icam-1 related noncoding rna in cancer stem cells ... · biology of human tumors icam-1–related...

Biology of Human Tumors

ICAM-1–Related Noncoding RNA in CancerStem Cells Maintains ICAM-1 Expression inHepatocellular CarcinomaWeixing Guo1, Shupeng Liu2, Yuqiang Cheng1, Lei Lu2, Jie Shi1, Guixia Xu2,Nan Li1, Kai Cheng2, Mengchao Wu1, Shuqun Cheng1, and Shanrong Liu2

Abstract

Purpose: Portal vein tumor thrombus (PVTT) is a major com-plication of hepatocellular carcinoma (HCC) and is associatedwith poor survival. Long noncoding RNAs (lncRNA) contribute toHCC metastasis, but whether and how lncRNAs affect PVTTdevelopment remains unclear. In the present study, a novel highlyexpressed lncRNA (ICAM-1–related, ICR)was identified in ICAM-1þ cancer stem cells (CSC) in HCC. This lncRNA regulated CSCproperties and contributed to PVTT development.

Experimental Design:We used microarray and bioinformat-ics analyses to identify differentially expressed lncRNAs. Real-time PCR and Western blotting were used to assess geneexpression in cell lines and tumors. Sphere formation assayswere performed to investigate stem cell properties of tumorcells in vitro. Retrospective and prospective studies were used toinvestigate the relationship between ICR expression and clinicaloutcomes.

Results: Compared with the corresponding primary tumors,PVTT expressed different lncRNAs and mRNAs, including theupregulated lncRNA ICR and ICAM-1. ICR regulated ICAM-1expression by increasing the stability of its mRNA through RNAduplex formation, whichmodulated the CSCproperties of ICAM-1þ HCC cells. ICR transcription in ICAM-1þ HCC cells wasregulated by Nanog, and inhibition of ICR in situ significantlyreduced ICAM-1 expression and ICAM-1þHCC cells in tumors invivo. Moreover, elevated ICR and ICAM-1 expression in tumorswas correlated with PVTT development and poor clinicaloutcomes.

Conclusions: Our study demonstrates that ICR specificallyregulates CSC properties of ICAM-1þ HCC cells and that ICRcontributes to PVTT development. Therefore, ICR may be apromising target for HCC therapy. Clin Cancer Res; 22(8); 2041–50.�2015 AACR.

IntroductionHepatocellular carcinoma (HCC) is the sixth most common

cancer and the second leading cause of cancer-related deathworldwide, although substantial progress has been made regard-ing its diagnosis and treatment (1). Approximately 90%of cancer-associatedmortality is due to tumormetastasis and relapse (2). InHCC, the poor prognosis is also related to metastasis (3). Portalvein tumor thrombus (PVTT), a special type ofHCCmetastases, isamajor complication ofHCC that is associatedwith poor survival(4). MRI and ultrasonography have demonstrated that approxi-mately 50% to 80% of HCC cases are accompanied by portal orhepatic vein invasion (5). Although growing evidence indicates

that noncoding RNA molecules are involved in HCC tumormetastasis and PVTT development (6–8), the mechanism under-lying PVTT formation remains largely unknown.

Long noncoding RNAs (lncRNA) are a type of RNA composedof 200 to thousands of nucleotides that do not produce proteins(9). Emerging data have indicated that lncRNAs play importantregulatory roles in diverse biologic cellular processes, includingtumorigenesis and cancer progression (10, 11). Moreover, inHCC, lncRNAs, such as Highly Upregulated in Liver Cancer(HULC), High Expression in HCC (HEIH), and MicrovascularInvasion inHCC (MVIH), were reported to be upregulated and topromote tumor microvascular invasion and metastasis (12–14).In addition, lncRNAs have also been reported to promote pro-liferation and stem cell–like properties of HCC cells (15). Cancerstemcells (CSC)are also inextricably involved in tumormetastasis(2, 16). In HCC, varied CSC subpopulations, such as EpCAMþ

cells and CD24þ cells, have been identified and found to correlatewith tumor incidence andmetastasis (17, 18). Although lncRNAsand CSCs have been found to be involved in tumor metastasis,whether and how they affect PVTT development remains unclear.

In the present study, we employed microarray analyses toinvestigate lncRNAs and mRNAs that are specifically highlyexpressed in PVTT tissues and the established PVTT cell line(CSQT-2; ref. 19). Among the differentially expressed molecules,further analyses revealed that an lncRNA had approximately 800bp complementary to the ICAM-1 mRNA sequence, thus termedas ICAM-1–related (ICR) lncRNA. ICR formed an RNA duplexwith ICAM-1 to maintain ICAM-1 expression, thereby regulating

1Eastern Hepatobiliary Surgery Hospital, Second Military Medical Uni-versity, Shanghai, China. 2Changhai Hospital, Second Military MedicalUniversity, Shanghai, China.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

W. Guo, S. Liu, and Y. Cheng contributed equally to this article.

Corresponding Authors: Shanrong Liu, Changhai Hospital, Second MilitaryMedical University, 168 Changhai Road, Shanghai 200433, China. Phone: 86-21-31162091; Fax: 86-21-31162091; E-mail: [email protected]; or ShuqunCheng, Eastern Hepatobiliary Surgery Hospital, Second Military Medical Uni-versity, 225 Changhai Road, Shanghai, China. Phone: 86-21-81875251; Fax: 86-21-65562400; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-14-3106

�2015 American Association for Cancer Research.

ClinicalCancerResearch

www.aacrjournals.org 2041

Cancer Research. by guest on August 24, 2020. Copyright 2015 American Association forhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

https://bloodcancerdiscov.aacrjournals.org

the stem cell properties of ICAM-1þ HCC cells, as ICAM-1 waspreviously observed to be aCSCmarker inHCC (20). In addition,retrospective and prospective analyses revealed that ICR expres-sion was closely associated with PVTT development and HCCpatients' prognosis.

