Human Cancer Biology

Locked Nucleic Acid In situ Hybridization Analysis of miR-21 Expression during Development Nobutake Yamamichi,1,2 Ryoichi Shimomura,3 Ken-ichi Inada,3 Kouhei Sakurai,1 Takeshi Haraguchi,1 Yuka Ozaki,1 Shuji Fujita,1 Taketoshi Mizutani,1 Chihiro Furukawa,1 Mitsuhiro Fujishiro,2 Masao Ichinose,4 Kazuya Shiogama,3 Yutaka Tsutsumi,3 Masao Omata,2 and Hideo Iba1

Abstract Purpose: To better understand microRNA miR-21 function in carcinogenesis, we ana- lyzed miR-21 expression patterns in different stages of colorectal cancer development using in situ hybridization (ISH). Experimental Design: Locked nucleic acid (LNA)/DNA probes and a biotin-free tyramide signal amplification system were used in ISH analyses of miRNA expression. Condi- tions for specific detection of miR-21 were determined using human cell lines and miR-21–expressing lentiviral vectors. Expression was determined in 39 surgically excised colorectal tumors and 34 endoscopically resected colorectal polyps. Results: In the surgical samples, miR-21 expression was much higher in colorectal cancers than in normal mucosa. Strong miR-21 expression was also observed in can- cer-associated stromal fibroblasts, suggesting miR-21 induction by cancer-secreted cytokines. Protein expression of PDCD4, a miR-21 target, was inversely correlated with miR-21 expression, confirming that miR-21 is indeed a negative regulator of PDCD4 in vivo. In the endoscopic samples, miR-21 expression was very high in malignant but was not elevated in nontumorigenic polyps. Precancerous also frequently showed miR-21 up-regulation. Conclusion: Using the LNA-ISH system for miRNA detection, miR-21 was detectable in precancerous adenomas. The frequency and extent of miR-21 expression increased dur- ing the transition from precancerous colorectal to advanced carcinoma. Expres- sion patterns of miR-21 RNA and its target, tumor suppressor protein PDCD4, were mutually exclusive. This pattern may have clinical application as a biomarker for colorec- tal cancer development and might be emphasized by self-reinforcing regulatory systems integrated with the miR-21 gene, which has been previously shown in cell culture.

Novel mechanisms of human gene regulation mediated by mi- Authors' Affiliations: 1Division of Host-Parasite Interaction, Department of croRNAs (miRNA) have recently been established (1). MiRNAs Microbiology and Immunology, Institute of Medical Science, and control gene expression at the posttranscriptional level by tar- 2Department of , Faculty of Medicine, University of geting mRNAs for translational repression or mRNA degrada- 3 Tokyo, Tokyo, Japan; First Department of Pathology, Fujita Health tion (1). Evidence is accumulating that many miRNAs are University School of Medicine, Aichi, Japan; and 4Second Department of Internal Medicine, Wakayama Medical College, Wakayama, Japan differentially regulated in normal development and cancers, Received 12/22/08; revised 3/30/09; accepted 3/31/09; published OnlineFirst and that deregulation of specific miRNAs may contribute to hu- 6/9/09. man cancer development and progression (1). To better under- Grant support: Grant-in-Aid for Scientific Research on Priority Areas from stand the molecular mechanisms that regulate miRNA gene the Ministry of Education, Cultures, Sports, Science, and Technology (MEXT) of Japan, and in part by Strategic cooperation to control emerging expression,our group has recently developed an algorithm to and reemerging infections funded by the Special Coordination Funds for predict the promoters of human miRNA genes that are likely Promoting Science and Technology of MEXT. to be involved in evolutionarily conserved regulatory systems The costs of publication of this article were defrayed in part by the payment of (2). Using this algorithm,we predicted 59 putative miRNA gene page charges. This article must therefore be hereby marked advertisement promoter regions,one of which was the miR-21 miRNA pro- in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Note: Supplementary data for this article are available at Clinical Cancer moter. Because high miR-21 expression levels have been re- Research Online (http://clincancerres.aacrjournals.org/). ported in various ,including breast cancer (3 –5), Requests for reprints: Hideo Iba, Division of Host-Parasite Interaction, (3,6,7), (8),hepatocel- Department of Microbiology and Immunology, Institute of Medical Science, lular carcinoma (9),brain tumor (10),leukemia (11,12),cer- University of Tokyo, 4-6-1, Shirokanedai Minato-ku, Tokyo 108-8639, Japan. Phone: 81-3-5449-5730; Fax: 81-3-5449-5449; E-mail: [email protected]. vical cancer (13),ovarian cancer (14),colorectal cancer (3,15), F 2009 American Association for Cancer Research. prostate cancer (3),lung cancer (3),and gastric cancer (3),we doi:10.1158/1078-0432.CCR-08-3257 concentrated on miR-21 analysis. Our biochemical analyses

www.aacrjournals.org 4009 Clin Cancer Res 2009;15(12) June 15, 2009 Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2009 American Association for Cancer Research. Human Cancer Biology

