HIF-1α/Ascl2/miR-200b regulatory feedback circuit modulated the epithelial-mesenchymal transition (EMT) in colorectal cancer cells
Abstract
We have reported that Achaete scute-like 2 (Ascl2) transcriptionally repressed miR-200 family members and affected the epithelial-mesenchymal transition (EMT)-mesenchymal-epithelial transition (MET) plasticity in colorectal cancer (CRC) cells. However, little is known about the regulation of the Ascl2/miR-200 axis. Here, we found that hypoxia inducible factor-1α (HIF-1α) mRNA levels were positively correlated with Ascl2 mRNA levels and inversely correlated with miR-200b in CRC samples. Mechanistically, we showed that Ascl2 was a down-
stream target of HIF-1α and had a critical role in the EMT phenotype induced by hypoxia or HIF-1α over- expression. Hypoxia or HIF-1α over-expression activated Ascl2 expression in CRC cells in a direct transcriptional mechanism via binding with the hypoxia-response element (HRE) at the proximal Ascl2 promoter. HIF-1α- induced Ascl2 expression repressed miR-200b expression to induce EMT occurrence. Furthermore, we found HIF-1α was a direct target of miR-200b. MiR-200b bound with the 3′-UTR of HIF-1α in CRC cells. HIF-1α/Ascl2/ miR-200b regulatory feedback circuit modulated the EMT-MET plasticity of CRC cells. Our results confirmed a novel HIF-1α/Ascl2/miR-200b regulatory feedback circuit in modulating EMT-MET plasticity of CRC cells, which could serve as a possible therapeutic target.
1. Introduction
Colorectal cancer (CRC) is the third most common malignancy worldwide [1], and metastasis is responsible for the majority of CRC- related mortality. The plasticity of epithelial and mesenchymal features involves the epithelial-mesenchymal transition (EMT) and its reverse process, mesenchymal-epithelial transition (MET). As a result, this pathway is one of the central mechanisms for the induction of invasion and tumor metastasis and is a process by which epithelial cells lose their polarity and are converted to a mesenchymal phenotype. Although there still is debate concerning the role of EMT in the metastasis of certain types of cancers, EMT is generally considered to be a crucial step in the metastatic cascades of many types of malignancies, including CRC [2–6].
Hypoxia inducible factor-1 (HIF-1) transcription complex consists of an oxygen-sensitive α-subunit (HIF-1α) and a constitutively expressed β-unit (HIF-1β), which is activated and stabilized by intratumoral hypoxia followed by stabilization and activation of the HIF-1α transcrip- tion complex [7–9]. This is one of the most important mechanisms that promotes metastasis and thereby increases tumor aggressiveness. HIF-1α is one of the most important transcriptional regulators, which pro- motes the EMT process and metastasis in malignancy by inducing the expression of certain EMT regulators, such as Snail, ZEB1, TWIST and so on [10–12]. However, whether hypoxia activates other critical EMT regulators in CRC, which may have differential and non-redundant roles, remains to be explored further.
Recent studies indicate that the miR-200 family plays critical roles in arresting cell migration and modulating epithelial-mesenchymal transition in a wide range of cancer cells [13–15]. The miR-200b, one member of miR-200 family which modulates EMT via miR-200b/ZEB1 feedback loop in breast cancer cells [16], has been shown to be down- regulated by Hypoxia/HIF-1α as a hypoxamiR in human microvascular
endothelial cells [17]. Thus, we hypothesized that hypoxia or HIF-1α could promote EMT via downregulating miR-200b in colorectal cancer. However, how HIF-1α mediates miR-200b is still unknown.
Recently, our group reported that Achaete Scute-like 2 (Ascl2) was a critical EMT regulator in CRC cells via suppressing miR-200b and med- iating its downstream EMT regulators and markers [14]. In HIF-deficient trophoblast stem (TS) cells, the expression of Mash2 (Ascl2), which is a transcription factor essential for spongiotrophoblast development, was abolished, and cells become syncytiotrophoblasts instead, suggesting that Ascl2 might be a HIF target gene [18,19]. Considering the important roles of both HIF-1α and miR-200b in the EMT process, we hypothesized that Ascl2 is involved in the HIF-1α-mediated regulation of miR-200b and EMT plasticity in CRC cells. In this study, we present the first the regulation of HIF-1 activity during extended periods of hypoxia and modulates the EMT-MET plasticity of CRC cells.
