Regulation of Colorectal Carcinoma Stemness, Growth, and Metastasis by an Mir-200C-Sox2–Negative Feedback Loop Mechanism

Published OnlineFirst March 21, 2014; DOI: 10.1158/1078-0432.CCR-13-2348

Clinical Cancer Human Cancer Biology Research

Regulation of Colorectal Carcinoma Stemness, Growth, and Metastasis by an miR-200c-Sox2–Negative Feedback Loop Mechanism

Yan-Xia Lu1, Li Yuan1, Xiao-Lei Xue2, Min Zhou1, Yan Liu1, Chao Zhang1,3, Jing-Ping Li4, Lin Zheng1, Min Hong1, and Xue-Nong Li1

Abstract Purpose: To elucidate a novel mechanism of miR-200c in the regulation of stemness, growth, and metastasis in colorectal carcinoma (CRC). Experimental Design: Quantitative reverse transcription PCR was used to quantify miR-200c expression in CRC cell lines and tissues. A luciferase assay was adopted for the target evaluation. The functional effects of miR-200c in CRC cells were assessed by its forced or inhibited expression using lentiviruses. Results: MiR-200c was statistically lower in CRC clinical specimens and highly metastatic CRC cell lines compared with their counterparts. Sox2 was validated as a target for miR-200c. The knockdown of miR-200c signiﬁcantly enhanced proliferation, migration, and invasion in CRC cell lines, whereas the upregulation of miR-200c exhibited an inverse effect. Moreover, rescue of Sox2 expression could abolish the effect of the upregulation of miR-200c. In addition, the reduction of miR-200c increased the expression of CRC stem cell markers and the sphere-forming capacity of CRC cell lines. Further study has shown that miR-200c and Sox2 reciprocally control their expression through a feedback loop. MiR-200c suppresses the expression of Sox2 to block the activity of the phosphoinositide 3-kinase (PI3K)–AKT pathway. Conclusion: Our ﬁndings indicate that miR-200c regulates Sox2 expression through a feedback loop and is associated with CRC stemness, growth, and metastasis. Clin Cancer Res; 20(10); 2631–42. 2014 AACR.

Introduction in some human tumors, governing tumorigenesis and Cancer cells are considered to be capable of unlimited metastases (4–6). However, the functions of stemness-asso- proliferation and self-renewal, similar to stem cells. More- ciated genes in control of CRC tumorigenesis and metas- over, a small number of cancer cells express stem cell tases remain elusive. miRNAs markers and possess the stem cell–like ability to sustain Growing evidence indicates that are aberrantly tumor growth, metastasis, and recurrence. Previous studies expressed in many human cancers and involved in the suggested that carcinoma cells most likely transform into initiation, development, and metastasis of cancers. Several miRNAs cancer stem cell (CSC)-like cells due to genetic alterations or are emerging as important regulators of the proliferation and metastases of CSCs. The reduction of miR-34a microenvironment changes (1–3). Stemness-associated þ genes such as Sox2, KLF4, Bmi1, and Oct4 are involved in in CD44 prostatic CSCs assists prostate cancer develop- the maintenance of the stemness of embryonic stem cells, ment and metastasis (7). The metastasis and maintenance and more recent studies have demonstrated their expression of cancer-initiating cells that rely on KLF4-Numb–like signaling are suppressed by miR-296 in lung cancer (8). Recent findings have noted a connection between miRNAs and CSCs in CRCs. MiR-93 is downregulated during the course 1 Authors' Affiliations: Department of Pathology, School of Basic Medical of colon CSC differentiation into colon cancer cells and Sciences, 2Department of Pathology, Nanfang Hospital, Southern Medical University, 3Department of Pathology, Sun Yat-Sen University Cancer depresses the proliferation of colon CSCs (9). MiR-328 Center, Guangzhou; and 4Department of Reproductive Endocrinology, maintains a CSC-like side population (SP) phenotype in Women's Hospital, Zhejiang University School of Medicine, Hangzhou, MiR-200c China CRCs (10). is the predominant member of the miR-200 family, which suppresses epithelial–mesenchymal Note: Supplementary data for this article are available at Clinical Cancer miR-200 Research Online (http://clincancerres.aacrjournals.org/). transition (EMT), and the downregulation of family members in some tumors promotes invasion and metas- Corresponding Author: Xue-Nong Li, Department of Pathology, School of miR-200c Basic Medical Sciences, Southern Medical University, Guangzhou 510515, tasis (11–15). The significance of in cancer bio- Guangdong Province, China. Phone: 86-20-6164-8720; Fax: 86-20-6164- logic processes is becoming apparent. Moreover, there has 8720; E-mail: [email protected] been no published data about the role of miR-200c in CRC doi: 10.1158/1078-0432.CCR-13-2348 stemness. Therefore, we examined its expression and roles 2014 American Association for Cancer Research. in stemness sustainment and metastasis in CRCs.

www.aacrjournals.org 2631

Lu et al.

