hypoxia causes downregulation of dicer in hepatocellular … · biology of human tumors hypoxia...

Biology of Human Tumors

Hypoxia Causes Downregulation of Dicer inHepatocellular Carcinoma, Which Is Required forUpregulation of Hypoxia-Inducible Factor 1a andEpithelial–Mesenchymal TransitionAhmed Atef Ibrahim1,2, Christian Schmithals1, Erik Kowarz3, Verena K€oberle1,Bianca Kakoschky1, Thomas Pleli1, Otto Kollmar4, Scarlett Nitsch1,5, Oliver Waidmann1,Fabian Finkelmeier1, Stefan Zeuzem1, Horst-Werner Korf6, Tobias Schmid7,Andreas Weigert7, Bernd Kronenberger1, Rolf Marschalek3, and Albrecht Piiper1

Abstract

Purpose: A role of Dicer, which converts precursor miRNAs tomature miRNAs, in the tumor-promoting effect of hypoxia iscurrently emerging in some tumor entities. Its role in hepatocel-lular carcinoma (HCC) is unknown.

Experimental Design: HepG2 and Huh-7 cells were stablytransfected with an inducible Dicer expression vector and wereexposed to hypoxia/normoxia. HepG2-Dicer xenografts wereestablished in nude mice; hypoxic areas and Dicer were detectedin HCC xenografts and HCCs from mice with endogenous hepa-tocarcinogenesis; and epithelial–mesenchymal transition (EMT)markers were analyzed by immunohistochemistry or by immu-noblotting. The correlation between Dicer and carbonic anhy-drase 9 (CA9), a marker of hypoxia, was investigated in resectedhuman HCCs.

Results: Hypoxia increased EMT markers in vitro and in vivoand led to a downregulation of Dicer in HCC cells. The levels of

Dicer were downregulated in hypoxic tumor regions in micewith endogenous hepatocarcinogenesis and in HepG2 xeno-grafts. In human HCCs, the levels of Dicer correlated inverselywith those of CA9, indicating that the negative regulation ofDicer by hypoxia also applies to HCC patients. Forced expres-sion of Dicer prevented the hypoxia-induced increase in hyp-oxia-inducible factor 1a (HIF1a), HIF2a, hypoxia-induciblegenes (CA9, glucose transporter 1), EMT markers, and cellmigration.

Conclusions: We here identify downmodulation of Diceras novel essential process in hypoxia-induced EMT in HCCand demonstrate that induced expression of Dicer counter-acted hypoxia-induced EMT. Thus, targeting hypoxia-induceddownmodulation of Dicer is a promising novel strategyto reduce HCC progression. Clin Cancer Res; 23(14); 3896–905.�2017 AACR.

IntroductionCancer mortality is largely attributable to resistance to therapy,

invasion, and metastasis (1). The underlying mechanisms arecomplex. Recent studies indicate that these phenomena involve

the acquisition of stem cell and metastatic properties by a molec-ular program encompassing the transition from an epithelial to amesenchymal phenotype (EMT; ref. 2). Hypoxia, a commonmicroenvironmental component in solid tumors due to theabnormal tumor vascularity, is a major driver of EMT, metastasis,and poor prognosis (3, 4). Thus, hypoxic regions in tumors areparticularly important for EMT, therapy resistance, and the prog-nosis of the patients. The main sensors of oxygen mediatingcellular adaptation to hypoxia are a subunits of hypoxia-induc-ible factors 1 and 2 (HIF1a, HIF2a), which become stabilized inthe absence of oxygen and then heterodimerize with HIF1bsubunits to induce hypoxia-responsive genes, including manyprometastatic and EMT-driving genes, extracellular matrix–mod-ifying enzymes, and secretory factors that facilitate the perme-ability of blood vessels and the growth in secondary organs (4–6).

Deregulation of miRNAs, evolutionarily conserved small RNAmolecules intricately involved in gene regulation, also plays animportant role in EMT,metastasis, and resistance to therapy (7, 8).miRNAs are formed from long primary miRNAs (pri-miRNAs),which become converted to precursor miRNAs (pre-miRNAs) bythe Drosha complex and further to mature miRNAs by Dicer (9).Individual miRNAs and miRNA families regulate multiple hall-marks of cancer such as cell proliferation, apoptosis, metastasis,

1Department of Medicine 1, University Hospital Frankfurt, Frankfurt, Germany.2The Immunology and Infectious Diseases Laboratory, Therapeutic ChemistryDepartment, The National Research Center, Dokki, Cairo, Egypt. 3Institute ofPharmaceutical Biology, Goethe-University of Frankfurt Biocenter, Frankfurt/Main, Germany. 4Department of General and Visceral Surgery, HELIOS Dr. HorstSchmidt-Kliniken, Wiesbaden, Germany. 5Department of Biology, TechnicalUniversity of Darmstadt, Darmstadt, Germany. 6Institute of Anatomy 2, Univer-sity Hospital Frankfurt, Frankfurt, Germany. 7Institute of Biochemistry I, Goethe-University Frankfurt, Frankfurt am Main, Germany.

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

C. Schmithals, Erik Kowarz, and V. K€oberle contributed equally to this article.

Corresponding Author:Albrecht Piiper, University Hospital Frankfurt, Theodor-Stern-Kai 7, D-60590 Frankfurt, Germany. Phone: 49-69-6301-87667; Fax: 49-69-6301-87689; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-16-1762

�2017 American Association for Cancer Research.

ClinicalCancerResearch

Clin Cancer Res; 23(14) July 15, 20173896

on October 18, 2020. © 2017 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst February 6, 2017; DOI: 10.1158/1078-0432.CCR-16-1762

http://crossmark.crossref.org/dialog/?doi=10.1158/1078-0432.CCR-16-1762&domain=pdf&date_stamp=2017-6-23

http://clincancerres.aacrjournals.org/

and angiogenesis, serving as oncogenes or tumor-suppressorgenes (10, 11). Many carcinomas show not only deregulation ofoncogenic and tumor-suppressivemiRNAs, but also global down-regulation of miRNAs (12, 13), which is associated with moreaggressive, invasive disease, distant recurrence, and poor overallsurvival (12, 14, 15). The procarcinogenic effect of reducedmiRNA biogenesis is underlined by the finding that members ofthe miRNA biogenesis pathway are haplo-insufficient tumorsuppressors (16–20). Thus,micewith conditionalDicer knockoutin hepatocytes spontaneously develop HCC, most likely in cellswith residual Dicer activity (21). However, up- and downregula-tion of Drosha and Dicer have been found in different tumorentities (22), indicating that reduced levels of the miRNA bio-genesis enzymes can explain reduced miRNA biogenesis only insome tumors.