2. Materials and methods
18s Forward: 5′-GGCCCTGTAATTGGAATGAGTC−3′ 146 bp Reverse: 5′- CCAAGATCCAACTACGAGCTT−3′miR-200b Forward: 5′- GGGGTAATACTGCCTGGT−3′ 63 bp Reverse: 5′- TGCGTGTCGTGGAGTC−3′
2.1. Cell culture and oxygen deprivation
Human colonic adenocarcinoma cell lines LS174T and SW480 were obtained from the Chinese Academy of Sciences Cell Bank of Type Culture Collection (Shanghai, China) and grown in McCoy’s 5 A U6 Forward: 5′- GCTTCGGCAGCACATATACTAAAAT−3′ Reverse: 5′- CGCTTCACGAATTTGCGTGTCAT−3′
2.2. Tissue samples
Colorectal cancer patients (n = 50) who were scheduled for colo- noscopy or surgical resection were analyzed in the study. Primary tumor samples and the corresponding non-cancerous matched tissues were obtained by biopsy or from surgical resection. The fresh specimens were immediately stored in liquid nitrogen for RNA extraction and further analysis of HIF-1α and Ascl2 mRNA levels and miR-200b ex-
pression levels using real-time PCR. This study was approved by the local clinical research ethics committee. All of the subjects provided informed consent before their colonoscopy or resection surgery.
2.3. Total RNA extraction and real-time PCR analysis
Total RNA containing the miRNA fraction was extracted from cells or tissues using a PureLink RNA kit (Promega, USA) following the manufacturer’s protocol. RT primer mix and specific reverse-transcribed primers for U6 and miR-200b synthesis were used for cDNA synthesis using the PrimeScript RT Reagent Kit (TaKaRa Biotechnology Co, Ltd. Dalian, China). The specific reverse-transcribed primers were: U6: CGCTTCACGAATTTGCGTGTCAT; and miR-200b: GTCGTATCCAGTGC GTGTCGTGAGTCGCAATTGCACTGATACGACTCATCAT. We used the SYBR Premix Ex Taq TM Green II (TaKaRa Biotechnology Co, Ltd.Dalian, China) for real-time PCR reactions, and the relative expressions Proteins were extracted from cells using SDS sample buffer con- taining a protease inhibitor mixture (Roche Applied Science). Protein extracts were incubated for 5 min at 95 °C prior to separation by SDS- PAGE. Following SDS-PAGE, the proteins were transferred onto PVDF membranes. The membranes were then blocked with 5% dry milk in TBST and incubated with a specific primary antibody overnight at 4 °C.The α-tubulin was used as a control. The detailed western blotting procedure was described previously. The primary antibodies used in this study were summarized in Table 2.
2.5. Establishment of stable HIF-1α knockdown or over-expression cells
LS174T cell lines were transfected with lentiviruses containing short hairpin RNA against HIF-1α (sh-HIF-1α/LS174T) or non-target control (sh-Ctr/LS174T) for HIF-1α knocked down. The sequences for targeting HIF-1α were as follows: siHIF-1α (1), forward: 5′-GCUGAUUUG UGAACCCAUUTT-3′ and reverse: 5′-AAUGGGUUCACAAAUCAGCTT-3′;siHIF-1α (2), forward: 5′-GCCGCUCAAUUUAUGAAUATT-3′ and re- verse: 5′-UAUUCAUAAAUUGAGCGGCTT-3′; and siHIF-1α (3), forward: 5′-GCCGAGGAAGAACUAUGAATT-3′ and reverse: 5′-UUCAUAGUUCUUCCUCGGCTT-3′. The siHIF-1α (1) for inhibiting HIF-1α expression was most efficient (not shown) and inserted into the lentivirus vector system (LV3 vector, GenePharma, Shanghai), and the non-specific control, forward: 5′-UUCUCCGAACGUGUCACGUTT-3′ and reverse: 5′- ACGUGACACGUUCGGAGAATT-3′; was also designed in these experi- ments. For stable HIF-1α over-expression in SW480 cells in normoxic conditions, site-directed mutagenesis of double point mutants (amino acids 402/564 in the HIF-1α CDS) was performed using the mutagenic primers (the mutated sites were underlined): p402a-F: ACTTTGCTGGCCGCAGCCGCTGG, p402a-R: CCAGCGGCTGCGGCCAGCAAAGT; p564a-F: GATGTTAGCTGCTTATATCCCAAT, p564a-R: ATTGGGATATAAGCAGCTAACATC, and the pcDNA3.1 vector containing 2481 bp HIF- 1α CDS region. These amino acids can be hydroxylated by proline-hy- droxylase-2 (PHD2), leading to degradation of endogenous HIF-1α. The double mutant HIF-1α (402/564) CDS region was inserted into the lentivirus vector system for complete viral packaging and titer measurements (GenePharma, Shanghai). SW480 cells were infected with the lentiviruses using the LV5 vector (GenePharma, Shanghai) con- taining mutated the HIF-1α (402/564) CDS region. After a 48 h infection, puromycin treatment at a concentration of 2 μg/ml was used to select for the positively-transduced cells. Successful knockdown or over-expression of HIF-1α was validated by real-time PCR and Western blot.