Transfection and lentiviral transduction Translational Relevance Lentiviral constructs expressing (Lenti-miR microRNA The acquisition of cancer-associated traits such as precursor clone collection; System Biosciences) or repres- metastasis and tumorigenicity seems to be primarily sing (miRZips lentiviral-based microRNA inhibition; System ascribed to the presence of cancer stem cell (CSC)-like Biosciences) miR-200c were packaged using the pPACKH1 cells, but the linkage mechanism has not been fully Lentivector Packaging Kit (System Biosciences) and were elucidated. Here, a novel mechanism of colorectal used to infect CRC cells to establish cells constitutively tumorigenesis associated with miR-200c was verified. expressing or repressing miR-200c, respectively. Then, the The miR-200c was statistically lower in colorectal carci- miR-200c-overexpressing cell lines were transfected with noma (CRC) clinical specimens and highly metastatic Sox2 vectors not containing its 30-untranslated region CRC cell lines (SW620 and Lovo) when compared (UTR). with their counterparts. We found that miR-200c and Sox2 reciprocally control their expression through a Quantitative reverse transcription PCR feedback loop not only modulating Sox2-induced Total RNA, including miRNA from the tissue samples and stemness but also mediating proliferation and metas- cultured cells, was extracted using TRIzol (Takara) accord- tasis through the phosphoinositide 3-kinase (PI3K)– ing to established protocols. The expression of miR-200c AKT signaling transduction pathway in CRCs. These and RNU6B (U6 snRNA, a reference gene) was analyzed by findings elucidate that miR-200c displayed the ability quantitative reverse transcription PCR (qRT-PCR) using the to regulate stemness, growth, and metastasis in CRCs, All-in-One miRNA qRT-PCR Detection Kit (GeneCopeia). and it could be considered a potential oncosuppressor The PrimeScript RT Reagent Kit (Perfect Real Time, Takara) for CRCs. was used to generate cDNA for the detection of Sox2, Bmi1, Oct4, CD133, CD166, and b-catenin mRNA. Their expression levels were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA. The relative mRNA or miR-200c C In addition, we identified Sox2 as a novel target of miR- levels were calculated using the comparative t DDC 200c. Together with Oct4 and NANOG, Sox2 encodes method ( t). Primer sequences for qRT-PCR are listed in homeodomain proteins that are required for the early Supplementary Table S1. development and propagation of undifferentiated embryonic stem cells (16). Sox2 is involved in the proliferation Western blotting or initiation of several cancers such as glioblastoma, Protein extracts were obtained using a lysis buffer and gastric, and breast cancer (17–19). Sox2 is essential for then quantified by a bicinchoninic acid (BCA) protein the transformation of foregut basal progenitor cells to the assay (KeyGen Biotech). Equivalent amounts of cell esophagus and forestomach and cooperates with inflam- lysates were separated using SDS-PAGE and transferred matory signaling pathways to induce the formation of to a polyvinylidene difluoride (PVDF) membrane (Roche carcinomas (20). Applied Sciences). The membrane was then blocked in In summary, miR-200c and its target may be involved in TBST solution containing 5% non–fat milk and incubated tumorigenesis and CSC-like cell transformation, but there is with the primary antibody at 4 C overnight, followed by no direct evidence that the network including miR-200c and the appropriate second antibody. The bands were visua- Sox2 is implicated in CRC stemness and metastasis. We lised using Pierce ECL Western Blotting Substrate therefore attempted to focus on the miR-200c-Sox2–related (Thermo Scientific). mechanism that transforms CRC stemness and modulates proliferation and metastasis. Proliferation, plate colony formation, cell migration, and invasion assays The proliferation, plate colony formation, migration, and Materials and Methods invasion of transfected CRC cells were determined as pre- Clinical specimens and cell lines viously described (21). Human CRC specimens were collected at Nanfang Hos- pital, Southern Medical University (Guangzhou, China), Soft agar assay and sphere assay with written consent. Surgically removed tissues were For soft assays, the base agar layer (1.32%) was pre- immediately frozen in liquid nitrogen and stored at 80C. pared in 6-well plates, and a single-cell solution of 3 The use of clinical materials for research purposes was 104 cells/mL was prepared in the top agar solution approved by the Southern Medical University Institutional (0.66%). For sphere assays, cells were cultured and sus- Board. The CRC cell lines SW480, SW620, HCT116, Lovo, pended in serum-free DMEM/F12 medium (Hyclone) and HT29 were obtained from the American Type Culture supplemented with basic fibroblast growth factor (20 Collection (ATCC) and authenticated according to the ng/mL), EGF (20 ng/mL), leukemia-inhibitory factor recommendation of the ATCC. All cells were cultured in (10 ng/mL; Invitrogen), and insulin (25 mg/mL; Sigma). RMPI-1640 medium containing 10% FBS (Hyclone) in 5% The number of colonies and colospheres was observed for CO2 at 37 C. more than 2 weeks.

2632 Clin Cancer Res; 20(10) May 15, 2014 Clinical Cancer Research

MiR-200c Regulates CRC Stemness, Growth, Metastasis by Sox2

Luciferase reporter assay were analyzed by one-way ANOVA or the independent- Luciferase reporter plasmids were generated by ligating samples t test. Relationships between miR-200c expres- oligonucleotides containing the wild-type (Wt) or mutant sion and clinicopathologic characteristics were tested (Mut) putative target site of the Sox2 30UTR into the psi- using Fisher exact test. Differences were considered sig- CHECK2 vectors (Promega). Cells were cotransfected with nificant if P < 0.05: , P < 0.05; , P < 0.01; , P < 0.001. the luciferase vectors and the pre-miR-200c or the negative control pre-miRNA, using Lipofectamine 2000 (Invitro- Results gen). After 48 hours, cells were lysed, and luciferase activity MiR-200c downregulation in CRC correlates with was determined using the Dual Luciferase Reporter Assay Kit metastatic status, tumor grade, and growth (Promega) according to the manufacturer’s instructions. To The average expression level of miR-200c was signifi- miR-200c generate an promoter vector, a 667-bp fragment cantly decreased in 30 of 34 CRC specimens (P < 0.001) containing the 2 binding sites of Sox2 was PCR-amplified compared with their normal counterparts (Fig. 1A and B), and inserted into a PGL3- luciferase reporter vector (Pro- with a 9.35-fold decrease in the colorectal cancer tissue mega). In addition, some mutation vectors related to the samples. Furthermore, to investigate the clinicopathologic Sox2-binding site were constructed. These PGL3-derived significance of miR-200c expression in patients with CRCs, Renilla vectors, the control pRL-TK plasmids (Promega), the median relative expression level of miR-200c in the 34 and Sox2-expressing vectors were cotransfected into CRC samples was recommended as the cutoff point for HEK293 or SW480 cells using Lipofectamine 2000 Reagent dividing miR-200c level into a low-expression group and a (Invitrogen). Luciferase activities were assayed as the afore- high-expression group. Correlation analysis showed that mentioned methods. The primer sequences used for PCR the miR-200c expression level was reversely correlated to amplification of plasmid construction are listed in Supple- tumor size (P < 0.05), serosal invasion (P < 0.05), lymph mentary Table S2. metastasis (P < 0.05), and tumor–node–metastasis (TNM) classification (P < 0.001; Supplementary Table S3). Rela- Chromatin immunoprecipitation tive miR-200c expression was significantly lower in highly Cells were cultured, fixed with 1% formaldehyde, metastatic CRC cell lines SW620 and Lovo, compared with washed, harvested, and lysed, followed by sonication to the low metastasis cell lines SW480, HT29, and HCT116 produce chromatin of primarily mononucleosomal size. (Fig. 1C). We reasoned that expression of miR-200c is Immunoprecipitation was performed overnight with anti- negatively correlated with the metastatic potential of CRCs Sox2 or anti-IgG antibodies. Protein–DNA complexes were and that miR-200c may suppress CRC growth. recovered using protein G agarose beads, washed, and then eluted. Cross-links were reversed at 65C overnight, and DNA was purified using reagents provided in the EZ-ChIP Sox2 is a novel target for miR-200c Chromatin Immunoprecipitation Kit (Millipore). The Potential targets of miR-200c were analyzed by bioinfor- immunoprecipitated DNA was amplified by PCR for matic algorithms. Sox2, Bmi1, and KLF4 were the predicted sequences containing Sox2-binding sites. targets of miR-200c in TargetScan, Pictar, and microRNA.org in unison. Bioinformatic analysis indicated a putative miR- Tumorigenesis and metastasis assay in vivo 200c target site in the Sox2 30UTR (Fig. 1D). MiRNA is Four- to 6-week-old athymic BALB/c nude mice were inversely correlated with its target genes in expression. The obtained from the Central Laboratory of Animal Science basal expression of miR-200c was inversely correlated with at Southern Medical University and housed in laminar flow Sox2 mRNA in the CRC cell lines (P ¼ 0.037, r ¼0.900), cabinets under specific pathogen-free conditions. All exper- whereas the Bmi1 and KLF4 mRNAs were not inversely imental procedures involving animals were performed in correlated with miR-200c in these cells (P ¼ 0.144, r ¼ accordance with animal protocols approved by the Animal 0.396; P ¼ 0.136, r ¼0.404; Fig. 2A). Accordingly, we Care and Use Committee of Southern Medical University. chose Sox2 as a preferential, predicted target of miR-200c. Xenograft tumors were generated by subcutaneous injection Sox2 expression in colorectal cancer (91.2%, 31 of 34) was of 2 106 cells. From the seventh day after injection, the size significantly more frequent than that in the corresponding of the tumor was measured as described previously (22). For normal tissue (Fig. 2B). Spearman correlation analyses orthotropic metastasis assays, nude mice were anesthetized, demonstrated that Sox2 and miR-200c were again inversely and their caeca were exteriorized by laparotomy. The sub- related in expression (P ¼ 0.01, r ¼0.870; Fig. 2C). We cutaneous tumors were cut into small masses and embed- also quantified the protein level of Sox2 in the CRC cell lines ded into the mesentery at the tail end of the caecum. The gut (Fig. 2D). was reposited to the abdominal cavity and subsequently There was scarcely any expression of miR-200c in the closed with surgical sutures. Six weeks later, the mice were HEK293 cells (23). First, we cotransfected the luciferase sacrificed, and all organs were resected for biopsy. plasmids Wt Sox2 30UTR or Mut Sox2 30UTR and pre-miR- 200c into the HEK293 cells and compared these cells with Statistical analysis cells transfected with negative control pre-miRNA.We Quantitative values of all experiments are expressed as the observed that the exogenous expression of miR-200c signif- mean SD. Differences among/between sample groups icantly decreased the luciferase activity of the Wt Sox2

