Let-7b inhibits human cancer phenotype by targeting cytochrome P450 epoxygenase 2J2

Abstract

Background: MicroRNAs (miRNAs) are small, noncoding RNA molecules of 20 to 22 nucleotides that regulate gene expression by binding to their 3' untranslated region (3'UTR). Increasing data implicate altered miRNA participation in the progress of cancer. We previously reported that CYP2J2 epoxygenase promotes human cancer phenotypes. But whether and how CYP2J2 is regulated by miRNA is not understood.

Methods and results: Using bioinformatics analysis, we found potential target sites for miRNA let-7b in 3'UTR of human CYP2J2. Luciferase and western blot assays revealed that CYP2J2 was regulated by let-7b. In addition, let-7b decreased the enzymatic activity of endogenous CYP2J2. Furthermore, let-7b may diminish cell proliferation and promote cell apoptosis of tumor cells via posttranscriptional repression of CYP2J2. Tumor xenografts were induced in nude mice by subcutaneous injection of MDA-MB-435 cells. The let-7b expression vector, pSilencer-let-7b, was injected through tail vein every 3 weeks. Let-7b significantly inhibited the tumor phenotype by targeting CYP2J2. Moreover, quantitative real-time polymerase chain reaction and western blotting were used to determine the expression levels of let-7b and CYP2J2 protein from 18 matched lung squamous cell cancer and adjacent normal lung tissues; the expression level of CYP2J2 was inversely proportional to that of let-7b.

Conclusions: Our results demonstrated that the decreased expression of let-7b could lead to the high expression of CYP2J2 protein in cancerous tissues. These findings suggest that miRNA let-7b reduces CYP2J2 expression, which may contribute to inhibiting tumor phenotypes.

Figure 1. Identification of CYP2J2 as a direct target of let-7b.

A, schematic representation of the predicted target sites of let-7b in the 3′UTR of CYP2J2. The 3′UTR of CYP2J2 was cloned into a luciferase reporter plasmid, termed pMIR/CYP2J2-3′UTR. A series of mutants carried mutated nucleotides in six potential binding sites for the let-7b seed region were generated based on wild type pMIR/CYP2J2-3′UTR. Mutants of pMIR/CYP2J2-3′UTR are constructed by mutating the complementary site (labeled by underline) in the let-7b seed region to their complementary bases. B, luciferase activity was analyzed in HepG2 cells 24 h after transfection with reporter plasmid pMIR/CYP2J2-3′UTR or pMIR (empty vector). C–D, full-length sequence CYP2J2-3′UTR containing mutant binding sites for the let-7b seed region were generated based on wild type pMIR/CYP2J2-3′UTR. We transfected these constructs into the cells (MDA-MB-435 and SK-MES-1) and analysed luciferase reporter activity. As we expected, let-7b did not affect luciferase activity of mutants compared with wild type. E, the six mutants were transfected into SK-MES-1 cells, in addition to let-7b or random let-7b. The luciferase activity of the six mutants was not repressed by let-7b. Renilla luciferase activities were used to normalize firefly luciferase activity. Columns, mean of three experiments; bars, SD. *, P<0.05.

Figure 2. Effect of exogenous let-7b on CYP2J2 expression and its enzymatic activity.

A, protein level of CYP2J2. HeLa, Tca-8113, MDA-MB-435, and SK-MES-1 cells were treated with let-7b or random let-7b (100 nM) for 48 h. The protein level of CYP2J2 was examined by western blot analysis. B, protein level of CYP2J2 was quantified by densitometry. Columns, mean of three experiments; bars, SD. *, P<0.05. C, stable metabolite 14, 15-DHET in HeLa, Tca-8113, and MDA-MB-435 cells were determined as described under Materials and Methods. Cells treated with exogenous hsa-let-7b produced fewer EETs than those treated with random let-7b. Points, mean of three experiments; bars, SD. *, P<0.05.

Figure 3. Influence of let-7b on cell proliferation and apoptosis.