feedback regulation. Indeed,it was recently reported that in Ras- Translational Relevance induced cellular transformation of a rat thyroid cell line,PDCD4 reduces miR-21 activity,at least partly,by suppressing AP-1 ac- To better understand microRNA (miRNA) function tivity (29). However,at this moment,there is even no clear in carcinogenesis, sensitive and reproducible meth- proof that PDCD4 is a target of miR-21 in vivo because of the ods for in situ hybridization (ISH) are needed. For ex- lack of efficient tools to detect miR-21 expression in a single-cell perimental analysis, we have chosen analysis of resolution. miR-21 expression during the process of colorectal As a first step toward understanding mechanisms regulating cancer development, and strictly determined condi- miR-21 expression in vivo,we examined miR-21 expression pat- tions for specific detection of miR-21 using several terns during the cancer development. Colorectal tumors were human cell lines and miR-21–expressing lentiviral chosen for the experimental analyses because the clinical and vectors. ISH analysis of miRNA expression levels histologic features of colorectal cancers are relatively simple was finally established with locked nucleic acid compared with other malignancies,and most of colorectal can- (LNA)/DNA probes and biotin-free tyramide signal cers are thought to develop from precancerous adenomas (30). amplification system, which is finally applicable to For this purpose,we established a sensitive and stable in situ formalin-fixed paraffin-embedded clinical samples. hybridization (ISH) method using formalin-fixed,paraffin- The established LNA-ISH system for miRNA detec- embedded (FFPE) tissues. Experimental results indicate that tion showed that elevation of miR-21 becomes de- the frequency and extent of miR-21 expression increase during tectable from precancerous adenomas, and the colorectal cancer progression from precancerous adenoma to extent and frequency of miR-21 expression increase advanced carcinoma. during the colorectal canceration from precancerous adenoma to advanced carcinoma. Mutually exclu- sive expression patterns between miR-21 RNA and Materials and Methods its target, tumor suppression protein PDCD4, in ade- nocarcinomas and precancerous adenomas will Human cell lines. HEK293 (originated from embryonic kidney), have a potential clinical application as a biomarker HeLa (cervical carcinoma),MDA-MB435 (breast ductal carcinoma), PA-1 (embryonic carcinoma),NCC-IT (embryonic carcinoma),A427 for colorectal cancer development. (non small cell lung carcinoma),G401 (rhabdoid tumor),SW620 (colorectal ),and AZ521 (gastric cancer) cells were maintained at 37°C in high-glucose DMEM supplemented with 10% FCS (Gibco/Invitrogen). – confirmed that the miR-21 promoter predicted by the algorithm Locked nucleic acid modified oligonucleotide probes. Locked nucleic – was correct (2). Transcription of primary miR-21 transcripts acid (LNA) modified oligonucleotide probes labeled with FITC at their 3′-ends were obtained from Molecular Biology of ThermoElectron (pri-mir-21) was initiated 30 bp downstream of the promoter GmbH. The sequences of miR-21 probe and the scramble control probe TATA box. Other evolutionally conserved regulatory elements for a negative control were 5′-TLcAALcATLcAGLtCTLgATLaAGLcTA-3′ present in the miR-21 promoter included the binding sites of and 5′-CLaTTLaATLgTCLgGALcAALcTCLaAT-3′,respectively. La,Lt,Lc, activator protein (AP-1),Ets/PU.1,SP1,CCAAT/enhancer bind- and Lg were LNA monomers corresponding to the bases A,T,C,and G, ing protein-α,nuclear factor I (NFI),serum response factor, respectively. p53,and signal transducer and activator of transcription 3. ISH with LNA-modified oligonucleotide probes. Five-micrometer-thin We further showed that phorbol 12-myristate 13-acetate activat- sections of FFPE tissues adhered to glass slides were deparaffinized in ed miR-21 transcription through the several AP-1 and Ets/PU.1 three consecutive xylene baths for 1 min each,followed by 1 min each in serial dilutions of ethanol (100%,100%,95%,95%) and three binding sites in the miR-21 promoter,and also showed that changes of diethyl pyrocarbonate–treated water. Slides were then im- miR-21 was induced during macrophage differentiation of mersed in 0.3% H2O2 for 30 min at room temperature,washed thrice HL-60 cells (16). Therefore,we hypothesized that increased with diethyl pyrocarbonate–treated water,digested with 400 μg/mL miR-21 expression,which has been validated in many cancers, proteinase K (Roche) at 37°C for 15 min,washed thrice with diethyl may reflect elevated tumor cell AP-1 activity. We also identified pyrocarbonate–treated water,submerged in 95% ethanol for 1 min, the negative transcriptional regulator nuclear factor I-B (NFIB) and air-dried completely. Slides were then hybridized in incubation μ – as a target for miR-21 regulation (16). Because the miR-21 pro- chambers overnight at 37°C in an oven,using 0.2 mol/L LNA modi- moter contains an NFIB binding site and NFIB efficiently sup- fied probes diluted with mRNA ISH solution (DAKO). After hybridiza- tion,slides were rinsed thrice in 0.5× SSC,washed for 30 min at 50°C in presses AP-1–dependent miR-21 transactivation,we concluded 0.5× SSC/0.1% Brij35 (Sigma),and rinsed twice in TBS. An anti-FITC that a double-negative feedback loop consisting of miR-21, horseradish peroxidase–conjugated antibody (DAKO,P5100) at 1:100 NFIB,and the miR-21 promoter could self-reinforce miR-21 dilution in TBS/1% bovine serum albumin was applied to the slides for expression (16). 60 min at room temperature,followed by three washes in TBS/0.1% Several other miR-21 targets have also been suggested such as Tween 20 (TBS-T). For amplification of antibody signals,FITC-conjugat- RECK (17), Sprouty2 (18), PTEN (8,9),TPM1 (19),and PDCD4 ed phenol (fluorescyl-tyramide,DAKO,K1497) was applied to the slides (20–23). Tumor suppressor PDCD4 is reported to suppress pro- for 30 min at room temperature,followed by three washes in TBS-T. tein synthesis (24,25) through cytoplasmic binding of eIF4A Finally,an anti-FITC antibody conjugated to horseradish peroxidase (DAKO,K1497) was added to the slides for 30 min at room tempera- and is also known to inhibit several transcription factors,in- β ture,followed by three washes in TBS-T. The reaction products were cluding AP-1 (26,27),SP1 (28),and -catenin (26). Because visualized using a 50 mg/dL 3,3′-diaminobenzidine tetrahydrochloride DNA binding sites for some of these factors are present in solution containing 0.003% hydrogen peroxide. miR-21 promoters,PDCD4 might also contribute to the Immunohistochemical staining. Deparaffinization,endogenous down-regulation of miR-21 expression through double-negative peroxidase inactivation,antigen retrieval of FFPE clinical tissues,and