Tumor cell–intrinsicmechanisms such asmutations inDicer aswell as the tumor microenvironment have been reported tocontribute to the downregulation ofmiRNA biogenesis in tumors(22–24). Recently, it has been found that hypoxia causes down-regulation of Dicer and Drosha and a global reduction of miRNAlevels in breast and ovarian cancer, thereby promoting EMT andtumor progression (25–27), indicating a link between the EMT-promoting effect of hypoxia and the deregulation of miRNAbiogenesis in these tumor entities. However, due to the manifoldand tumor-specific influence of hypoxia and the different mem-bers of the miRNA biogenesis pathway and vice versa, the cross-talk between miRNA biogenesis and hypoxic signaling is stillrelatively unexplored, and the targets and methods of interven-tion in a potential therapeutic setting remain to be established.

HCCs, ranking third among cancer-related mortality world-wide, contain extensive hypoxic regions in addition to well-perfused areas and deregulated and globally downregulatedmiRNAs andDicer (12, 28–30).Whether tumor hypoxia regulatesDicer and whether Dicer plays a role in hypoxic response in HCCare currently unknown. The data of the present study clearlydemonstrate that hypoxia is associated with downregulation ofDicer in humanHCCs,which always correlatedwith upregulationof HIF1a, EMT, and cell migration. By genetic complementationof Dicer, all hypoxia-driven effects were alleviated, indicating thatDicer expression acts dominant over hypoxia. Thus, preventing

downmodulation of Dicer under hypoxic conditions is a prom-ising strategy to impair tumor progression in HCC.

Materials and MethodsCell culture conditions and stable transfection of Dicer

HepG2 andHuh-7 cells were obtained from the American TypeCulture Collection (ATCC) and RIKEN BioResource Center,respectively (31). Both cell lines were authenticated in 2016[Deutsche Sammlung von Mikroorganismen und Zellkulturen(DSMZ)]. The cells were grown in DMEM (Invitrogen) supple-mented with 10% FBS (Invitrogen) and 1% penicillin/strepto-mycin (Invitrogen) and maintained at 37�C in a humidifiedincubator with 5% CO2.

For hypoxia treatments, the cells were incubated in a hypoxicworkstation with 1% O2, 94% N2, and 5% CO2 (Invivo2 400;Ruskinn Technology). Prior to hypoxia treatment, the cells wereseeded in normoxia and grown to a confluence of 60% to 70%.Used medium was replaced by fresh medium 2 hours beforeexposure of the cells to hypoxia for the indicated durations.

For overexpression of Dicer, we obtained human Dicer-con-taining plasmids as a gift from W. Filipowicz (FMI, Switzerland).The Dicer fragment was cloned into a tetracycline-inducibleSleeping Beauty transposon system as described previously(32). HepG2 and Huh-7 cells were transfected with the recom-binant Dicer plasmids using Lipofectamine reagent (ThermoScientific). Stable transfectants were selected with 2 mg/mL ofpuromycin (Sigma) andwere tested for induction of Dicer expres-sion upon treatment of the cells with 1 mg/mL of doxycycline(Sigma).

In vivo mouse models of HCCAll animal experiments were approved by the local animal care

committee and were in agreement with the German legal require-ments. Male TGFa/c-myc bi-transgenic mice were generated bycrossing homozygous metallothionein/TGFa and albumin/c-myc single transgenic mice in CD13B6CBA background (33,34). Hepatocarcinogenesis was induced by ZnCl2 via the drinkingwater. HCCs were detected by contrast-enhanced MRI (31).

To generate nudemice bearingHepG2 xenografts, NMRI Foxn1nude mice (Harlan Laboratories B.V.) were used as describedrecently (31). Five- to 6-week-old animals were anesthetized, and5� 106 cells in 100 mL PBS were injected subcutaneously into thelateral flanks of the mice. The animals were examined for tumordevelopment twice per week. When tumors reached 200 mm3,mice were fed a doxycycline-containing diet (1 g/kg, Ssniff for 7days or left untreated fed a control diet (Ssniff).

For immunohistochemical staining of hypoxic regions in thetwo HCC mouse models, mice were intraperitoneally injectedwith pimonidazole (60 mg/kg, Hypoxyprobe-1, Hypoxyprobe),which forms protein adducts under hypoxia that can be detectedimmunochemically. Ninety minutes later, the mice were sacri-ficed and the liver containing the tumors or the tumor xenograftswere excised and were either directly frozen in optimal cuttingtemperature (Tissue-Tek) media to prepare frozen sections, orsnap-frozen for lysate preparation for protein or RNA analysis.

Human tumor samplesHCC specimens were collected from 24 patients who under-

went anatomic liver resection (segmentectomy or subsegmen-tectomy; Supplementary Table S1). Surgical HCC specimens

Translational Relevance

Tumor hypoxia and deregulated miRNA biogenesis aremajor mechanisms to increase metastasis and resistance totherapy of carcinomas by promoting epithelial–mesenchymaltransition (EMT) and thereby govern tumor aggressiveness.Weshow here for the first time that the hypoxic tumor microen-vironment causes downregulation of Dicer in hepatocellularcarcinoma (HCC). This is required for hypoxia-induced upre-gulation of the major cellular oxygen sensors, hypoxia-induc-ible factor 1a (HIF1a) and HIF2a, and their EMT-promotingeffect in HCC cells, in HCC mouse models, and also in HCCsfrom patients. Importantly, we provide first evidence that aDicer rescue prevents the hypoxia-induced upregulation ofHIF1a, HIF2a, EMT, and cell migration. Thus, targeting hyp-oxia-induced downmodulation of Dicer is a promising novelstrategy to restrict selectively progression and metastasis ofHCC.

Dicer in Hypoxia-Induced EMT in HCC

www.aacrjournals.org Clin Cancer Res; 23(14) July 15, 2017 3897




and corresponding normal liver tissue were collected frompatients who underwent surgical resection between 2003 and2007. All patients provided informed consent for tissue pro-curement, which was approved by the local ethics committee ofthe €Arztekammer des Saarlandes (154/02). Frozen tumor sam-ples (approximately 0.2 mg each) were used for total RNA andprotein extraction.

Presentation and statistical analysis of the dataData are expressed as mean � SD or SEM as indicated. Com-

parisons were performed with the Mann–Whitney U (Kruskal–Wallis) or Student t test. P < 0.05 was considered significant. ThePearson correlation test was used to test the correlation betweenDicer and CA9 mRNA levels. Data were analyzed using the BiASsoftware for Windows (version 9.11; Epsilon-Verlag).