2.6. Migration and invasion assays
The in vitro cell migration and invasion ability was examined by trans-well assays. Cells (1×105) in 200 µL of McCoy’s 5 A medium containing 1% FBS were seeded onto the upper chamber with 8-µm pore filters (Corning, USA), and the lower chamber was filled with 600 µL of culture medium containing 20% FBS as a chemoattractant. The cells were uncoated for the migration assay, or coated with a thin layer of BD Matrigel Matrix (BD Biosciences, Sparks, MD) for the in- vasion assay according to the manufacturer’s protocol. After incubation, cells that had migrated and invaded were fixed with 4% paraf- ormaldehyde and subsequently stained with 0.1% crystal violet for assessment. For each experimental group, a minimum of three random fields of stained cells were photographed by a camera connected to a phase contrast microscope and counted.
2.7. siRNA knockdown for Ascl2 expression analysis
Ascl2 siRNA or nontarget control was transfected into HIF-1α- overexpressed SW480 cells using the FuGene HD Transfection Reagent
(Promega, USA). The siRNA sequences were designed and synthesized by GenePharma Co. Ltd (Shanghai, China). The sequences of the two Ascl2 siRNA were as follows: siAscl2-1, Forward: 5′-GCGUGAAGCUGGUGAACUUTT-3′, Reverse: 5′-AAGUUCACCAGCUUCACGCTT-3′; siRNA-Ascl2-2, Forward: 5′-UCGACUUCUCCAGCUGGUUTT-3′,Reverse: 5′-AACCAGCUGGAGAAGUCGATT-3′. The cells were used to examine Ascl2 expression after 24 h of transfection.
2.8. Transfection of miR-200b inhibitor and miR-200b mimic
For the functional test of miR-200b in vitro, LS174T and SW480 cells were transiently transfected with a miR-200b inhibitor versus the negative control or a miR-200b mimic versus the negative control fol- lowing the previously described manufacturer’s protocol (GenePharma Co. Ltd. Shanghai, China).
2.9. Luciferase reporter assays
Transcriptional activity of the Ascl2 promoter and miR-200b pro- moter were examined by luciferase reporter assay as described in our previous report. In brief, for generating a Ascl2 promoter construct, the fragments (− 628 to + 620 bp, −82 to + 620 bp and + 484 to + 620 bp) located 5′-upstream of the Ascl2 coding regions (CDS) were cloned in the Kpn I/Nhe I sites of the pGL3-Basic vector and the re-
combinant construct served as the wild-type construct. The Ascl2 pro- moter recombinant constructs, which harbor mutations in the HRE1 element and/or HRE2 element via PCR based site-directed mutagenesis, served as the mutated constructs. The primer sequences used for am- plification and mutation are summarized in Table 3. The pGL3-Basic construct containing the miR-200b promoter (−1528 to −13 bp) was kindly supplied by Gregory J. Goodall. Twenty-four hours after trans- fection of the Ascl2 promoter wild-type luciferase reporters or mutated constructs, the transfected cells were subject to either normoxic or hypoxic conditions for 16 h. For detecting the activity of the miR-200b promoter, HIF-1α-overexpressing SW480 cells and control cells were transfected with siAscl2 versus its control and then transfected with the miR-200b promoter construct 24 h later. To examine the negative regulatory role of miR-200b on HIF-1α, a 300 bp region containing either putative wild-type (GCAGTA) or mutated miR-200b binding sites (CCTGAA) of the HIF-1α 3′-UTR, which were synthesized by the Gen- ePharma Co. Ltd (Shanghai, China), were sub-cloned into the dual-luciferase miRNA target expression vector, pmirGLO (Promega, USA). Twenty-four hours after transfection of the miR-200b mimic or negative control into LS174T cells using FuGene HD (Promega, USA), these cells were further transfected with the pmirGLO constructs containing the wild-type or mutated HIF-1α 3′-UTR binding sequences. To confirm
whether miR-200b targeted HIF-1α suppressed the activity of Ascl2 promoter, miR-200b inhibitor and negative control were transfected into SW480 cells using FuGene HD. The pGL3-Basic vector containing the Ascl2 promoter (− 628 to + 620 bp) was transfected into SW480 cells 24 h later, and luciferase reporter assays were performed to determine the luciferase signals as described.