Materials and MethodsPatients and design

A total of 617 HCC patients who underwent partial hepatec-tomy at Eastern Hepatobiliary Surgery Hospital from 2001 to2011 were included according to the guidelines as previouslydescribed (21). A total of 245HCC patients with PVTT from 2001to 2004 (Supplementary Table S1) were used for the trainingcohort. Another 372 patients from 2010 to 2011were recruited ina prospective study and used as the validation cohort (Supple-mentary Table S2). The study was approved by the Ethics Com-mittee of Eastern Hepatobiliary Surgery Hospital of Second Mil-itary Medical University. Written informed consent was obtainedfrom all patients according to the regulations of the committee.

Cell linesThe human PVTT cell line CSQT-2 was established in our

laboratory (19). Huh7 and Hep3B were obtained from the Jap-anese Collection of Research Bioresources and the AmericanTissue Culture Collection, respectively. The cell lines were main-tained in high-glucose DMEM (Gibco BRL) supplemented with10% FBS (Gibco BRL), 100 mg/mL penicillin G, and 50 mg/mLstreptomycin (Gibco BRL) at 37�C in a humidified atmospherecontaining 5% CO2. At the beginning of this study, all the 3 celllines were authenticated through testing the expression of a-feto-protein and cytokeratin 19 by flow cytometry as what we didbefore in our previously study (20).

Statistical analyses, primers (Supplementary Table S3), anti-bodies (Supplementary Table S4), microarray data (GSE43630)analysis, and other materials and methods were listed in Supple-mentary Materials and Methods in detail.

ResultsICR was upregulated in PVTT tissues and CSQT-2 cells

To investigate lncRNAs that contribute to PVTT development,cancerous tissues were collected from 3 HCC patients with PVTT

and then assessed using lncRNAmicroarrays to screen for lncRNAsthat were specifically expressed in PVTT. Many lncRNAs wereidentified as differentially expressed between PVTT and the

Figure 1.Elevated expression of lncRNA ICR in PVTT tissues and CSQT-2 cells. A,hierarchical clustering analysis of differentially expressed lncRNAs (left) andmRNAs (right) between PVTT tissues (PVTT) and the correspondingprimary tumor tissues (Tumor). B, representative image of RT-PCR productsof 50-RACE. The major PCR product is indicated by the arrow (left). Thesequence of the 50-end of ICR is also shown (right). C, DNAMAN analysisrevealed a poly(A) tail at the 30-ends of the ICR sequence. D, real-time PCRanalysis of ICR expression in tissues and cell lines used in microassay. Theerror bars represent the SD of data obtained in at least three independentexperiments. �� , P < 0.01.

Translational Relevance

Portal vein tumor thrombus (PVTT) is amajor complicationof hepatocellular carcinoma (HCC) and is significantly asso-ciated with poor survival. PVTT is considered as a special typeof HCC metastasis, and we previously found that specificinhibition of ICAM-1 reduced HCC incidence and metastasis.In the present study, we report the regulation of ICAM-1 by theICAM-1–related long non-coding RNA (ICR) in PVTT, thecorrelation of ICRwith PVTT development, and the inhibitionof ICAM-1 expression in HCC cells via ICR downregulation.The inhibition of tumor cell migration and growth by ICRdownregulation suggested that ICRmaybe a therapeutic targetfor the treatment of metastatic HCC. The findings reportedhere suggest a promising strategy for the treatment of PVTT inHCC.

Guo et al.

Clin Cancer Res; 22(8) April 15, 2016 Clinical Cancer Research2042



corresponding primary tumors (Fig. 1A, left), indicating that thesignificantly altered expression of these lncRNAsmay be involvedin the pathogenesis of PVTT. In addition, many differentiallyexpressed mRNAs were also identified in the microarrayanalysis between PVTT and the corresponding primary tumors(Fig. 1A, right). To verify the altered molecules, CSQT-2 cellsand Hep3B cells were also assessed via microarray analyses.Similarly to the results obtained using tissues, many lncRNAsand mRNAs were found to be differentially expressed betweenthe two cell lines (Supplementary Fig. S1). Among the dif-ferentially expressed RNAs, 32 lncRNAs (Supplementary TableS5) and 26 mRNAs (Supplementary Table S6) were found inboth tissues and cell lines. These RNAs may be candidatemolecules involved in PVTT carcinogenesis. For further iden-tification, the bio-software Blat was used to perform bioin-formatic analyses to identify lncRNA–mRNA pairs amongthese altered RNAs, as lncRNAs can regulate gene expressionbased on sequence complements (22). Among the 11lncRNA–mRNA pairs, lnc24236-ICAM-1 mRNA had an 812-base pair (bp) continuous complementary sequence, whichwas the longest complementary sequence (SupplementaryTable S7 and Supplementary Fig. S2). This observation indi-cates that lnc24236 may be involved in ICAM-1 regulation.Because ICAM-1 was highly expressed in HCC stem cells andplays pivotal roles in HCC carcinogenesis and metastasis,lnc24236 was thus selected for further analysis and termedICR in the present study.