Analytic methods are described in the Supplement.

ResultsHypoxia downregulates Dicer in HCC

It is currently unknown whether Dicer plays a role in hypoxicresponse in HCC and whether tumor hypoxia plays a role in theobserved global depression of miRNAs. To investigate whether

tumor hypoxia influences the expression of Dicer in HCC, westudied the levels of Dicer and of carbonic anhydrase 9 (CA9),which is reliably upregulated under hypoxic conditions and cantherefore serve as marker of hypoxia, in a set of surgically resectedhumanHCCs. The levels ofDicer andCA9 correlated negatively atthe mRNA (Fig. 1A) and protein (Fig. 1B) level in the HCCs frompatients, suggesting a negative relation between hypoxia andDicer expression in human HCCs.

To investigate whether tumor hypoxia influences the levels ofDicer in HCC, we compared the levels of Dicer in HCC cell linesunder normoxic and hypoxic conditions. As illustrated in Fig. 1Cand D, the Dicer levels were lower in HepG2 as well as in Huh-7cells cultured under hypoxic conditions for 24 hours than those inthe normoxic cells.

Hypoxia may induce apoptosis and thereby cause nonspecificdownregulation of Dicer. However, inHepG2 cells, we found thathypoxia treatment for 24 hours that repressed Dicer did notinduce apoptosis as revealed by the detection of cleaved cas-pase-3 fragments, whereas the Plk1 inhibitor BI 2536 increasedcaspase-3 cleavage (Supplementary Fig. S1). Similar data wereobtained in Huh-7 cells. Examination of the time course ofhypoxia-induced downmodulation of Dicer in Huh-7 cellsshowed significantly reduced Dicer protein levels already after

Figure 1.

Dicer correlates negativelywith hypoxia in HCCs from patients and is downregulated under hypoxic conditions in HepG2 and Huh-7 cells.A,RNAwas extracted from24 surgical human HCC samples, and the mRNA levels of Dicer and CA9 were determined by quantitative RT-PCR. The Ct values were related to those of b-actin.Pearson correlation between the relative Dicer and CA9 mRNA levels was determined using BiAS software (r ¼ –0.41, P ¼ 0.048). B, Protein was extractedfrom the human HCC samples, and equal amounts of protein were analyzed for Dicer, CA9, and b-actin expression by immunoblotting. The band intensities weredetermined by densitometry. Mean � SE of the relative Dicer protein levels (normalized by b-actin) in CA9-positive (n ¼ 7) and CA9-negative HCCs (n ¼ 3) aredisplayed. The asterisk indicates a significant difference (� , P < 0.05). C and D, HepG2 (C) or Huh-7 (D) cells were cultured under normoxic (N) or hypoxic (H) (1%oxygen) conditions for 24 hours. Subsequently, the cell lysates were analyzed for Dicer and b-actin by immunoblotting. The numbers below each blot represent theratio of Dicer/b-actin relative to the normoxic condition. The blots representing three independent experiments each were analyzed densitometrically. Thedensity ratio of Dicer/b-actin obtained in lysates fromnormoxic cellswas set to 1. Thedata representmean�SE.E andF, Time course of the effect of hypoxia onDicerprotein (E) and mRNA (F) levels in Huh-7 (E) and HepG2 cells (F) under hypoxic conditions at the indicated time points. Data are normalized to the correspondingnormoxic condition. C–F, The asterisks represent significant differences as compared with the normoxic condition (� , P < 0.05).

Ibrahim et al.

Clin Cancer Res; 23(14) July 15, 2017 Clinical Cancer Research3898




4 hours of hypoxia treatment, and the Dicer levels remainedlowered during the entire period of observation of 48 hours (Fig.1E). Moreover, we found that hypoxia suppressed Dicer-mRNAlevels in HepG2 cells (Fig. 1F). In Huh-7 cells, we observed asmaller downregulation of Dicer-mRNA of 20% after 24 hours.Altogether, these data support the idea that hypoxia downmo-dulates Dicer levels in HCC.

Next, we examined the relation between Dicer and tumorhypoxia in endogenously formed HCCs in TGFa/c-myc mice.The HCCs in the animals were identified by gadolinium ethox-ybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)–enhanced MRI (31, 33). Similar to HCCs of patients (29, 30), thelevels of Dicer were downregulated in these HCCs as comparedwith the surrounding nontumorous liver tissues (Fig. 2A).

An altered ratio of precursor to mature miRNA is characteristicfor altered Dicer levels (25, 27). To investigate whether Dicerdownregulation is relevant for miRNA maturation in HCC, westudied the levels of precursor and mature miR-122, whichaccounts for approximately 70% of all miRNAs in the liver andis a major suppressor of hepatocarcinogenesis (35, 36), in theHCC and the surrounding liver tissue in TGFa/c-myc mice. Asillustrated in Fig. 2B, the amount of the mature miR-122 wasdecreased, whereas the level of precursormiR-122 and the ratio ofprecursor to mature miR-122 were increased in the HCCs ascompared with the surrounding normal liver tissue, indicatingthat the reduced Dicer levels in the HCCs indeed led to thecorresponding changes in the levels of this major hepatictumor-suppressive miRNA.

Figure 2.

Reduced levels of Dicer and miR-122 in HCC as compared with the surrounding nontumorous liver tissue in TGFa/c-mycmice and antagonistic localization betweenDicer and hypoxia in HCCs from TGFa/c-myc mice and HepG2-bearing nude mice. A, Pairs of tissue samples from HCC (T) and surrounding normal livertissue (L) were obtained from TGFa/c-myc mice and were analyzed for the expression of Dicer and b-actin by immunoblotting. Left, a representative blotwith one pair of liver and tumor tissue. The numbers below the blot represent the ratio of Dicer/b-actin in the tumor (T) relative to the liver (L) sample. Right,densitometric analyses of the Dicer and b-actin bands reveal a significant difference (n ¼ 6; � , P < 0.05) between the Dicer content of liver and tumor. B, Levelsof pre-miR-122 andmaturemiR-122 and their ratio in HCC (T) and normal liver (L) tissue fromTGFa/c-mycmice as revealed by quantitative RT-PCR. The levels in livertissue were set to 1. The asterisks represent a significant difference as compared with the normoxic condition (� , P < 0.05). C and D, TGFa/c-myc mice withradiologically verified HCC (C) and nude mice bearing HepG2 xenografts (D) were injected with pimonidazole (i. p.). Ninety minutes later, the animals weresacrificed and the HCCs were excised from the livers. Cryosections were stained with anti-Dicer (red) and anti-pimonidazole (green) and Hoechst (not shown) andwere imaged by confocal laser scanning immunofluorescence microscopy (scale bars, 30 mm). E, Quantification of the levels of Dicer in areas of high and lowpimonidazole staining in xenografted HepG2 tumors. A threshold for the pimonidazole fluorescence signal (green) was set above the background to differentiatepimonidazole-positive (hypoxic) from pimonidazole-negative (normoxic) areas. Using Image J software, the mean Dicer fluorescence intensities in 5 to 7different pimonidazole-negative areas and in 5 to 7 different pimonidazole-positive areas within each tumor section (each tumor section being from a differentnude mouse) were measured and averaged to obtain representative Dicer fluorescence intensity per normoxic and hypoxic area. Data are presented as meanDicer fluorescence intensity � SE of 7 mice. The asterisk represents a significant difference between the two groups (� , P < 0.05).