2.10. Chromatin immunoprecipitation (ChIP) assays
ChIP assays were performed using a ChIP assay kit (Millipore, 17408) following the protocols as described previously. Briefly, soluble chromatin was prepared from the sh-Ascl2/LS174T, sh-HIF-1α/LS174T,lentivirus-HIF-1α/SW480 (lv-HIF-1α/SW480) cells and their negative
control cells. The sheared chromatin was immunoprecipitated with 10 μg of anti-HIF-1α antibody or anti-Ascl2 antibody. Quantitative analysis of the immunoprecipitated DNA was performed using real-time PCR with two pairs of primers flanking the hypoxia-response elements of the Ascl2 promoter, or five pairs of primers flanking the E-box elements of miR-200b promoter. The ChIP-quantitative PCR data were analyzed using the2-△△Ct method in which the immunoprecipitated sample Ct value was normalized with the input DNA Ct value. The primer sequences are summarized in Table 3.
2.11. Statistical analysis
SPSS 13.0 software was used for all statistical analyses. The differ- ences were deemed significant when P < 0.05 and very significant when P < 0.01. 3. Results 3.1. Correlation between HIF-1α, Ascl2 and miR-200b expression levels in CRC samples To verify whether the HIF-1α expression level in human colorectal cancer was related to Ascl2 and miR-200b expression, quantitative real- time PCR was used to quantify the HIF-1α, Ascl2 mRNA expression levels and miR-200b levels in 50 human colorectal cancer samples. HIF- 1α and Ascl2 mRNA expression were higher in colon cancer samples, whereas miR-200b expression was lower in colon cancer samples when compared with that in the peri-cancerous mucosa (Fig. 1A–C). The HIF- 1α mRNA expression level was inversely correlated with miR-200b (P = 0.022), and the HIF-1α mRNA expression was correlated with Ascl2 expression (P = 0.045) (Fig. 1D and E) in human CRC samples. 3.2. Expression of HIF-1α, Ascl2, and EMT markers dynamically change during hypoxia in CRC cells, and hypoxia-induced EMT is partially reversed by Ascl2 interference Hypoxia induces several types of tumor cells to transform from epithelium to mesenchyme, both morphologically and biologically. The tumors include pancreatic, breast, ovarian, colon and hepatocellular cancers [20,21]. HIF (heterodimer composed of HIF1α and HIF1β subunits) regulation of Mash2 (Ascl2) promotes spongiotrophoblast differentiation [18,19]. To explore the effects of hypoxia on CRC cells, we observed the expression of HIF-1α, Ascl2, and EMT markers in LS174T and SW480 CRC cells in normoxia or hypoxia. Real-time PCR and western blot analysis showed that hypoxia (1% O2) increased the mRNA and protein expression levels of HIF-1α, Ascl2 and mesenchymal markers (N-cadherin and ZEB1), whereas it reduced the mRNA and protein expression levels of epithelial marker (E-cadherin) (Fig. 2A–C). The increased Ascl2 corresponded with HIF-1α over-expression due to hypoxia, but it was unknown whether Ascl2 mediated the HIF-1α- modulated EMT process. We produced Ascl2 knockdown in LS174T cells and observed whether this could reverse the EMT process induced by hypoxia. As shown in Fig. 2D and F, Ascl2 knockdown led to the decrease of Ascl2, N-cadherin and ZEB1 mRNA and protein expression,and the increase of E-cadherin mRNA and protein expression. The hy- poxia-induced epithelial-mesenchymal transition process was atte- nuated by Ascl2 knockdown in LS174T cells, which was supported by the reduction of N-cadherin and ZEB1 mRNA and protein expression in sh-Ascl2/LS174T cells compared with sh-Ctr/LS174T cells, which were both cultured in 1% O2. Hypoxia (1% O2) rescued Ascl2 expression and EMT progress in shRNA-Ascl2/LS174T cells compared with normoxia (20% O2), which was supported by the increase of Ascl2, N-cadherin and ZEB1 mRNA and protein expression and the reduction of E-cad- herin mRNA and protein expression. 