Characterization of HULC, a Novel Gene With Striking Up-Regulation in Hepatocellular Carcinoma, as Noncoding RNA
2006; Elsevier BV; Volume: 132; Issue: 1 Linguagem: Inglês
10.1053/j.gastro.2006.08.026
ISSN1528-0012
AutoresKatrin Panzitt, Marisa M.O. Tschernatsch, Christian Guelly, Tarek Moustafa, Martin Stradner, Heimo Strohmaier, Charles R. Buck, Helmut Denk, Renée Schroeder, Michael Trauner, Kurt Zatloukal,
Tópico(s)MicroRNA in disease regulation
ResumoBackground & Aims: Recent studies have highlighted the role of noncoding RNAs (ncRNAs) in carcinogenesis, and suggested that this class of genes might be used as biomarkers in cancer. We searched the human genome for novel genes including ncRNAs related to hepatocellular carcinoma (HCC). Methods: An HCC-specific gene library was generated and screened for deregulated genes with 46 HCCs, 4 focal nodular hyperplasias, and 7 cirrhoses utilizing cDNA arrays. Sequencing of library clones identified a novel ncRNA as the most up-regulated gene in HCC. This gene was also cloned from different monkeys and characterized by quantitative RT-PCR, Northern blot analysis and in situ hybridization. Structural and functional studies included comparative sequence and protein expression analyses, quantitative RT-PCR of polysomal preparations, and siRNA-mediated knockdown experiments. Results: The most up-regulated gene in HCC named highly up-regulated in liver cancer (HULC) was characterized as a novel mRNA-like ncRNA. HULC RNA is spliced and polyadenlyated, and resembles the mammalian LTR transposon 1A. It does not contain substantial open reading frames, and no native translation product was detected. HULC is present in the cytoplasm, where it copurifies with ribosomes. siRNA-mediated knockdown of HULC RNA in 2 HCC cell lines altered the expression of several genes, 5 of which were known to be affected in HCC, suggesting a role for HULC in post-transcriptional modulation of gene expression. Conclusions: HULC is the first ncRNA with highly specific up-regulation in HCC. Because HULC was detected in blood of HCC patients, a potential use as novel biomarker can be envisaged. Background & Aims: Recent studies have highlighted the role of noncoding RNAs (ncRNAs) in carcinogenesis, and suggested that this class of genes might be used as biomarkers in cancer. We searched the human genome for novel genes including ncRNAs related to hepatocellular carcinoma (HCC). Methods: An HCC-specific gene library was generated and screened for deregulated genes with 46 HCCs, 4 focal nodular hyperplasias, and 7 cirrhoses utilizing cDNA arrays. Sequencing of library clones identified a novel ncRNA as the most up-regulated gene in HCC. This gene was also cloned from different monkeys and characterized by quantitative RT-PCR, Northern blot analysis and in situ hybridization. Structural and functional studies included comparative sequence and protein expression analyses, quantitative RT-PCR of polysomal preparations, and siRNA-mediated knockdown experiments. Results: The most up-regulated gene in HCC named highly up-regulated in liver cancer (HULC) was characterized as a novel mRNA-like ncRNA. HULC RNA is spliced and polyadenlyated, and resembles the mammalian LTR transposon 1A. It does not contain substantial open reading frames, and no native translation product was detected. HULC is present in the cytoplasm, where it copurifies with ribosomes. siRNA-mediated knockdown of HULC RNA in 2 HCC cell lines altered the expression of several genes, 5 of which were known to be affected in HCC, suggesting a role for HULC in post-transcriptional modulation of gene expression. Conclusions: HULC is the first ncRNA with highly specific up-regulation in HCC. Because HULC was detected in blood of