To study the local distribution of hypoxia and Dicer in theseHCCs, TGFa/c-mycmice with HCC according to Gd-EOB-DTPA–enhanced MRI were injected intraperitoneally with pimonida-zole, which forms protein adducts at oxygen concentrationsbelow 2%. Ninety minutes later, the mice were sacrificed and thetumors were excised. Slices of the HCCs were stained with anti-Dicer and anti-pimonidazole and analyzed by confocal fluores-cence microscopy (CLSM). As shown in Fig. 2C, hypoxic tumorregions showed reduced Dicer levels. Similar results wereobtained in nude mice with HepG2 xenografts. Again, the local-ization ofDicer correlated negativelywith that of hypoxia (Fig. 2Dand E). Together, these data revealed an inverse correlationbetweenDicer and hypoxia in the tumors in different HCCmousemodels.

Tumor hypoxia leads to an increase in EMT markers in HCCInternalization of E-cadherin from the cell surface and

downregulation of its expression are hallmarks of EMT, whichis known to be enhanced by tumor hypoxia (37, 38). Accord-ingly, exposure of HepG2 cells to hypoxia for 72 hours led to aloss of E-cadherin from the cell surface and the appearance ofE-cadherin spots within the cells (Fig. 3A). Moreover, E-cad-herin was downregulated in hypoxic cultured cells (Fig. 3B).Concomitantly, hypoxia induced the expression of the mes-enchymal markers vimentin, N-cadherin, Snail1 (Fig. 3B andC), and of b-catenin (Fig. 3C). To investigate the local distri-bution of E-cadherin and hypoxia in HCC in vivo, nude micewith HepG2 xenografts were injected with pimonidazole,followed by costaining of pimonidazole and E-cadherin intumor slices. As illustrated in Fig. 3D, hypoxic regions in thetumors showed lower levels of E-cadherin as compared with

normoxic parts. Similarly, staining of tumor slices with CA9,an endogenous marker of hypoxia, and E-cadherin revealedantagonistic distribution of E-cadherin and CA9 immunoreac-tivity (Supplementary Fig. S2). These data indicate that withinHCCs, hypoxic regions also correlate negatively with E-cad-herin expression.

Forced expression of Dicer suppresses hypoxia-inducedupregulation of HIFa and EMT markers in HCC

To investigate the role of Dicer in hypoxic response in HCC,we generated HepG2 and Huh-7 cells stably expressing adoxycycline-inducible sleeping beauty vector encoding forDicer (Supplementary Fig. S3A and S3B). These cells showedlong-term expression of the Dicer upon their incubation withdoxycycline (Supplementary Fig. S3C), and this effect was alsoobserved in HCC xenografts from doxycycline-treated mice(Supplementary Fig. S3D).

Hypoxia-induced suppression of Dicer does not depend onHIF (25, 27). To investigate whether hypoxia-induced down-regulation of Dicer affects HIF1a and HIF2a, we examinedthe effect of Dicer induction on HIFa levels in HepG2 andHuh-7 cells under normoxic and hypoxic conditions. Theinduction of Dicer strongly inhibited the hypoxia-inducedupregulation of HIF1a and HIF2a levels in both Huh-7 (Fig.4A) and HepG2 cells (Fig. 4B), indicating that hypoxia-induced upregulation of HIF1a and HIF2a is suppressed byDicer.

Next, we studied the effect of Dicer induction on the levelsof other HIF-dependent genes under hypoxic and normoxicconditions in Huh-7 cells. Hypoxia increased the levels of CA9(Fig 4C) and glucose transporter 1 (GLUT1, Fig. 4D), whereas the

Figure 3.

Tumor hypoxia increases EMTmarkersin HCC in vitro and in vivo. HepG2cells were cultured under normoxic(N) or hypoxic (H) conditions for72 hours. A, The cells were stainedfor E-cadherin (red) byimmunocytochemistry and Hoechst(blue) and were imaged by CLSM(scale bars, 10 mm).B andC, Lysates ofthe cellswere analyzed for E-cadherin,vimentin, N-cadherin, Snail1,b-catenin, and b-actin byimmunoblotting. The numbers beloweachblot indicate the relative intensityof each protein band normalized byb-actin. D, Nude mice bearing HepG2xenografts were i.p. injected withpimonidazole. Ninety minutes later,the animals were sacrificed and thetumors were removed. Cryosectionsof the tumors were stained withanti-pimonidazole (green) andanti-E-cadherin (red) and wereimaged by CLSM (scale bar, 30 mm).

Ibrahim et al.





induction of Dicer expression suppressed their upregulationunder hypoxia.

A number of miRNAs can regulate HIF1a. We examined thelevels of some of these miRNAs in hypoxic and normoxic HepG2

cells with andwithout induction of Dicer. As illustrated in Fig. 4E,miR-18a, an miRNA that has been shown to regulate HIF1a inHCC cells (39), revealed significant downregulation by hypoxia,whereas hypoxia had no significant effect onmiR-18a levels upon

Figure 4.

Induction of Dicer expressionsuppresses hypoxia-induced increasein HIF1a, HIF2a, HIF-dependent genes,and miR-18a in HCC cells. A and B,Stable Dicer-transgenic Huh-7 (A) orHepG2 cells (B) were treated withdoxycycline (Dox) or vehicle for 24hours and were then cultured undernormoxic (N) or hypoxic (H)conditions for 24 hours. Lysates of thecells were immunoblotted andanalyzed for Dicer, HIF1a, HIF2a, andb-actin as indicated. C and D,Doxycycline-induced and noninducedDicer-transgenic Huh-7 cells werecultured under normoxic (N) orhypoxic (H) (1%oxygen) conditions for1 to 3 days. Subsequently, the celllysateswere analyzed for CA9 (C, after1 day) and GLUT1 (D, after 3 days)expression by immunoblotting usingb-actin as loading control. Thenumbers below each blot indicate therelative intensity of each protein bandnormalized by b-actin. E, HepG2 cellswith induced overexpression (OE) ofDicer or without induction of Dicerwere exposed to normoxia or hypoxiafor 24 hours. Thereafter, the levels ofmiR-18a were determined by qRT-PCR. Data are mean � SE normalizedto the normoxic condition (�,P <0.05).






induction of Dicer expression. Thus, miR-18a is a candidate tomediate the effects of hypoxia-induced downregulation of Diceron HIF1a.

