Skip to main content
NCBI home page
Journal of Molecular Cell Biology logoLink to Journal of Molecular Cell Biology
. 2012 Aug 29;5(1):3–13. doi: 10.1093/jmcb/mjs049

A regulatory circuit of miR-148a/152 and DNMT1 in modulating cell transformation and tumor angiogenesis through IGF-IR and IRS1

Qing Xu 1, Yue Jiang 2,3, Yu Yin 1,4, Qi Li 1, Jun He 2, Yi Jing 2, Yan-Ting Qi 2, Qian Xu 1, Wei Li 1, Bo Lu 5, Stephen S Peiper 2, Bing-Hua Jiang 1,2,*, Ling-Zhi Liu 2,*
PMCID: PMC3570052  PMID: 22935141

Abstract

Dysregulation of microRNAs is a common feature in human cancers, including breast cancer (BC). Here we describe the epigenetic regulation of miR-148a and miR-152 and their impact on BC cells. Due to the hypermethylation of CpG island, the expression levels of both miR-148a and miR-152 (miR-148a/152) are decreased in BC tissues and cells. DNMT1, the DNA methyltransferase 1 for the maintenance methylation, is aberrantly up-regulated in BC and its overexpression is responsible for hypermethylation of miR-148a and miR-152 promoters. Intriguingly, we found that DNMT1 expression, which is one of the targets of miR-148a/152, is inversely correlated with the expression levels of miR-148a/152 in BC tissues. Those results lead us to propose a negative feedback regulatory loop between miR-148a/152 and DNMT1 in BC. More importantly, we demonstrate that IGF-IR and IRS1, often overexpressed in BC, are two novel targets of miR-148a/152. Overexpression of miR-148a or miR-152 significantly inhibits BC cell proliferation, colony formation, and tumor angiogenesis via targeting IGF-IR and IRS1 and suppressing their downstream AKT and MAPK/ERK signaling pathways. Our results suggest a novel miR-148a/152-DNMT1 regulatory circuit and reveal that miR-148a and miR-152 act as tumor suppressors by targeting IGF-IR and IRS1, and that restoration of miR-148a/152 expression may provide a strategy for therapeutic application to treat BC patients.

Keywords: miR-148a, miR-152, DNMT1, IGF-IR, IRS1, breast cancer, tumor angiogenesis

Introduction

Breast cancer (BC) is the most common cancer and the second leading cause of cancer death among women in the USA (Brooks et al., 2009). It is well demonstrated that genetic alterations, such as specific gene amplifications, deletions, mutations, and chromosome rearrangements, are associated with BC development (Campan et al., 2006; Mirza et al., 2007; Sui et al., 2007). Recent findings have indicated that epigenetic modifications may constitute to BC development (Birgisdottir et al., 2006; Visvanathan et al., 2006). Multiple genes such as DNA repair gene BRCA1, MGMT, apoptotic gene APC, and HIC1 are hypermethylated in BC tissue compared with non-cancerous tissue (Jin et al., 2001).

miRNAs are small endogenous non-coding RNAs composed of ∼19–24 nucleotides that bind to imperfect sequence homology sites of mRNA and recruit the RNA-induced silencing complex, causing either degradation or inhibition of protein translation, thus effectively silencing their mRNA targets. Generally, one gene can be repressed by multiple miRNAs and one miRNA may repress multiple target genes, which results in the formation of complex regulatory feedback networks (Bartel, 2004; Calin et al., 2005). Recent study has revealed that multiple miRNAs are deregulated in BC such as miR-125b, miR-145, miR-21, and miR-155 using genome-wide approaches (Iorio et al., 2005). Like protein-coding genes, DNA sequences encoding miRNAs may undergo aberrant DNA methylation, leading to miRNA up-regulation (through DNA hypomethylation) or down-regulation (through DNA hypermethylation) in human cancers. For example, previous studies have shown that hsa-miR-9 is significantly decreased due to hypermethylation in promoter region in clear cell renal cell carcinomas and BC (Lehmann et al., 2008; Hildebrandt et al., 2010). Frequent silencing of miR-34b and miR-34c is also associated with CpG islands hypermethylation in various cancers (Kozaki et al., 2008; Toyota et al., 2008). Moreover, aberrant hypermethylation of the miR-148a-coding region occurs early in human pancreatic carcinogenesis and leads to the down-regulation of miR-148a expression (Hanoun et al., 2010). Recent reports show that high degree of methylation of the miR-152 CpG is found in MLL-rearranged acute lymphoblastic leukemia (Stumpel et al., 2011). However, the roles of miR-148a and miR-152 in BC development are unknown, and the mechanism of miR-148a/152 down-regulation in BC remains to be elucidated.

In the present study, we found that miR-148a and miR-152 are down-regulated in BC cells and tumors. We will ask several important questions in this study: (i) whether DNA methylation is involved in the down-regulation of miR-148a/152 expression; (ii) whether the expression of miR-148a/152 is associated with clinicopathologic features in patients with BC; (iii) whether there is a circular regulation loop between miR-148a/152 and DNMT1; (iv) what are the potential direct targets of miR-148a/152 that may be associated with cancer development; (v) what are the roles of miR-148a and miR-152 in cell proliferation, colony formation, and tumor angiogenesis; and (vi) what signaling pathways are involved in miR-148a- and miR-152-modulated colony formation and tumor angiogenesis. The answer of these questions would provide new insights into the molecular mechanism of BC development and angiogenesis as well as provide new strategy for BC diagnostics and treatment in the future.

Results

miR-148a and miR-152 are down-regulated in BC cell lines and tumor tissues

The expression levels of miR-148 family members (miR-148a, miR-148b, and miR-152) in MCF-7, T47D, MDA-MB-231 BC cells and in immortalized breast epithelial MCF-10A cells were determined by TaqMan real-time polymerase chain reaction (PCR). The results showed that the levels of miR-148a and miR-152, but not miR-148b, were dramatically down-regulated in BC cells when compared with MCF-10A cells (Figure 1A). We further analyzed the expression levels of miR-148a and miR-152 in 22 pairs of BC tissue specimens and matched adjacent normal breast tissues. Consistent with the results in cell lines, the expression levels of miR-148a and miR-152 in BC tissues were significantly decreased when compared with the matched normal tissues (Figure 1B). Clinicopathologic features of BC patient are listed in Table 1. Lower miR-148a and miR-152 expression levels were detected in 31 (67%) and 35 (76%) of 46 BC cases, respectively, and were significantly correlated with the BC grade. Significant down-regulation levels of miR-148a and miR-152 were also observed in 17 BC cases with lymph node-metastasis (LN)-positive patients when compared with 27 LN-negative patients. There was no significant correlation between the expression levels of miR-148a/152 and patient ages, tumor histological and subtypes, tumor sizes, or status of ER, PR, and HER2 in BC tissues. To investigate the location of miR-148a and miR-152 in breast tissues, we performed in situ hybridization using digoxigenin-labeled locked nucleic acid (LNA)-miRNA probes to detect the expression of miR-148a/152 in breast tissue specimens. The data showed that the expression levels of miR-148a/152 in normal breast tissues were high, with positive signals predominantly in luminal and less intense or no signal in myoepithelial cells in both ductal and lobular structures. However, little expressions of miR-148a/152 were detected in the BC tumor sections, consistent with our observation that miR-148a and miR-152 are down-regulated in BC tissues (Figure 1C). The down-regulation of miR-148a/152 expression was also found in human ovarian cancer cell lines (Supplementary Figure S1A), indicating that the deregulation of miR-148a/152 may be a common hallmark in different human cancers.