HCC patients, a potential use as novel biomarker can be envisaged. Noncoding RNAs (ncRNAs) have emerged as a new class of functional transcripts in eukaryotic cells, and were grouped into 3 subclasses according to their number of nucleotides.1Mattick J.S. Non-coding RNAs: the architects of eukaryotic complexity.EMBO Rep. 2001; 2: 986-991Crossref PubMed Scopus (626) Google Scholar, 2Costa F.F. Non-coding RNAs: new players in eukaryotic biology.Gene. 2005; 357: 83-94Crossref PubMed Scopus (282) Google Scholar The growing class of micro-RNAs (mRNAs) (21–25 nt) has been related to cell differentiation and cancer in recent publications.3Lu J. Getz G. Miska E.A. Alvarez-Saavedra E. Lamb J. Peck D. Sweet-Cordero A. Ebert B.L. Mak R.H. Ferrando A.A. Downing J.R. Jacks T. Horvitz H.R. Golub T.R. MicroRNA expression profiles classify human cancers.Nature. 2005; 435: 834-838Crossref PubMed Scopus (8237) Google Scholar, 4Bentwich I. Avniel A. Karov Y. Aharonov R. Gilad S. Barad O. Barzilai A. Einat P. Einav U. Meiri E. Sharon E. Spector Y. Bentwich Z. Identification of hundreds of conserved and nonconserved human microRNAs.Nat Genet. 2005; 37: 766-770Crossref PubMed Scopus (1566) Google Scholar, 5Jopling C. Yi M. Lancaster A. Lemon S. Sarnow P. Modulation of hepatitis C virus RNA abundance by a liver-specific microRNA.Science. 2005; 309: 1577-1581Crossref PubMed Scopus (2140) Google Scholar Small ncRNAs with a length of 100–200 nt are commonly found as translational repressors, and long ncRNAs (>10.000 nt) are involved in gene silencing.2Costa F.F. Non-coding RNAs: new players in eukaryotic biology.Gene. 2005; 357: 83-94Crossref PubMed Scopus (282) Google Scholar According to their structural features, these 3 subclasses of heterogeneous transcriptional units can be further subcategorized. mRNA-like ncRNAs, for example, usually lack extensive open reading frames (ORF) and are, therefore, difficult to predict from genomic sequences. In general, they are more than 200 nt long and, in most of the cases, they are spliced and polyadenylated.6Erdmann V.A. Szymanski M. Hochberg A. de Groot N. Barciszewski J. Collection of mRNA-like non-coding RNAs.Nucleic Acids Res. 1999; 27: 192-195Crossref PubMed Scopus (63) Google Scholar Recent studies imply ncRNAs in the regulation of gene expression by a variety of mechanisms such as RNA interference, gene silencing, imprinting, and DNA demethylation, indicating that this novel class of transcripts plays a central role in development and cell differentiation,6Erdmann V.A. Szymanski M. Hochberg A. de Groot N. Barciszewski J. Collection of mRNA-like non-coding RNAs.Nucleic Acids Res. 1999; 27: 192-195Crossref PubMed Scopus (63) Google Scholar although to most ncRNAs no specific function has been ascribed. Increasing evidence relates changes in expression levels of ncRNAs to complex diseases such as cancer.7Tam W. Ben-Yehuda D. Hayward W.S. Bic, a novel gene activated by proviral insertions in avian leukosis virus-induced lymphomas, is likely to function through its non-coding RNA.Mol Cell Biol. 1997; 3: 1490-1502Google Scholar, 8Ji P. Diederichs S. Wang W. Boing S. Metzger R. Schneider P.M. Tidow N. Brandt B. Buerger H. Bulk E. Thomas M. Berdel W.E. Serve H. Muller-Tidow C. MALAT-1, a novel non-coding RNA, and thymosin beta4 predict metastasis and survival in early-stage non-small cell lung cancer.Oncogene. 2003; 22: 8031-8041Crossref PubMed Scopus (1799) Google Scholar, 9Chen W. Bocker W. Brosius J. Tiedge H. Expression of neural BC200 RNA in human tumours.J Pathol. 1997; 183: 345-351Crossref PubMed Scopus (174) Google Scholar, 10Li J. Witte D.P. Van Dyke T. Askew D.S. Expression of the putative proto-oncogene His-1 in normal and neoplastic tissues.Am J Pathol. 