Examination of the expression and distribution of E-cadherinwith andwithout induction ofDicer expression revealed amarkedreduction in hypoxia-induced internalization (Fig. 5A) anddownmodulation of E-cadherin (Fig. 5B) upon induction ofDicer

expression in HepG2 cells. The induction of Dicer expression alsoreduced the hypoxia-induced upregulation of b-catenin, Snail1(Fig. 5C), and N-cadherin (Fig. 5D) in HepG2 cells. As upregula-tion of vimentin was stronger in Huh-7 cells as compared withHepG2 cells, we examined the vimentin levels in normoxic andhypoxic Huh-7 cells with and without induction of Dicer expres-sion. As shown in Fig. 5E, Dicer suppressed hypoxia-induced

Figure 5.

Induction of Dicer prevents hypoxia-induced EMT markers in HCC cells andin HepG2 xenografts. A and B, StableDicer-transgenic HepG2 cells weretreated with doxycycline (Dox) toinduce overexpression of Dicer (OEDicer) or vehicle for 24 hours andwerethen cultured under normoxic (N) orhypoxic (H) (1%oxygen) conditions for72 hours. Subsequently, the cells weresubjected to immunocytochemicalstaining of E-cadherin (red) andimaged by CLSM (scale bars, 10 mm;A), or cell lysates were analyzed byimmunoblotting with anti–E-cadherinand anti–b-actin (B). C and D,Doxycycline-treated or vehicle-treated Dicer-transgenic HepG2 cellswere cultured under normoxic (N) orhypoxic (H) (1%oxygen) conditions for1 to 3 days. Expression of b-catenin (C,after 1 day), Snail1 (C, 1 after 1 day), andN-cadherin (D, after 3 days) wasdetermined by immunoblotting.b-Actin was used as endogenousreference. E, Huh-7 cells were culturedunder normoxic (N) or hypoxic (H) (1%oxygen) conditions for 72 hours.Subsequently, the cell lysates wereanalyzed for vimentin and b-actin byimmunoblotting. F, Nude micexenografted with Dicer-transgenicHepG2 tumors were treated withdoxycycline through the diet (1 g/kg)for 7 days to induce overexpression ofDicer (OE Dicer) or left untreated fed acontrol diet followed by the i. p.injection of pimonidazole. Ninetyminutes later, the tumors wereharvested and frozen sections of thetumorswere double-stainedwith anti-pimonidazole (red) and anti-E-cadherin (green) and were imaged byCLSM (scale bars, 30 mm).

Ibrahim et al.





increase in vimentin in these cells. Thus, Dicer upregulationsuppressed hypoxia-induced induction of EMT in HCC cells.

To investigate whether Dicer suppresses tumor hypoxia-induced EMT in vivo, nudemice xenograftedwithDicer-transgenicHepG2 tumors were treated with doxycycline through the diet for7 days orwere fedwith a control diet. Costaining of pimonidazoleand E-cadherin in tumor slices from such mice revealed thathypoxic regions in control HepG2 xenografts without inducedDicer expression were associated with loss of E-cadherin expres-sion or with E-cadherin internalization and loss at the cell–cellcontacts, comparedwithnormoxic areas (Fig. 5F). Interestingly, E-cadherin expression or membranous localization was preservedin hypoxic regions in HepG2 xenografts with induced Dicerexpression (Fig. 5F). These data indicate that Dicer overexpressionprevents hypoxia-induced EMT in HepG2 xenografts in vivo andfurther support the notion that Dicer acts as a dominant suppres-sor for EMT induced by hypoxia in the tumor microenvironment.

As EMT plays a key role in metastasis and Dicer counteractedhypoxia-induced EMT, we next studied the effect of Dicer upre-gulation on cell migration. In agreement with an inhibitory effectof Dicer on hypoxia-induced EMT, Dicer upregulation inhibitedcell migration under hypoxic conditions (Fig. 6).

DiscussionThe present study demonstrates that hypoxia-induced down-

regulationofDicer serves as keymechanism in themaintenance ofthe hypoxic response in HCC and that prevention of hypoxicsuppression of Dicer not only alleviates hypoxia-induced upre-gulation of HIF1a and HIF2a and other key hypoxia-responsive/HIF target genes, but also inhibits hypoxia-induced metastaticphenotypes such as EMT and increased cell motility. Thus, Dicersuppression may underlie part of the known association ofhypoxia with metastasis and poor outcome in HCC patients, andpreventing downmodulation of Dicer in hypoxic tumor regionsmay impair HIFa-dependent EMT, triggering metastasis andresistance to therapy.

Specifically, we found that tumor hypoxia and Dicer inverselylocalize in HCCs of mice with endogenous hepatocarcinogenesisor in HepG2 xenografts. Moreover, there was a negative correla-tion between the expression of Dicer and CA9, a marker ofhypoxia, in HCCs from patients. In vitro experiments showedthat chronic hypoxia decreased Dicer levels in HepG2 and Huh-7cells. These data indicate that hypoxia in the tumor microenvi-ronment is a key contributor toDicer downregulation inHCCandthat this is relevant for patients. Previous studies revealedhypoxia-

induced downregulation of Dicer in endothelial and pulmonaryartery smooth muscle cells (25) as well as in multiple organs ofmice subjected to hypoxic conditions. Very recent studies showedrepressionofDicer andDrosha andmiRNAbiogenesis byhypoxiain breast and ovarian cancers (26, 27). The two research groupsalso showed that low Dicer levels correlate with markers ofhypoxia in clinical samples from breast and ovarian cancer.Although monoallelic loss of Dicer and individual Dicer regula-tors, including Tap63, miR-103/107, miR-630, and VHL (16, 17,20, 25, 40), have been implicated in suppressing Dicer andmiRNA biogenesis in some cancer entities (lung, breast, andovarian cancers), Dicer and other members of the miRNA bio-genesis pathway are not downregulated inmany cancers (12, 22),rendering it unclear how the global miRNA downregulation isaccomplished in these cancer entities. A recent study indicates thatphosphorylation of exportin-5 by extracellular signal–regulatedkinase inhibits nuclear export of pre-miRNA, which could con-tribute to reduced miRNA biogenesis in tumors (41). Neverthe-less, the negative correlation between tumor hypoxia and Dicerlevels we and others found in different types of cancer indicatesthat tumor hypoxia plays a major role in downregulating Dicerand miRNA biogenesis in cancer.