Figure 1.

Figure 1

Open in a new tab

Down-regulation of miR-148a and miR-152 expression in breast cancer cells and tumors. (A) Levels of miR-148a, miR-148b, and miR-152 expression in MCF-10A, MCF-7, T47D, and MDA-MB-231 (MB-231) cells were determined by Taqman qRT–PCR assay, and normalized to the U6 levels using the 2−ΔΔCT method. Results represent the mean ± SD from three independent experiments. * and ** indicate significant difference at P < 0.05 and <0.01, respectively. (B) Relative expression levels of miR-148a and miR-152 in 22 pairs of BC tumors (BCT) and normal breast tissues (NBT) were determined by Taqman qRT–PCR assay, and normalized to the U6 levels. (C) The expressions of miR-148a, miR-152, and U6 in normal breast tissues and BC tumor sections were analyzed by in situ hybridization. The staining for miR-148a and miR-152 was mainly observed in luminal epithelial cells, some staining in myoepithelial cells in both ductual and lobular structures, but not in tumor tissues (×400).

Table 1.

Clinicopathologic features of BC patients and the levels of miR-148a and miR-152 expression in the cancer tissues.

Characteristic Number of patients % Molecular expression level
miR-148a P-value miR-152 P-value
All patients 46 100
Age at diagnosisa 0.71 0.73
 ≤50 years 26 57 0.39 (0.08–1.3) 0.3 (0.07–1.7)
 >50 years 20 43 0.43 (0.06–1.6) 0.34 (0.04–1.1)
Gradeb 0.02* 0.05*
 I, well-differentiated 6 13 0.77 (0.36–1.6) 0.6 (0.15–1.4)
 II, moderately differentiated 26 57 0.45 (0.03–1.7) 0.38 (0.03–1.6)
 III, poorly differentiated 14 30 0.17 (0.01–0.6) 0.19 (0.01–0.6)
Tumor histologicalb 0.24 0.66
 Ductal carcinoma in situ 7 15 0.47(0.17–0.9) 0.37 (0.12–10.7)
 Invasive ductal carcinoma 33 72 0.3 (0.06–1.5) 0.28 (0.03–1.2)
 Invasive lobular carcinoma 6 13 0.23 (0.02–0.6) 0.23 (0.06–0.7)
Tumor subtypea 0.6 0.36
 Luminal A 21 46 0.39 (0.1–0.84) 0.35 (0.04–1.4)
 Luminal B 11 24 0.41 (0.16–0.98) 0.34 (0.04–1.6)
 Her2-enriched 7 15 0.52 (0.03–1.2) 0.47 (0.16–0.9)
 Triple-negative 7 15 0.31 (0.08–1.1) 0.15 (0.02–0.55)
ER statusa 0.22 0.4
 Negative 21 46 0.48 (0.01–1.2) 0.39 (0.02–1.4)
 Positive 25 54 0.35 (0.02–1.4) 0.3 (0.01–1.6)
PR statusa 0.45 0.8
 Negative 20 43 0.36 (0.03–0.9) 0.32 (0.07–1.2)
 Positive 26 57 0.44 (0.04–1.4) 0.35 (0.03–1.3)
HER2/neu statusa 0.68 0.51
 Negative 29 63 0.4 (0.12–1.5) 0.31 (0.03–1.5)
 Positive 17 37 0.43 (0.3–0.8) 0.39 (0.2–0.9)
Tumor sizea 0.17 0.12
 ≤2 cm 19 41 0.5 (0.14–1.2) 0.41 (0.02–1.7)
 >2 cm 27 59 0.35 (0.03–1.6) 0.27 (0.01–1.6)
Lymph node-metastasisa 0.048* 0.02*
 Negative 27 59 0.48 (0.13–1.6) 0.43 (0.07–1.6)
 Positive 17 37 0.27 (0.03–0.9) 0.17 (0.01–0.6)
Open in a new tab

ER, estrogen receptor; PR, progesterone receptor. aEvaluated with the Mann–Whitney U-test. bEvaluated with the Kruskal–Wallis test. *Significant difference.

Down-regulation of miR-148a and miR-152 is due to DNA hypermethylation in BC

The methylation of gene promoter often occurs in carcinogenesis, resulting in reduced or lost expression of the methylated gene (Wilson et al., 2007). We hypothesize that DNA methylation is responsible for the miR-148a/152 down-regulation in BC. To verify this hypothesis, we first analyzed the genomic DNA sequence spanning of miR-148a and miR-152 genes, and found both of these genes have large amount of CpG-rich regions (CpG islands) in their promoter regions. Subsequently, we performed methylation-specific PCR (MSP) analysis to detect the methylation status of the promoter regions of miR-148a and miR-152. Hypermethylation of CpG islands in miR-148a and miR-152 promoters were found in T47D and MDA-MB-231 cells when compared with MCF-10A cells (Figure 2A). In addition, bisulfite sequencing results showed that the average methylation levels of miR-148a and miR-152 promoter are high in MDA-MB-231 (79.5% and 84.2%), followed by T47D (42% and 60.3%) and are low in MCF-10A cells (22.7% and 46.3%) (Figure 2B). Combined with miR-148a/152 expression pattern in these three cells, these results suggested that the higher the methylation rate in the miR-148a/152 promoter the lower the expression level of miR-148a/152. Similarly, in BC tissues, 28 out of 36 (77.8%) and 30 out of 36 (83.3%) CpG islands were predominantly methylated, while only 5 out of 22 (22.7%) and 7 out of 22 (31.8%) CpG islands were methylated in normal tissues (data no shown). To test whether lower levels of miR-148a and miR-152 expression in different grades and LN status may be associated with aberrant methylation, we randomly picked six breast tumor tissues with different grades, LN status, and one normal specimen as control to analyze DNA methylation. The results revealed that the promoters of miR-148a and miR-152 in six tumor tissues were highly methylated with different degrees when compared with the normal breast tissue. The methylation rates were higher in LN-positive cases when compared with LN-negative cases, and were increased with higher grade, with the highest rates in Grade 3 of LN-positive BC tissue (Figure 2C). To test the functional relevance of DNA methylation, we found that demethylation treatment by 5-Aza-dC dramatically restored both pri-miR-148a/152 and miR-148a/152 expression in T47D and MDA-MB-231 cells (Figure 2D and E), indicating that hypermethylation is responsible for the silencing of miR-148a and miR-152 expression.

Figure 2.

