Glucose Metabolism in Cancer
2003; Elsevier BV; Volume: 278; Issue: 17 Linguagem: Inglês
10.1074/jbc.m300608200
ISSN1083-351X
AutoresAshish Goel, Saroj P. Mathupala, Peter L. Pedersen,
Tópico(s)Genetic Syndromes and Imprinting
ResumoOne of the “signature” phenotypes of highly malignant, poorly differentiated tumors, including hepatomas, is their remarkable propensity to utilize glucose at a much higher rate than normal cells, a property frequently dependent on the marked overexpression of type II hexokinase (HKII). As the expression of the gene for this enzyme is nearly silent in liver tissue, we tested the possibility that DNA methylation/demethylation events may be involved in its regulation. Initial studies employing methylation restriction endonuclease analysis provided evidence for differential methylation patterns for the HKII gene in normal hepatocytes and hepatoma cells, the latter represented by a highly glycolytic model cell line (AS-30D). Subsequently, sequencing following sodium bisulfite treatment revealed 18 methylated CpG sites within a CpG island (−350 to +781 bp) in the hepatocyte gene but none in that of the hepatoma. In addition, treatment of a hepatocyte cell line with the DNA methyltransferase inhibitors, 5′-azacytidine and 5′-aza-2′-deoxycytidine, activated basal expression levels of HKII mRNA and protein. Finally, stably transfecting the hepatocyte cell line with DNA demethylase also resulted in activating the basal expression levels of HKII mRNA and protein. These novel observations indicate that one of the initial events in activating the HKII gene during either transformation or tumor progression may reside at the epigenetic level. One of the “signature” phenotypes of highly malignant, poorly differentiated tumors, including hepatomas, is their remarkable propensity to utilize glucose at a much higher rate than normal cells, a property frequently dependent on the marked overexpression of type II hexokinase (HKII). As the expression of the gene for this enzyme is nearly silent in liver tissue, we tested the possibility that DNA methylation/demethylation events may be involved in its regulation. Initial studies employing methylation restriction endonuclease analysis provided evidence for differential methylation patterns for the HKII gene in normal hepatocytes and hepatoma cells, the latter represented by a highly glycolytic model cell line (AS-30D). Subsequently, sequencing following sodium bisulfite treatment revealed 18 methylated CpG sites within a CpG island (−350 to +781 bp) in the hepatocyte gene but none in that of the hepatoma. In addition, treatment of a hepatocyte cell line with the DNA methyltransferase inhibitors, 5′-azacytidine and 5′-aza-2′-deoxycytidine, activated basal expression levels of HKII mRNA and protein. Finally, stably transfecting the hepatocyte cell line with DNA demethylase also resulted in activating the basal expression levels of HKII mRNA and protein. These novel observations indicate that one of the initial events in activating the HKII gene during either transformation or tumor progression may reside at the epigenetic level. type II hexokinase methylation-sensitive restriction endonuclease DNA methyltransferase 5′-azacytidine 5′-aza-2′-deoxycytidine DNA demethylase 3-(cyclohexylamino)propanesulfonic acid reverse transcriptase N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine One of the most common biochemical phenotypes of highly malignant, poorly differentiated cancer cells is their capacity to metabolize glucose at elevated rates (1Warburg O. The Metabolism of Tumors. Arnold Constable, London1930Google Scholar, 2Aisenberg A.C. The Glycolysis and Respiration of Tumors. Academic Press, New York1961Google Scholar, 3Pedersen P.L. Prog. Exp. Tumor Res. 1978; 22: 190-274Crossref PubMed Google Scholar). This aberrant metabolism serves well the goal of the cancer cell to proliferate both by maintaining a constant supply of energy even when oxygen levels decrease