3.3. miR-200b expression and MET induced by Ascl2 knockdown was perturbed by miR-200b inhibitor under hypoxic conditions As we previously demonstrate, Ascl2 inhibits miR-200 family members and the miR-200 family members are important in speci- fying an epithelial state [14]. To elucidate the specific role of miR- 200b in hypoxia (1% O2) response in CRC cells, quantitative real- time PCR experiments were performed, which demonstrated a sig- nificant reduction of miR-200b in hypoxic conditions in LS174T and SW480 cells (Fig. 3A). Ascl2 knockdown in LS174T cells led to miR- 200b increase, whereas increased miR-200b was attenuated by hy- poxia (Fig. 3B). These results indicated that hypoxia-induced miR- 200b reduction was mediated by Ascl2. Real-time PCR analysis de- monstrated the endogenous miR-200b expression level was down- regulated after transfection of miR-200b inhibitor in SW480 cells compared with untransfected and negative control (NC) samples (Fig. 3C). The mRNA and protein levels of EMT markers were de- tected after transfection with miR-200b inhibitor or NC inhibitor under normoxia (20% O2) or hypoxia (1% O2) in sh-Ascl2/LS174T cells versus sh-Ctr/LS174T cells. Hypoxia reduced E-cadherin and increased N-cadherin and ZEB1 mRNA and protein expression in sh- Ctr/LS174T cells. Compared with sh-Ctr/LS174T cells under hy- poxia, E-cadherin, N-cadherin and ZEB1 mRNA and protein expres- sion in sh-Ascl2/LS174T cells under hypoxia had the opposite effect, which were further reversed by transfection with miR-200b inhibitor under hypoxia (1% O2) in sh-Ascl2/LS174T cells (Fig. 3D). To further observe the cellular behavior changes under hypoxia, Ascl2 inter- ference and miR-200b inhibition trans-well cell migration and in- vasion assays were performed after transfection with miR-200b in- hibitor into sh-Ascl2/LS174T cells versus sh-Ctr/LS174T cells under hypoxia condition. The sh-Ctr/LS174T cells under hypoxia (1% O2) migrated and invaded significantly more than normoxia (20% O2). The migration and invasion ability of sh-Ascl2/LS174T cells under hypoxia was significantly inhibited when compared with sh-Ctr/ LS174T cells under hypoxia but was significantly enhanced by fur- ther transfection with miR-200b inhibitor under hypoxia (1% O2) in sh-Ascl2/LS174T cells (Fig. 3E and F). 3.4. HIF-1α interference reduced Ascl2, increased miR-200b expression and suppressed EMT, whereas the enforced double-mutant HIF-1α expression caused opposite effect To confirm whether Ascl2 and miR-200b aberrant expression under hypoxia is caused by HIF-1α, we produced HIF-1α-interfered and enforced double-mutant HIF-1α expressing CRC cells. As shown in Fig. 4A–C, Ascl2, miR-200b and EMT markers in sh-HIF-1α/ LS174T or sh-Ctr/LS174T cells under nomoxia (20% O2) or hypoxia (1% O2) were analyzed by real-time PCR and western blot. HIF-1α knockdown in LS174T cells inhibited their HIF-1α and Ascl2 ex- pression, increased miR-200b levels, and inhibited EMT progress. These cells exhibited increased mRNA and protein levels of E-cad- herin, which is considered an epithelial marker, as well as decreased mRNA and protein levels of N-cadherin and ZEB1, which are con- sidered mesenchymal markers, compared with sh-Ctr/LS174T cells in both nomoxia (20% O2) and hypoxia (1% O2) conditions. Real-time PCR and western blot were performed to analyze HIF-1α, Ascl2,miR-200b and EMT marker expression in stable double-mutant HIF- 1α over-expressed SW480 cells in normoxia (20% O2). The double- mutant HIF-1α over-expression in normoxia produced more HIF-1α, inhibited miR-200b, induced Ascl2 expression and promoted EMT progress, which exhibited decreased mRNA and protein levels of E- cadherin and increased mRNA and protein levels of N-cadherin and ZEB1 (Fig. 4D–F). 