Figure 2

Open in a new tab

Epigenetic modification of miR-148a and miR-152 genes in BC cells and tumors. (A) MSP analyses of miR-148a and miR-152 gene promoter in MCF-10A, T47D, and MB-231 cells. U, unmethylated status; M, methylated status. (B) Results of bisulfite sequence of miR-148a and miR-152 promoter regions in MCF-10A, T47D, and MB-231 cells. (C) Genomic bisulfite sequencing of miR-148a and miR-152 in BC patients with different grades and LN status (Sample ID, grade levels, and LN status were indicated). A normal breast tissue was included as non-cancerous control. The schemata represent maps of the CpG rich regions of miR-148a and miR-152. The CpG sites are indicated by vertical ticks. Each row represents a single clone for each individual genomic. Open and filled squares represent unmethylated and methylated CpG sites, respectively. (D) T47D and MB-231cells were treated with 5-Aza-dC treatment for 5 days. The methylation of miR-148a and miR-152 promoter in the cells was analyzed using MSP. (E) The BC cells were treated without or with 5-Aza-dC as indicated. Pri-miR-148a/152 and miR-148a/152 expression levels were measured by qRT–PCR. The graphs show the mean ± SD of the relative levels from three replications.

DNMT1 up-regulation is responsible for the hypermethylation of miR-148a and miR-152 gene promoters

Previous studies have shown that DNMT1 overexpression contributes to gene promoter hypermethylation and is associated with the malignant potential and poor prognosis of human cancer (Bernardino et al., 1997; Soares et al., 1999). In our study, we found that DNMT1 expression was strongly increased in BC tumors and cell lines (Figures 3A and 4A). Consistent with western blot results, the immunostaining signals of DNMT1 were weak in normal tissues, mainly located in the cytoplasm of epithelial cells; while the signals in tumor tissues were much stronger, especially localized to the nuclei of the cells (Supplementary Figure S2A). Spearman's rank correlation analysis showed that the expression levels of DNMT1 and miR-148a or miR-152 in 40 BC specimens were inversely correlated: correlation of DNMT1 with miR-148a is −0.394 (P = 0.012), and DNMT1 with miR-152 is −0.371 (P = 0.018) (Figure 3B). To further explore the role of DNMT1 in regulating expression of miR-148a and miR-152, we silenced de novo DNMT1 expression in the cells using siRNA against DNMT1. DNMT1 knockdown abolished the hypermethylation of miR-148a and miR-152 genes (Figure 3C), and induced up-regulation of both pri-miR-148a/152 and miR-148a/152 expression (Figure 3D). It has been reported that miR-148a and miR-152 directly target DNMT1 in lupus and hepatocellular carcinoma (Huang et al., 2010; Pan et al., 2010). By using immunoblotting and miRNA-target luciferase activity assay as showed in Figure 3E and Supplementary Figure S2B and C, we confirmed that DNMT1 is a direct target of miR-148a/152 in BC cells. Collectively, DNMT1 overexpression directly results in the hypermethylation of miR-148a and miR-152 genes, leading to the down-regulation of miR-148a and miR-152 expression: an interesting new regulatory circuit between DNMT1 and miR-148a/152 in BC cells.

Figure 3.

Figure 3

Open in a new tab

DNMT1 was involved in the hypermethylation of miR-148a and miR-152 genes. (A) Relative DNMT1 expression levels in 22 matched BCT and NBT tissues were determined by western blot. Data are presented as fold changes of BCT to adjacent NBT. (B) Scatter plots of the Spearman correlation coefficient corresponding to an inverted monotonic trend between relative DNMT1, and miR-148a (P = 0.0012), or miR-152 (P = 0.018) expression levels in BC tumors. (C) MSP analyses for miR-148a and miR-152 methylation in T47D and MB-231 cells after siDNMT1 transfection. U, unmethylated status; M, methylated status. (D) The BC cells were transfected with siRNA against DNMT1 (siDNMT1) and scramble siRNA (SCR). Total RNAs from the cells were used to analyze both pri-miR-148a/152 and miR-148a/152 using qRT–PCR. The mean ± SD from three replications (*P < 0.05 and **P < 0.01). (E) The T47D and MB-231 cells were transfected with miR-148 and miR-152 precursors, or anti-miR-148 and anti-miR-152 inhibitors, or the negative control microRNA precursor or inhibitor as indicated. The expression of DNMT1 and β-actin in the cells was analyzed by immunoblotting 72 h after the transfection.

Figure 4.

Figure 4

Open in a new tab

miR-148a and miR-152 directly targeted IGF-IR and IRS1 in BC cells. (A) Representative of western blot images showing expression levels of DNMT1, IGF-IR, IRS1, and β-actin in BC cells and in six paired normal breast tissues (N) and breast tumor tissues (T). β-actin was used as an internal loading control. (B) IGF-IR and IRS1 expression levels were analyzed by immunoblotting, and the signal intensities were quantified and presented as fold change of BCT to NBT in 22 paired samples. (C) Scatter plots of the Spearman correlation coefficient analysis of relative IGF-IR and IRS1 expression levels with miR-148a and miR-152 expressions levels in BC tumors. (D) The alignment of miR-148a and miR-152 putative binding sites in human IGF-IR and IRS1 3′-UTR regions to show complementary pairing of miR-148a or miR-152 with IGF-IR and IRS1 wild-type (WT) and mutant (Mut) 3′-UTR reporter constructs. (E) The reporter constructs containing the wild-type and mutant (Mut) IGF-IR or IRS1 3′-UTR regions were cotransfected into the cells with miR-148a, miR-152, or scramble miRNA precursors and β-gal plasmid. The relative luciferase/β-gal activities were analyzed in the cells 48 h after the transfection. All experiments were performed in triplicate. Bars indicate relative luciferase activities ± SD. *P < 0.05. (F) Relative levels of IGF-IR, p-IGF-IR, IRS1, and p-IRS1 expression in T47D and MB-231 cells were analyzed by immunoblotting 72 h after transfection with miR-148a, miR-152, or miR-Scr; or with anti-miR-148a, anti-miR-152, or anti-miR-Scr.

IGF-IR and IRS1 are two direct targets of miR-148a and miR-152

IGF-IR and IRS1 proteins were reported to be overexpressed in more than 40% of BC (Chang et al., 2002; Sehat et al., 2010; Tamimi et al., 2011). By using western blot and immunohistochemical staining, we also observed that the expression levels of IGF-IR and IRS1 were significantly up-regulated in BC tissues and cell lines compared with matched normal breast tissues and MCF-10A, respectively (Figure 4A and B, and Supplementary Figure S2A). There are significantly inverse correlations between IGF-IR and miR-148a (correlation = −0.379, P = 0.016), and miR-152 (correlation = −0.415, P = 0.01), respectively, in the BC tumor tissues (Figure 4C). Similarly, the correlations of IRS1 with miR-148a (correlation = −0.33, P = 0.05) and with miR-152 (correlation = −0.483, P = 0.008) are also statistically significant (Figure 4C). Those impressing inverse correlation indicating a relationship might also exist between miR-148a/152 and IGF-IR/IRS1. By using bioinformatics analysis, we found that IGF-IR and IRS1 were two potential targets of miR-148a and miR-152. To confirm these targets, reporter constructs were made to contain the putative binding sites of IGF-IR or IRS1 3′-UTR regions, or with three nucleotide substitute in their 3′-UTR regions (Mut) as indicated in Figure 4D. Overexpression of miR-148a or miR-152 inhibited both IGF-IR and IRS1 wild type, but not mutant (Mut) reporter activities (Figure 4E), demonstrating that miR-148a and miR-152 can specifically target IGF-IR and IRS1 3′-UTR regions by binding to their putative sequences. Forced expression of miR-148a or miR-152 repressed both IGF-IR and IRS1 protein expression; whereas blockade of endogenous miR-148a or miR-152 using antisense inhibitors increased both IGF-IR and IRS1 expression levels in BC cells (Figure 4F). Similarly, effects of miR-148a/152 on targeting IGF-IR and IRS1 could also been observed in ovarian cancer cells (data not shown). As phosphorylated IGF-IR and IRS1 demonstrate the activity of these proteins, we observed the similar effect of miR-148a/152 or anti-148a/152 on p-IGF-IR and p-IRS1 protein levels (Figure 4F). These results demonstrate that miR-148a/152 directly target IGF-IR and IRS1 by binding their 3′-UTRs, thus suppressing protein expression. Although the endogenous levels of miR-148a/152 are low in T47D and MDA-MB-213 cells, anti-miR-148a/152 inhibitors exhibit strong effects on the expression of IGF-IR and IRS1, showing that the low levels of miR-148a/152 in these cells still have inhibitory biological function on their target expression.