and by providing enhanced levels of biosynthetic precursors. Thus, the transformation/progression process that ultimately leads to the high glycolytic tumor phenotype provides the tumor with a metabolic advantage over its normal tissue of origin. Significantly, we have demonstrated in earlier studies the essential role that hexokinase plays in sustaining the high glycolytic tumor phenotype (4Bustamante E. Pedersen P.L. Proc. Natl. Acad. Sci. U. S. A. 1977; 74: 3735-3739Crossref PubMed Scopus (342) Google Scholar, 5Bustamante E. Morris H.P. Pedersen P.L. J. Biol. Chem. 1981; 256: 8699-8704Abstract Full Text PDF PubMed Google Scholar), particularly the type II isoform that becomes markedly elevated in rapidly growing, highly malignant hepatomas (6Nakashima R.A. Paggi M.G. Scott L.J. Pedersen P.L. Cancer Res. 1988; 48: 913-919PubMed Google Scholar,7Mathupala S.P. Rempel A. Pedersen P.L. J. Biol. Chem. 1995; 270: 16918-16925Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar). These experimental observations are dramatic considering that liver normally expresses glucokinase (type IV “highKm” hexokinase), whereas the type II “lowKm” form is nearly silent (7Mathupala S.P. Rempel A. Pedersen P.L. J. Biol. Chem. 1995; 270: 16918-16925Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar). In contrast, within a poorly differentiated hepatoma, the expression of HKII1 may be elevated more than 100-fold (6Nakashima R.A. Paggi M.G. Scott L.J. Pedersen P.L. Cancer Res. 1988; 48: 913-919PubMed Google Scholar), whereas the type IV enzyme is undetectable (6Nakashima R.A. Paggi M.G. Scott L.J. Pedersen P.L. Cancer Res. 1988; 48: 913-919PubMed Google Scholar, 8Parry D.M. Pedersen P.L. J. Biol. Chem. 1983; 258: 1094-10912Abstract Full Text PDF Google Scholar). Thus, in the transformation/progression process the genetic machinery has been directed to completely down-regulate the expression of type IV hexokinase and markedly up-regulate that of HKII. The major advantages of doing this are 2-fold (9Pedersen P.L. Mathupala S. Rempel A. Geschwind J.F. Ko Y.H. Biochim. Biophys. Acta. 2002; 1555: 14-20Crossref PubMed Scopus (301) Google Scholar), one of which is to enhance the glycolytic rate. This role is served optimally by HKII as it has a high affinity for ATP and binds to outer mitochondrial membrane porin (10Nakashima R.A. Mangan P.S. Colombini M. Pedersen P.L. Biochemistry. 1986; 25: 1015-1021Crossref PubMed Scopus (187) Google Scholar) where it has more ready access to ATP for phosphorylating glucose (11Arora K.K. Pedersen P.L. J. Biol. Chem. 1988; 263: 17422-17428Abstract Full Text PDF PubMed Google Scholar) and is less sensitive to both product inhibition (4Bustamante E. Pedersen P.L. Proc. Natl. Acad. Sci. U. S. A. 1977; 74: 3735-3739Crossref PubMed Scopus (342) Google Scholar) and proteolytic degradation (12Rose I.A. Warms J.V.B. Arch. Biochem. Biophys. 1982; 213: 625-634Crossref PubMed Scopus (26) Google Scholar). The second advantage is that, by binding to the mitochondria, HKII acts as an antiapoptotic factor (13Pastorino J.G. Shulga N. Hoek J.B. J. Biol. Chem. 2002; 277: 7610-7618Abstract Full Text Full Text PDF PubMed Scopus (561) Google Scholar), thus protecting the cancer cells against death signals and promoting their immortality. In a program designed to elucidate the molecular basis for the marked activation of HKII in rapidly growing hepatomas, we have employed the AS-30D cell line growing in ascites form in the peritoneal cavity of rats. This is a hepatocellular carcinoma line derived originally from a solid liver tumor induced by feeding rats the carcinogen dimethylaminoazobenzene (14Chang J.P. Gibley Jr., C.W. Ichinoe K. Cancer Res. 1967; 27: 2065-2071PubMed Google Scholar, 15Smith D.F. Walborg Jr., E.F. Chang J.P. Cancer Res. 1970; 30: 2306-2309PubMed Google Scholar). This cell line exhibits the high glycolytic phenotype characteristic of aggressive tumors (4Bustamante E. Pedersen P.L. Proc. Natl. Acad. Sci. U. S. A. 