3.5. HIF-1α interference and enforced double-mutant HIF-1α expression in CRC cells altered Ascl2 binding to the miR-200b promoter and miR-200b transcription To determine whether Ascl2 mediated miR-200b transcription in response to HIF-1α or hypoxia, ChIP assays were performed to examine the binding of Ascl2 to the miR-200b promoter in sh-HIF-1α/LS174T and lv-HIF-1α/SW480 cells versus negative control cells. There were eleven E-boxes residing at the proximal promoter of miR-200b, and five regions were selected for ChIP assays using an anti-Ascl2 antibody (Fig. 5A). The ChIP3-5 experiment indicated that the binding of Ascl2 to its potential binding site (E-box) in the miR-200b promoter in sh-HIF- 1α/LS174T cells was significantly reduced compared with control cells under both normoxia and hypoxia condition. Hypoxia in sh-Ctr/LS174T cells significantly increased Ascl2 binding to its potential binding sites in the miR-200b promoter based on the ChIP 3–5 experiment, but hy- poxia in sh-HIF-1α/LS174T cells only increased Ascl2 binding to its potential binding site in the miR-200b promoter based on the ChIP5 experiment (Fig. 5B). On the contrary, HIF-1α overexpression enhanced Ascl2 binding to miR-200b promoter under normoxia (Fig. 5C). To further confirm whether HIF-1α regulated miR-200b expression via Ascl2, we performed Ascl2 interference experiments in lv-HIF-1α/ SW480 cells, Ascl2 expression in mRNA and protein levels in lv-HIF- 1α/SW480 cells transiently transfected with si-Ascl2 were significantly reduced (Fig. 5D) and si-Ascl2-2 was used for further experiments. The lv-HIF-1α/SW480 and its control cells were transfected with the pGL3-Basic construct containing the miR-200b promoter (−1528/−13 bp), which was kindly supplied by Gregory J. Goodall. The miR-200b pro- moter activity in lv-HIF-1α/SW480 cells was significantly lower than the control cells. Ascl2 transient interference led to a dramatic increase of miR-200b promoter activity in both lv-HIF-1α/SW480 and letivirus- Ctr/SW480 (lv-Ctr/SW480) cells (Fig. 5E). The miR-200b expression levels, which were inhibited by HIF-1α over-expression in lv-HIF-1α/ SW480 cells, were significantly increased by si-Ascl2 transfection in both lv-HIF-1α/SW480 and lv-Ctr/SW480 cells compared with si-Ctr transfection (Fig. 5F). These studies suggested that Ascl2 plays a critical role in HIF-1α-mediated miR-200b inhibition in CRC cells. 3.6. Ascl2 is a direct downstream regulator of HIF-1α in colorectal cancer cells To demonstrate whether Ascl2 is regulated directly by HIF-1α, we analyzed a cohort of transcription factor response elements located within the − 628 bp to + 620 bp region of the human Ascl2 gene using promoter analysis. Two putative hypoxia response elements (HREs) were identified in the proximal promoter of the Ascl2 gene (Fig. 6A). Two promoter deletion-luciferase constructs and three different site- directed HRE mutants in the Ascl2 promoter (− 628/+ 620 bp) were generated to identify the sites of transcriptional regulation within the human Ascl2 promoter that responded to HIF-1α interference or en- forced double-mutant HIF-1α expression under normoxia and hypoxia condition. A significant decrease in Ascl2 promoter (− 628/+ 620 bp; and − 82/+ 620 bp) activity was observed in sh-HIF-1α/LS174T cells compared with sh-Ctr/LS174T cells under normoxia and hypoxia con- dition. On the contrary, a significant increase in the Ascl2 promoter (− 628/+ 620 bp; and − 82/+ 620 bp) activity was observed in lv-HIF- 1α/SW480 compared with lv-Ctr/SW480 cells under normoxic condi- tions. These results suggested that HIF-1α was a critical factor for Ascl2 transcription in CRC cells. The site-directed HRE1 mutagenesis and double mutation of HRE1 and HRE2 in the Ascl2 promoter (− 628/+ 620 bp) led to reduced promoter activity in sh-HIF-1α/LS174T cells compared with sh-Ctr/LS174T cells under normoxia and hypoxia conditions. HRE2 mutation in the Ascl2 promoter (− 628/+ 620 bp) only resulted in reduced promoter activity in sh-HIF-1α/LS174T cells com- pared with sh-Ctr/LS174T cells under hypoxic conditions, but not under normoxia. HRE1 mutation and double HRE1/HRE2 mutation prevented to HRE2 in the Ascl2 promoter under hypoxia and HIF-1α binding to HRE2 was decreased in sh-HIF-1α/LS174T cells compared with sh-Ctr/ LS174T cells (Fig. 6E). HIF-1α over-expression led to a significant in- crease in HIF-1α binding to HRE1 and HRE2 in lv-Ctr/SW480 cells compared with their control cells (Fig. 6F). These results demonstrated that HIF-1α activated Ascl2 transcription via direct binding to HRE1/ HRE2 in the Ascl2 proximal promoter. 