1997; 150: 1297-1305PubMed Google Scholar For PCGEM,11Srikantan V. Zou Z. Petrovics G. Xu L. Augustus M. Davis L. Livezey J.R. Connell T. Sesterhenn I.A. Yoshino K. Buzard G.S. Mostofi F.K. McLeod D.G. Moul J.W. Srivastava S. PCGEM1, a prostate-specific gene, is overexpressed in prostate cancer.Proc Natl Acad Sci U S A. 2000; 97: 12216-12221Crossref PubMed Scopus (286) Google Scholar and DD3,12Bussemakers M.J. van Bokhoven A. Verhaegh G.W. Smit F.P. Karthaus H.F. Schalken J.A. Debruyne F.M. Ru N. Isaacs W.B. DD3: a new prostate-specific gene, highly overexpressed in prostate cancer.Cancer Res. 1999; 59: 5975-5979PubMed Google Scholar for example, a tumor-associated overexpression in prostate cancer was found, implicating these ncRNAs in prostate tumorigenesis.13Petrovics G. Zhang W. Makarem M. Street J.P. Connelly R. Sun L. Sesterhenn I.A. Srikantan V. Moul J.W. Srivastava S. Elevated expression of PCGEM1, a prostate-specific gene with cell growth-promoting function, is associated with high-risk prostate cancer patients.Oncogene. 2004; 23: 605-611Crossref PubMed Scopus (220) Google Scholar BC200 RNA overexpression has recently been correlated with the progression of breast tumors and proposed as a new molecular marker for breast carcinomas.14Iacoangeli A. Lin Y. Morley E.J. Muslimov I.A. Bianchi R. Reilly J. Weedon J. Diallo R. Bocker W. Tiedge H. BC200 RNA in invasive and preinvasive breast cancer.Carcinogenesis. 2004; 25: 2125-2133Crossref PubMed Scopus (136) Google Scholar Increased expression of the MALAT-1 gene indicates a worse clinical outcome in lung cancer patients, and thus further emphasizes the potential role of ncRNAs in tumorigenesis.8Ji P. Diederichs S. Wang W. Boing S. Metzger R. Schneider P.M. Tidow N. Brandt B. Buerger H. Bulk E. Thomas M. Berdel W.E. Serve H. Muller-Tidow C. MALAT-1, a novel non-coding RNA, and thymosin beta4 predict metastasis and survival in early-stage non-small cell lung cancer.Oncogene. 2003; 22: 8031-8041Crossref PubMed Scopus (1799) Google Scholar Hepatocellular carcinoma (HCC) is one of the leading causes of cancer-related deaths worldwide, and affects approximately half a million people yearly.15Parkin D.M. Bray F. Ferlay J. Pisani P. Estimating the world cancer burden: Globocan 2000.Int J Cancer. 2001; 94: 153-156Crossref PubMed Scopus (3311) Google Scholar As a multifactorial genetic and epigenetic disease with a complex etiology, most HCCs originate on the basis of long-term liver injury typically caused by chronic hepatitis C or B virus infections, alcoholic liver disease, aflatoxin exposure, or a variety of inherited metabolic diseases.16Thorgeirsson S.S. Grisham J.W. Molecular pathogenesis of human hepatocellular carcinoma.Nat Genet. 2002; 31: 339-346Crossref PubMed Scopus (1269) Google Scholar The majority of these chronic liver diseases lead to the development of liver cirrhosis, which promotes HCC formation. Besides the highly malignant HCC, several other tumor types also occur in the liver. The most common benign tumor-like lesion in the liver is focal nodular hyperplasia (FNH), which is characterized by polyclonal proliferation of hepatocytes and immature ductular cells that are considered to be related to liver progenitor cells.17Roskams T. De Vos R. Desmet V. “Undifferentiated progenitor cells” in focal nodular hyperplasia of the liver.Histopathology. 