www.aacrjournals.org Clin Cancer Res; 20(10) May 15, 2014 2633

Lu et al.

Figure 1. MiR-200c downregulation in CRC correlates with growth and metastasis. A, qRT-PCR analysis of miR-200c expression in 34 paired human colorectal cancer tissues; miR-200c expression was expressed relative to the matched adjacent normal tissues. B, comparison of miR-200c abundance in 34 paired primary CRC tissues (T) with paired adjacent normal tissues (N). C, qRT-PCR analysis of miR-200c expression in CRC cell lines. Experiments were performed 3 times. Data, mean SD. D, schematic diagram of Sox2–30UTR. Sequences were compared between the mature miR-200c and the wild-type (Wt) or mutant (Mut) putative target sites in the 30UTR of Sox2 mRNA. , P < 0.01; , P < 0.001.

30UTR but not the Mut Sox2 30UTR (P < 0.001; Fig. 2E). To (P < 0.05; Supplementary Fig. S1B), and scramble was determine whether endogenous miR-200c levels regulated used as a control. We observed an inverse change in Sox2 Sox2 expression, SW620 and HCT116 cells were transfected mRNA and protein expression when the level of the miR- with the Wt Sox2 30UTR or Mut Sox2 30UTR. The luciferase 200c was altered (P < 0.05; Supplementary Fig. S1C and activity of Wt Sox2 30UTR was lower in the HCT116 cells (P < S1D). These results suggest that miR-200c regulates Sox2 0.001; Fig. 2E), but the luciferase activity of Mut Sox2 30UTR expression through mRNA degradation and translation had no signiﬁcant difference in HCT116 and SW620 cells. silencing. The cotransfection of miR-200c/Sox2 (not con- Therefore, Sox2 is a direct target of miR-200c, and the taining its 30UTR) rescued Sox2 expression (Supplemen- negative regulation of Sox2 by miR-200c is due to the tary Fig. S1C). miR-200c–binding sites in the Sox2 30UTR. A recent publi- The upregulation of miR-200c restrained cell proliferation cation reports that the miR-200 family promotes murine compared with control cells, as measured by CCK-8 cell neural progenitor cell-cycle exit and differentiation by proliferation assays (P < 0.05; Fig. 3A) and plate colony directly targeting Sox2 (24). formation assay (P < 0.001; Fig. 3B). However, reducing miR-200c in CRC cell lines yielded the opposite effect (P < MiR-200c represses Sox2 to suppress CRC growth and 0.05; Fig. 3A and P < 0.001; Supplementary Fig. S2A). This invasion in vitro evidence indicates that miR-200c could impede the prolif- The miR-200c expression lentiviruses were stably trans- eration of CRC cells in vitro. duced into SW480 and SW620 cells generating sub-cell Thesoftagarassaysshowedthatoverexpressionof lines SW480/miR-200c and SW620/miR-200c,whichper- miR-200c caused a marked reduction of anchorage-inde- sistently overexpressed mature miR-200c (P < 0.05; Sup- pendent growth ability, as indicated by the reduction in plementary Fig. S1A); their controls were infected with colony number and volume in soft agar (P < 0.001; lentiviruses containing empty vectors. Meanwhile, miR- Fig. 3C). Inversely, the clonogenicity in soft agar was 200c was permanently knocked down in SW480 and observably increased in SW480 and HCT116 cells deplet- HCT116 by the anti-miR-200c lentiviruses producing the ed of endogenous miR-200c in contrast with scramble- SW480/zip-200c and HCT116/zip-200c sub-cell lines transfected cells (P < 0.001; Supplementary Fig. S2B).

2634 Clin Cancer Res; 20(10) May 15, 2014 Clinical Cancer Research

MiR-200c Regulates CRC Stemness, Growth, Metastasis by Sox2

A 200 800 800 Sox2 Bmi1 KLF4 180

160 140

120 600 600 20 Relative expression Relative expression Relative expression

0 00 SW480 SW620 Lovo HCT116 HT29 SW480 SW620 Lovo HCT116 HT29 SW480 SW620 Lovo HCT116 HT29 B 40 1N 1T 2N 2T 3N 3T 4T4N Sox2 30 Tubulin

20 5N 5T 6N 6T 7N 7T 8T8N Sox2 10 Tubulin

0 Fold change of Sox2 (T/N) of Sox2 Fold change 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

CE1.5 Control miR-200c 35 3.5 SoX2 3 30 miR-200c 1.0 25 2.5

20 2 0.5 15 1.5

10 1 Relative luciferase activity Relative luciferase 0.0 5 0.5 Wt Mut

Relative expression of Sox2 Relative expression 0 HEK293 0 of miR-200c Relative expression 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 2627 28 29 3031 32 33 34 50 Wt

40 Mut D SW480 SW620 Lovo HCT116 HT29 30

Sox2 20

10 Tubulin

Relative luciferase activity Relative luciferase 0 HCT116 SW620

Figure 2. Sox2 is a direct target of miR-200c. A, relative expression of Sox2, Bmi1, and KLF4 in CRC cell lines determined by qRT-PCR. B, expression of Sox2 in CRC tissues. Left, qRT-PCR analysis of Sox2 expression in 34 paired human colorectal cancer tissues and Sox2 expression was expressed relative to the matched adjacent normal tissues. Right, Western blot analyses of Sox2 expression in 8 paired CRC tissues. C, Spearman correlation analysis showed a negative relationship between the miR-200c expression level and the Sox2 mRNA in 34 cases of CRC tissue samples. D, Western blot analyses of Sox2 in CRC cell lines. E, relative luciferase activity of the Wt Sox2 30UTR or Mut Sox2 30UTR in HEK293 cells (top) transfected with pre-miR-200c (miR-200c) or negative control pre-miRNA (Control) and in SW620 and HCT116 cells (bottom). Values, mean SD of 3 independent assays. , P < 0.01; , P < 0.001.