3.7. miR-200b targeted HIF-1α and further modulated Ascl2 transcription Hypoxia maintains HIF-1α protein expression because proline-hy- droxylase-2 (PHD2) and HIF-1α subunit inhibitor (HIF-1AN) are in- active under low oxygen tension [22]. Despite HIF-1α mRNA expres- sion was significantly increased in LS174T and SW480 cells (Fig. 2A), the mechanism of increased HIF-1α mRNA by hypoxia was unclear. Therefore, we hypothesized and investigated whether miR-200b played a regulatory feedback role and targeted HIF-1α expression. HIF-1α mRNA levels in SW480 cells transfected with miR-200b inhibitor were significantly increased compared with its negative control (NC in- hibitor) under normoxia and hypoxia (Fig. 7A), HIF-1α protein levels in luciferase reporter assay (Fig. 7E). The pmiRGLO vector was fused with wild-type or mutated HIF-1α 3′-UTR, which contains a putative miR- 200b target sites, and was transfected into LS174T cells, which were also transfected with miR-200b mimic or its negative control. The lu- ciferase reporter assays demonstrated that the miR-200b mimic sig- nificantly suppressed luciferase activity of the wild-type HIF-1α 3′-UTR compared with NC mimic. This suppressive effect was substantially abolished when the putative miR-200b binding site was mutated, re- vealing that miR-200b bound to the 3′-UTR of HIF-1α in CRC cells and that HIF-1α is a bona fide target of miR-200b in human CRC cells (Fig. 7F).Taken our present findings and the previous report [14] together, we suggest a novel mechanism linking the HIF-1α/Ascl2/miR-200b regulatory feedback circuit to the modulation of EMT-MET plasticity of colon cancer cells (Fig. 8). 4. Discussion Our data show that Ascl2 is a downstream target of HIF-1α and has a critical role in regulating EMT phenotypes induced by hypoxia or HIF- 1α over-expression. Although different EMT-related factors had been determined to be regulated by HIF-1α, our results indicate that HIF-1α- induced Ascl2 expression represses miR-200b to induce EMT occur- rence, suggesting that it has a differential and nonredundant role. We propose that hypoxia or HIF-1α induces EMT through the direct activation of Ascl2 expression, which may represent an early step and a critical mechanism causing hypoxia-induced tumor progression and metastasis. Furthermore, HIF-1α was also a direct target of miR-200b and the miR-200b/HIF-1α negative feedback loop modulates the EMT- MET plasticity of CRC cells. Ascl2 plays a critical role in intestinal stem cells and CRC progenitor cells [23,24]. Ascl2 is also over-expressed in colorectal cancer and has the potential to affect stem and progenitor cells during liver metastasis, resulting in the self-renewal of cancer stem cells [25–27], and in gastric cancer and lung squamous cell carcinoma, resulting in accelerated cell growth, increasing tumor resistance to 5-FU, which serves as an in- dependent prognostic indicator [28,29]. Our previous findings sug- gested that Ascl2 is upregulated in CD133+ CRC progenitor cells and maintains its self-renewability via a miR-302-dependent mechanism [24]. Ascl2 is a downstream target of the Wnt signaling pathway and the canonical Wnt/β-catenin signaling pathway is known to be essential for tumorigenesis and progression of CRC [30,31]. However, Ascl2 regulation by other signaling pathways is undefined. In this paper, we found that Ascl2 was regulated by HIF-1α under hypoxic conditions. Hypoxia is a common feature of all solid tumors and plays an essential role in tumor occurrence and development [32]. The hypoxia micro- environment could be found in CRC because of the imbalance between the oxygen supply and consumption