To ensure that ICR was a nontranslated transcript, we clonedthe full-length transcript of ICR via 50 rapid amplification of thecDNA ends (50 RACE), even though a poly(A) tail was detected atthe 30-end of the ICR sequence via a DNAMAN analysis (Fig. 1Band C). We then calculated the coding probability (CP) of ICRusing the Coding Potential Assessment Tool (23). The CP of ICRwas 0.046, whichwas statistically associatedwith noncoding RNA(CP < 0.364; Supplementary Fig. S3). To verify ICR expression inthe tumor tissues and cell lines, real-time PCR was performedusing PVTT tissues (PVTT) and the corresponding primary HCCtumor tissues (Tumor) and cell lines (CSQT-2 andHep3B), whichwere used in the microarray analysis. The expressions of ICR inPVTT and CSQT-2 were found higher than that in the correspond-ing control (Fig. 1D), which was consistent with the result fromthe microarray analysis. In addition, ICAM-1 expression wasevaluated in these tissues and cell lines. Elevated expressions ofICAM-1 in PVTT and CSQT-2 were also observed (SupplementaryFig. S1B), indicating a potential relationship between ICAM-1 andICR.

ICR/ICAM-1 expressionswere involved in theprognosis ofHCCpatients with PVTT in the training cohort

We further examined whether ICR/ICAM-1 expression levelscorrelated with the outcome of PVTT patients after hepatectomy.The expression patterns of ICR and ICAM-1 in humanHCC tumortissues were firstly assessed using ISH. A small population of cellsin the tumor tissues stained double positive for ICR (red) andICAM-1 (green), although some ICAM-1þ cells stained negativefor ICR (arrows; Fig. 2A; Supplementary Fig. S4A). The sameexpression pattern was found in PVTT tissue via staining, furtherconfirming their expression correlation (Supplementary Fig.S4B). Then, real-time PCR was used to investigate the expressionlevel of ICR and ICAM-1 in tumor tissues and paired PVTT tissuesfrom a training cohort (cohort 1), which included 245 HCC

patients with PVTT. The expression levels of both ICR andICAM-1 were higher in PVTT tissues (PVTT) than those in thecorresponding tumor tissues (Tumor; Fig. 2B). Moreover, ICRexpression in PVTT was linearly correlated with ICAM-1 expres-sion (Fig. 2C; r¼ 0.625; P < 0.001). To examined the relationship

Figure 2.The association of ICR/ICAM-1 expression and prognosis in the trainingcohort. A, a representative image from immunofluorescent staining analysisof ICAM-1 and ISH analysis for ICR expression in HCC tumor tissues.The arrow indicated cells expressing ICAM-1 (green) but not ICR (red); scalebar, 10 mm. B, real-time PCR analysis of ICR (left) and ICAM-1 (right)expression in PVTT tissues (PVTT) and corresponding primary tumor tissues(Tumor). The error bars represent the SD of data obtained in at least threeindependent experiments. �� , P < 0.01. C, linear correlation between ICR andICAM-1 was observed. DCt values were used to measure gene expression,which was normalized to b-actin expression levels. D, Kaplan–Meier'sanalyses of correlations between OS of 245 HCC patients and ICR (left) orICAM-1 (right) expression level.

ICR in CSCs Maintains ICAM-1 Expression in HCC

www.aacrjournals.org Clin Cancer Res; 22(8) April 15, 2016 2043



between ICR/ICAM-1 expression and the clinical outcomes ofthese HCC patients, the patients were classified into a high-ICRgroup (n ¼ 123) and a low-ICR group (n ¼ 122) and a high-ICAM-1 group (n ¼ 125) and a low-ICAM-1 group (n ¼ 120)according to the expression of ICR and ICAM-1, respectively, intheir PVTT tissues (the median values were used as the cutoffpoints). The Kaplan–Meier analysis showed that the patientswith higher ICR or higher ICAM-1 in the PVTT tissues hadreduced overall survival (OS) than the corresponding low-expression groups (Fig. 2D). These results suggested theinvolvement of ICR and ICAM-1 in PVTT development andindicated that ICR and ICAM-1 could serve as prognosticpredictors for HCC patients with PVTT.

ICR regulated ICAM-1 expression in HCC cellsWe first showed that ICR expression was correlated with

ICAM-1. Next, we investigated whether ICR regulated ICAM-1 expression. Three siRNAs targeting ICR at noncomplementarysequences were generated and transfected into CSQT-2 cells. Allthree siRNAs reduced ICR expression significantly, among

which siRNA1 reduced ICR expression to the most extent(Supplementary Fig. S5A). When siRNA1 was transfected intoCSQT-2, Hep3B, and Huh7 cells, ICAM-1 expression wasmarkedly reduced at the mRNA and protein levels (Fig. 3A;Supplementary Fig. S5B, left). Transfection of siRNA3 alsodisplayed the reduction of ICAM-1 mRNA and protein in thethree cell lines, further verifying the downregulation of ICAM-1by ICR reduction (Supplementary Fig. S5C). Moreover, theoverexpression of ICR due to transfection of the cell lines withpcDNA3.1-ICR (ICR) plasmids (Supplementary Fig. S5D) ledto a significant upregulation of ICAM-1 mRNA and proteinexpression (Fig. 3A; Supplementary Fig. S5B, right). Next, weexamined the underlying mechanism of ICAM-1 regulation byICR. As shown in Fig. 3B (top) and Supplementary Fig. S2,approximately 800 bp of the ICR sequence (nucleotide posi-tions 2404–3223 bp) was nearly complementary to the ICAM-1mRNA transcript (nucleotide positions 3246–2431 bp). Severalstudies have demonstrated that certain lncRNAs regulated tar-get genes by binding to DNA or protein (9). Therefore, wehypothesized that ICR may upregulate ICAM-1 expression by