miR-148a and miR-152 inhibit PI3K/AKT and MAPK/ERK signaling pathways and HIF-1α and VEGF expression via targeting IGF-IR and IRS1

The signalings through the phosphatidylinositol-3 kinase (PI3K)/AKT and mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinases (ERK) cascades are frequently hyperactivated in cancer cells and play critical roles in controlling cell metabolism, proliferation, survival, and motility (Vivanco and Sawyers, 2002; Hennessy et al., 2005). To investigate whether miR-148a/152 inhibit PI3K/AKT and MAPK/ERK pathways via targeting IGF-IR and IRS1, we performed IGF-IR/IRS1 loss- and gain-of-function experiments in BC cells. The knockdown of endogenous IGF-IR or IRS1 significantly down-regulated IGF-IR or IRS1 as well as p-IGF-IR and p-IRS1 levels, which led to the inhibition of AKT and ERK phosphorylation, while the total AKT and ERK protein levels were not changed (Figure 5A). Forced expression of miR-148a/152 in BC cells had similar effect to decrease p-AKT and p-ERK levels as IGF-IR and IRS1 siRNAs (Figure 5B). In addition, the blocking of miR-148a or miR-152 expression induced AKT and ERK1/2 activation (Figure 5C). The effect was similar to that in the cells with the overexpression of IGF-IR or IRS1 (Figure 5D). These results show that miR-148a/152 suppress PI3K/AKT and MAPK/ERK signaling pathways via targeting IGF-IR and IRS1 in BC cells.

Figure 5.

Figure 5

Open in a new tab

miR-148a and miR-152 inhibited PI3K/AKT and MAPK/ERK signaling pathways by targeting IGF-IR and IRS1. T47D and MDA-MB-231 (MB-231) cells were transfected with siRNAs against IGF-IR (siIGF-IR), IRS1 (si-IRS1), or scrambled siRNA (si-Scr) (A); with pre-miR-148a, pre-miR-152, or Scr (B); or with anti-miR-148a, anti-miR-152, or Scr (C). The protein or RNA samples were harvested 72 h after the transfection. (D) The cells were infected with retrovirus carrying IGF-IR (RV-IGF-IR), IRS1 (RV-IRS1), or SCR (RV-Scr), and cultured for 48 h after the infection. The total proteins were prepared and analyzed by immunoblotting using antibodies against IGF-IR, p-IGF-IR, IRS1, p-IRS1, p-AKT, AKT, p-ERK, ERK, HIF-1α, and β-actin. Total RNAs were prepared and analyzed for VEGF mRNA expression levels using quantitative PCR and GAPDH mRNA levels as internal control (right panel). All experiments were performed in triplicate. *P < 0.05; **P < 0.01.

The expression of hypoxia-inducible factor 1α (HIF-1α) and vascular endothelial growth factor (VEGF) is important in tumor angiogenesis (Jiang et al., 2000; Semenza, 2011). In this study, we also demonstrated that miR-148a or miR-152 inhibited HIF-1α and VEGF through repression of IGF-IR and IRS1 expression (Figure 5A and B), while the blockade of miR-148a and miR-152 function or overexpression of IGF-IR and IRS1 increased HIF-1α and VEGF expression in BC cells (Figure 5C and D).

miR-148a and miR-152 inhibit cell proliferation, colony formation, and tumor angiogenesis

To investigate the biological functions of miR-148a and miR-152 as potential tumor suppressors in BC cells, we established T47D and MDA-MB-231 cell lines stably expressing scramble miRNA, miR-148a, or miR-152. Increased expression levels of miR-148a or miR-152 were verified by qRT–PCR (Supplementary Figure S3A). Since binding activity of miRNA to RNA and structural features of these complexes are crucial for gene silencing, the miRNA–RNA duplexes pairing pattern indicated that miRNAs might regulate other miRNA expression. The overexpression of let-7 is known to affect miR-107 expression (Chen et al., 2011). To test whether the overexpression of miR-148a/152 affects other miRNA expression, we randomly analyzed expression levels of eight different miRNAs in the cells stably expressing miR-148a or miR-152, and showed that the overexpression of miR-148a or miR-152 did not significantly affect other miRNA expression levels (Supplementary Figure S3B). The levels of p-AKT, p-ERK, and HIF-1α were also significantly attenuated in the stable cell lines expressing miR-148a or miR-152 (Supplementary Figure S4). The overexpression of miR-148a or miR-152 greatly decreased cell proliferation at different time points (Figure 6A) and significantly inhibited colony formation (Figure 6B). To test whether IGF-IR and IRS1 expression are sufficient to mediate miR-148a/152-inhibited cell growth and colony formation, we established T47D and MDA-MB-231 stable cell lines expressing IGF-IR or IRS1 using IGF-IR and IRS1 cDNAs lacking their 3′-UTR regions, which makes them resistant to miR-148a and miR-152 targeting in the constructs. Interestingly, forced expression of IGF-IR or IRS1 cDNAs partially or completely restored miR-148a- or miR-152-inhibited cell proliferation (Figure 6C) and colony formation (Figure 6D). These results strongly suggest that miR-148a and miR-152 down-regulation affect cell growth and colony formation via IGF-IR and IRS-1 overexpression.

Figure 6.