1996; 28: 291-299Crossref PubMed Scopus (104) Google Scholar Based on the increasing evidence for a relationship between ncRNAs and tumor formation and progression, we performed a genome-wide search for novel transcripts, which are associated with the molecular pathogenesis of HCC, but not properly represented in gene collections and commercially available platforms. Utilizing HCC-specific gene libraries and cDNA microarrays, we identified a novel ncRNA as the most up-regulated gene in the HCCs investigated and named it HULC (highly up-regulated in liver cancer). Due to its striking expression pattern, we performed a detailed characterization of this first ncRNA specifically associated with HCC. Human tissue samples were retrieved from the biobank at the Institute of Pathology, Medical University of Graz, Austria. Tissues obtained from surgically resected tumors and adjacent non-neoplastic tissue or from explanted livers were either snap frozen in methyl butane precooled with liquid nitrogen within 20 minutes after operation or fixed in phosphate-buffered (pH 7.4) 4% formaldehyde solution and embedded in paraffin. The study was approved by the ethics committee of the Medical University of Graz, Austria. Total RNA was isolated from 3 HCC samples and from 3 non-neoplastic liver samples using Trizol reagent (Invitrogen, Carlsbad, CA) and pooled. PolyA+-RNA was prepared using Oligotex mRNA spin columns (Qiagen, West Sussex, UK). cDNA synthesis and subtraction were performed using the PCR-Select cDNA Subtraction Kit (BD-Biosciences-Clontech, Mountain View, CA). Subtracted cDNAs were cloned into pCRII (Invitrogen) and transformed into Escherichia coli XL-1Blue (Invitrogen). For microarray generation, 2304 colonies containing 960 up- and 960 down-regulated cDNAs plus 384 normalized colonies (where the subtractive step was omitted) were selected from the HCC cDNA library. Additionally, 3514 IMAGE cDNA clones representing genes with well-characterized roles in signaling, cell cycle regulation, apoptosis, tissue remodeling, angiogenesis, and immune reactions were purchased from the RZPD (German Resource Center for Genome Research, Heidelberg, Germany). Four plant clones, cab, tim, xcp2, and rbcl (Stratagene, La Jolla, CA) were included as internal standards for quality control. All 5832 individual clones were PCR-amplified and processed according to standard protocols18Hegde P. Qi R. Abernathy K. Gay C. Dharap S. Gaspard R. Hughes J.E. Snesrud E. Lee N. Quackenbush J. A concise guide to cDNA microarray analysis.BioTechniques. 2000; 29: 548-550Crossref PubMed Scopus (794) Google Scholar and printed onto Corning CMT-GAPS™ aminosilane coated glass microscope slides (Corning Life Sciences, New York, NY) using a GMS 417 Arrayer (Genetic MicroSystems, Woburn, MA). Frozen tissue sections were dissected under the microscope. Fifty to 100 mg of tissue were used for RNA isolation with Trizol reagent (Invitrogen). Microarray hybridization experiments included a non-neoplastic liver pool (n = 3) as reference. In a direct labeling reaction, 20 μg of total RNA of each sample were spiked with different amounts of Arabidopsis mRNAs (Stratagene). The RNAs were fluorescently labeled with either Cy 3 (reference) or Cy 5 dCTP (test sample) by reverse transcription. Probes were hybridized to the chip at 42°C for 16 to 20 hours18Hegde P. Qi R. Abernathy K. Gay C. Dharap S. Gaspard R. Hughes J.E. Snesrud E. Lee N. Quackenbush J. A concise guide to cDNA microarray analysis.BioTechniques. 2000; 29: 548-550Crossref PubMed Scopus (794) Google Scholar and microarrays were scanned using a GMS 418 Array Scanner (Genetic MicroSystems). Fluorescent images were analyzed by ImaGene 4.1 and 4.2 software (BioDiscovery, El Segundo, CA). Photomultiplier tube and laser value for scanning were calibrated over all spots in both the Cy 3 and Cy 5 channels. After local background correction by using GeneSight 3.0 software (BioDiscovery), ratios were calibrated by applying normalization factors calculated from the mean intensities over all spots. According to their spike-in ratios, the normalization drift in the 2 fluorescence channels