Our data indicate that the mechanism of hypoxia-inducedsuppression of Dicer at least partially involves downregulationof Dicer at themRNA level. However, Dicer is also regulated at thetranslational and protein level (25). Although it is unknownwhether these events are regulated by hypoxia, it is evident thatthe different mechanisms that contribute to the regulation ofDicer complicate the elucidation of the mechanisms involved inDicer downmodulation under hypoxic conditions.

HIF1a and HIF2a play key roles in the hypoxic induction ofEMT and metastasis in HCC and other cancer entities (5, 37, 42,43). HIF-1 has been found to induce EMT by upregulation ofSnail1 in HCC (37). In the present study, rescue of Dicer expres-sion in hypoxic HCC cells disrupted the hypoxia-induced EMTphenotype, consisting of an increase in HIF1a, HIF2a, and themesenchymal markers N-cadherin, vimentin, Snail1, and b-cate-nin and decrease in the expression of the epithelial marker E-cadherin. Moreover, Dicer upregulation inhibited the hypoxia-induced increase in cell motility. Interestingly, rescuing Dicerexpression prevented the EMT phenotype, including decrease orloss of E-cadherin at the cell–cell contacts, in hypoxic tumorregions in HCC-xenografted nude mice. Together, these dataindicate that preventing Dicer downregulation in hypoxic tumorareas is a valid strategy to reduce HIF signaling and the EMTphenotype in HCC. Remarkably, in breast cancer, the relation

Figure 6.

Dicer upregulation decreased cellmotility under hypoxic conditions.Overexpression of Dicer (OE Dicer)was induced in Dicer-transgenicHepG2 cells with doxycycline or not.The confluent cells were wounded andcultured under hypoxic or normoxicconditions for 72 hours. Then, thewound areas were assessed. Woundclosure areawas calculated by (woundarea of 0 hour – wound area of72 hours), which was then normalizedto normoxia. Data are presented asmean � SE of n ¼ 3; � , P < 0.05.






between HIF and Dicer appears to differ from HCC, as hypoxicdownregulation of Dicer does not affect HIFa levels, althoughdownregulation of Dicer promoted EMT (27). Thus, the hypoxia-induced Dicer downregulation regulates HIF and the HIF-depen-dent genes in a cell/tumor type–dependent manner.

Interestingly, we found that hypoxic suppression of Dicer isalso required for maintaining hypoxia-induced upregulation ofb-catenin in HCC cells, which was found recently to contribute tohypoxia-induced EMT and metastasis in an orthotopic HCCmouse model independently of HIF-1 (44). Thus, Dicer may alsosuppress HIF-independent hypoxic responses.

miR-122, accounting formost of themiRNAspresent in normalliver and downregulated in HCC (45, 46), might be an importanteffector of Dicer downregulation in hypoxic regions. Knockout ofthis miRNA in hepatocytes triggers hepatocarcinogenesis similarto the loss of Dicer (21, 35, 36). miR-122 has been shown tosuppress HCC cell proliferation, HIF1a, EMT, and intrahepaticmetastasis (47, 48). We here provide evidence that reduced Dicerlevels in HCC led to reduced miR-122 concentrations and aconcomitant increase of the ratio of pre-miR-122 to maturemiR-122. Recent data indicate that miR-122 also directly targetsHIF1a, vimentin (47), Wnt1, and thereby the Wnt/b-cateninpathway (49). Thus, it is feasible to assume that the effects ofhypoxia-induced downmodulation of Dicer on hepatocarcino-genesis, EMT, and tumor progression could bemediated at least inpart by reducing miR-122 levels. We also found that miR-18a, anmiRNA implicated in the regulation of HIF1a (39), was down-regulated byhypoxia, and this effect was partially rescued byDiceroverexpression. Because HIF1a upregulates the expression ofSnail1 (37), hypoxic suppression of Dicer may modulate Snail1and EMT by downregulation of miR-18a in HCC. Nevertheless, itis possible that other miRNAs, such as miR-200 (27), are alsoinvolved in EMT induced by Dicer downmodulation in HCC.

Hypoxic downregulation of Dicer is HIF-independent asreported in different normal and cancer cells (25, 27, 50), whereasDrosha suppression in hypoxic cells appears to be highly depen-dent on HIF1a (28). This suggests that rescuing Dicer expressionin hypoxic cells may lead to increased Drosha expression in suchcells by inhibiting HIF1a. This indicates the important role ofDicer not only in HCC, but also in other cancer types showingHIF-dependent hypoxic suppression of other miRNA-processingproteins.

Altogether, our study is in line with data in other tumor entitiesshowing that the HIF-independent downmodulation of Dicer is a

novel mechanism by which tumor hypoxia can affect the malig-nant traits of tumors and thus provides a clear mechanistic linkbetween hypoxia-Dicer/miRNA biogenesis and tumor aggres-siveness in HCC. Future studies aiming to identify how thedownregulation of Dicer is executed upon hypoxia will help totranslate these findings into new translational approaches for thetreatment of solid cancers.

Disclosure of Potential Conflicts of InterestO. Waidmann is a consultant/advisory board member for Bayer. No poten-

tial conflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: A.A. Ibrahim, V. K€oberle, O. Waidmann, S. Zeuzem,A. Weigert, B. Kronenberger, R. Marschalek, A. PiiperDevelopment of methodology: A.A. Ibrahim, C. Schmithals, E. Kowarz,S. NitschAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.):A.A. Ibrahim, C. Schmithals, V. K€oberle, B. Kakoschky,O. Kollmar, S. Nitsch, F. Finkelmeier, H.-W. Korf, T. Schmid, A. Weigert,B. Kronenberger, A. PiiperAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): A.A. Ibrahim, T. Pleli, O. Kollmar, O. Waidmann,F. Finkelmeier, S. Zeuzem, H.-W. Korf, A. Weigert, B. Kronenberger, A. PiiperWriting, review, and/or revision of the manuscript: A.A. Ibrahim,C. Schmithals, V. K€oberle, O. Waidmann, F. Finkelmeier, S. Zeuzem,H.-W. Korf, T. Schmid, A. Weigert, B. Kronenberger, R. Marschalek, A. PiiperAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): B. Kakoschky, T. Pleli, B. KronenbergerStudy supervision: C. Schmithals, O. Kollmar, O. Waidmann, S. Zeuzem,B. Kronenberger, A. Piiper

AcknowledgmentsWe thank Snorri Thorgeirsson (National Cancer Institute, NIH, Bethesda) for

providing the mice transgenic for c-myc and TGFa, and Witold Filipowicz(FriedrichMiescher Institute, Basel, Switzerland) and Petr Svoboda (Institute ofMolecular Genetics, Prague, Czech Republic) for kindly donating the cDNA forDicer.