Figure 6

Open in a new tab

miR-148a and miR-152 inhibited cell proliferation, colony formation, and tumor angiogenesis, while the overexpression of IGF-IR or IRS1 could partly or completely restore the inhibitory effect of miR-148a and miR-152 in the cells. The T47D and MB-231 cells stably expressing Scr, miR-148a, or miR-152 were used to perform cell proliferation assay (A) and soft agar colony assay (B). The representative colonies were shown in the top panel. T47D and MB-231 cells stably expressing Scr, miR-148a, and miR-152 were infected with retrovirus carrying IGF-IR, IRS1, or Scr. Cell proliferation (C) and soft agar colony assay (D) were analyzed as above. (E) These stable T47D cells above (2 × 106 cells, 15 μl) were mixed in 1:1 ratio with Matrigel, and implanted onto the CAM of a 9-day old chicken embryo. After 4 days of implantation, the number of blood vessels was counted from six replicate CAMs, and normalized to that of the scramble control group as relative angiogenesis responses. The representative plugs from each group were shown. Scale, 2 mm. * or #P < 0.05, ** or ##P < 0.01. * or ** indicates significant difference when compared with that of the Scr; # or ## indicates significant difference when compared with the miR-148a or miR-152 group. (F) The diagram summarizes our findings. miR-148a and miR-152 expression levels are down-regulated in BC cells due to the DNA methylation of their promoters induced by DNMT1 overexpression, and miR-148a/152 in turn inhibit DNMT1 expression. IGF-IR and IRS1 are two known oncoproteins as direct targets of miR-148a/152, and down-regulation of miR-148a/152 increases IGF-IR and IRS1 expression levels in BC cells and tissues, which activates AKT and ERK signaling pathways to increase HIF-1 and VEGF expression, and tumor growth and angiogenesis.

To further study the effects of miR-148a and miR-152 on tumor angiogenesis, we showed that the overexpression of either miR-148a or miR-152 in BC cells was sufficient to inhibit tumor angiogenesis, while the overexpression of IGF-IR or IRS1 partially or completely restored the inhibitory effect of miR-148a and miR-152 in angiogenesis (Figure 6E). These results further support that IGF-IR and IRS1 are functional relevant targets of miR-148a and miR-152 in BC development, and reveal a new mechanism of IGF-IR and IRS1 induction in BC cells via suppression of miR-148a and miR-152.

Discussion

In recent years, there have been increasing interests in the role of epigenetic modifications in the etiology of human diseases (Wolffe and Matzke, 1999; Egger et al., 2004). For instance, aberrant hypermethylation of CpG islands of tumor suppressor genes and the resulting transcriptional silencing are associated with malignant transformation in cancer (Balaguer et al., 2010). The DNA methylation is carried out by DNMTs. Three main types of DNMTs (DNMT1, DNMT3A, and DNMT3B) are involved in genomic DNA methylation. DNMT1 exhibits a strong preference for hemimethylated over unmethylated DNA and maintains DNA methylation; DNMT3A and DNMT3B are responsible for establishing de novo DNA methylation (Bestor, 2000). DNMTs are ubiquitously expressed in normal human tissues (Robertson et al., 1999). In cancer, they may be overexpressed in various tumor types, e.g. in leukemia, colorectal cancer, ovarian cancer, prostate cancer, and BC (Ahluwalia et al., 2001; Mizuno et al., 2001; Karpf and Matsui, 2005; Roll et al., 2008). In the present study, we demonstrated that overexpression of DNMT1 led to the hypermethylation of miR-148a and miR-152 genes. Meanwhile, the levels of DNMT1, a direct target of miR-148a and miR-152, were inhibited by miR-148a and miR-152 overexpression. The DNMT1 and miR-148a/152 form a negative feedback loop that is disrupted in BC cells. It is interesting to note that the expression of DNMT1 in tumors tissues is mainly localized in the nucleus, where it may act to hypermethylate the miR-148a/152 gene promoters. Our results further demonstrated that there was an inverse correlation between DNMT1 and miR-148a or miR-152, indicating that the negative feedback loop observed in the cultured cells also exists in BC tumor tissues.

In gastric cancer, lower expression levels of miR-148a/152 were observed to be associated with increased tumor sizes and tumor stages (Chen et al., 2010; Zheng et al., 2011). Our study showed that the down-regulation of miR-148a and miR-152 is associated with tumor grades and LN status in BC tissues caused by the methylation of miR-148a and miR-152 promoters.

IGF-IR and IRS1 are frequently up-regulated in BC tissues. Aberrant expression of IGF-IR, via interactions with the adaptor protein IRS (Sun et al., 1991; De Meyts and Whittaker, 2002), can activate multiple downstream signaling cascades, including PI3K/AKT and MAPK/ERK signaling pathways (Berns et al., 2007; Park and Davidson, 2007), which mediate key mechanisms underlying tumor growth and progression. IRS-1 is constitutively activated in a variety of solid tumors including BC. The overexpression of IRS-1 individually in the mammary gland of mice is sufficient to promote mammary hyperplasia, tumorigenicity, and metastasis (Dearth et al., 2006), while the blockade of IRS-1 signaling in BC dramatically reduced cancer cell growth (del Rincon et al., 2004). In our study, we demonstrated a novel link between IGF-IR/IRS1 up-regulation and miR-148a/152 suppression in BC tissues. We showed that miR-148a/152 could suppress both IGF-IR/IRS1 protein expressions through binding to their 3′-UTR regions. The forced expression of either miR-148a or miR-152 was sufficient to inhibit both IGF-IR and IRS1 protein levels, suggesting that miR-148a and miR-152 share the same targets with similar effects. Moreover, we observed a significant inverse correlation of IGF-IR or IRS1 with miR-148a or miR-152 expression levels in the BC tissues, which provide strong evidence that IGF-IR/IRS1 may also be targets of miR-148a and miR-152 in vivo.

The functional effects of ectopic overexpression of miR-148a or miR-152 were similar to the effects of IGF-IR or IRS1 knockdown by siRNAs on BC cells. The miR-148a/152-mediated IGF-IR and IRS1 down-regulation is accompanied by the attenuation of IGF-IR/IRS1 phosphorylation and AKT/ERK activation, which results in the inhibition of cell proliferation, colony formation in vitro, and tumor angiogenesis in vivo. Meanwhile, restoration of IGF-IR or IRS1 expression in miR-148a- or miR-152-overexpressing cell lines can partially or completely restore miR-148a- and miR-152-inhibited cell proliferation, colony formation, and tumor angiogenesis; and restore these miRNA inhibitory effects on p-AKT and p-ERK, as well as HIF-1α and VEGF expression levels. Notably, in the restoration experiment, IRS1 yielded better cell colony formation results than IGF-IR. These findings are consistent with the previous study showing that IRS1 plays an important role in promoting cell transformation (La Rocca et al., 2009), and also suggest that IGF-IR and IRS1 are functional targets of miR-148a and miR-152. miR-148a promotes apoptosis of colorectal cancer by targeting Bcl-2, and miR-148a/152 regulate immune homeostasis by targeting CaMKIIa (Liu et al., 2010). It is likely that miR-148a and miR-152 may have different functional targets in different types of cancers which warrant further investigation.

Taken together, this report shows for the first time that (i) a new negative regulatory circuit between DNMT1 and miR-148a/152 in BC cells; (ii) the down-regulated expression levels of miR-148a/152 due to DNA methylation of their promoters are inversely correlated with tumor grades and LN statue in BC tissues; (iii) IGF-IR and IRS1 are two direct targets of miR-148a and miR-152 with biological function relevance to BC development; and (iv) miR-148a and miR-152 inhibit cell proliferation, colony formation, and tumor angiogenesis through targeting IGF-IR and IRS1 and inhibiting AKT and ERK signaling pathways (Figure 6F).