Figure 3.ICR regulated ICAM-1 expression in tumor cells. A, real-time PCR and Western blotting analyses of ICAM-1 expression in tumor cells with downregulation of ICR(siRNA1; left top and bottom plots) or upregulation of ICR (ICR; right top and bottom plots). B, RPA performed on RNA samples from CSQT-2 cells. Theschematic diagram of the RNA duplex formed between ICR and ICAM-1 (top). Amplification plots of overlapping (overlapping) or nonoverlapping (nonoverlapping)regions of ICAM-1 mRNA after degradation by RNase AþT (bottom). Undigested, undigested RNA group; digested, digested RNA group. C, analysis of thestability of ICAM-1 transcripts from CSQT-2 cells with downregulation of ICR (siRNA1) or Hep3B cells with upregulation of ICR (ICR) by real-time PCR. The mRNAlevels were measured after incubation with a-amanitin (50 mmol/L) for 6 hours and relative to mRNA levels at time 0 (when a-amanitin was added). D,analysis of ICAM-1 protein from CSQT-2 cells with downregulation of ICR (siRNA1) or Hep3B cells with upregulation of ICR (ICR) by Western blotting 6 hoursafter incubation with a-amanitin (50 mmol/L). E, Transwell assays performed using CSQT-2 cells with downregulation of ICR (siRNA1) or Hep3B cells withupregulation of ICR (ICR). The error bars represent the SD of data obtained in at least three independent experiments. � , P < 0.05 and �� , P < 0.01.

Guo et al.




partially binding to ICAM-1 mRNA and increasing its stability.To investigate the possibility of RNA duplex formation, anRNase protection assay (RPA) was performed on RNA isolatedfrom CSQT-2 cells. Subsequent real-time PCR analysis revealedthat the overlapping portions of ICR and ICAM-1 were pro-tected from degradation, indicating that ICR and ICAM-1indeed form an RNA duplex (Fig. 3B, bottom plots and Sup-plementary Fig. S6). We further assessed the stability of ICAM-1mRNA via real-time PCR after ICR downregulation or upregu-lation in vitro. The results demonstrated that the stability ofICAM-1 mRNA was decreased by the downregulation of ICRusing siRNA1 and increased by the upregulation of ICR using avector expressing a complementary sequence (Fig. 3C). Thecorresponding changes in ICAM-1 protein were also observedby the following Western blotting analysis (Fig. 3D). BecauseICAM-1 expression was correlated with cell metastasis, we alsoinvestigated the metastatic capability of cells with modulatedICR expression by transwell migration assay. As shown in Fig.3E and Supplementary Fig. S7, migration was significantlydecreased when ICR was downregulated by siRNA1 transfection(siRNA1) and increased when ICR was upregulated by over-expression vector transfection (ICR), suggesting that ICR

increased the migratory capability of HCC cells by upregulatingICAM-1 expression.

Collectively, our data demonstrated that ICR regulated ICAM-1expression by increasing the stability of its mRNA through RNAduplex formation and then promoted HCC cells metastasis.

ICR modulated the CSC properties of ICAM-1þ HCC cellsBecause ICAM-1þHCC cells were previously identified as HCC

CSCs (20), we next investigated whether ICR was involved inmaintaining the CSC properties of ICAM-1þ cells due to itsregulation of ICAM-1 expression in HCC cells. We first verifiedthat ICRwas only expressed in the ICAM-1þ cell subpopulation ofHCCs. ICAM-1þ and ICAM-1�HCC cells were sorted inHCC cells(Huh7 and Hep3B) and primary tumor tissues (n ¼ 6), and ICRexpression was assessed by real-time PCR. Significantly elevatedICR expressionwas observed in ICAM-1þ cells fromboth cell linesand tumors compared with the corresponding ICAM-1� cells(Fig. 4A). To determine whether ICR is involved in maintainingthe CSC properties of ICAM-1þ cells, lentiviruses containingeither the shRNA expression vector targeting ICR (LV-ICR-shRNA1) or themock sequence expression vector (LV-ICR-mock)were constructed and used to infect Huh7 or Hep3B cells. Four

Figure 4.ICR modulated the CSC properties ofICAM-1þ HCC cells. A, real-time PCRanalysis of ICR expression in ICAM-1þ orICAM-1� tumor cells sorted from celllines (Huh7 and Hep3B) and tumortissue primary cultures (Tumors). B,flow cytometry analysis of ICAM-1þ cellsin HCC cell lines (Huh7 and Hep3B) withLV-ICR-shRNA1 (ICR-shRNA1) orLV-ICR-mock (ICR-mock) infection. C, arepresentative image of spheres formedby ICAM-1þ tumor cells from Huh7 cellsinfected by LV-ICR-mock (ICR-mock) orLV-ICR-shRNA1 (ICR-shRNA1; scale bar,100 mm). D, analysis of ICAM-1expression in Huh7 tumors usingimmunofluorescent staining (left; scalebar, 50 mm), real-time PCR, andWestern blotting (right). ICR-shRNA1:LV-ICR-shRNA1 infection; ICR-mock:LV-ICR-mock infection. The error barsrepresent the SD of data obtained in atleast three independent experiments.�� , P < 0.01.





days later, comparedwith LV-ICRmock infection, LV-ICR-shRNA1infection reduced thenumber of ICAM-1þ cells, as assessedbyflowcytometry (Fig. 4B; Supplementary Fig. S8A). Subsequently, asphere formation assay was performed to investigate the CSCproperties of ICAM-1þ cells in vitro. The number of hepatosphereswas reduced in cultures infected with LV-ICR-shRNA1 comparedwith cultures infected with LV-ICR-mock (Fig. 4C; SupplementaryFig. S8B), indicating that ICR was involved in the self-renewal ofICAM-1þ cells in vitro. The reduction of hepatospheres induced byICR downregulationwas further verified by another shRNA expres-sion vector targeting ICR (LV-ICR-shRNA3) transfection (Supple-mentary Fig. S8C).We then assessed thepossible role of ICR in vivo.Huh7 cells were injected subcutaneously into nude mice to estab-lish amousemodel. First, ICAM-1 expression inHuh7 tumors wasassessed 7 days after tumor cell implantation by fluorescencemicroscopy, and immunofluorescent staining found aminor pop-ulation of CSC tumor cells expressing ICAM-1 (Fig. 4D, left andSupplementary Fig. S9). Then, LV-ICR-shRNA1 and LV-ICR-mockwere intratumorally administered to these nude mice as describedin Materials and Methods. As shown in Supplementary Fig. S8D,the administration of LV-ICR-shRNA1 successfully downregulatedICRexpression comparedwith theLV-ICR-mock injection. ICAM-1mRNA and protein expressions were also significantly reduced invivo (Fig. 4D, right). Flow cytometry analysis revealed that the