was evaluated. Genes were hierarchically clustered using Genesis software (http://genome.tugraz.at). Statistical significance of defined subgroups was tested by the Mann–Whitney U-test. The same test was used to compute P values for testing groups with PCR. In Figures 1C and D, 6, and 7 the error bars represent the standard errors of mean (SEM).Figure 6Polysomal profile of HULC. (A) Polysomal fractions (fr) were isolated after sucrose gradient centrifugation and loaded onto a denaturing (1× MOPS) RNA gel. The first fractions (1–5) on the gel demonstrate RNA not bound to ribosomes, whereas subsequent fractions represent ribosome bound RNA (fractions 6–18). (B) Representative ribosome fractions were examined by quantitative RT-PCR and relative concentrations of HULC and GAPDH RNA were calculated from total RNA standard curves. Both genes showed elevated expression levels in the ribosome-bound fractions starting with fraction 13, where the first complete ribosomes were observed.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 7Efficiency of siRNA-mediated HULC knockdown in the Hep3B and HepG2 hepatoma cell lines. Cells were treated with 2 different siRNA oligonucleotides and 2 different control siRNAs for 24 hours. HULC RNA expression levels from 2 biological replicates were monitored by quantitative RT-PCR in triplicates, revealing the indicated knockdown efficiencies relative to the nonsilencing control (=1). Error bars present the standard errors of mean (±SEM, n = 6). *P < .01, **P < .05 versus nonsilencing siRNA control (=1), Student’s t-test.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The Smart RACE (rapid amplification of cDNA ends) cDNA Amplification kit (BD-Biosciences-Clontech) was used for the construction of 5′ and 3′ RACE libraries from polyA RNA, from total RNA extracted from a pool of HCC tissue and from total RNA primed with multiple gene-specific primers. The nucleotide sequence for the HULC region has been deposited into the Genbank data base with the accession number AY914050. Ten micrograms of total RNA were separated on a 1.2% MOPS gel and transferred to a Hybond-N+ membrane (GE Healthcare-Amersham Biosciences, Little Chalfont, UK) by capillary blot. The HULC probe corresponding to nucleotides 99–425 (GenBank accession #AY914050) was radioactively labeled with [α-32Xu X.R. Huang J. Xu Z.G. Qian B.Z. Zhu Z.D. Yan Q. Cai T. Zhang X. Xiao H.S. Qu J. Liu F. Huang Q.H. Cheng Z.H. Li N.G. Du J.J. Hu W. Shen K.T. Lu G. Fu G. Zhong M. Xu S.H. Gu W.Y. Huang W. Zhao X.T. Hu G.X. Gu J.R. Chen Z. Han Z.G. Insight into hepatocellular carcinogenesis at transcriptome level by comparing gene expression profiles of hepatocellular carcinoma with those of corresponding non-cancerous liver.Proc Natl Acad Sci U S A. 2001; 98: 15089-15094Crossref PubMed Scopus (325) Google ScholarP]dCTP (Amersham Biosciences) using the Rediprime II random prime labeling kit (Amersham Biosciences). ULTRAhyb buffer (Ambion, Cambridgeshire, UK) was used for both prehybridization and hybridization, and signals were detected by autoradiography. For the preparation of peripheral blood mononuclear cells from blood samples, for which informed consent was given, 5 mL of peripheral venous blood was collected in 5 mL vacutainer ethylenediamine tetraacetic acid (EDTA) tubes (BD Pharma, Le Pont-De-Claix, France). Samples were diluted 1:1 in balanced salt solution and subjected to centrifugation in a Ficoll Paque Plus (Amersham Biosciences) gradient in a Sorvall RC-5B centrifuge for 40 minutes at 1400 rpm and 18°C. RNA from cell pellets prepared from the interphase by addition of 6 mL balanced salt solution and centrifugation for 10 minutes at 700 rpm and 18°C was isolated using the RNeasy Kit (Qiagen). For the extraction of total RNA from frozen tissue samples, 4-μm cryosections were dissected under the microscope. Fifty to 100 mg of tissue were used for RNA isolation with Trizol reagent (Invitrogen) according to the manufacturer’s recommendations. Total RNA from paraffin-embedded samples was prepared as described elsewhere.19Hrzenjak A. Moinfar F. Tavassoli F. Strohmeier B. Kremser M.