Materials and methods

Cell lines and antibodies

Human embryonic kidney cell line 293T (HEK293T), BC cell lines (MCF-7, T47D, MDA-MB-231), and mammary epithelia cell (MCF10A), were obtained from American Type Culture Collection. Ovarian surface epithelial cell lines (IOSE 397, IOSE 386) and ovarian cancer cell lines (A2780, OVCAR3) were described previously (Xia et al., 2007). The cells were cultured at 37°C in an incubator with 5% CO2. Antibodies against IGF-IR, IRS1, p-IGF-IR, p-IRS1, ERK, p-ERK, AKT, and p-AKT were purchased from Cell Signaling Technology. Antibodies against HIF-1α and DNMT1 were from BD Biosciences, and Abcam, respectively. Cell lines were treated with 5 mM of 5-Aza-2′-deoxycytidine (5-Aza-dC) (Sigma) for 120 h, replacing the drug each day.

Tissue samples

A collection of 68 breast tissues analyzed in this study, which included 22 pairs of human breast tumor samples and adjacent normal tissues and 24 breast tumor samples, were from the tissue bank of the first affiliated hospital with Nanjing Medical University and the first affiliated hospital with Anhui Medical University. All tumor samples were collected immediately after the surgical removal, and snap-frozen in liquid nitrogen. None of the patients received preoperative chemotherapy. All patients provided written informed consent and the experimental procedures were approved by the Institutional Review Board of the Nanjing Medical University and Anhui Medical University. Before preparing RNA, DNA, and protein samples, sections were stained with H&E for histological diagnosis and tumor cell evaluation. Only those cases with >70% tumor cell population in the section were used in this study. The diagnosis and histological grade of all the cases were independently confirmed by two pathologists based on World Health Organization (WHO) classification. Clinicopathologic features of patient are presented in Table 1.

DNA methylation analysis

Genomic DNAs were modified with sodium bisulfite using the EpiTect Kit (Qiagen), and analyzed by MSP. For bisulfate-sequencing PCR, amplified bisulfate PCR products were subcloned into the pBluescript KS(±) vector. Eight independent clones for each sample were picked, and the T7 primers were used to sequence inserted fragments. The primers used are summarized in Supplementary Table S1.

In situ hybridization

The slides were treated according to the manufactory instruction and hybridized with 10 pmol probe (LNA-modified and DIG labeled oligonucleotide; Exiqon) complementary to miR-148a, miR-152, or U6. After incubation with anti-DIG-HRP Fab fragments conjugated to horseradish peroxidase, the hybridized probes were detected by incubating with 3, 3′-diaminobenzidine solution and nuclei were counterstained with hematoxylin Carazzi. A positive control (U6, Exiqon) was included for each hybridization procedure.

Targetsearch program

We developed a new algorithm ‘Targetsearch’ by taking the advantage of the combination of publicly available search engines: microRNA, TargetScan, FindTar 3, and RNA 22 to obtain the putative targets (Xu et al., 2012). In brief, Targetsearch miRNA target prediction program requires microRNA binding sites with at least six matching seed-pairing, miRNA:target heteroduplexes structure (which could be separated by 0–3 bp non-matching sequence) and certain free energy for the microRNA binding to be predicted targets of miRNAs.

Chicken chorioallantoic membrane assay

Tumor angiogenesis was analyzed by chicken chorioallantoic membrane (CAM) assay as described previously (Brooks et al., 1998). In brief, white fertilized chicken eggs were incubated at 37°C under conditions of constant humidity. An artificial air sac was created over a region containing small blood vessels in the CAM. The cells were implanted onto the CAM. Angiogenesis responses were analyzed 4 days after the implantation, and the numbers of total blood vessels were counted by two observers in a double-blind manner.

Statistical analysis

The results were analyzed using the SPSS 13 statistical software. Quantitative variables were analyzed using Student's test, Wilcoxon test, and Mann–Whitney U-test. The correlations between miRNA expression levels and clinicopathologic parameters were analyzed using the Mann–Whitney U-test when comparing the differences between two groups, and using the Kruskal–Wallis test when comparing the differences among three or more groups. The correlations were analyzed using Spearman's rank test. The differences were considered significant at P < 0.05.

Supplementary material

Supplementary material is available at Journal of Molecular Cell Biology online.

Funding

This work was supported in part by the National Key Basic Research Program of China (2011CB504003), by National Natural Science Foundation of China (81071642 and 30871296), and by National Cancer Institute, NIH (R01CA109460).

Conflict of interest: none declared.

Acknowledgements

We thank all the subjects for their participation in this study. We also thank Faton Agani (Department of Pathology, Anatomy and Cell Biology, Thomas Jefferson University, Philadelphia, PA, USA) for revising the manuscript.