number of ICAM-1þ cells was markedly reduced in the tumorsfrom mice treated with LV-ICR-shRNA1 (ICR-shRNA1) comparedwith the mice treated with LV-ICR-mock (ICR-mock; Supplemen-tary Fig. S8E). In parallel, treatment with LV-ICR-shRNA1 (ICR-shRNA1) slowed the tumor growth (Supplementary Fig. S8F).Collectively, these results demonstrate that ICR was involved inmodulating ICAM-1þ CSCs by regulating ICAM-1 expression invitro and in vivo.

ICR was regulated by Nanog in ICAM-1þ tumor cellsHaving determined that ICR was involved in maintaining the

CSC properties of HCC CSCs by regulating ICAM-1, we sought toelucidate the mechanism controlling ICR expression in CSCs.Bioinformatic analysis was performed to identify transcriptionfactor binding sites in the ICRpromoter. Because of the crucial roleof Nanog in both ICAM-1þ CSC maintenance and ICAM-1expression (20) and the existence of two Nanog binding sites inthe DNA sequence (-5 kb) upstream of ICR (Fig. 5A, left), weproposed that Nanog may also regulate ICR expression in CSCs.Chromatin immunoprecipitation experiments were then per-formed with CSCs enriched from tumor cell lines and primarycultures of HCC tissues to determine whether Nanog bound tothese sites. As shown in Fig. 5A (right) and Supplementary Fig.S10, Nanog bound to site 1 in the ICR promoter, indicating that

Figure 5.ICR was regulated by Nanog in tumorcells. A, analysis of binding sites ofNanog in the 5-kb genomic sequenceupstream (�5 kb) of the ICR genebased on bioinformatic analysis (left)and chromatin immunoprecipitationassays with spheres from tumor celllines (Huh7 and Hep3B) and primarycultures of HCC tissues (Tumors; right).Site 1, potential binding site of Nanog; Nsite, nonbinding site of Nanog. B and C,real-time PCR analysis of ICRexpression in GFPþ cells in Huh7 orHep3B cellswith upregulation of Nanog(Nanog; B) or decreased Nanogexpression (N-shRNA1; C). D, schematicrepresentationof thepathwaybywhichNanog-regulated ICR transcriptionaffects ICAM-1 expression and CSCmaintenance. The error bars representthe SD of data obtained in at least threeindependent experiments. �� , P < 0.01.

Guo et al.




Nanog may regulate ICR transcription by binding to the ICRpromoter.

To further confirm that ICR expressionwas regulated byNanog,the pIRES2-EGFP-nanog plasmid was transfected into Huh7 andHep3B cells. After validating the Nanog overexpression inducedby pIRES2-EGFP-nanog transfection (Supplementary Fig. S11A),real-time PCR revealed that ICR expression was upregulatedapproximately 7- or 4-fold in cells transfected with pIRES2-EGFP-nanog (Nanog) compared with cells transfected with anempty pIRES2-EGFP vector (EGFP; Fig. 5B). Moreover, we alsoexamined whether ICR expression can be reduced by downregu-lating Nanog expression in tumor cells. Because there is a minorpopulation of Nanogþ cells in cell lines (<10%), we sorted theNanogþ cells from Huh7 and Hep3B cells using the plasmid pH-nanog-promoter-EGFP (pH-NP-EGFP), which expresses GFPunder the control of the human nanog promoter (–446 to þ50bp). After validating elevated Nanog expression (SupplementaryFig. S11B), GFPþ cells were transfected with the pLKO-nanog-shRNA1 plasmid. At 24 hours after transfection, ICR expressionwas reduced and accompanied by the downregulation of Nanogexpression (N-shRNA1; Fig. 5C; Supplementary Fig. S11C). Thisresult was further verified by transfection of another plasmid

expressing shRNA targeting Nanog (pLKO-nanog-shRNA2; Sup-plementary Fig. S11C and S11D). Collectively, these resultsindicated that ICR transcription was directly regulated by Nanogin CSCs and that an ICR-mediated pathway was involved in CSCmaintenance (Fig. 5D).

Relationship between ICR/ICAM-1 expression and theclinicopathologic features of HCC patients in validation cohort

To further verify the clinical significance of ICR and ICAM-1 inHCC, a prospective study including 372 HCC patients with orwithout PVTT fromcohort 2was performed as a validation cohort.According to the auxiliary (CT, MRI, and B-USG) and pathologicexaminations aswell as observations during surgery, the 372HCCpatients were divided into two groups: HCC without PVTT (n ¼284) and HCC with PVTT (n¼ 88). Then, expressions of ICR andICAM-1 were examined in the tumor tissues (Tumor) and pairedperitumoral (PT) tissues using real-time PCR. As shown in Fig. 6A,the expression levels of ICR and ICAM-1 were higher in tumortissues than in PT tissues in both groups. And the expression ofICR and ICAM-1 in tumor tissues from HCC patients with PVTTwas strikingly higher than those fromHCCpatients without PVTT(Fig. 6A).We then analyzed the relationshipbetween ICR/ICAM-1