Grant SupportThis study was supported by German academic exchange service (DAAD),

Scholari foundation, GRK 1172 (DFG), Deutsche Krebshilfe (grant 111950).The costs of publication of this articlewere defrayed inpart by the payment of

page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received July 11, 2016; revised January 18, 2017; accepted January 19, 2017;published OnlineFirst February 6, 2017.

References1. Valastyan S, Weinberg RA. Tumor metastasis: Molecular insights and

evolving paradigms. Cell 2011;147:275–92.2. Tsai JH, Yang J. Epithelial-mesenchymal plasticity in carcinomametastasis.

Genes Dev 2013;27:2192–206.3. Vaupel P, Mayer A. Hypoxia in cancer: Significance and impact on clinical

outcome. Cancer Metastasis Rev 2007;26:225–39.4. Wong CC, Kai AK, Ng IO. The impact of hypoxia in hepatocellular

carcinoma metastasis. Front Med 2014;8:33–41.5. Lu X, Kang Y. Hypoxia and hypoxia-inducible factors: Master regulators of

metastasis. Clin Cancer Res 2010;16:5928–35.6. Semenza GL.Oxygen sensing, hypoxia-inducible factors, and disease path-

ophysiology. Annu Rev Pathol 2014;9:47–71.7. Raza U, Zhang JD, Sahin O. MicroRNAs: Master regulators of

drug resistance, stemness, and metastasis. J Mol Med 2014;92:321–36.

8. Tang J, Li Y, Wang J, Wen Z, Lai M, Zhang H. Molecular mechanisms ofmicroRNAs in regulating epithelial-mesenchymal transitions in humancancers. Cancer Lett 2016;371:301–13.

9. Kim VN, Han J, Siomi MC. Biogenesis of small RNAs in animals. Nat RevMol Cell Biol 2009;10:126–39.

10. Calin GA, Croce CM. MicroRNA–cancer connection: The beginning of anew tale. Cancer Res 2006;66:7390–94.

11. He L, Thomson JM, Hemann MT, Hernando-Monge E, Mu D, Goodson S,et al. A microRNA polycistron as a potential human oncogene. Nature2005;435:828–33.

12. Lu J, GetzG,Miska EA, Alvarez-Saavedra E, Lamb J, PeckD, et al.MicroRNAexpression profiles classify human cancers. Nature 2005;435:834–8.

13. Thomson JM, Newman M, Parker JS, Morin-Kensicki EM, Wright T,Hammond SM. Extensive post-transcriptional regulation of microRNAsand its implications for cancer. Genes Dev 2006;20:2202–7.

Ibrahim et al.





14. Grelier G, VoirinN, Ay AS, CoxDG, Chabaud S, Treilleux I, et al. Prognosticvalue of Dicer expression in human breast cancers and associationwith themesenchymal phenotype. Br J Cancer 2009;101:673–83.

15. Merritt WM, Lin YG, Han LY, Kamat AA, Spannuth WA, Schmandt R, et al.Dicer, Drosha, and outcomes in patients with ovarian cancer. N Engl J Med2008;359:2641–50.

16. Kumar MS, Lu J, Mercer KL, Golub TR, Jacks T. Impaired microRNAprocessing enhances cellular transformation and tumorigenesis. Nat Genet2007;39:673–7.

17. MartelloG, Rosato A, Ferrari F,Manfrin A, CordenonsiM,Dupont S, et al. AmicroRNA targeting dicer formetastasis control. Cell 2010;141:1195–207.

18. Melo SA, Ropero S, Moutinho C, Aaltonen LA, Yamamoto H, Calin GA,et al. A TARBP2 mutation in human cancer impairs microRNA processingand DICER1 function. Nat Genet 2009;41:365–70.

19. Melo SA,MoutinhoC, Ropero S, CalinGA, Rossi S, Spizzo R, et al. A geneticdefect in exportin-5 traps precursor microRNAs in the nucleus of cancercells. Cancer Cell 2010;18:303–15.

20. Su X, Chakravarti D, Cho MS, Liu L, Gi YJ, Lin YL, et al. TAp63 suppressesmetastasis through coordinate regulation of Dicer and miRNAs. Nature2010;467:986–90.

21. Sekine S, Ogawa R, Ito R, Hiraoka N, McManus MT, Kanai Y, et al.Disruption of Dicer1 induces dysregulated fetal gene expression andpromotes hepatocarcinogenesis. Gastroenterology 2009;136:2304–15.

22. Lin S,GregoryRI.MicroRNAbiogenesis pathways in cancer.Nat RevCancer2015;15:321–33.

23. Shen J, Xia W, Khotskaya YB, Huo L, Nakanishi K, Lim SO, et al. EGFRmodulates microRNA maturation in response to hypoxia through phos-phorylation of AGO2. Nature 2013;497:383–7.

24. Wu C, So J, Davis-Dusenbery BN, Qi HH, Bloch DB, Shi Y, et al. Hypoxiapotentiates microRNA-mediated gene silencing through posttranslationalmodification of Argonaute2. Mol Cell Biol 2011;31:4760–74.

25. Ho JJ, Metcalf JL, YanMS, Turgeon PJ, Wang JJ, ChalsevM, et al. Functionalimportance of Dicer protein in the adaptive cellular response to hypoxia. JBiol Chem 2012;287:29003–20.

26. Rupaimoole R, Wu SY, Pradeep S, Ivan C, Pecot CV, Gharpure KM, et al.Hypoxia-mediated downregulation of miRNA biogenesis promotestumour progression. Nat Commun 2014;5:5202.

27. van denBeucken T, Koch E, ChuK, Rupaimoole R, Prickaerts P, AdriaensM,et al. Hypoxia promotes stem cell phenotypes and poor prognosis throughepigenetic regulation of DICER. Nat Commun 2014;5:5203.

28. Hernandez-Gea V, Toffanin S, Friedman SL, Llovet JM. Role of the micro-environment in the pathogenesis and treatment of hepatocellular carci-noma. Gastroenterology 2013;144:512–27.