1977; 74: 3735-3739Crossref PubMed Scopus (342) Google Scholar) and contains markedly elevated levels of both HKII mRNA (7Mathupala S.P. Rempel A. Pedersen P.L. J. Biol. Chem. 1995; 270: 16918-16925Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar) and the expressed enzyme bound to the outer mitochondrial membrane (4Bustamante E. Pedersen P.L. Proc. Natl. Acad. Sci. U. S. A. 1977; 74: 3735-3739Crossref PubMed Scopus (342) Google Scholar, 11Arora K.K. Pedersen P.L. J. Biol. Chem. 1988; 263: 17422-17428Abstract Full Text PDF PubMed Google Scholar). From this cell line we have isolated the HKII promoter (4.3 kb) and shown that it is quite promiscuous in its activation response to a number of physiological agents or conditions (7Mathupala S.P. Rempel A. Pedersen P.L. J. Biol. Chem. 1995; 270: 16918-16925Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar, 16Mathupala S.P. Heese C. Pedersen P.L. J. Biol. Chem. 1997; 272: 22776-22780Abstract Full Text Full Text PDF PubMed Scopus (209) Google Scholar, 17Rempel A. Mathupala S.P. Pedersen P.L. FEBS Lett. 1996; 385: 233-237Crossref PubMed Scopus (47) Google Scholar, 18Mathupala S.P. Rempel A. Pedersen P.L. J. Biol. Chem. 2001; 276: 43407-43412Abstract Full Text Full Text PDF PubMed Scopus (306) Google Scholar). These include hypoxia, glucose, dibutyryl cAMP, a phorbol ester, mutated p53, and the opposing hormones insulin and glucagon. Furthermore, fluorescencein situ hybridization analysis showed that the HKII gene is located on a single rat chromosome where it is amplified at least 5-fold without noticeable chromosomal aberrations or rearrangements (19Rempel A. Mathupala S.P. Griffin C.A. Hawkins A.L. Pedersen P.L. Cancer Res. 1996; 56: 2468-2471PubMed Google Scholar). Finally, we have sequenced the normal rat liver promoter and found that it is about 99% identical to the AS-30D hepatoma promoter (GenBankTM accession number AY082375), rendering it unlikely that liver versus hepatoma differences in HKII expression are related to differences in the nucleotide sequence of the two promoters. Although the above studies demonstrated that a combination of gene amplification and transcriptional events contribute significantly to the marked expression of HKII in highly glycolytic hepatoma cells, they fail to explain why the expression of the enzyme is nearly silent in normal liver (7Mathupala S.P. Rempel A. Pedersen P.L. J. Biol. Chem. 1995; 270: 16918-16925Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar). These findings, and the recent progress in the study of the role of epigenetic factors in the silencing and activation of genes (20Attwood J.T. Yung R.L. Richardson B.C. Cell. Mol. Life Sci. 2002; 59: 241-257Crossref PubMed Scopus (331) Google Scholar, 21Bird A. Genes Dev. 2002; 16: 6-21Crossref PubMed Scopus (5448) Google Scholar, 22Leonhardt H. Cardoso M.C. J. Cell. Biochem. 2000; 35 (suppl.): 78-83Crossref Google Scholar), led us to generate a working hypothesis. Stated simply, our hypothesis envisions that methylation/demethylation events may be involved in regulating HKII gene expression in hepatocytes and highly malignant hepatomas. The results of experiments described below provide substantial support for this working hypothesis. Rats (Sprague-Dawley, female) were obtained from Charles River Breeding Laboratories. Their care and experimental use was approved by and conducted in accordance with the guidelines of The Johns Hopkins University Animal Care and Use Committee. Rat hepatocytes, freshly prepared by the collagenase perfusion method (23Freshney R.I. Culture of Animal Cells: A Manual of Basic Technique. 