Clin Cancer Res 2009;15(12) June 15, 20094010 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2009 American Association for Cancer Research. Mir-21 Up-Regulation during Colorectal Carcinogenesis immunostaining with anti-PDCD4 (ab51495,Abcam) antibody were tyramide signal amplification system. Using this method, done as described previously (31). The immunostained sections were 3,3′-diaminobenzidine staining of miR-21 was visible in the evaluated independently by two pathologists in conjunction with the cytoplasm (brown; Fig. 1A). H&E-stained sections from the same lesions. To confirm the specificity of miR-21 detection and more pre- Lentivirus vectors. For the polymerase II–driven (SV40 promoter) cisely quantify miR-21 expression levels,we produced two len- vector,a part of pri-miR-21 sequence (from −6 bp to +65 bp of the 5′-end of mature miR-21 sequence) was inserted between the BamHI tiviral vectors carrying a portion of pri-miR-21 (including the and EcoRI sites of pLSP (22) to generate pLSP-pre-mir21_short. For entire pre-miR-21) driven by the SV40 promoter (polymerase the polymerase III–driven (U6 promoter) vector,the mouse U6 pro- II–driven pLSP-pre-mir21_short) or the mouse U6 promoter moter and a part of the pri-miR-21 sequence (from −57 bp to +115 (polymerase III–driven pLSP-mU6-pre-mir21_long; Supple- bp of the 5′-end of mature miR-21 sequence) were inserted into the mentary Fig. S2). HEK293 cells were transduced with pLSP- BamHI and EcoRI sites of pLSP to generate pLSP-mU6-pre-mir21_long. pre-mir21_short,pLSP-mU6-pre-mir21_long,and pLSP-shGFP Sequences of synthetic oligonucleotide pairs for shCre (targeting Cre- (control),respectively,and stable transductants were selected recombinase of P1phage) and shU3-12 (targeting part of MuLV-LTR) for further analyses. Expression of miR-21 in these established were described previously. They were inserted into pLPS as described above to generate pLPS-shCre and pLSP-shU3-12 (32),respectively,and cell lines was analyzed (Supplementary Fig. S3),and LNA-ISH used for the negative controls. Vesicular stomatitis virus-G pseudotyped for miR-21 was done. As shown in Fig. 1B,HEK293 cells trans- lentiviral vectors were produced using the ViraPower Lentiviral Expres- duced with miR-21–expressing vectors exhibited a clear cyto- sion System (Invitrogen) according to the manufacturer's instructions. plasmic staining when compared with control cells,and Virus transduction and protein analysis. HEK293 cells were trans- pLSP-mU6-pre-mir21_long–transduced cells showed denser duced with vesicular stomatitis virus-G pseudotyped vectors at multi- staining than pLSP-pre-mir21_short–transduced cells. From plicity of infection of 1 to 2 and selected with puromycin for 1 wk. these results,we concluded that our LNA-ISH could detect About 1 wk after the end of selection,cells were disrupted for total pro- miR-21 in a semiquantitative manner. tein preparation. Western blotting analysis with anti-PDCD4 and anti- actin antibodies was done as described previously (22). High miR-21 expression levels were observed not only in cancer Colorectal tissue samples. Thirty-four endoscopically resected colo- cells but also in cancer-associated fibroblasts from colorectal FFPE rectal polyps and 39 surgically excised colorectal tumors were selected tissues. We first performed the LNA-ISH for miR-21 using the from a list of patients with colorectal lesions who underwent endoscop- surgically excised advanced colorectal cancer tissues. In the 34 ic or surgical operation in 2006. All tissue samples were banked at the slides examined,precancerous adenomatous lesions were Fujita Health University School of Medicine,Aichi,Japan. For the en- found in five cases: three were separate polyps and two were doscopically resected samples,the patients were between 35 and 83 y adenomatous masses adjacent to malignant adenocarcinoma. old (mean age,58.8 ± 12.0 y) and included 23 males and 11 females. Including these five lesions,we performed the LNA-ISH for For the surgically excised samples,the patients were between 41 and 83 y old (mean age,65.7 ± 9.4 y) and included 23 males and 16 fe- miR-21 on 39 lesions (Supplementary Table S1). The expres- males. Among the 35 cases of surgically removed colorectal cancer, sion of miR-21 in adenocarcinoma was much higher than the the clinical tumor stage distribution was stage I in 1 patient,stage II expression in normal mucosa (Fig. 2; Table 1). In the surgical in 16 patients,stage III in 15 patients,and stage IV in 2 patients,accord- samples,however,increased miR-21 expression was barely de- ing to the International Union Against Cancer classification. This study tectable in precancerous adenomas. Unexpectedly,up-regula- was approved by the institutional ethical review board for human tion of miR-21 was observed not only in malignant cells but investigation at Fujita Health University. also in the stromal fibroblasts adjacent to the tumor (Fig. 2B; Supplementary Fig. S4A; Supplementary Table S1). Overexpres- Results sion of miR-21 was never observed in fibroblasts far from the tumor mass (Fig. 2B). When an equivalent LNA probe but in Detection of miR-21 expression levels by ISH with LNA-modified which the miR-21 oligonucleotide sequence has been scram- oligonucleotide probes. To screen for miR-21–positive and miR- bled was used instead,this control probe showed no significant 21–negative control cultures for ISH,levels of primary tran- staining in tumor regions,normal tissue,and the stromal fibro- script of miR-21 (pri-miR-21) and mature miR-21 were deter- blasts adjacent to the tumor (Supplementary Fig. S5B and D), mined (Supplementary Fig. S1A and B). Expression level showing clear contrast to the staining by the miR-21 probe in patterns of pri-miR-21 and mature miR-21 were very similar, the sequential FFPE colorectal tissue sections (Supplementary suggesting that there were no significant rate-limiting steps in Fig. S5A and C). All these results suggest that nonmalignant the processing of miR-21 production among these cell lines. stromal fibroblasts adjacent to tumors might induce miR-21 ex- Among the nine cell lines tested,we used malignant HeLa pression due to factors secreted from the nearby tumors. and MDA-MB435 cells as positive controls and nonmalignant Up-regulation of miR-21 was frequently observed in precancer- HEK293 cells as a negative control. ous adenomas but never in nontumorigenic polyps. We next an- A LNA-modified probe was designed in which every third alyzed miR-21 expression levels in endoscopically resected DNA nucleotide was substituted with a corresponding LNA colorectal polyps that were <15 mm in diameter. Among the monomer,which was subsequently used for ISH with LNA- 34 samples examined,28 cases were tumorigenic and 6 cases modified oligonucleotide probes (LNA-ISH). With fluorescent were nontumorigenic (Table 2; Supplementary Table S2). The microscopic observation,the FITC-labeled probe clearly stained ISH signal intensities of these endoscopically resected adeno- the cytoplasm of HeLa and MDA-MB435 cells,whereas mas were far stronger than those of the five surgically excised HEK293 cell cytoplasmic staining was much weaker (Fig. 1A). adenomas (Supplementary Table S1; Table 1). Importantly, Because mature miRNAs are present in the cytoplasm (33),the the RNA preservation of endoscopically resected tissues was probe most likely detected the mature miR-21 signals. To apply much better than that of surgical excised tissues. This difference this LNA-ISH to clinical FFPE samples,we also attempted could be due to a longer time before immersion in formalin for to detect miR-21 by combining LNA-ISH with the biotin-free the surgical samples (1-3 hours) compared with that for