Figure 6.Relationship between ICR/ICAM-1expression and the clinicopathologicfeatures of 372 HCC patients in thevalidation cohort. A, real-time PCRanalysis of ICR (top) and ICAM-1(bottom) expression in tumor tissues(Tumor) and corresponding peritumoraltissues (PT) from HCC patients with orwithout PVTT. The error bars representthe SD of data obtained in at least threeindependent experiments. � , P < 0.05;�� , P < 0.01. B, Kaplan–Meier's analysis ofcorrelations between overall survive(top) or DFS (bottom) of 372 HCCpatients and ICR/ICAM-1 expressionlevel. C, ROC analysis of the potentialprognostic indicators determined bymultivariate analysis. P < 0.05 for all. A,the area under the curve; FPF, false-positive fraction; TPF, true-positivefraction.





expression and the clinicopathologic features of HCC patients incohort 2.We found that higher expression levels of ICRand ICAM-1 were associated with larger tumor size (P ¼ 0.0004 and P ¼0.0109, respectively), higher rate of intrahepatic metastasis (P ¼0.0003 and P ¼ 0.0263, respectively), and the more advancedtumor–node–metastasis (TNM) stage (P¼ 0.001 and P¼ 0.0009,respectively; Supplementary Table S8). In addition, high ICRexpression was also related to high AFP levels (P ¼ 0.0224) andintercurrent PVTT (P ¼ 0.001; Supplementary Table S8). Thesefindings provided further evidence for the relationship betweenICR/ICAM-1 expression and PVTT, which was revealed by thestudies conducted in cohort 1, and further clarified the closecorrelation between ICR/ICAM-1 and the clinicopathologic fea-tures of HCC patients.

Furthermore, we investigated the association between ICR/ICAM-1 and HCC patients' prognosis. Univariate analysisrevealed that the expression of ICR/ICAM-1 together with thepresence of PVTT, tumor size, tumor encapsulation, intrahepaticmetastasis, and TNM stage were significantly associated with bothdisease-free survive (DFS) and OS of HCC patients (Supplemen-tary Table S9). These factors were further assessed by multivariateanalysis using the Cox proportional hazards model. And theresults showed that ICR was an independent risk factor of pooroutcome of HCC patients (Supplementary Table S9).

Given the close link between ICR-ICAM-1 expression andHCCprognosis, we further analyzed their prognostic significance inHCC. Patients (n ¼ 372) in cohort 2 were classified into fourgroups according to the combined expression of ICR/ICAM-1 intheir tumor tissues: group I (n¼ 123), high ICR and high ICAM-1;group II (n ¼ 63), high ICR and low ICAM-1; group III (n ¼ 61),low ICR and high ICAM-1; and group IV (n ¼ 125), low ICRand low ICAM-1.As shown inFig. 6B, theOSandDFS rates amongthe four groups were significantly different. The 3-year OS andDFS rates were 77.6% and 42.9% for group IV, respectively, andonly 36.7% and 4.9% for group I, respectively. The potentialprognostic indicators suggested by multivariate analysis wereevaluated by ROC analysis. All the factors can serve as predictorsof death and recurrence (Fig. 6C, P < 0.05 for all). For patientdeath, the areas under the curve (AUC) values for ICR and ICAM-1were 0.701 [95% confidence interval (CI), 0.648–0.754; P <0.001] and 0.697 (95% CI, 0.643–0.750; P < 0.001), respectively.For recurrence, the AUC valueswere 0.765 (95%CI, 0.712–0.817;P < 0.001) and 0.773 (95% CI, 0.7192–0.828; P < 0.001),respectively (Fig. 6C). ROC analysis revealed that the expressionlevels of ICR and ICAM-1 are promising prognostic indicators fordeath and recurrence in HCC and exhibited stronger predictivepotential than AFP levels and tumor size, which were also includ-ed in the assessment.

DiscussionLncRNAs are reported to have important epigenetic regulatory

roles in diverse biologic cellular processes, including tumorigen-esis and cancer metastasis. In this study, we found differentlyexpressed lncRNAs in PVTTwith primary HCC tissues. Among thedifferently expressed lncRNAs, ICR was highly expressed in PVTT.Further analyses found that ICR regulated ICAM-1 expression byincreasing the stability of ICAM-1 mRNA through RNA duplexformation. In addition, ICR modulated the CSC properties ofICAM-1þ HCC cells in vitro and in vivo. And ICR expression inICAM-1þ cells was regulated by the stemness-related protein

Nanog. Moreover, we demonstrated that ICR expression wascorrelated with clinical PVTT incidence, aggressive tumor behav-ior, and poor clinical outcomes of HCC patients.

PVTT, as a special type of HCC metastasis, was previouslyreported to have different miRNAs expression patterns com-pared with primary HCC tissues (7). In the present study, wedemonstrated that the expression of lncRNAs in PVTT differedfrom the corresponding tumors. This result explains why PVTTexhibits unique molecular characteristics and why PVTT dis-plays different phenotypes, such as high mobility, with primaryHCC tumors. In particular, we found that lncRNA ICR, highlyexpressed in PVTT, upregulated ICAM-1 expression, promotedHCC cells metastasis, and modulated the CSC properties ofICAM-1þ HCC cells. These findings may explain why PVTTdisplays high mobility and correlates with poor prognosis,since both ICAM-1 and CSCs were correlated with tumormetastasis. Obviously, there are many other lncRNAs differ-ently expressed between PVTT and tumors, indicating that ICRis not the only lncRNA influencing the PVTT characteristics. Inaddition, one study reported that RNA polymerase II subunit 5(RPB5)–mediating protein contributed to PVTT formation bymaintaining tumorigenesis and metastases capacity via pro-moting IL6 production (6). Similarly to other biologic process-es, PVTT formation also involves complex molecular networks,including miRNAs, lncRNAs, and proteins.