Therefore, miR-200c is involved in the regulation of migration and invasion of HCT116 and SW480 cells, tumorigenesis in CRC cells. compared with scramble-transfected cells (P < 0.001; Migration and invasion assays showed that miR-200c Supplementary Fig. S2C and S2D).Therefore, miR-200c overexpression markedly repressed the motility and inva- dramatically attenuated the migratory and invasive abil- siveness of SW480 and SW620 cells as compared with ities of CRC cells. their control cells (P < 0.001; Fig. 3D and E). However, As an indication of the function of miR-200c in CRC cell the knockdown of miR-200c signiﬁcantly enhanced the lines, we attempted to determine the mechanisms of its

www.aacrjournals.org Clin Cancer Res; 20(10) May 15, 2014 2635

Lu et al.

Blank Blank Blank Black A Control Control 1.5 1.5 2.0 Scramble 2.0 Scramble miR-200c miR-200c zip-200c zip-200c miR-200c/Sox2 miR-200c/Sox2 1.5 1.5 1.0 1.0 1.0 1.0

OD value 0.5 OD value 0.5 OD value 0.5 OD value 0.5

0.0 0.0 0.0 0.0 1d 2d 3d 4d 5d 6d 1d 2d 3d 4d 5d 6d 1d 2d 3d 4d 5d 6d 1d 2d 3d 4d 5d 6d SW480 SW620 HCT116 SW480

miR-200c Blank Control miR-200c /Sox2 BCmiR-200c Blank Control miR-200c /Sox2

SW480 SW480

SW620 SW620

25 200 Blank Blank Control Control 20 150 miR-200c miR-200c 15 100 miR-200c/Sox2 miR-200c/Sox2 10 50

Colony number Colony 5

0 rate (%) Colony-forming 0 SW480 SW620 SW480 SW620

miR-200c miR-200c DEBlank Control miR-200c /Sox2 Blank Control miR-200c /Sox2

SW480 SW480

SW620 SW620

60 60 Blank Blank Control Control 40 miR-200c 40 miR-200c miR-200c/Sox2 miR-200c/Sox2 20 20

0 cells of invaded Numbers Numbers of migrated cells Numbers SW480 SW620 SW480 SW620

Figure 3. MiR-200c represses Sox2 to suppress CRC growth and metastasis in vitro. A and B, effects of miR-200c and miR-200c/Sox2 on cell proliferation were determined by CCK-8 cell proliferation assay (A) and colony formation assay (B). C, tumorigenicity of miR-200c and miR-200c/Sox2 was evaluated by soft agar assay. Bottom, the quantiﬁcation of the colony-forming rate. D and E, impact of miR-200c and miR-200c/Sox2 on cell migration (D)/invasion (E) across a Transwell chamber. Bottom, the quantiﬁcation of the number of migrated (D)/invaded (E) cells. Error bars, mean SD from 3 independent experiments. , P < 0.001.

effects. Further analyses revealed that Sox2 overexpression decreased the growth rate, colony formation, and migration increased CRC cell proliferation, migration, and invasion of SW620 (25). The rescue of Sox2 expression almost (Supplementary Fig. S3). However, the knockdown of Sox2 completely restored the proliferation, tumorigenesis,

2636 Clin Cancer Res; 20(10) May 15, 2014 Clinical Cancer Research

MiR-200c Regulates CRC Stemness, Growth, Metastasis by Sox2

migration, and invasion in SW480/miR-200c and SW620/ miR-200c/Sox2 groups (P < 0.01). Tumors in mice injected miR-200c cells (Fig. 3). These results indicate that Sox2 is with SW620/miR-200c/Sox2 cells showed no significant one of the significant functional targets of miR-200c in CRC difference from SW620/control cells (P > 0.05; Fig. 4A and cells. B). Immunostaining confirmed that the cell proliferation index Ki-67 and the CSC biomarkers b-catenin, CD133, and MiR-200c inhibits tumor growth and metastasis by CD44 were downregulated by miR-200c and restored by downregulating Sox2 in vivo Sox2 (Supplementary Fig. S4A–S4C). To assess the effect of miR-200c on tumor growth in vivo, To test the effect of miR-200c on the metastasis of CRC SW620/control, SW620/miR-200c, and SW620/miR-200c/ in vivo, tiny tumor masses of subcutaneous tumors were Sox2 were injected subcutaneously on the hind limb of transplanted into the mouse cecal subserosa. Sixty percent nude mice. Tumor growth in the SW620/miR-200c group (3 of 5) of mice in the SW620/control group or the was slower than that in the SW620/control and the SW620/ SW620/miR-200c/Sox2 group had hepatic metastatic

AB 800 Control ) 3 miR-200c Control 600 miR-200c/Sox2

400 miR-200c 200

miR-200c (mm volume Tumor 0 /Sox2 7 9 11 14 17 20 (d)

C Control miR-200c miR-200c/Sox2

Figure 4. MiR-200c suppresses tumor growth and metastasis of CRC in vivo by targeting Sox2. A, effects of miR-200c and miR-200c/ Sox2 on subcutaneous tumor generation (n ¼ 6). B, tumor sizes were measured on the indicated days (d). Data, mean tumor volume SD. C, representative images of the whole body, hepatic metastases, and hematoxylin and eosin (H&E) stains of metastatic livers of mice in the orthotropic metastasis assay (n ¼ 5 per group). Scale bar, 100 mm. , P < 0.01; , P < 0.001.

www.aacrjournals.org Clin Cancer Res; 20(10) May 15, 2014 2637

Lu et al.

lesions, and the hepatic metastatic lesions in the 2 groups MiR-200c depletion increases the stemness of CRC cells showed no signiﬁcant differences. However, no mice in It has been reported that Sox2 is vital for the mainte- the SW620/miR-200c group had hepatic metastatic nance of the self-renewal of embryonic stem cells and lesions (Fig. 4C). The number of hepatic metastatic neural stem cells (NSC) as well as the tumorigenicity of lesions in mice of the SW620/miR-200c group was obvi- glioblastoma tumor-initiating cells (17, 26). In other ously reduced compared with the SW620/control group words, Sox2 plays an integral role in maintaining the (P < 0.01), whereas enhanced expression of Sox2 stemness of adult cells as well as CSCs. Thus, we hypoth- increased the number of hepatic metastases of mice and esized that Sox2 regulator miR-200c may have a connec- blocked the inhibitory effects of miR-200c (Supplemen- tion with stemness maintenance in CRC cells. Loss of tary Fig. S4D). No metastatic nodules were discovered in miR-200c results in the acquisition of stemness, which is the other organs in any group. These results collectively characterized by an increased stem cell sphere-forming indicate that miR-200c upregulation profoundly sup- capacity (Supplementary Fig. S5A) and an increased presses tumor growth and metastasis in vivo and is partly expression of CRC stem cell markers such as CD166, reversed by Sox2. CD133, and b-catenin(Fig.5A).Mostessentially,lossof