www.aacrjournals.org 4011 Clin Cancer Res 2009;15(12) June 15, 2009 Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2009 American Association for Cancer Research. Human Cancer Biology

Fig. 1. ISH with FITC-labeled LNA-modified probe (LNA-ISH) to detect miR-21 expression levels in formalin-fixed human cell lines. A, top, combined images of cytoplasmic miR-21 highlighted with FITC signal (green) and nuclear DNA counterstained with 4′,6-diamidino-2-phenylindole (DAPI; blue). Bottom, FITC in the probes were detected by the biotin-free tyramide signal amplification system. MiR-21 is finally stained with 3,3′-diaminobenzidine (DAB; brown), whereas the nuclei were counterstained with hematoxylin (blue). B, LNA-ISH detecting miR-21 of HEK293 cells, which were transduced with miR-21 expression vectors (pLSP-pre-mir21_short and pLSP-mU6-pre-mir21_long). Combined images of cytoplasmic miR-21 detected with FITC signal (green) and nuclear DNA counterstained with 4′,6-diamidino-2-phenylindole (blue). HEK293 cells transduced with pLSP-shGFP were used as a negative control. endoscopic ones (1-10 minutes) and also to the the much lon- that PDCD4 protein is up-regulated when cells were transduced ger formalin fixation time for surgical tissues (24-96 hours) with lentivirus vectors expressing newly developed decoy RNAs compared with the fixation time for endoscopic ones that specifically inhibit miR-21 function (22). We here tested (<6 hours). whether HEK293 cells,which show marginal miR-21 expression Expression of miR-21 in nontumorigenic lesions was not (Fig. 1),reduce endogenous PDCD4 expression by stable expres- elevated above levels detected in normal colorectal mucosa sion of exogenous miR-21. By comparing PDCD4 protein levels (Fig.3A).Onthecontrary,increasedmiR-21expressionwas in cells transduced with miR-21 expression lentivirus vector frequently observed in both precancerous adenomas and ade- (pLSP-mU6-pre-mir21_long) and those with control vectors nocarcinomas (Fig. 3B and C). As shown in Table 2 and (or untransduced cells) by Western blotting,we have observed Fig. 3,miR-21 expression was obviously higher in malignant clear reduction in the steady-state levels of PDCD4 by exoge- adenocarcinoma than in precancerous adenoma. From these re- nous miR-21 expression (Fig. 2C). sults,we concluded that elevation of miR-21 expression accom- We next performed ISH for miR-21 and immunostaining of panies colorectal tumor development from precancerous PDCD4 using sequential sections obtained from surgically re- adenoma to advanced carcinoma. We also expect that early sected colorectal cancer tissues. As shown in Fig. 4A,expression detection of miR-21 up-regulation may have potential clinical of PDCD4 was high in normal tissues (right) but was nondetect- application as a new biomarker for colorectal tumorigenesis. able in colorectal cancer cells. Areas with abundant miR-21 ex- MiR-21 and PDCD4 expression show mutually exclusive pression wrapped over the areas with low PDCD4 expression. patterns in the areas around colorectal cancer tissues. It has been Even when endoscopically resected samples were used,mutually reported that PDCD4 is a target of miR-21 mostly by transiently exclusive expression patterns were frequently observed between introducing miR-21 RNA or antisense oligonucleotides for miR-21 and PDCD4 in both malignant adenocarcinoma miR-21 exogenously (20–23). Recently,we have also shown (Figs. 4B; Supplementary Fig. S6A) and precancerous adenoma