2nd Ed. Alan R. Liss, Inc., New York1987Google Scholar), were kindly provided by Dr. Anna Mae Diehl, Department of Medicine, The Johns Hopkins University School of Medicine. The normal rat liver (clone 9) cells (American Type Tissue Culture Collection) were grown in 90% DMEM/Ham's F-12 (1:1) with 15 mm HEPES, pH 7.5, l-glutamine, and 10% fetal bovine serum at 37 °C in a humidified atmosphere with 5% CO2. The clone 9 cells were maintained in the exponential growth phase at all times with subculture every 48 h at 1:5 dilution. AS-30D hepatoma cells were grown in the peritoneal cavity of female Sprague-Dawley rats (100–150 g) and were harvested from the ascites fluid 6–7 days post-transplantation as described earlier (6Nakashima R.A. Paggi M.G. Scott L.J. Pedersen P.L. Cancer Res. 1988; 48: 913-919PubMed Google Scholar). The primers (Table I) used in sodium bisulfite sequencing and RT-PCR experiments were synthesized by Invitrogen.Table IOligonucleotides for sodium bisulfite sequencingSequence no.Primer nameSequence 5′–3′1-aBoldface letters indicate primer bases corresponding to cytosine/uracil conversions.OrientationTarget sizeAnnealGenomic position1-bGenBank™ accession number U19605. 1UMF4011TTAGTTTTTTAATATTTTTTGGGTTSense189504013–4037 2UMR4203TACTCCAACACCCAAACAAAAntisense189504201–4182 3TBSUMFTTTGGGTGTTGGAGTAGTSense274484186–4203 4TBSUMRCCTTATCAAAATCACTAAAAAntisense274484459–4440 5BUMFGAAGGTAGGAGTATTATSense170484424–4440 6BUMRCTACTAAAAAACACCTAAAAntisense170484593–4575 7CUMFAAGGAGGAAAATTTGTTTTSense265504630–4648 8CUMRCCTTCTACACTTAATTTTAAAntisense265504894–4875 9DUMFGTTTAATTAAAATTAAGTGTSense271504869–488810DUMRAAAACTTTAAAAATAAAAAAAATAAntisense271505139–51161-a Boldface letters indicate primer bases corresponding to cytosine/uracil conversions.1-b GenBank™ accession number U19605. Open table in a new tab The genomic DNA was obtained from freshly isolated rat hepatocytes and AS-30D hepatoma cells using a genomic DNA isolation kit (Qiagen) according to the manufacturer's protocol. The rat hepatocyte and hepatoma genomic DNA (30 μg) were digested to completion with different methylation-sensitive restriction enzymes, BstUI,HhaI, HpaII, EagI, and ClaI (New England Biolabs), according to the manufacturer's protocol. The digested DNA was fractionated on a 1% agarose gel and subsequently depurinated (0.25 n HCl) for 20 min, denatured (1.5m NaCl, 0.5 m NaOH) for 30 min, and neutralized (0.5 m Tris-Cl, pH 8.0, 1.5 m NaCl) for 30 min. The gel was subsequently soaked in 10× SSC (1.5 m NaCl, 0.15 m sodium citrate, pH 7.0) for 30 min and transferred overnight onto a nylon membrane using TurboblotterTM rapid downward transfer system (Schleicher & Schuell). On the following day, the DNA was fixed onto the membrane by UV cross-linking and hybridized with the 4.3-kb HKII promoter and associated first exon and first intron (7Mathupala S.P. Rempel A. Pedersen P.L. J. Biol. Chem. 1995; 270: 16918-16925Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar). The full-length HKII promoter (7Mathupala S.P. Rempel A. Pedersen P.L. J. Biol. Chem. 1995; 270: 16918-16925Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar) was used to prepare the probe usingGene Images random prime labeling module (Amersham Biosciences) according to the manufacturer's protocol. The blot was prehybridized with buffer (5× SSC, 0.1% w/v SDS, 5% w/v dextran sulfate) at 60 °C for 30 min. The blot was hybridized with a heat-denatured labeled HKII probe (10 ng/ml) at 60 °C for 16–18 h. After stringency washes at 60 °C (once with 1× SSC, 0.1% SDS; once with 0.5× SSC, 0.1% SDS), the methylation-sensitive restriction fragments of the HKII promoter were detected using Gene Images CDP-Star detection module according to the manufacturer's protocol. Sodium bisulfite deaminates unmethylated cytosine to uracil in single-stranded DNA under conditions where the 5-methylcytosine remains nonreactive. Thus, all cytosine residues remaining after PCR amplification and sequencing represent cytosines that were methylated in the original DNA sequence. The genomic DNA was isolated from freshly isolated rat hepatocyte and AS-30D hepatoma cells using a genomic DNA isolation kit (Qiagen) according to the manufacturer's protocol. DNA (10 μg) was digested to completion by BglII (New England Biolabs) at 37 °C and purified using a Wizard