References

  1. Ahluwalia A., Hurteau J.A., Bigsby R.M., et al. DNA methylation in ovarian cancer. II. Expression of DNA methyltransferases in ovarian cancer cell lines and normal ovarian epithelial cells. Gynecol. Oncol. 2001;82:299–304. doi: 10.1006/gyno.2001.6284. doi:10.1006/gyno.2001.6284. [DOI] [PubMed] [Google Scholar]
  2. Balaguer F., Link A., Lozano J.J., et al. Epigenetic silencing of miR-137 is an early event in colorectal carcinogenesis. Cancer Res. 2010;70:6609–6618. doi: 10.1158/0008-5472.CAN-10-0622. doi:10.1158/0008-5472.CAN-10-0622. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bartel D.P. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell. 2004;116:281–297. doi: 10.1016/s0092-8674(04)00045-5. doi:10.1016/S0092-8674(04)00045-5. [DOI] [PubMed] [Google Scholar]
  4. Bernardino J., Roux C., Almeida A., et al. DNA hypomethylation in breast cancer: an independent parameter of tumor progression? Cancer Genet. Cytogenet. 1997;97:83–89. doi: 10.1016/s0165-4608(96)00385-8. doi:10.1016/S0165-4608(96)00385-8. [DOI] [PubMed] [Google Scholar]
  5. Berns K., Horlings H.M., Hennessy B.T., et al. A functional genetic approach identifies the PI3K pathway as a major determinant of trastuzumab resistance in breast cancer. Cancer Cell. 2007;12:395–402. doi: 10.1016/j.ccr.2007.08.030. doi:10.1016/j.ccr.2007.08.030. [DOI] [PubMed] [Google Scholar]
  6. Bestor T.H. The DNA methyltransferases of mammals. Hum. Mol. Genet. 2000;9:2395–2402. doi: 10.1093/hmg/9.16.2395. doi:10.1093/hmg/9.16.2395. [DOI] [PubMed] [Google Scholar]
  7. Birgisdottir V., Stefansson O.A., Bodvarsdottir S.K., et al. Epigenetic silencing and deletion of the BRCA1 gene in sporadic breast cancer. Breast Cancer Res. 2006;8:R38. doi: 10.1186/bcr1522. doi:10.1186/bcr1522. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Brooks P.C., Silletti S., von Schalscha T.L., et al. Disruption of angiogenesis by PEX, a noncatalytic metalloproteinase fragment with integrin binding activity. Cell. 1998;92:391–400. doi: 10.1016/s0092-8674(00)80931-9. doi:10.1016/S0092-8674(00)80931-9. [DOI] [PubMed] [Google Scholar]
  9. Brooks J., Cairns P., Zeleniuch-Jacquotte A. Promoter methylation and the detection of breast cancer. Cancer Causes Control. 2009;20:1539–1550. doi: 10.1007/s10552-009-9415-y. doi:10.1007/s10552-009-9415-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Calin G.A., Ferracin M., Cimmino A., et al. A MicroRNA signature associated with prognosis and progression in chronic lymphocytic leukemia. N. Engl. J. Med. 2005;353:1793–1801. doi: 10.1056/NEJMoa050995. doi:10.1056/NEJMoa050995. [DOI] [PubMed] [Google Scholar]
  11. Campan M., Weisenberger D.J., Laird P.W. DNA methylation profiles of female steroid hormone-driven human malignancies. Curr. Top. Microbiol. Immunol. 2006;310:141–178. doi: 10.1007/3-540-31181-5_8. doi:10.1007/3-540-31181-5_8. [DOI] [PubMed] [Google Scholar]
  12. Chang Q., Li Y., White M.F., et al. Constitutive activation of insulin receptor substrate 1 is a frequent event in human tumors: therapeutic implications. Cancer Res. 2002;62:6035–6038. [PubMed] [Google Scholar]
  13. Chen Y., Song Y., Wang Z., et al. Altered expression of MiR-148a and MiR-152 in gastrointestinal cancers and its clinical significance. J. Gastrointest. Surg. 2010;14:1170–1179. doi: 10.1007/s11605-010-1202-2. doi:10.1007/s11605-010-1202-2. [DOI] [PubMed] [Google Scholar]
  14. Chen P.S., Su J.L., Cha S.T., et al. miR-107 promotes tumor progression by targeting the let-7 microRNA in mice and humans. J. Clin. Invest. 2011;121:3442–3455. doi: 10.1172/JCI45390. doi:10.1172/JCI45390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. De Meyts P., Whittaker J. Structural biology of insulin and IGF1 receptors: implications for drug design. Nat. Rev. Drug Discov. 2002;1:769–783. doi: 10.1038/nrd917. doi:10.1038/nrd917. [DOI] [PubMed] [Google Scholar]
  16. Dearth R.K., Cui X., Kim H.J., et al. Mammary tumorigenesis and metastasis caused by overexpression of insulin receptor substrate 1 (IRS-1) or IRS-2. Mol. Cell. Biol. 2006;26:9302–9314. doi: 10.1128/MCB.00260-06. doi:10.1128/MCB.00260-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. del Rincon S.V., Guo Q., Morelli C., et al. Retinoic acid mediates degradation of IRS-1 by the ubiquitin-proteasome pathway, via a PKC-dependant mechanism. Oncogene. 2004;23:9269–9279. doi: 10.1038/sj.onc.1208104. [DOI] [PubMed] [Google Scholar]
  18. Egger G., Liang G., Aparicio A., et al. Epigenetics in human disease and prospects for epigenetic therapy. Nature. 2004;429:457–463. doi: 10.1038/nature02625. doi:10.1038/nature02625. [DOI] [PubMed] [Google Scholar]
  19. Hanoun N., Delpu Y., Suriawinata A.A., et al. The silencing of microRNA 148a production by DNA hypermethylation is an early event in pancreatic carcinogenesis. Clin. Chem. 2010;56:1107–1118. doi: 10.1373/clinchem.2010.144709. doi:10.1373/clinchem.2010.144709. [DOI] [PubMed] [Google Scholar]
  20. Hennessy B.T., Smith D.L., Ram P.T., et al. Exploiting the PI3K/AKT pathway for cancer drug discovery. Nat. Rev. Drug Discov. 2005;4:988–1004. doi: 10.1038/nrd1902. doi:10.1038/nrd1902. [DOI] [PubMed] [Google Scholar]
  21. Hildebrandt M.A., Gu J., Lin J., et al. Hsa-miR-9 methylation status is associated with cancer development and metastatic recurrence in patients with clear cell renal cell carcinoma. Oncogene. 2010;29:5724–5728. doi: 10.1038/onc.2010.305. doi:10.1038/onc.2010.305. [DOI] [PubMed] [Google Scholar]
  22. Huang J., Wang Y., Guo Y., et al. Down-regulated microRNA-152 induces aberrant DNA methylation in hepatitis B virus-related hepatocellular carcinoma by targeting DNA methyltransferase 1. Hepatology. 2010;52:60–70. doi: 10.1002/hep.23660. doi:10.1002/hep.23660. [DOI] [PubMed] [Google Scholar]
  23. Iorio M.V., Ferracin M., Liu C.G., et al. MicroRNA gene expression deregulation in human breast cancer. Cancer Res. 2005;65:7065–7070. doi: 10.1158/0008-5472.CAN-05-1783. doi:10.1158/0008-5472.CAN-05-1783. [DOI] [PubMed] [Google Scholar]
  24. Jiang B.H., Zheng J.Z., Aoki M., et al. Phosphatidylinositol 3-kinase signaling mediates angiogenesis and expression of vascular endothelial growth factor in endothelial cells. Proc. Natl Acad. Sci. USA. 2000;97:1749–1753. doi: 10.1073/pnas.040560897. doi:10.1073/pnas.040560897. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Jin Z., Tamura G., Tsuchiya T., et al. Adenomatous polyposis coli (APC) gene promoter hypermethylation in primary breast cancers. Br. J. Cancer. 2001;85:69–73. doi: 10.1054/bjoc.2001.1853. doi:10.1054/bjoc.2001.1853. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Karpf A.R., Matsui S. Genetic disruption of cytosine DNA methyltransferase enzymes induces chromosomal instability in human cancer cells. Cancer Res. 2005;65:8635–8639. doi: 10.1158/0008-5472.CAN-05-1961. doi:10.1158/0008-5472.CAN-05-1961. [DOI] [PubMed] [Google Scholar]