Clin Cancer Res 2009;15(12) June 15, 20094012 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2009 American Association for Cancer Research. Mir-21 Up-Regulation during Colorectal Carcinogenesis

Fig. 2. A, LNA-ISH detection of miR-21 in the colorectal adenocarcinoma (left) and normal colorectal mucosa (right) on the same slide. Bar, 200 μm. B, LNA-ISH detection of miR-21 in colorectal tissue from another patient; colorectal adenocarcinoma (left) and normal colorectal mucosa (right) on the same slide. *, fibroblasts around the cancer; **, fibroblasts around the normal gland. FFPE colorectal tissues were obtained from surgical operation. Bar, 200 μm. C, expression of PDCD4 and actin proteins in HEK293 cells transduced with miR-21 expression lentiviral vectors (pLSP-mU6-pre-mir21_long) or two control viral vectors (pLSP-shCre and pLSP-shU3-12). Total protein samples were isolated 2 wk after the transduction and analyzed by Western blotting. Untreated HEK293 cells (nontransduction) were also used as a control.

(Figs. 4C; Supplementary Fig. S6B). These results support the hy- of endoscopic mucosal resection (Fig. 3; Table 2). Even in this pothesis that PDCD4 is an in vivo target of miR-21 and further early stage,we observed that PDCD4,a target of miR-21, suggest that an early increase in miR-21 expression during colo- was concomitantly reduced in miR-21-up-regulated regions rectal tumorigenesis results in a decrease in PDCD4 expression. (Fig. 4C). These results indicate the importance of up-regula- tion of miR-21 and down-regulation of PDCD4 as diagnostic biomarkers of colorectal carcinogenesis. Discussion Recently,extensive analysis on miRNA profiles that are associated with prognosis and therapeutic outcome in colon MiR-21 expression and cancer development. MiR-21 expres- adenocarcinoma was reported (15). Importantly,more ad- sion has been reported to be one of the best hit miRNA in many vanced tumors expressed higher levels of miR-21 using either profiling experiments designed to detect up-regulated miRNA microarray data from the test cohort or the quantitative reverse in human cancer including colorectal carcinoma (3). Our ISH transcription-PCR data from the validation cohort. By analysis analysis on the colorectal carcinomas clearly detected high level on pooled cohorts,they further showed that high miR-21 expression of miR-21 in most of them (Fig. 2; Table 1),and im- expression is associated with a poor prognosis in either stage portantly,we further showed that this miR-21 increase can be frequently detected from the adenoma stage in the section

Table 2. Summary of the association between high level expression of miR-21 expression and Table 1. Summary of miR-21 expression levels in the histologic features of endoscopically resected surgically excised colorectal adenocarcinoma colorectal polyps (malignant tumor) and ( with cancerous potential) Histologic features of Up- Total colorectal polyps regulation Histologic features of Evaluation of miR-21 of miR-21 colorectal tumors expression ++ + − 5 4 3 2 1 Total Tumorigenic Noninvasive carcinoma 7 01 8 Adenocarcinoma 61086434 Adenoma 821020 (malignant tumor) (high or low grade) Adenoma (benign tumor 0 0014 5 Nontumorigenic Hyperplastic 003 3 with cancerous potential) polyp Juvenile 001 1 Peutz-Jegher's 0022 NOTE: Evaluations of the miR-21 expression based on the LNA- (hamartomatous) ISH staining compared with normal colorectal mucosa on the same slides. Values assigned to the staining (from 1 to 5) were NOTE: Pathohistologic diagnoses were judged by the Vienna clas- decided as follows: 5, most tumor cells (>80%) show much stron- sification of gastrointestinal epithelial neoplasia. Evaluations of the ger staining than normal epithelial cells; 4, most tumor cells intensity of miR-21 staining by LNA-ISH were decided as follows: (>80%) show stronger staining than normal epithelial cells; 3, a ++, cells in the polyp show much stronger staining compared with part of tumor cells (∼20-80%) show stronger staining than normal normal epithelial cells on the same slide; +, cells in the polyp show epithelial cells; 2, some tumor cells (more than 20%) show slight- stronger staining compared with normal epithelial cells on the ly stronger staining than normal epithelial cells; 1, almost all tu- same slide; −, cells in the polyp show the same staining intensity mor cells show same staining intensity as normal epithelial cells. as normal epithelial cells on the same slide.

www.aacrjournals.org 4013 Clin Cancer Res 2009;15(12) June 15, 2009 Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2009 American Association for Cancer Research. Human Cancer Biology

Fig. 3. LNA-ISH detecting miR-21 with FFPE colorectal tissues obtained from endoscopic mucosal resection. Bar, 500 μm. A, nontumorous colorectal tissues. Photos from left to right are normal colorectal mucosa, , , and Peutz-Jagher's hamartomatous polyp. B, tumorous but nonmalignant colorectal lesions. Photos from left to right are serrated adenoma (negative for miR-21), tubular adenoma (negative for miR-21), tubular adenoma (miR-21 positive in the edge), and tubular adenoma (positive for miR-21). C, tumorous and malignant colorectal lesions. In each photo, adenocarcinoma in adenoma, strongly positive for miR-21, is shown.