Blank Blank A HCT116 SW480 Scramble Scramble 6 10 zip-200c zip-200c 8 4 Blank Scramblezip-200c Blank Scramblezip-200c 6

CD133 4 2 b-Catenin 2 Relative expression Relative expression Tubulin 0 0 b-Catenin CD133 CD166 b-Catenin CD133 CD166 HCT116 SW480 BC HEK293 SW480 Sox2 TSS miR-200c 1.5 0.5

0.4 1.0 A B +1 0.3 (–1648 ~ –1643) (–1326 ~ –1321) 0.2 0.5 WT-Luc 0.1

Mut-Luc A 0.0 0.0 Relative luciferase activity Relative luciferase Mut-Luc B activity Relative luciferase Mut-Luc AB PGL3-Basic PGL3-Basic PGL3-WT-Luc PGL3-WT-Luc PGL3-Mut-LucAPGL3-Mut-LucB PGL3-Mut-LucAPGL3-Mut-LucB PGL3-Mut-LucAB PGL3-Mut-LucAB

D SW620 HCT116 SW480

Marker Input IgG Anti-Sox2 Input IgG Anti-Sox2 Marker Input IgG Anti-Sox2 200 bp 200 bp A A 100 bp 100 bp 200 bp 200 bp B B 100 bp 100 bp

Figure 5. MiR-200c and Sox2 reciprocally control their expression through a feedback loop. A, Western blotting and qRT-PCR analysis of stem cell markers. B, schematic representation of the promoter regions of miR-200c with the putative Sox2 TFBSs (A and B) and the structure of the wild-type (WT-Luc) and TFBS- mutant (Mut-LucA, Mut-LucB, and Mut-LucAB) luciferase reporters driven by the promoter. C, relative luciferase activity of the indicated promoter vectors in HEK293 and SW480 cells transfected with Sox2 plasmids. D, ChIP assay using antibodies against Sox2 in CRC cells; PCR gel showing ampliﬁcation of Sox2- binding sites A and B. , P < 0.01; , P < 0.001.

2638 Clin Cancer Res; 20(10) May 15, 2014 Clinical Cancer Research

MiR-200c Regulates CRC Stemness, Growth, Metastasis by Sox2

miR-200c causes SW480 and HCT116 cells to reverse icantly bound to TFBS A within the miR-200c promoter differentiate to CSC-like cells, represented by the cells (Fig. 5D), indicating a direct association of Sox2 with the transitioning from spindle-shaped cells to round cells miR-200c promoter. These findings establish a negative (Supplementary Fig. S5B). feedback loop controlling Sox2 and miR-200c expression that regulates CRC stemness. This negative feedback loop MiR-200c and Sox2 reciprocally control their mechanism was verified once again by the increase in expression through a feedback loop miR-200c in the Sox2-overexpressing cells, causing sup- Overexpressing Sox2 in CRC cells accompanied a pression of Sox2 (Supplementary Fig. S5D). decrease of miR-200c (Supplementary Figs. S1A and S1C and S5C). To further investigate the potential rela- MiR-200c is a PI3K–AKT signaling pathway regulator in tionship between miR-200c and Sox2, we relied on bio- CRC informatics. While analyzing a 2-kb region upstream of Sox2 cooperates or interacts with some genes to govern the transcription start site (TSS) of miR-200c using UCSC, embryonic stem cell differentiation (16, 27). KIT is a Sox2 TESS, and TFSEARCH, we noted that there were 2 Sox2 enhancer–interacting gene assayed by 4C-seq (circular chro- transcription factor–binding sites (TFBS) located within mosome conformation capture) in human embryonic stem the miR-200c promoter. For convenience, the 2 TFBSs cells (28). Previous research indicated that KIT promotes were named A and B (Fig. 5B). A reduction of the wild- proliferation and invasion by activating the PI3K–AKT type miR-200c promoter luciferase activity was observed pathway in some human carcinomas (29, 30). For this upon upregulation of Sox2 in the HEK293 and SW480 reason, we wondered whether miR-200c participates in the cell lines (P < 0.001), and a similar effect was observed growth and metastasis of CRCs through the PI3K–Akt when B was mutated alone (P < 0.001). However, Sox2 pathway. The upregulation of miR-200c restrained the phos- overexpression did not result in further reduction of the phorylation of PI3K and AKT (p-PI3K and p-AKT) com- luciferase activity when A was mutated alone or A and B pared with control and blank cells; the restrained of p-PI3K were mutated together (Fig. 5C). These data indicate that and p-AKT by miR-200c was restored by Sox2 (Fig. 6A). On Sox2 binds to specific promoter TFBS of miR-200c and the other hand, the downregulation of miR-200c caused inhibits transcription. The chromatin immunoprecipi- the opposite effect. Treatment with the PI3K–Akt inhibitor taion (ChIP) analyses revealed that Sox2 is most signif- LY294002 could abrogate the activities of p-PI3K and p-AKT

A SW480 SW620 B SW480 HCT116

miR-200c miR-200c zip-200c/ zip-200c/ Blank Control miR-200c /Sox2 Blank Control miR-200c /Sox2 Blank Scramble zip-200c LY294002 Blank Scramble zip-200c LY294002 p-PI3K p-PI3K T-PI3K T-PI3K p-AKT p-AKT T-AKT T-AKT GAPDH GAPDH

C miR-200c CD133

Sox2 CD166 Stemness

b-Catenin PI3K P P Activational phosphorylation Activation Inhibition AKT P

Proliferation and metastasis

Figure 6. MiR-200c is a PI3K–AKT signaling pathway regulator in CRC. A, Western blot analyses of p-PI3K, T-PI3K, p-AKT, and T-AKT in the cells as indicated. B, Western blot analyses of indicated proteins in SW480 and HCT116 knockdown of miR-200c, followed by the treatment of LY294002. GAPDH served as the loading control. C, proposed model of how miR-200c regulates CRC proliferation and metastasis through regulation of Sox2.

www.aacrjournals.org Clin Cancer Res; 20(10) May 15, 2014 2639

Lu et al.