  27. Kozaki K., Imoto I., Mogi S., et al. Exploration of tumor-suppressive microRNAs silenced by DNA hypermethylation in oral cancer. Cancer Res. 2008;68:2094–2105. doi: 10.1158/0008-5472.CAN-07-5194. doi:10.1158/0008-5472.CAN-07-5194. [DOI] [PubMed] [Google Scholar]
  28. La Rocca G., Shi B., Badin M., et al. Growth inhibition by microRNAs that target the insulin receptor substrate-1. Cell Cycle. 2009;8:2255–2259. doi: 10.4161/cc.8.14.9026. doi:10.4161/cc.8.14.9026. [DOI] [PubMed] [Google Scholar]
  29. Lehmann U., Hasemeier B., Christgen M., et al. Epigenetic inactivation of microRNA gene hsa-mir-9-1 in human breast cancer. J. Pathol. 2008;214:17–24. doi: 10.1002/path.2251. doi:10.1002/path.2251. [DOI] [PubMed] [Google Scholar]
  30. Liu X., Zhan Z., Xu L., et al. MicroRNA-148/152 impair innate response and antigen presentation of TLR-triggered dendritic cells by targeting CaMKIIalpha. J. Immunol. 2010;185:7244–7251. doi: 10.4049/jimmunol.1001573. doi:10.4049/jimmunol.1001573. [DOI] [PubMed] [Google Scholar]
  31. Mirza S., Sharma G., Prasad C.P., et al. Promoter hypermethylation of TMS1, BRCA1, ERalpha and PRB in serum and tumor DNA of invasive ductal breast carcinoma patients. Life Sci. 2007;81:280–287. doi: 10.1016/j.lfs.2007.05.012. doi:10.1016/j.lfs.2007.05.012. [DOI] [PubMed] [Google Scholar]
  32. Mizuno S., Chijiwa T., Okamura T., et al. Expression of DNA methyltransferases DNMT1, 3A, and 3B in normal hematopoiesis and in acute and chronic myelogenous leukemia. Blood. 2001;97:1172–1179. doi: 10.1182/blood.v97.5.1172. doi:10.1182/blood.V97.5.1172. [DOI] [PubMed] [Google Scholar]
  33. Pan W., Zhu S., Yuan M., et al. MicroRNA-21 and microRNA-148a contribute to DNA hypomethylation in lupus CD4+ T cells by directly and indirectly targeting DNA methyltransferase 1. J. Immunol. 2010;184:6773–6781. doi: 10.4049/jimmunol.0904060. doi:10.4049/jimmunol.0904060. [DOI] [PubMed] [Google Scholar]
  34. Park B.H., Davidson N.E. PI3 kinase activation and response to Trastuzumab Therapy: what's neu with herceptin resistance? Cancer Cell. 2007;12:297–299. doi: 10.1016/j.ccr.2007.10.004. doi:10.1016/j.ccr.2007.10.004. [DOI] [PubMed] [Google Scholar]
  35. Robertson K.D., Uzvolgyi E., Liang G., et al. The human DNA methyltransferases (DNMTs) 1, 3a and 3b: coordinate mRNA expression in normal tissues and overexpression in tumors. Nucleic Acids Res. 1999;27:2291–2298. doi: 10.1093/nar/27.11.2291. doi:10.1093/nar/27.11.2291. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Roll J.D., Rivenbark A.G., Jones W.D., et al. DNMT3b overexpression contributes to a hypermethylator phenotype in human breast cancer cell lines. Mol. Cancer. 2008;7:15. doi: 10.1186/1476-4598-7-15. doi:10.1186/1476-4598-7-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sehat B., Tofigh A., Lin Y., et al. SUMOylation mediates the nuclear translocation and signaling of the IGF-1 receptor. Sci. Signal. 2010;3:ra10. doi: 10.1126/scisignal.2000628. doi:10.1126/scisignal.2000628. [DOI] [PubMed] [Google Scholar]
  38. Semenza G.L. Oxygen sensing, homeostasis, and disease. N. Engl. J. Med. 2011;365:537–547. doi: 10.1056/NEJMra1011165. doi:10.1056/NEJMra1011165. [DOI] [PubMed] [Google Scholar]
  39. Soares J., Pinto A.E., Cunha C.V., et al. Global DNA hypomethylation in breast carcinoma: correlation with prognostic factors and tumor progression. Cancer. 1999;85:112–118. doi:10.1002/(SICI)1097-0142(19990101)85:1&lt;112::AID-CNCR16&gt;3.0.CO;2-T. [PubMed] [Google Scholar]
  40. Stumpel D.J., Schotte D., Lange-Turenhout E.A., et al. Hypermethylation of specific microRNA genes in MLL-rearranged infant acute lymphoblastic leukemia: major matters at a micro scale. Leukemia. 2011;25:429–439. doi: 10.1038/leu.2010.282. doi:10.1038/leu.2010.282. [DOI] [PubMed] [Google Scholar]
  41. Sui M., Huang Y., Park B.H., et al. Estrogen receptor alpha mediates breast cancer cell resistance to paclitaxel through inhibition of apoptotic cell death. Cancer Res. 2007;67:5337–5344. doi: 10.1158/0008-5472.CAN-06-4582. doi:10.1158/0008-5472.CAN-06-4582. [DOI] [PubMed] [Google Scholar]
  42. Sun X.J., Rothenberg P., Kahn C.R., et al. Structure of the insulin receptor substrate IRS-1 defines a unique signal transduction protein. Nature. 1991;352:73–77. doi: 10.1038/352073a0. doi:10.1038/352073a0. [DOI] [PubMed] [Google Scholar]
  43. Tamimi R.M., Colditz G.A., Wang Y., et al. Expression of IGF1R in normal breast tissue and subsequent risk of breast cancer. Breast Cancer Res. Treat. 2011;128:243–250. doi: 10.1007/s10549-010-1313-1. doi:10.1007/s10549-010-1313-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Toyota M., Suzuki H., Sasaki Y., et al. Epigenetic silencing of microRNA-34b/c and B-cell translocation gene 4 is associated with CpG island methylation in colorectal cancer. Cancer Res. 2008;68:4123–4132. doi: 10.1158/0008-5472.CAN-08-0325. doi:10.1158/0008-5472.CAN-08-0325. [DOI] [PubMed] [Google Scholar]
  45. Visvanathan K., Sukumar S., Davidson N.E. Epigenetic biomarkers and breast cancer: cause for optimism. Clin. Cancer Res. 2006;12:6591–6593. doi: 10.1158/1078-0432.CCR-06-2001. doi:10.1158/1078-0432.CCR-06-2001. [DOI] [PubMed] [Google Scholar]
  46. Vivanco I., Sawyers C.L. The phosphatidylinositol 3-kinase AKT pathway in human cancer. Nat. Rev. Cancer. 2002;2:489–501. doi: 10.1038/nrc839. doi:10.1038/nrc839. [DOI] [PubMed] [Google Scholar]
  47. Wilson A.S., Power B.E., Molloy P.L. DNA hypomethylation and human diseases. Biochim. Biophys. Acta. 2007;1775:138–162. doi: 10.1016/j.bbcan.2006.08.007. [DOI] [PubMed] [Google Scholar]
  48. Wolffe A.P., Matzke M.A. Epigenetics: regulation through repression. Science. 1999;286:481–486. doi: 10.1126/science.286.5439.481. doi:10.1126/science.286.5439.481. [DOI] [PubMed] [Google Scholar]
  49. Xia C., Meng Q., Liu L.Z., et al. Reactive oxygen species regulate angiogenesis and tumor growth through vascular endothelial growth factor. Cancer Res. 2007;67:10823–10830. doi: 10.1158/0008-5472.CAN-07-0783. doi:10.1158/0008-5472.CAN-07-0783. [DOI] [PubMed] [Google Scholar]
  50. Xu Q., Liu L.Z., Qian X., et al. MiR-145 directly targets p70S6K1 in cancer cells to inhibit tumor growth and angiogenesis. Nucleic Acids Res. 2012;40:761–774. doi: 10.1093/nar/gkr730. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zheng B., Liang L., Wang C., et al. MicroRNA-148a suppresses tumor cell invasion and metastasis by downregulating ROCK1 in gastric cancer. Clin. Cancer Res. 2011;17:7574–7583. doi: 10.1158/1078-0432.CCR-11-1714. doi:10.1158/1078-0432.CCR-11-1714. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Molecular Cell Biology are provided here courtesy of Oxford University Press

RESOURCES