II or stage III colon cancer patients,indicating its potential as signal amplification system was used to detect the nuclear pro- a prognostic biomarker (15). tein Brm,which is difficult to detect by immunohistologic LNA-ISH combined with biotin-free tyramide signal amplifica- methods (37). In the present study,we applied this method tion system is a useful technique for determination of miRNA to LNA-ISH and were able to sensitively detect miR-21 expres- expression levels in FFPE tissues. The LNA-modified oligonucle- sion. Because detection of nonspecific signals is not infrequent otide would be one of the most sensitive probes currently avail- when using tyramide amplification,we confirmed that staining able for miRNA detection (34,35). Nevertheless,it is quite by the scramble control probe (Supplementary Fig. S5B) as well difficult to detect miRNAs by ISH,especially in FFPE clinical as staining unrelated to the LNA/DNA probe (Supplementary tissues. It has been reported that the LNA-ISH technique could Fig. S4B) was only rarely detected. detect some miRNA species in FFPE samples (36). However,us- High-level expression of miR-21 in cancer-associated fibroblasts ing LNA-modified probes alone or LNA-ISH combined with a may be induced by secreting factors originating from cancer cells. universal immunoenzyme polymer method (Histofine Simple It is very interesting that the stromal fibroblasts around tumors Stain MAX PO_MULTI purchased from Nichirei),we were not frequently express miR-21 at high levels. In most cases from the able to detect miR-21 expression in FFPE samples (data not present study,these cancer-associated fibroblasts showed strong shown). In our previous study (31),the biotin-free tyramide miR-21 expression compared with distant normal fibroblasts

Clin Cancer Res 2009;15(12) June 15, 20094014 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2009 American Association for Cancer Research. Mir-21 Up-Regulation during Colorectal Carcinogenesis and normal epithelial cells (Supplementary Table S1). In some had elevated miR-21 expression,which was nondetectable in cases,the fibroblast staining intensities were even more intense normal colorectal epithelium (Fig. 4). In progressive colorectal than those of adjacent malignant cells (Supplementary cancers,almost all cells expressed miR-21,whereas PDCD4 was Fig. S4A). Therefore,we hypothesized that this is a non– cell- almost undetectable. In summary of our ISH analysis,expres- autonomous phenomenon and that cytokines secreted from sion of miR-21 RNA and PDCD4 protein showed mutually ex- the adjacent malignant tumors might contribute to miR-21 in- clusive patterns in colorectal epithelial cells. These observations duction. In this regard,it has been reported that interleukin-6 support that PDCD4 is a good target of miR-21 in vivo. Because levels are elevated in the cancer-associated fibroblasts around a PDCD4 is a potent tumor suppresser,miR-21 may perform on- colon cancer,in serum,and in tumor tissues from colorectal cogenic functions,at least in part,through down-regulation of cancer patients (38,39). It is also noteworthy that interleu- PDCD4. kin-6 induces the transcription of miR-21 in multiple myeloma We have previously shown that in human cell culture sys- cells through mediation of signal transducer and activator of tems,a double-negative feedback loop operates between transcription-3 activity (12). Importantly,signal transducer miR-21 and its target protein,NFIB,through the miR-21 pro- and activator of transcription-3 binding sites are present just moter,which has the binding site of this negative transcrip- upstream of the transcriptional start site in the miR-21 promot- tional regulator (16). This means that the miR-21 gene er. Therefore,interleukin-6 is a candidate for miR-21 induction integrates a system that self-reinforces its own expression. in these cancer-associated fibroblasts. Whereas in adult rats NFIB mRNA is highly expressed (41), MiR-21 RNA and PDCD4 protein expression patterns were we are currently not able to perform specific immunohisto- mutually exclusive in colorectal epithelial cells. PDCD4 was chemical staining due to the absence of a specific anti-NFIB highly expressed in normal colorectal epithelium,but PDCD4 antibody that is applicable to FFPE clinical samples and is expression was often reduced in precancerous colorectal regions non–cross-reactive with other NFI family members. Therefore, (Fig. 4). This observation is consistent with a recent report that direct evidence of NFIB involvement in vivo remains to be normal mucosa showed strong nuclear PDCD4,which was sig- established. Using rat cell culture system,Dr. Verde's group nificantly reduced in adenomas (40). Our ISH analyses further very recently indicated that PDCD4 suppressed miR-21 pro- indicated that cells with reduced PDCD4 expression frequently moter activity,at least in part,by inhibiting AP-1 activity