29. Kitagawa N, Ojima H, Shirakihara T, Shimizu H, Kokubu A, Urushidate T,et al. Downregulation of the microRNA biogenesis components and itsassociation with poor prognosis in hepatocellular carcinoma. Cancer Sci.2013;104:543–51.

30. Wu JF, ShenW, Liu NZ, Zeng GL, Yang M, Zuo GQ, et al. Down-regulationof Dicer in hepatocellular carcinoma. Med Oncol 2011;28:804–9.

31. Schmithals C, K€oberle V, Korkusuz H, Pleli T, Kakoschky B, Augusto EA,et al. Improving drug penetrability with iRGD leverages the therapeuticresponse to sorafenib and doxorubicin in hepatocellular carcinoma. Can-cer Res 2015;75:3147–54.

32. Kowarz E, L€oscher D, Marschalek R. Optimized Sleeping Beauty transpo-sons rapidly generate stable transgenic cell lines. Biotechnol J 2015;10:647–53.

33. Haupenthal J, Bihrer V, Korkusuz H, Kollmar O, Schmithals C, Kriener S,et al. Reduced efficacy of the Plk1 inhibitor BI 2536 on the progression ofhepatocellular carcinoma due to low intratumoral drug levels. Neoplasia2012;14:410–9.

34. Murakami H, Sanderson ND, Nagy P, Marino PA, Merlino G, Thor-geirsson SS. Transgenic mouse model for synergistic effects of nuclearoncogenes and growth factors in tumorigenesis: Interaction of c-mycand transforming growth factor alpha in hepatic oncogenesis. CancerRes 1993;53:1719–23.

35. Hsu SH, Wang B, Kota J, Yu J, Costinean S, Kutay H, et al. Essentialmetabolic, anti-inflammatory, and anti-tumorigenic functions of miR-122 in liver. J Clin Invest 2012;122:2871–83.

36. Tsai WC, Hsu SD, Hsu CS, Lai TC, Chen SJ, Shen R, et al. MicroRNA-122plays a critical role in liver homeostasis and hepatocarcinogenesis. J ClinInvest 2012;122:2884–97.

37. Zhang L, Huang G, Li X, Zhang Y, Jiang Y, Shen J, et al. Hypoxia inducesepithelial-mesenchymal transition via activation of SNAI1 by hypoxia-inducible factor -1a in hepatocellular carcinoma. BMC Cancer 2013;13:108.

38. Zijl FV, Zulehner G, Petz M, Schneller D, Kornauth C, Hau M, et al.Epithelial-mesenchymal transition in hepatocellular carcinoma. FutureOncol 2009;5:1169–79.

39. Jiang Y, Zhu Y, Wang X, Gong J, Hu C, Guo B, et al. Temporal regulation ofHIF-1 and NF-kB in hypoxic hepatocarcinoma cells. Oncotarget 2015;6:9409–19.

40. Rupaimoole R, Ivan C, Yang D, Gharpure KM, Wu SY, Pecot CV, et al.Hypoxia-upregulated microRNA-630 targets Dicer, leading to increasedtumor progression. Oncogene 2016;35:4312–20.

41. Sun HL, Cui R, Zhou J, Teng KY, Hsiao YH, Nakanishi K, et al. ERKactivation globally downregulates miRNAs through phosphorylatingExportin-5. Cancer Cell 2016;30:723–36.

42. Zhang Q, Bai X, Chen W, Ma T, Hu Q, Liang C, et al. Wnt/b-cateninsignaling enhances hypoxia-induced epithelial-mesenchymal transition inhepatocellular carcinoma via crosstalk with hif-1a signaling. Carcinogen-esis 2013;34:962–73.

43. Yang J, Zhang X, Zhang Y, Zhu D, Zhang L, Li Y, et al. HIF-2a promotesepithelial-mesenchymal transition through regulating Twist2 binding tothe promoter of E-cadherin in pancreatic cancer. J Exp Clin Cancer Res2016;35:26.

44. Liu L, Zhu XD, Wang WQ, Shen Y, Qin Y, Ren ZG, et al. Activation ofbeta-catenin by hypoxia in hepatocellular carcinoma contributes toenhanced metastatic potential and poor prognosis. Clin Cancer Res2010;16:2740–50.

45. Coulouarn C, Factor VM, Andersen JB, DurkinME, Thorgeirsson SS. Loss ofmiR-122 expression in liver cancer correlates with suppression of thehepatic phenotype and gain of metastatic properties. Oncogene 2009;28:3526–36.

46. Lagos-Quintana M, Rauhut R, Yalcin A, Meyer J, Lendeckel W, Tuschl T.Identification of tissue-specific microRNAs from mouse. Curr Biol2002;12:735–9.

47. Csak T, Bala S, Lippai D, Satishchandran A, Catalano D, Kodys K, et al.microRNA-122 regulates hypoxia-inducible factor-1 and vimentin in hepa-tocytes and correlates with fibrosis in diet-induced steatohepatitis. Liver Int2015;35:532–41.

48. Tsai WC, Hsu PW, Lai TC, Chau GY, Lin CW, Chen CM, et al.MicroRNA-122, a tumor suppressor microRNA that regulates intra-hepatic metastasis of hepatocellular carcinoma. Hepatology 2009;49:1571–82.

49. Wang N, Wang Q, Shen D, Sun X, Cao X, Wu D. Downregulation ofmicroRNA-122 promotes proliferation,migration, and invasion of humanhepatocellular carcinoma cells by activating epithelial-mesenchymal tran-sition. Onco Targets Ther 2016;9:2035–47.

50. Bandara V, Michael MZ, Gleadle JM. Hypoxia represses microRNA bio-genesis proteins in breast cancer cells. BMC Cancer 2014;14:533.






2017;23:3896-3905. Published OnlineFirst February 6, 2017.Clin Cancer Res Ahmed Atef Ibrahim, Christian Schmithals, Erik Kowarz, et al. Transition

Mesenchymal− and EpithelialαHypoxia-Inducible Factor 1Carcinoma, Which Is Required for Upregulation of Hypoxia Causes Downregulation of Dicer in Hepatocellular

Updated version

10.1158/1078-0432.CCR-16-1762doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2017/02/04/1078-0432.CCR-16-1762.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/23/14/3896.full#ref-list-1

This article cites 50 articles, 9 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/23/14/3896To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-16-1762

http://clincancerres.aacrjournals.org/content/suppl/2017/02/04/1078-0432.CCR-16-1762.DC1

http://clincancerres.aacrjournals.org/content/23/14/3896.full#ref-list-1

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/23/14/3896


hypoxia causes downregulation of dicer in hepatocellular … · biology of human tumors hypoxia...

Documents