-L. Zatloukal K. Denk H. JAZF1/JJAZ1 gene fusion in endometrial stroma sarcomas.J Mol Diagn. 2005; 7: 388-395Abstract Full Text Full Text PDF PubMed Scopus (74) Google Scholar Using SuperScript II RNase H-reverse transcriptase (Invitrogen), cDNA was synthesized from 1 μg of total RNA prepared from at least 3 independent samples per tissue, including the liver samples from the cDNA array. Applied Biosystems’ Universal Taqman mastermix or SYBRgreen mastermix were used with 40 ng of template cDNA per reaction, and samples were analyzed in triplicates. Standard curves were generated in duplicate with 50, 10, 2, and 0.4 ng cDNA, and the values were normalized against GAPDH and/or β-actin. The following primers were used: HULC forward primer: 5′-atctgcaagccaggaagagtc-3′; HULC probe: 5′-FAM-ccagaccatgcaggaactctgatcgtggac-TAMRA-3′; HULC reverse primer: 5′-cttgcttgatgctttggtctgt-3′. GAPDH forward primer: 5′-ccacatcgctcagacaccat-3′; GAPDH probe: 5′-FAM-caaatccgttgactccgaccttca-TAMRA-3′; GAPDH reverse primer: 5′-accaggcgcccaatacg-3′; β-actin forward primer: 5′-aaggccaaccgcgagaagat-3′; β-actin probe: 5′-FAM-ccatgtacgttgctatccaggctgtgctatcc-TAMRA-3′; β-actin reverse primer: 5′-gtcaccggaatccatcacga-3′ (all primers were obtained from MWG biotech, Ebersberg, Germany). After normalization to the housekeeping gene, RNA quantities were shown as fold overexpression. Radioactive in situ hybridization was performed as previously described.20Fickert P. Trauner M. Fuchsbichler A. Stumptner C. Zatloukal K. Denk H. Cytokeratins as targets for bile acid-induced toxicity.Am J Pathol. 2002; 160: 491-499Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar The radioactive probe was generated by in vitro transcription (antisense transcript: nucleotide 95–455 of HULC, GenBank accession #AY914050; sense transcript: nucleotide 92–426, GenBank accession #AY914050). Slides were covered in photographical emulsion K2 (Ilford Ltd., Cheshire, United Kingdom) and stored at 4°C. After 10 days of exposure, the slides were developed with a Kodak D19 developer and counterstained with hematoxylin. Hep3B cells (ATTC #HB-8064) were lysed in NP-40 buffer (0.5% NP-40, 10 mmol/L Tris–HCl pH 8.0, 140 mmol/L NaCl, 1.5 mmol/L MgCl2), supplemented with 480 U/mL RNasin, 150 μg/mL cycloheximide, 20 mmol/L DTT, and 1 mmol/L PMSF. After pelleting the nuclei, mitochondria and membrane particle samples were spun on a 15%–40% sucrose gradient in an ultracentrifuge (Sorvall OTD65B) for 2 hours at 38,000 rpm. Fractions (620 μL) were taken from the gradient and phenol–chloroform extracted. Following LiCl precipitation, RNA preparations were monitored on agarose gels. cDNA was synthesized from 1 μg RNA and subjected to quantitative RT-PCR analysis. Cells were lysed in RIPA buffer (50 mmol/L Tris pH 7.6, 150 mmol/L NaCl, 1% NP-40, 0.5% deoxycholate, 0.1% SDS) containing 2 μg/mL leupeptide/pepstatin/aprotinine (Roche, Basel, Switzerland), 1 mmol/L DDT and 1 mmol/L PMSF. Protein lysate, 15 μg per lane, was separated on 10%–20% Tris-glycine gels and transferred onto a Hybond-P+ membrane (Amersham Biosciences) with a semidry blotter (Amersham Biosciences). Membranes were incubated sequentially with the primary and secondary antibodies for 1 hour each. The following antibodies were used: anti-His (Dianova, Hamburg, Germany), anti-FLAG (Sigma Diagnostics, St. Louis, MO), anti-GFP (Zymed, South