Fig. 4. Images from sequential FFPE colorectal tissue sections in which LNA-ISH for miR-21 RNA (left) and immunostaining for PDCD4 protein (right) were done. A, a surgically resected colorectal adenocarcinoma. The border of the colorectal cancer (left half) and normal colorectal mucosa (right half)is apparent in the two photos. Bar, 200 μm. B, a containing adenocarcinoma and adenoma obtained from endoscopic mucosal resection. In the two photos, evidence of malignant change is apparent in the left half. Bar, 500 μm. C, a nonmalignant adenomatous colorectal polyp obtained by endoscopic mucosal resection. Adenomatous change is apparent at the edge of the colorectal mucosa. Bar, 500 μm.

www.aacrjournals.org 4015 Clin Cancer Res 2009;15(12) June 15, 2009 Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2009 American Association for Cancer Research. Human Cancer Biology

(29). Therefore,these double-negative feedback loops operat- scription factors by PDCD4 could significantly advance our ing through the miR-21 promoter may contribute to the self- understanding of cancer progression. Therefore,elucidation reinforced expression of miR-21. This feedback could further of the molecular mechanisms involved in the entire regulatory lead to the mutually exclusive expression patterns observed network formed by miR-21 is important for understanding the between miR-21 RNA and PDCD4 protein in colorectal can- initial stages of colorectal carcinogenesis. cer. It also remains to be determined which transcriptional factors that are normally inhibited by PDCD4 are crucial for Disclosure of Potential Conflicts of Interest miR-21 gene induction in colorectal cancer. The molecular mechanisms involved in functional suppression of these tran- No potential conflicts of interest were disclosed.