by inhibiting miR-200c (Fig. 6B). No change in the total vitro and in vivo. Knockdown of miR-200c strongly increased protein amounts of PI3K and AKT (T-PI3K and T-AKT) was the CRC stemness phenotype. We propose that Sox2 may be observed under any circumstance. In brief, the PI3K–AKT necessary to maintain a stemness phenotype or that miR- pathway has been considered relevant to the maintenance 200c influences stemness in a direct modulation manner. of CRC stemness, growth, and metastasis. Cancer stem cell theory posits that a fraction of cancer cells maintain typical stemness properties, including the capac- ities of self-renewal, differentiation, and tumorigenicity, Discussion and are thought to be crucial for the maintenance of In this study, we evaluated the expression of miR-200c in uncontrolled tumor proliferation as well as metastasis. We CRC cell lines and clinical tissues. The findings indicate that propose that suppressing miR-200c, leading to an increased a lower level of miR-200c is significantly associated with the expression of Sox2, maintains the stemness of CRC cells. On growth and metastasis of CRC as well as a more advanced the basis of these observations, we speculate that miR-200c stage. MiR-200c abnormalities have been found in various restrains cell proliferation and metastasis by eliminating a tumors. Expression of miR-200c was significantly lower in stem cell phenotype. Our findings are also supported by esophageal adenocarcinoma in comparison to Barrett data showing that ZEB1 promotes tumor cell dissemination esophageal epithelium (31). MiRNA microarray analyses and tumorigenicity by suppressing stemness-inhibiting of noninvasive and invasive bladder urothelial carcinoma miRNAs, including miR-200c, miR-203, and miR-183 (39). indicated that miR-200c was reduced in invasive lesions Increasing evidence indicates that miRNAs play crucial (32). These data may indicate that downregulation of miR- roles in carcinoma cell proliferation and metastasis. Previ- 200c has potential as a cancer progression and metastasis ous studies showed that miR-200c–inhibited CRC cell biomarker. Overexpression of miR-200c and miR-141 in PC- migration and invasion correlated with loss of the EMT 3 cells impaired proliferation and survival (33). AC3Eþ phenotype (37). In the present study, we found that the cells expressing miR-200 had obviously less tumorigenic miR-200c-Sox2 negative feedback loop mechanism was potential in vivo (34). Increasing miR-200c in the highly involved in the regulation of CRC cell proliferation and metastatic MDA-MB-231 cells displayed remarkably metastasis. Sox2, occupying the promoters of miR-200c,in reduced motility (35). These previous observations turn decreased expression of miR-200c; such a loop might depicted that miR-200c is tumor repressor. Our data are well lead to the reactivation of Sox2. In a previous report, a consistent with previous evidence. However, Liu and col- miR-200c-Sox2 feedback loop contributed to cell-cycle exit leagues reported that miR-200c was upregulated in non– and the neuronal differentiation of neural stem/progenitor small cell lung cancer (NSCLC) tissues compared with cells (24). During induced pluripotent stem cell (iPSC) adjacent normal lung tissues (36). This inconsistent expres- reprogramming in mouse embryonic fibroblasts, Oct4 spe- sion phenotype of miR-200c in tumors may be due to cifically activates the mir-141/200c cluster and Sox2 specif- carcinoma heterogeneity. In the context of tumor develop- ically activates the mir-200a/b/429 cluster by binding to ment and progression, tumor cells display phenotypic and their promoters (41). Altogether, miR-200c might be regu- functional heterogeneity that might result in differential lated by a distinct set of transcriptional activations/repres- expression of some genes in different types of tumors or in sions of Sox2. We noted a reciprocal regulation of miR-200c different areas in the same tumor. and Sox2 in adult tumor cells paradigms that might be Functionally, inhibition of miR-200c reinforced prolifer- different from the pluripotent/multipotent stem/progeni- ation, migration, and invasion in CRC cells. Correspond- tor cell situation. ingly, overexpression of miR-200c attenuated proliferation, In the EMT process, miR-200s enhance E-cadherin expres- migration, and invasion. Furthermore, overexpression of sion by directly targeting ZEB1 and ZEB2, which encode miR-200c displayed significant suppression of growth and transcriptional repressors of E-cadherin (11, 12, 42, 43). metastasis in vivo. Our data provide evidence to support the Apart from EMT, the inflammatory signaling harnessed by role of miR-200c in suppressing CRC growth and metastasis. miR-200c is helpful in epithelial cell transformation and However, our findings have yielded a contradiction to a mammary cell tumorigenesis (44). The PI3K–Akt signaling recent study on the role of miR-200c in CRC cell growth; the pathway controls fundamental cellular processes such as miR-200c expression levels in the SW480 and HCT1116 cell cell survival, growth, proliferation, cell repair, cell migra- lines in our study did not coincide with the data reported tion, and angiogenesis and is constitutively activated in previously (37). These contrary findings may be due, in part, several cancer types, including CRCs. The Sox2 enhancer- to the different source organization and empirical method interacting gene KIT could activate multiple downstream models. With regard to the role of miR-200c in tumor signaling cascades, including the PI3K–AKT pathway (45, growth, several recent studies have reported its conflicting 46). Consequently, we speculated that miR-200c would roles (23, 38–40). negatively suppress Sox2 to inhibit the activation of the Sox2 was known to be a stem cell factor and was identified PI3K–AKT pathway. Here, we reported that overexpression as a target of miR-200c. In our study, Sox2 enhanced CRC of miR-200c results in low levels of p-PI3K and p-AKT and cell proliferation, migration, and invasion. MiR-200c that this phenomenon could be rescued by Sox2. Luo and reduced the expression of Sox2, which caused the restora- colleagues have reported that the miR-20a/miR-200c-PTEN- tion of the growth and metastasis inhibited by miR-200c in AKT axis regulates EMT-associated CSC enrichment of

