serum E2levels were strongly and negatively correlatedwith miR-31and miR-135b expression, but not for miR-155expression in cancer tissue.We also found that in cancer tissue, hMLH1mRNA levels were negatively correlated withthat of miR-155and miR-135b expression, whereas ER-β mRNA levels were positivelycorrelated with serum E2concentrations. Interestingly, only miR-135b has a significantcorrelation with serum E2levels, ER-β and hMLH1expression. Our previous study showedthat in normal individuals, only when serum E2levels were higher than45pg/ml, a strongpositive correlation of E2with hMLH1gene expression was observed; and there was nosignificant correlation between hMLH1expression and serum E2levels when E2levelswere lower (E2,＜45pg/ml). In the current study of18CRC patients, only two patients’serum E2levels were higher than45pg/ml (a male, E2=69.16pg/ml; a femail, E2=70.18pg/ml).4. Inhibition of miR-135b expression in COLO205cells, can significantly induced thecells to apoptosis. HEK-293cells were co-transfected with luciferase constructs containingthe wild-type (WT) or mutant (MUT) target site of the LATS23′UTR and miR-135b mimicor miRNA NC for24h. The pGL3-vector was used as a negative control. The luciferaseexpression from the wild type LATS23′UTR was reduced to40%of that of control vectorand mutant. Conclusions1. E2inhibited the expressions of miRNAs, and upregulate hMLH1expression inCOLO205cells, which could be reversed by E2antagonist ICI182.780. The expression ofmiR-135b was significantly higher in CRC tissues than in normal tissues, and inverselycorrelated with serum E2level and ER-β mRNA expression in CRC patients’ cancertissues.2. There was no significant correlation between hMLH1expression and serum E2levelin CRC patients, when more patients’ serum E2levels were lower than45pg/ml.Therefore,the E2regulatory effects on MMR through ER-β and miRNA expression may be animportant event that occurs early in tumorigenesis, and during cancer progression, adecreased estrogen and/or reduced ER-β expression result in the increase of miR-31,miR-155and miR-135b, which contributes to MMR instability and eventually colorectalcancer carcinogenesis.3. Estrogen might induce apoptosis through the inhibition of miR-135b, whichcontribute to an increase in miR-135b target gene, LATS2(Large Tumor Suppressor2),expression and activity.