References 1. Chen CZ. MicroRNAs as oncogenes and tumor expression triggered by AP-1 is sustained 29. Talotta F, Cimmino A, Matarazzo MR, et al. An suppressors. N Engl J Med 2005;353:1768–71. through a double-negative feedback mechanism. autoregulatory loop mediated by miR-21 and 2. Fujita S, Iba H. Putative promoter regions of J Mol Biol 2008;378:492–504. PDCD4 controls the AP-1 activity in RAS trans- miRNA genes involved in evolutionarily con- 17. Gabriely G, Wurdinger T, Kesari S, et al. Micro- formation. Oncogene 2009;28:73–84. served regulatory systems among vertebrates. RNA 21 promotes glioma invasion by targeting 30. Leslie A, Carey FA, Pratt NR, Steele RJ. The Bioinformatics 2008;24:303–8. matrix metalloproteinase regulators. Mol Cell colorectal adenoma-carcinoma sequence. Br 3. Volinia S, Calin GA, Liu CG, et al. A microRNA Biol 2008;28:5369–80. J Surg 2002;89:845–60. expression signature of human solid tumors 18. Sayed D, Rane S, Lypowy J, et al. MicroRNA- 31. Yamamichi N, Inada K, Ichinose M, et al. defines cancer gene targets. Proc Natl Acad Sci 21 targets Sprouty2 and promotes cellular out- Frequent loss of Brm expression in gastric U S A 2006;103:2257–61. growths. Mol Biol Cell 2008;19:3272–82. cancer correlates with histologic features and 4. Iorio MV, Ferracin M, Liu CG, et al. MicroRNA 19. Zhu S, Si ML, Wu H, Mo YY. MicroRNA- differentiation state. Cancer Res 2007;67: gene expression deregulation in human breast 21 targets the tumor suppressor gene tropo- 10727–35. cancer. Cancer Res 2005;65:7065–70. myosin 1 (TPM1). J Biol Chem 2007;282: 32. Haraguchi T, Mizutani T, Yamamichi N, Ito T, 5. Si ML, Zhu S, Wu H, Lu Z, Wu F, Mo YY. MiR-21- 14328–36. Minoguchi S, Iba H. SiRNAs do not induce mediated tumor growth. Oncogene 2007;26: 20. Frankel LB, Christoffersen NR, Jacobsen A, RNA-dependent transcriptional silencing of ret- 2799–803. Lindow M, Krogh A, Lund AH. Programmed cell rovirus in human cells. FEBS Lett 2007;581: 6. Lee EJ, Gusev Y, Jiang J, et al. Expression pro- death 4 (PDCD4) is an important functional tar- 4949–54. filing identifies microRNA signature in pancreat- get of the microRNA miR-21 in breast cancer 33. Kim VN. MicroRNA biogenesis: coordinated ic cancer. Int J Cancer 2007;120:1046–54. cells. J Biol Chem 2008;283:1026–33. cropping and dicing. Nat Rev Mol Cell Biol 7. Roldo C, Missiaglia E, Hagan JP, et al. Micro- 21. Asangani IA, Rasheed SA, Nikolova DA, et al. 2005;6:376–85. RNA expression abnormalities in pancreatic en- MicroRNA-21 (miR-21) post-transcriptionally 34. Neely LA, Patel S, Garver J, et al. A single-mole- docrine and acinar tumors are associated with down-regulates tumor suppressor Pdcd4 and cule method for the quantitation of microRNA distinctive pathologic features and clinical be- stimulates invasion, intravasation and metasta- gene expression. Nat Methods 2006;3:41–6. havior. J Clin Oncol 2006;24:4677–84. sis in colorectal cancer. Oncogene 2008;27: 35. Kloosterman WP, Wienholds E, de Bruijn E, 8. Meng F, Henson R, Lang M, et al. Involvement 2128–38. Kauppinen S, Plasterk RH. In situ detection of of human micro-RNA in growth and response 22. Haraguchi T, Ozaki Y, Iba H. Vectors expres- miRNAs in animal embryos using LNA-modified to chemotherapy in human cholangiocarcino- sing efficient RNA decoys achieve the long- oligonucleotide probes. Nat Methods 2006;3: ma cell lines. Gastroenterology 2006;130: term suppression of specific microRNA 27–9. 2113–29. activity in mammalian cells. Nucleic Acids 36. Nelson PT, Baldwin DA, Kloosterman WP, 9. Meng F, Henson R, Wehbe-Janek H, Ghoshal K, Res. 2009;37:e43. Kauppinen S, Plasterk RH, Mourelatos Z. RAKE Jacob ST, Patel T. MicroRNA-21 regulates ex- 23. Lu Z, Liu M, Stribinskis V, et al. MicroRNA-21 and LNA-ISH reveal microRNA expression and pression of the PTEN tumor suppressor gene in promotes cell transformation by targeting the localization in archival human brain. RNA 2006; human hepatocellular cancer. Gastroenterology programmed cell death 4 gene. Oncogene 2008; 12:187–91. 2007;133:647–58. 27:4373–9. 37. Reisman DN, Sciarrotta J, Bouldin TW, 10. Chan JA, Krichevsky AM, Kosik KS. MicroRNA- 24. Palamarchuk A, Efanov A, Maximov V, Aqeilan Weissman BE, Funkhouser WK. The expression 21 is an antiapoptotic factor in human glioblas- RI,CroceCM,PekarskyY.Aktphosphorylates of the SWI/SNF ATPase subunits BRG1 and toma cells. Cancer Res 2005;65:6029–33. and regulates Pdcd4 tumor suppressor protein. BRM in normal human tissues. Appl Immuno- 11. Fulci V, Chiaretti S, Goldoni M, et al. Quantita- Cancer Res 2005;65:11282–6. histochem Mol Morphol 2005;13:66–74. tive technologies establish a novel microRNA 25. Yang HS, Jansen AP, Komar AA, et al. 38. Nakagawa H, Liyanarachchi S, Davuluri RV, profile of chronic lymphocytic leukemia. Blood The transformation suppressor Pdcd4 is a novel et al. Role of cancer-associated stromal fibro- 2007;109:4944–51. eukaryotic translation initiation factor 4A binding blasts in metastatic colon cancer to the liver 12. Loffler D, Brocke-Heidrich K, Pfeifer G, et al. protein that inhibits translation. Mol Cell Biol and their expression profiles. Oncogene 2004; Interleukin-6 dependent survival of multiple my- 2003;23:26–37. 23:7366–77. eloma cells involves the Stat3-mediated induc- 26. Wang Q, Sun Z, Yang HS. Down-regulation of 39. Esfandi F, Mohammadzadeh Ghobadloo S, tion of microRNA-21 through a highly tumor suppressor Pdcd4 promotes invasion Basati G. Interleukin-6 level in patients with colo- conserved enhancer. Blood 2007;110:1330–3. and activates both β-catenin/Tcf and AP-1- rectal cancer. Cancer Lett 2006;244:76–8. 13. Lui WO, Pourmand N, Patterson BK, Fire A. dependent transcription in colon carcinoma 40. Mudduluru G, Medved F, GrobholzR, et al. Patterns of known and novel small RNAs in cells. Oncogene 2008;27:1527–35. Loss of programmed cell death 4 expression human cervical cancer. Cancer Res 2007;67: 27. Yang HS, Jansen AP, Nair R, et al. A novel trans- marks adenoma-carcinoma transition, correlates 6031–43. formation suppressor, Pdcd4, inhibits AP-1 trans- inversely with phosphorylated protein kinase B, 14. Iorio MV, Visone R, Di Leva G, et al. MicroRNA activation but not NF-κB or ODC transactivation. and is an independent prognostic factor in re- signatures in human ovarian cancer. Cancer Res Oncogene 2001;20:669–76. sected colorectal cancer. Cancer 2007;110: 2007;67:8699–707. 28. Leupold JH, Yang HS, Colburn NH, Asangani 1697–707. 15. Schetter AJ, Leung SY, Sohn JJ, et al. Micro- I, Post S, Allgayer H. Tumor suppressor Pdcd4 41. Osada S, Matsubara T, Daimon S, et al. Ex- RNA expression profiles associated with prog- inhibits invasion/intravasation and regulates pression, DNA-binding specificity and tran- nosis and therapeutic outcome in colon urokinase receptor (u-PAR) gene expression scriptional regulation of nuclear factor 1 adenocarcinoma. JAMA 2008;299:425–36. via Sp-transcription factors. Oncogene 2007;26: family proteins from rat. Biochem J 1999;342: 16. Fujita S, Ito T, Mizutani T, et al. miR-21 gene 4550–62. 189–98.

Clin Cancer Res 2009;15(12) June 15, 20094016 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2009 American Association for Cancer Research. Locked Nucleic Acid In situ Hybridization Analysis of miR-21 Expression during Colorectal Cancer Development

Nobutake Yamamichi, Ryoichi Shimomura, Ken-ichi Inada, et al.

Clin Cancer Res 2009;15:4009-4016.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/15/12/4009

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2009/06/15/1078-0432.CCR-08-3257.DC1

Cited articles This article cites 40 articles, 17 of which you can access for free at: http://clincancerres.aacrjournals.org/content/15/12/4009.full#ref-list-1

Citing articles This article has been cited by 14 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/15/12/4009.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 Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/15/12/4009. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from clincancerres.aacrjournals.org on September 28, 2021. © 2009 American Association for Cancer Research.