San Francisco, CA), swine antirabbit HRP conjugated (Dako, Glostrup, Denmark), rabbit antimouse HRP conjugated (Dako), rabbit anti-HULC peptide antibodies (a: 0.15 mg/mL; b: 0.1 mg/mL) to the putative peptide sequences: (a) PREDTARPQSASQE, (b) DTARPQSASQEESSR. Signals were detected with the ECL kit (GE Healthcare-Amersham Biosciences). Hep3B cells were cultured in DMEM medium (Invitrogen, Gibco cat #41965-039) supplemented with 1% Penicillin/Streptomycin (Invitrogen, Gibco, cat #15140-122), 1% L-glutamine (Invitrogen, Gibco cat #25030-024), and 10% fetal calf serum (Gibco, cat #10270-106) at 37°C in 95% H2O and 5% CO2. Plasmids were transfected with Lipofectamine reagent or Lipofectamine 2000 reagent (Invitrogen). 4 × 105 cells/well HepG2 (ATTC #HB-8065) or 2 × 105 cells Hep3B (ATTC #HB-8064), respectively, were seeded into 6-well plates the day before transfection in DMEM medium containing 10% fetal calf serum. Chemically synthesized siRNA oligonucleotides were purchased from Qiagen-Xeragon (Germantown, MD) and designed using an siRNA design algorithm applying stringent homology analysis (https://www1.qiagen.com/Products/GeneSilencing/CustomSiRna/CustomSiRnaOrder.aspx). Two siRNA oligonucleotides targeting 2 different sections of the HULC RNA were used for the experiments (target sequence: 5′-aatctgcaagccaggaagagt-3′ for HULC siRNA 1 and 5′-aacctccagaactgtgatcca-3′ for HULC siRNA 2). Controls involved a scrambled (target sequence: 5′-aaccactgccttgatccgaaa-3′) and a nonsilencing control siRNA (#1022076, target sequence: 5′-aattctccgaacgtgtcacgt-3′) as negative controls and Lamin A+C siRNA (#1022050; target sequence 5′-aactggacttccagaagaaca-3′) as positive control. Transfection was carried out using Oligofectamine (Invitrogen) according to the manufacturer’s recommendations. Briefly, cells were washed prior to transfection with DMEM without additives and cultured in 800 μL DMEM per well without additives. For transfection, 10 μL of a 20 mmol/L siRNA oligonucleotide stock (or 10 μL siRNA resuspension buffer for mock transfection controls) diluted in 175 μL Opti-MEM Reduced Serum Medium (Invitrogen) per well were complexed by addition of 2.7 μL Oligofectamine and 12 μL Opti-MEM Reduced Serum Medium. After 6 hours, fetal calf serum concentration was adjusted to 10%. Cells were harvested 24 hours after transfection. Cells of a 6-well plate were pooled and subjected to RNA isolation using Trizol reagent (Invitrogen). One microgram of total RNA was reversely transcribed and the degree of knock down was determined by quantitative RT-PCR. After RNeasy clean-up (Qiagen), samples were subjected to microarray hybridization. The quality of RNA prepared from siRNA-treated cells was evaluated using the Agilent Bioanalyzer 2100 (Agilent Technologies, Palo Alto, CA). Samples with a RNA Integrity Number value above 9 were subjected to labeling (35 μg of total RNA per sample) following the Applied Biosystems Chemiluminescent Labeling protocol (Applied Biosystems, Foster City, CA). The directly labeled cDNA product was hybridized to Human Genome Survey Microarrays v02 (Applied Biosystems) covering approximately 28,000 genes according to the manufacturer’s instructions. Microarrays were analyzed using the AB1700 Chemiluminescent Microarray Analyzer. For each cell line and for each knockdown experiment 2 biological replicates were analyzed. Array images were processed using the Applied Biosystems 1700 Array Scanner software. Spot normalization and statistical analysis were used to determine significantly (P < .05) up- or down-regulated genes affected by siRNA-mediated knockdown. Expression values were normalized across arrays by quantile normalization using R Script and the Bioconductor software (http://www.bioconductor.org/). Significantly up-regulated genes (P < .05) were subjected to functional classification using PANTHER software (Applied Biosystems) to evaluate pathways and biological processes affected by the siRNA knock-down. For a genome-wide gene expression profiling approach, we generated HCC-specific cDNA libraries by subtractive suppressive hybridization.21Wang Z. Brown D.D. A gene expression screen.Proc Natl Acad Sci U S A. 1991; 88: 11505-11509Crossref PubMed Scopus (297) Google Scholar Clones from the HCC cDNA library in conjunction with well-known cancer-related and control genes were used for the production of an HCC-specific cDNA microarray containing a total of 6912 cDNA clones. Tissue samples of 46 HCCs, 4 FNHs, 7 cirrhoses, and 2 non-neoplastic livers were compared with a pool of non-neoplastic liver samples (Figure 1A; full dataset will be provided as supplemental material on our website). The most prominent cluster of up-regulated genes contained a previously unidentified EST together with transforming growth factor alpha and glutamate-ammonia ligase, 2 genes that are known to be highly up-regulated during the pathogenesis of human HCC (Figure 1B).16Thorgeirsson S.S. Grisham J.W. Molecular pathogenesis of human hepatocellular carcinoma.Nat Genet. 2002; 31: 339-346Crossref PubMed Scopus (1269) Google Scholar, 22Christa L. Simon M.T. Flinois J.P. Gebhardt R. Brechot C. Lasserre C. Overexpression of glutamine synthetase in human primary liver cancer.Gastroenterology. 1994; 106: 1312-1320Abstract PubMed Google Scholar, 23Santoni-Rugiu E. Jensen M.R. Thorgeirsson S.S. Disruption of the pRb/E2F pathway and inhibition of apoptosis are major oncogenic events in liver constitutively expressing c-myc and transforming growth factor alpha.Cancer Res. 1998; 58: 123-134PubMed Google Scholar Up-regulated 33-fold on average (32.7 ± 5.0, P = .016) over the non-neoplastic liver pool in 76% of HCCs, and represented by 3 independent HCC-specific library clones, this EST was named HULC. HULC expression levels were also, but to a lesser extent, up-regulated (17.1 ± 2.7-fold) in FNH (P = .188). In contrast to the situation in liver tumors, HULC was only slightly up-regulated in cirrhotic liver tissue (3.3 ± 0.7-fold, P < .001) (Figure 1C). To determine whether HULC is commonly overexpressed in neoplastic tissues of various organs, a broad range of normal tissues and their corresponding carcinomas/sarcomas were tested for HULC RNA expression by quantitative RT-PCR (Figure 1D). HULC was barely detectable in most of the normal tissues and not significantly elevated in the majority of the corresponding neoplastic tissues. The slight increase of HULC RNA levels in prostate carcinoma, astrocytoma, and sarcoma, however, was never as striking as in HCC, where the mean HULC RNA expression level exceeded all other normal and neoplastic tissues tested at least by a factor of 12. The quantitative RT-PCR results mirrored the cDNA microarray data, although the levels of up-regulation were generally lower (9.2 ± 1.5-fold, P < .005 versus 32.7 ± 5.0, P = .016-fold in the cDNA microarray). Nevertheless, these results demonstrate a highly specific up-regulation of HULC RNA expression levels in HCC. In situ hybridization revealed the specific and strong expression of HULC RNA in the cytoplasm of HCC, whereas it was not detected in tumor stroma and non-neoplastic liver cells (Figure 2). Northern blot analysis identified an HULC transcript with a length of approximately 500 nucleotides. Again, HULC RNA levels were markedly higher in HCC than in non-neoplastic liver (Figure 3A), which is consistent with our cDNA microarray and quantitative RT-PCR data. To clone a full-length gene and to localize the H
Referência(s)