A, proliferation assay via Cell-Light™ EdU DNA Cell Proliferation Kit showing decreased proliferation rate of MDA-MB-435 and SK-MES-1 cells treated with let-7b compared to the cells treated with random let-7b or control. The blue-stained cells were stained by Hoechst, and the red were EdU add-in cells. EdU-positive cells were calculated as (EdU add-in cells/Hoechst-stained cells)×100%. Columns, mean of three experiments; bars, SD. *, P<0.05. B, percentage of apoptotic cells was increased in MDA-MB-435 and SK-MES-1 cells treated with let-7b. Apoptosis measured by annexin V-FITC (x-axis) and propidium iodide staining (y-axis). Percentages of apoptotic cells (percentage of cells in the upper-right quadrant (annexin V-positive, PI-negative) plus cells in the low-right quadrant (annexin V-positive, PI-positive) in total cell number) are given under the relevant graph. Columns, mean of three experiments; bars, SD. *, P<0.05. C–D, percentages of EdU-positive cells and apoptotic cells were analyzed in HeLa and Tca-8113 cells transfected with let-7b or random let-7b. E–F, MDA-MB-435 and SK-MES-1 cells transfected with let-7b were treated with C26 (specific CYP2J2 inhibitor, 10 µM) or 14,15-EET (250 nM). 48 h later, percentages of EdU-positive cells and apoptotic cells were analyzed. Columns, mean of three experiments; bars, SD. *, P<0.05. G, let-7b overexpression influenced expression of tumor-related genes in MDA-MB-35 cells. Let-7b overexpression in MDA-MB-435 cells significantly downregulated PI3K, pAkt, and pERK but increased Bax and nm-23 expression. The data shown were repeated three times.

Figure 4. Relationship between CYP2J2 protein and let-7b expression in lung cancer and adjacent nontumor tissues.

A, expression levels of CYP2J2 in lung cancerous (C) and adjacent nontumor tissues (N) was measured by Western blot analysis. B, comparison between the expression levels of let-7b in lung cancerous and adjacent nontumor tissues (n = 18). Although changes of −ΔΔCT values [−(ΔCT_{non-tumor tissue}−ΔCT_{cancer tissue})] are relatively mild (from -1.85 to 5.4, 2.6±1.27), the fold changes of let-7b expression between 18 paired human lung squamous cancer and adjacent nontumor tissues are significant (from 0.277 to 42.22, 6.06±2.43). C, relationship between CYP2J2 protein and let-7b expression in lung cancer and paired adjacent nontumor tissues. Expression level of CYP2J2 protein was increased in 13 of 18 sets of lung cancer compared with the adjacent normal tissues (the fold change>1). As expected, let-7b levels were commonly reduced in these tumors compared with the adjacent normal tissues (the fold change<1). D–E, expression levels of CYP2J2 and let-7b in breast cancerous and adjacent nontumor tissues (n = 4). Real-time RT-PCR was used to determine mature let-7b levels. The U6 snRNA expression level was used to normalize the relative let-7b level. Columns, mean of three experiments; bars, SD. *, P<0.05.

Figure 5. Effect of let-7b on tumor growth.

MDA-MB-435 cells were injected subcutaneously into the right flank of nude mice to generate the mouse model of breast cancer. Two weeks later, mice received the let-7b treatment randomly. The let-7b expression vector (pSilencer-let-7b plasmid) was injected into mice through a tail vein at a dose of 4 mg/kg body weight every 3 weeks. A, the x-axis was labeled as the days of pSilencer-let-7b treatment. Tumor volume was measured weekly and calculated as TV (mm³)  =  length×width²×0.5236. B, the measurement of 14,15-DHET level in nude mice urine was performed by ELISA according to the manufacturer’s instructions. Points, mean of three experiments; bars, SD. *, P<0.05. C, average tumor weight and body weight of control and let-7b treatment groups after growth for 8 weeks. Columns, mean; bars, SD. *, P<0.05 versus control. D–E, the expression of the mature let-7b in the tumors and primary organs was validated by real-time RT-PCR. Columns, mean of three experiments; bars, SD. *, P<0.05. F, Western blot analysis showed alteration of the expression level of CYP2J2 protein and tumor-related genes of tumor samples.

Figure 6. Let-7b overexpression inhibits tumor metastasis.

A, average number of spleen metastases for each group (n = 6). Columns, mean; bars, SD. *, P<0.05 versus control. B, average weight of axillary lymph nodes for each group. Columns, mean; bars, SD. *, P<0.05 versus control. C–D, hematoxylin and eosin staining of sections of tumor (C1, C2) and spleen (D1, D2). E–F, TUNEL staining of tumor (E1, E2) and spleen (F1, F2) sections.