in proliferating tumors [33]. Ample evidence indicates that hypoxia-inducible factors (HIFs) play an important role in the pathogenesis and pathophysiology of CRC. HIF-1α regulated Snail, ZEB1 or TWIST under hypoxic condition was reported to regulate the EMT-MET plasticity of cancerous cells and peritoneal mesothelial cells in the literature [10–12,34], our in vitro and in vivo experiments had provided evidence of a molecular mechanism that linked HIF-1α pathways to Wnt/β-catenin signaling. An interaction was found between Ascl2, a Wnt target gene, and HIF-1α, implying an un- derlying effect from HIF-1α, besides its effect on β-catenin and T-cell factor-4 [35,36]. Given the complex mechanism underlying this cross- talk, further efforts should be made to investigate the network of Wnt/ β-catenin and hypoxia signaling pathways. The HIF-1α-Ascl2 axis might be novel and critical mechanism to understand the crosstalk between Wnt/β-catenin and hypoxia signaling. The miR-200b, a member of the miR-200 family, is important for specifying an epithelial or mesenchymal state not only during em- bryonic development but also during tumorigenesis. The miR-200b contributes to the regulation of the plasticity between epithelial and mesenchymal features. In most cases, the transcriptional regulation of miR-200b is not fully understood. MiR-200 family members can be activated by a series of transcription factors, such as p53/p63/p73 family, Smad3 and KLF5 [37–39]. On the contrary, how miR-200b is transcriptionally repressed is largely undetermined. A double-negative feedback loop controls ZEB1-SIP1 and miR-200 family expression, regulates cellular phenotype and has direct relevance to the role of these factors in tumor progression [16]. Our previous report confirmed that Ascl2 transcriptionally represses miR-200 family expression and modulates EMT-MET plasticity [14]. We first found that HIF-1α repressed miR-200b expression via Ascl2 via a direct transcriptional mechanism and further regulated the EMT-MET plasticity of CRC cells. Our present findings expanded our previous knowledge of the Ascl2/ miR-200b axis in the regulation of EMT-MET plasticity of CRC cells, hypoxia and hypoxia-induced HIF-1α modulation of the Ascl2/miR- 200b axis, and the regulation of EMT-MET plasticity of CRC cells. This novel finding provided further mechanistic and therapeutic choices for CRC patients. HIF-1 activity is tightly regulated through the oxygen-dependent degradation of the HIF-1α subunit. Despite the central importance of hydroxylases in regulating HIF-1α, many additional factors, including cytokines, growth factors, and histone deacetylase inhibitors have also been shown to control HIF-1α expression, even under normoxic con- ditions [40–42]. However, mechanisms controlling HIF-1α mRNA level and their role in cellular responses during hypoxia have not been well understood. Our data indicated that HIF-1α mRNA was significantly increased under hypoxic condition (Fig. 2). Little is known about the dynamics of HIF-1α mRNA processing in tumor cells during hypoxic conditions. A number of different hypoxamiRs that affect HIF-1α mRNA expression either positively or negatively have been identified, in- cluding miR-20a, miR-20b, miR-130a, miR-130b, miR-155, miR-199a, miR-200b, miR-200c, miR-210, miR-424, and miR-429 [43,44]. HIF mRNA is a direct target for only four of these miRNAs: miR-20a, miR- 20b, miR-199a and miR-429 [45]. The only other miRNA that has been identified in a negative regulatory loop with HIF-1α is miR−155 in gut epithelial cells [46]. Our present experiments provide evidence of a miR-200b negative regulatory loop with HIF-1α in CRC cells and gave an explanation of HIF-1α mRNA elevation under simultaneous hypoxic conditions. In a conclusion, we confirmed that the HIF-1α/Ascl2/miR-200b regulatory feedback circuit was present in the carcinogenesis of CRC and functioned as a MK-8617 regulator of EMT-MET plasticity of CRC cells.