The correlation of ICR expression level with PVTT developmentand patients' prognosis demonstrated that ICR may be a novelbiomarker of PVTT development and a prognostic predictor forHCC patients. Major vascular invasion is one of the most impor-tant factors contributing to the poor prognosis of HCC patients,and therapies for these patients still remain controversial. Resec-tionmay be the only therapeutic option that offers a possible curefor HCC patients with PVTT, though there is a high rate ofrecurrence after surgery (24). Among HCC patients who under-went hepatectomy, the PVTT status is correlated with the prog-nosis (24). Thus, factors associated with PVTT may aid in thedetermination of the appropriate surgical strategy. Our findingsthat ICR expression was correlated with PVTT development(Fig. 6) indicated the possible role of ICR in assessing the necessityof surgery, especially for liver transplantation. Patients with lowICR expression may be suitable for further surgical therapy. Ofnote, although ICR and ICAM-1 are located in overlapping posi-tions in chromosome 19, they are transcribed from differentstrands and only about 800 bp overlapping segment located atthe 50end of ICR and at 30end of ICAM-1. So ICR may be a singlegene and not antisense lncRNA.

Our study is the first to demonstrate the heterogeneousexpression of an lncRNA in tumor cells. Various lncRNAs havebeen correlated with diverse diseases, including cancer (25).Dysregulation of lncRNAs induces carcinogenesis and increasescancer invasiveness and metastasis (26–28). The effects ofaltered lncRNA expression on tumor cell biology have beenprimarily investigated in whole tumor cells and tissues (27).However, tumor cells are heterogeneous in tumor tissues, andadvancements in experimental technologies have revealed vary-ing heterogeneity in tumor cells, including cellular morphol-ogy, tumor histology, and genetic abnormalities (29). Verylittle is known regarding heterogeneous lncRNA expression. Inour study, we observed distinct ICR expression levels in ICAM-1þ and ICAM-1� tumor cells in Huh7 and Hep3B cells andclinical HCC tissues, further confirming that the different


Guo et al.



subpopulations may have distinct gene expression profiles andcell features in tumor tissues. Furthermore, ICR expression wascorrelated with aggressive tumor behavior and poor clinicaloutcomes, indicating the clinical significance of heterogeneousICR expression in HCC tissues. These findings suggest thatfurther attention should be paid to heterogeneous lncRNAexpression in tumor development.

In summary, we demonstrated that the lncRNA ICR,which is regulated by Nanog, played a pivotal role in CSCmaintenance and PVTT development by elevating ICAM-1expression. The findings presented here indicate the crucialroles of lncRNAs in CSC maintenance and PVTT develop-ment and provide insight into lncRNA function. In addi-tion, we demonstrated that ICR expression level was asso-ciated with HCC metastasis and HCC patient's prognosis,suggesting that ICR may be a new prognostic indicator forHCC and may provide a specific target for the treatment ofmetastatic HCC.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: W. Guo, S. Liu, M. Wu, S. Cheng, S. LiuDevelopment of methodology:W. Guo, Y. Cheng, J. Shi, N. Li, S. Cheng, S. LiuAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.):W. Guo, S. Liu, Y. Cheng, L. Lu, G. Xu, N. Li, S. Cheng

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): W. Guo, S. Liu, Y. Cheng, J. Shi, N. Li, K. Cheng,S. ChengWriting, review, and/or revision of the manuscript:W. Guo, S. Liu, Y. Cheng,S. Cheng, S. LiuAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): M. Wu, S. ChengStudy supervision: S. Cheng, S. Liu

Grant SupportThis work was supported by the China National Funds for Distinguished

Young Scientists (Nos. 81125018 and 81425019); Chang Jiang Scholars Pro-gram (2012) of China Ministry of Education; the grants of the Science Fund forCreative Research Groups (No. 81221061); the State Key Project on Diseases ofChina (2012zx10002016016003); the National Key Basic Research Program"973 project" (No. 2015CB554000); The China Natural Science FoundationProgram (Nos. 81301876 and 81302116); The New Excellent Young Talents ofShanghai Municipal Health Bureau (No. XQY 2013112); The ExcellentYoung Scholar of Second Military Medical University; Shanghai Science andTechnology Committee Program (Nos. 12ZR1439600 and 13ZR1409200);The New Excellent Talents Program of Shanghai Municipal Health Bureau(No. XBR2011025); Shanghai Science and Technology Committee (No.134119a0200); and SMMU Innovation Alliance for Liver Cancer Diagnosisand Treatment (2012).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

ReceivedDecember 3, 2014; revisedNovember 16, 2015; acceptedDecember9, 2015; published OnlineFirst December 14, 2015.

References1. McGlynn KA, Petrick JL, London WT. Global epidemiology of hepatocel-

lular carcinoma: an emphasis on demographic and regional variability.Clin Liver Dis 2015;19:223–38.

2. Chaffer CL, Weinberg RA. A perspective on cancer cell metastasis.Science 2011;331:1559–64.

3. Ryu SH, Jang MK, Kim WJ, Lee D, Chung YH. Metastatic tumor antigen inhepatocellular carcinoma: golden roads toward personalized medicine.Cancer Metastasis Rev 2014;33:965–80.

4. Shi J, Lai EC, Li N, Guo WX, Xue J, Lau WY, et al. Surgical treatment ofhepatocellular carcinoma with portal vein tumor thrombus. Ann SurgOncol 2010;17:2073–80.

5. Llovet JM, Bruix J. Novel advancements in the management of hepatocel-lular carcinoma in 2008. J Hepatol 2008;48:S20–37.