2640 Clin Cancer Res; 20(10) May 15, 2014 Clinical Cancer Research

MiR-200c Regulates CRC Stemness, Growth, Metastasis by Sox2

epithelial ovarian cancer (EOC) cells (47). In this axis, miR- therapeutic target for the inhibition of CRC growth and 200c expression was significantly higher in EOC spheroid metastasis. cells, which produce large numbers of CSC cells, and activated the PI3K–AKT pathway. In contrast, we presented Disclosure of Potential Conflicts of Interest evidence indicating that the downregulation of miR-200c No potential conflicts of interest were disclosed. confers the characteristics of CSC cells and stimulates the PI3K–AKT pathway in CRCs. The PI3K–AKT pathway can be Authors' Contributions triggered through various mechanisms, including increas- Conception and design: Y.-X. Lu, X.-N. Li ing the expression of its positive regulators such as EGF Development of methodology: M. Zhou, Y. Liu, C. Zhang, M. Hong Acquisition of data (provided animals, acquired and managed patients, receptor (EGFR) and RAS and decreasing the expression of provided facilities, etc.): Y.-X. Lu, L. Yuan, M. Zhou, Y. Liu, M. Hong its negative regulators such as PTEN. Our data indicate that Analysis and interpretation of data (e.g., statistical analysis, biosta- the induction of the PI3K–AKT pathway was dependent on tistics, computational analysis): Y.-X. Lu, J.-P. Li, X.-N. Li Writing, review, and/or revision of the manuscript: Y.-X. Lu, M. Zhou, the increased expression of Sox2. PI3K activity was inhibited J.-P. Li, L. Zheng, X.-N. Li by LY294002 in the enterocyte-like differentiation pro- Administrative, technical, or material support (i.e., reporting or orga- gram of colon CSC cells, indicating that PI3K signaling is, nizing data, constructing databases): Y.-X. Lu, X.-L. Xue, M. Zhou, C. Zhang, X.-N. Li in part, necessary for the maintenance of colon CSC cells Study supervision: X.-N. Li (48). In this regard, it is reasonable to assume that PI3K– AKT pathway activation is relevant to the maintenance of miR-200c Grant Support CRC stemness, growth, and metastasis when is This work is supported by the National Natural Science Foundation of suppressed. China (Nos 81272758 and 81302158) and the Science and Technology In summary, in addition to functioning as a tumor Project of Guangdong Province (2009B030803041). The costs of publication of this article were defrayed in part by the suppressor by blocking the PI3K-Akt signaling pathway payment of page charges. This article must therefore be hereby marked through an miR-200c-Sox2 negative feedback loop mecha- advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate nism, miR-200c also acts as a stemness inhibitor (Fig. 6C). this fact. miRNA Although -based therapeutics are still in their infan- Received August 27, 2013; revised February 26, 2014; accepted March 12, cy, our findings indicate that miR-200c may be a promising 2014; published OnlineFirst March 21, 2014.

References 1. Visvader JE. Cells of origin in cancer. Nature 2011;469:314–22. E-cadherin transcriptional repressors ZEB1 and ZEB2. J Biol Chem 2. Chaffer CL, Weinberg RA. A perspective on cancer cell metastasis. 2008;283:14910–4. Science 2011;331:1559–64. 13. Korpal M, Kang Y. The emerging role of miR-200 family of microRNAs 3. Chaffer CL, Marjanovic ND, Lee T, Bell G, Kleer CG, Reinhardt F, et al. in epithelial-mesenchymal transition and cancer metastasis. RNA Biol Poised chromatin at the ZEB1 promoter enables breast cancer cell 2008;5:115–9. plasticity and enhances tumorigenicity. Cell 2013;154:61–74. 14. Roybal JD, Zang Y, Ahn YH, Yang Y, Gibbons DL, Baird BN, et al. miR- 4. Masui S, Nakatake Y, Toyooka Y, Shimosato D, Yagi R, Takahashi 200 inhibits lung adenocarcinoma cell invasion and metastasis by K, et al. Pluripotency governed by Sox2 via regulation of Oct3/4 targeting Flt1/VEGFR1. Mol Cancer Res. 2011;9:25–35. expression in mouse embryonic stem cells. Nat Cell Biol 2007;9: 15. Gibbons DL, Lin W, Creighton CJ, Rizvi ZH, Gregory PA, Goodall GJ, 625–35. et al. Contextual extracellular cues promote tumor cell EMT and 5. Lengerke C, Fehm T, Kurth R, Neubauer H, Scheble V, Muller F, et al. metastasis by regulating miR-200 family expression. Genes Dev Expression of the embryonic stem cell marker SOX2 in early-stage 2009;23:2140–51. breast carcinoma. BMC Cancer 2011;11:42. 16. Boyer LA, Lee TI, Cole MF, Johnstone SE, Levine SS, Zucker JP, et al. 6. Jia X, Li X, Xu Y, Zhang S, Mou W, Liu Y, et al. SOX2 promotes Core transcriptional regulatory circuitry in human embryonic stem tumorigenesis and increases the anti-apoptotic property of human cells. Cell. 2005;122:947–56. prostate cancer cell. J Mol Cell Biol 2011;3:230–8. 17. Gangemi RM, Griffero F, Marubbi D, Perera M, Capra MC, Malatesta P, 7. Liu C, Kelnar K, Liu B, Chen X, Calhoun-Davis T, Li H, et al. The et al. SOX2 silencing in glioblastoma tumor-initiating cells causes stop microRNA miR-34a inhibits prostate cancer stem cells and metastasis of proliferation and loss of tumorigenicity. Stem Cells 2009;27:40–8. by directly repressing CD44. Nat Med 2011;17:211–5. 18. Leis O, Eguiara A, Lopez-Arribillaga E, Alberdi MJ, Hernandez-Garcia 8. Vaira V, Faversani A, Martin NM, Garlick DS, Ferrero S, Nosotti M, et al. S, Elorriaga K, et al. Sox2 expression in breast tumours and activation Regulation of lung cancer metastasis by Klf4-Numb-like signaling. in breast cancer stem cells. Oncogene 2012;31:1354–65. Cancer Res 2013;73:2695–705. 19. Tian T, Zhang Y, Wang S, Zhou J, Xu S. Sox2 enhances the tumor- 9. Yu XF, Zou J, Bao ZJ, Dong J. miR-93 suppresses proliferation and igenicity and chemoresistance of cancer stem-like cells derived from colony formation of human colon cancer stem cells. World J Gastro- gastric cancer. J Biomed Res 2012;26:336–45. enterol 2011;17:4711–7. 20. Liu K, Jiang M, Lu Y, Chen H, Sun J, Wu S, et al. Sox2 cooperates with 10. Xu XT, Xu Q, Tong JL, Zhu MM, Nie F, Chen X, et al. MicroRNA inflammation-mediated Stat3 activation in the malignant transforma- expression profiling identifies miR-328 regulates cancer stem cell-like tion of foregut basal progenitor cells. Cell Stem Cell 2013;12:304–15. SP cells in colorectal cancer. Br J Cancer 2012;106:1320–30. 21. Zhou C, Liu G, Wang L, Lu Y, Yuan L, Zheng L, et al. MiR-339–5p 11. Kong D, Li Y, Wang Z, Banerjee S, Ahmad A, Kim HR, et al. miR-200 regulates the growth, colony formation and metastasis of colorectal regulates PDGF-D-mediated epithelial-mesenchymal transition, cancer cells by targetizng PRL-1. PLoS One 2013;8:e63142. adhesion, and invasion of prostate cancer cells. Stem Cells 2009;27: 22. Cho SJ, Kim JS, Kim JM, Lee JY, Jung HC, Song IS. Simvastatin 1712–21. induces apoptosis in human colon cancer cells and in tumor xeno- 12. Korpal M, Lee ES, Hu G, Kang Y. The miR-200 family inhibits epithelial- grafts, and attenuates colitis-associated colon cancer in mice. Int mesenchymal transition and cancer cell migration by direct targeting of J Cancer 2008;123:951–7.

www.aacrjournals.org Clin Cancer Res; 20(10) May 15, 2014 2641

Lu et al.

23. Shimono Y, Zabala M, Cho RW, Lobo N, Dalerba P, Qian D, et al. poor prognosis in patients with lung cancer. Med Oncol 2012; Downregulation of miRNA-200c links breast cancer stem cells with 29:618–26. normal stem cells. Cell 2009;138:592–603. 37. Hur K, Toiyama Y, Takahashi M, Balaguer F, Nagasaka T, Koike J, et al. 24. Peng C, Li N, Ng YK, Zhang J, Meier F, Theis FJ, et al. A unilateral MicroRNA-200c modulates epithelial-to-mesenchymal transition negative feedback loop between miR-200 microRNAs and Sox2/E2F3 (EMT) in human colorectal cancer metastasis. Gut 2013;62:1315–26. controls neural progenitor cell-cycle exit and differentiation. J Neurosci 38. Yu J, Ohuchida K, Mizumoto K, Sato N, Kayashima T, Fujita H, et al. 2012;32:13292–308. MicroRNA, hsa-miR-200c, is an independent prognostic factor in 25. Fang X, Yu W, Li L, Shao J, Zhao N, Chen Q, et al. ChIP-seq and pancreatic cancer and its upregulation inhibits pancreatic cancer functional analysis of the SOX2 gene in colorectal cancers. OMICS invasion but increases cell proliferation. Mol Cancer 2010;9:169. 2010;14:369–84. 39. Wellner U, Schubert J, Burk UC, Schmalhofer O, Zhu F, Sonntag A, 26. Fong H, Hohenstein KA, Donovan PJ. Regulation of self-renewal and et al. The EMT-activator ZEB1 promotes tumorigenicity by repressing pluripotency by Sox2 in human embryonic stem cells. Stem Cells stemness-inhibiting microRNAs. Nat Cell Biol 2009;11:1487–95. 2008;26:1931–8. 40. Peng F, Jiang J, Yu Y, Tian R, Guo X, Li X, et al. Direct targeting of 27. Okumura-Nakanishi S, Saito M, Niwa H, Ishikawa F. Oct-3/4 and Sox2 SUZ12/ROCK2 by miR-200b/c inhibits cholangiocarcinoma tumour- regulate Oct-3/4 gene in embryonic stem cells. J Biol Chem igenesis and metastasis. Br J Cancer 2013;109:3092–104. 2005;280:5307–17. 41. Wang G GX, Hong W, Liu Q, Wei T, Lu C, Gao L, et al. A Unilateral 28. Gao F, Wei Z, An W, Wang K, Lu W. The interactomes of POU5F1 and negative feedback loop between miR-200 microRNAs and Sox2/E2F3 SOX2 enhancers in human embryonic stem cells. Sci Rep 2013;3: controls neural progenitor cell-cycle exit and differentiation. Proc Natl 1588. Acad Sci U S A 2013;110:2858–63. 29. Toki T, Kanezaki R, Adachi S, Fujino H, Xu G, Sato T, et al. The key role 42. Gregory PA, Bert AG, Paterson EL, Barry SC, Tsykin A, Farshid G, et al. of stem cell factor/KIT signaling in the proliferation of blast cells from The miR-200 family and miR-205 regulate epithelial to mesenchymal Down syndrome-related leukemia. Leukemia 2009;23:95–103. transition by targeting ZEB1 and SIP1. Nat Cell Biol 2008;10:593–601. 30. Yasuda A, Sawai H, Takahashi H, Ochi N, Matsuo Y, Funahashi H, et al. 43. Burk U, Schubert J, Wellner U, Schmalhofer O, Vincan E, Spaderna S, Stem cell factor/c-kit receptor signaling enhances the proliferation and et al. A reciprocal repression between ZEB1 and members of the miR- invasion of colorectal cancer cells through the PI3K/Akt pathway. Dig 200 family promotes EMT and invasion in cancer cells. EMBO Rep Dis Sci 2007;52:2292–300. 2008;9:582–9. 31. Smith CM, Watson DI, Leong MP, Mayne GC, Michael MZ, Wijnhoven 44. Rokavec M, Wu W, Luo JL. IL6-mediated suppression of miR-200c BP, et al. miR-200 family expression is downregulated upon neoplastic directs constitutive activation of inflammatory signaling circuit driving progression of Barrett's esophagus. World J Gastroenterol 2011;17: transformation and tumorigenesis. Mol Cell 2012;45:777–89. 1036–44. 45. Kosmider O, Denis N, Lacout C, Vainchenker W, Dubreuil P, Moreau- 32. Wszolek MF, Rieger-Christ KM, Kenney PA, Gould JJ, Silva Neto B, Gachelin F. Kit-activating mutations cooperate with Spi-1/PU.1 over- Lavoie AK, et al. A MicroRNA expression profile defining the invasive expression to promote tumorigenic progression during erythroleuke- bladder tumor phenotype. Urol Oncol 2011;29:794–801 e1. mia in mice. Cancer Cell 2005;8:467–78. 33. Vallejo DM, Caparros E, Dominguez M. Targeting Notch signalling by 46. Wandzioch E, Edling CE, Palmer RH, Carlsson L, Hallberg B. Activation the conserved miR-8/200 microRNA family in development and cancer of the MAP kinase pathway by c-Kit is PI-3 kinase dependent in cells. EMBO J 2011;30:756–69. hematopoietic progenitor/stem cell lines. Blood 2004;104:51–7. 34. Liu YN, Yin JJ, Abou-Kheir W, Hynes PG, Casey OM, Fang L, et al. MiR- 47. Luo X, Dong Z, Chen Y, Yang L, Lai D. Enrichment of ovarian cancer 1 and miR-200 inhibit EMT via Slug-dependent and tumorigenesis via stem-like cells is associated with epithelial to mesenchymal transition Slug-independent mechanisms. Oncogene 2013;32:296–306. through an miRNA-activated AKT pathway. Cell Prolif 2013;46: 35. Park SM, Gaur AB, Lengyel E, Peter ME. The miR-200 family deter- 436–46. mines the epithelial phenotype of cancer cells by targeting the E- 48. Vermeulen L, Todaro M, de Sousa Mello F, Sprick MR, Kemper K, cadherin repressors ZEB1 and ZEB2. Genes Dev 2008;22:894–907. Perez Alea M, et al. Single-cell cloning of colon cancer stem cells 36. Liu XG, Zhu WY, Huang YY, Ma LN, Zhou SQ, Wang YK, et al. High reveals a multi-lineage differentiation capacity. Proc Natl Acad Sci U S expression of serum miR-21 and tumor miR-200c associated with A 2008;105:13427–32.

2642 Clin Cancer Res; 20(10) May 15, 2014 Clinical Cancer Research

Regulation of Colorectal Carcinoma Stemness, Growth, and Metastasis by an miR-200c-Sox2−Negative Feedback Loop Mechanism

Yan-Xia Lu, Li Yuan, Xiao-Lei Xue, et al.

Clin Cancer Res 2014;20:2631-2642. Published OnlineFirst March 21, 2014.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-13-2348

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2014/03/21/1078-0432.CCR-13-2348.DC1

Cited articles This article cites 48 articles, 14 of which you can access for free at: http://clincancerres.aacrjournals.org/content/20/10/2631.full#ref-list-1

Citing articles This article has been cited by 4 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/20/10/2631.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/20/10/2631. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.