6. Zhang J, Pan YF, Ding ZW, Yang GZ, Tan YX, Yang C, et al. RMP promotesvenous metastases of hepatocellular carcinoma through promoting IL-6transcription. Oncogene 2015;34:1575–83.

7. Liu S, Guo W, Shi J, Li N, Yu X, Xue J, et al. MicroRNA-135a contributes tothe development of portal vein tumor thrombus by promoting metastasisin hepatocellular carcinoma. J Hepatol 2012;56:389–96.

8. Zhang J, Yang Y, Yang T, Yuan S, Wang R, Pan Z, et al. Double-negativefeedback loop between microRNA-422a and forkhead box (FOX)G1/Q1/E1 regulates hepatocellular carcinoma tumor growth and metastasis.Hepatology 2015;61:561–73.

9. Ponting CP, Oliver PL, ReikW. Evolution and functions of long noncodingRNAs. Cell 2009;136:629–41.

10. Yang L, Lin C, Jin C, Yang JC, Tanasa B, Li W, et al. lncRNA-dependentmechanisms of androgen-receptor-regulated gene activation programs.Nature 2013;500:598–602.

11. Kogo R, Shimamura T, Mimori K, Kawahara K, Imoto S, Sudo T, et al. Longnoncoding RNAHOTAIR regulates polycomb-dependent chromatinmod-ification and is associatedwith poor prognosis in colorectal cancers. CancerRes 2011;71:6320–6.

12. Yang F, Zhang L, Huo XS, Yuan JH, Xu D, Yuan SX, et al. Long noncodingRNA high expression in hepatocellular carcinoma facilitates tumor growththrough enhancer of zeste homolog 2 in humans. Hepatology 2011;54:1679–89.

13. Panzitt K, TschernatschMM, Guelly C,Moustafa T, StradnerM, StrohmaierHM, et al. Characterization of HULC, a novel gene with striking up-regulation in hepatocellular carcinoma, as noncoding RNA. Gastroenter-ology 2007;132:330–42.

14. Yuan SX, Yang F, Yang Y, Tao QF, Zhang J, Huang G, et al. Long noncodingRNA associated with microvascular invasion in hepatocellular carcinomapromotes angiogenesis and serves as a predictor for hepatocellular carci-noma patients' poor recurrence-free survival after hepatectomy. Hepatol-ogy 2012;56:2231–41.

15. Wang F, Yuan JH, Wang SB, Yang F, Yuan SX, Ye C, et al. Oncofetal longnoncoding RNA PVT1 promotes proliferation and stem cell-like propertyof hepatocellular carcinoma cells by stabilizing NOP2. Hepatology 2014;60:1278–90.

16. Visvader JE, Lindeman GJ. Cancer stem cells: current status and evolvingcomplexities. Cell Stem Cell 2012;10:717–28.

17. Lee TK, Castilho A, Cheung VC, Tang KH, Ma S, Ng IO. CD24(þ)liver tumor-initiating cells drive self-renewal and tumor initiationthrough STAT3-mediated NANOG regulation. Cell Stem Cell 2011;9:50–63.

18. Yamashita T, Ji J, Budhu A, Forgues M, Yang W, Wang HY, et al.EpCAM-positive hepatocellular carcinoma cells are tumor-initiatingcells with stem/progenitor cell features. Gastroenterology 2009;136:1012–24.

19. Wang T, Hu HS, Feng YX, Shi J, Li N, Guo WX, et al. Characterisation of anovel cell line (CSQT-2) with high metastatic activity derived from portalvein tumour thrombus of hepatocellular carcinoma. Br J Cancer 2010;102:1618–26.

20. Liu S, Li N, Yu X, Xiao X, Cheng K, Hu J, et al. Expression ofintercellular adhesion molecule 1 by hepatocellular carcinoma stemcells and circulating tumor cells. Gastroenterology 2013;144:1031–41e10.

21. McShane LM, AltmanDG, Sauerbrei W, Taube SE, GionM, Clark GM, et al.Reporting recommendations for tumor marker prognostic studies. J ClinOncol 2005;23:9067–72.

22. Faghihi MA, Modarresi F, Khalil AM, Wood DE, Sahagan BG, Morgan TE,et al. Expression of a noncoding RNA is elevated in Alzheimer's disease and





drives rapid feed-forward regulation of beta-secretase. Nat Med 2008;14:723–30.

23. Wang L, ParkHJ,Dasari S,WangS, Kocher JP, LiW.CPAT:Coding-PotentialAssessment Tool using an alignment-free logistic regressionmodel.NucleicAcids Res 2013;41:e74.

24. Peng ZW, Guo RP, Zhang YJ, Lin XJ, Chen MS, Lau WY. Hepatic resectionversus transcatheter arterial chemoembolization for the treatment ofhepatocellular carcinoma with portal vein tumor thrombus. Cancer 2012;118:4725–36.

25. Nagano T, Fraser P. No-nonsense functions for long noncoding RNAs. Cell2011;145:178–81.

26. Spizzo R, Almeida MI, Colombatti A, Calin GA. Long non-coding RNAsand cancer: a new frontier of translational research? Oncogene 2012;31:4577–87.

27. TingDT, LipsonD, Paul S, Brannigan BW, Akhavanfard S, Coffman EJ, et al.Aberrant overexpression of satellite repeats in pancreatic and other epi-thelial cancers. Science 2011;331:593–6.

28. Gupta RA, Shah N, Wang KC, Kim J, Horlings HM, Wong DJ, et al. Longnon-coding RNA HOTAIR reprograms chromatin state to promote cancermetastasis. Nature 2010;464:1071–6.

29. Kreso A, Dick JE. Evolution of the cancer stem cell model. Cell Stem Cell2014;14:275–91.


Guo et al.



icam-1 related noncoding rna in cancer stem cells ... · biology of human tumors icam-1–related...

Documents