DNA Clean-Up System (Promega). The bisulfite reaction was carried out for 16–18 h at 50 °C, pH 5.0, on 1 μg ofBglII-digested genomic DNA from either rat hepatocytes or AS-30D cells using CpGenomeTM DNA modification kit (Intergen) according to the manufacturer's instructions. The modified DNA was finally eluted in 50 μl of TE (10 mm Tris, 0.1 mm EDTA, pH 7.5) and stored at −20 °C for up to 1 month. PCR amplifications were performed using the HotStarTaqTM PCR kit (Qiagen). Sodium bisulfite-treated DNA (100 ng) was amplified in a 50-μl reaction mix containing 200 μm each of the four dNTPs, 30 pmol of each primer, 1.5 mm MgCl2, 1× PCR buffer, 1× Q solution, and 2.5 units of HotStarTaq DNA polymerase (Qiagen). All reagents used were those supplied with the kit. The sequences of strand-specific primers containing the modified cytosine bases together with the annealing temperature used for the amplification of sodium bisulfite-treated DNA are summarized in Table I. The general hotstart thermal cycler program used for all the reactions was as follows: 95 °C for 15 min × 1 cycle; 94 °C for 1 min, 48 or 50 °C for 1 min, 72 °C for 1 min × 40 cycles; 72 °C for 10 min × 1 cycle. The PCR fragments amplified from rat hepatocyte and AS-30D hepatoma modified DNA were cloned using pCR®2.1 TA Cloning® kit (Invitrogen) according to the manufacturer's instructions. The positive clones were sequenced in the Biosynthesis and Sequencing Facility, Department of Biological Chemistry, The Johns Hopkins University School of Medicine. To test the efficiency of bisulfite conversion, the modified DNA was PCR-amplified using modified primers specific for HKII and digested with the restriction enzymes ApoI (R↓AATT↑Y) orTsp509I (↓AATT↑) (New England Biolabs) that cut only modified DNA. These restriction enzyme sites are only generated when cytosine residues are modified to thymidine residues. Subsequently, the efficiency of bisulfite conversion is assessed by complete digestion of a PCR fragment by ApoI orTsp509I. Clone 9 hepatocyte cells that predominantly express high Km glucokinase were used in this study. These cells were seeded at a density of 5 × 105 cells/100-mm dish and maintained in DMEM/Ham's F-12 (1:1) (Invitrogen) and 10% fetal bovine serum as detailed before. The test populations of cells were treated with either 2.5 and 5 μm 5′azaC (Sigma) or 2.5 and 5 μm 5′azadC (Sigma). The cells were harvested after 96 and 120 h of drug treatment, and total RNA was isolated using RNeasy® kit (Qiagen) according to the manufacturer's instructions. RT-PCR was performed using TITANIUMTM One-Step RT-PCR kit (Clontech) according to the manufacturer's protocol. Total RNA (1 μg) from each test sample was used for multiplex RT-PCR in a 50-μl reaction mixture containing 40 mm Tricine, 20 mm KCl, 3 mmMgCl2, 0.2 mm dNTPs, 20 units of recombinant RNase inhibitor (Promega), and 20 pmol each of the HKII-specific primers: HKRTF (5′-GTGTGCTCCGAGTAAGGGTGAC-3′, sense, position 469–490 of HKII cDNA) and HKRTR (5′-CGGTTCGGATGTCATTGAGTG-3′, antisense, position 1023 to 1003 of HKII cDNA), 5 pmol of rat β-actin-specific primers for an internal control: RACTBF (5′-ATATCGCTGCGCTCGTCGTC-3′, sense, position 11–30 of rat β-actin cDNA) and RACTBR (5′-ATCCTGTCAGCGATGCCTGG-3′, antisense, position 938 to 919 of rat β-actin cDNA), and 1× RT-TITANIUMTM TaqEnzyme mix (containing MMLV-RT mutant, TITANIUMTMTaqDNA polymerase and TaqStart antibody). The PCR cycling parameters were 50 °C for 1 h × 1 cycle; 94 °C for 5 min × 1 cycle; 94 °C for 30 s, 65 °C for 30 s, 68 °C for 1 min × 30 cycles; 68 °C for 2 min × 1 cycle. PCR products were electrophoresed on 2% agarose gels, stained with ethidium bromide, and photographed. Clone 9 cells were seeded at a density of 2 × 106 cells/150-mm dish and treated with DNMT inhibitors: 5′azaC (2.5 and 5 μm) and 5′azadC (2.5 and 5 μm) for 120 h as described earlier. Total cell lysate (100 μg) from each test sample was separated by 10% SDS-PAGE. Subsequently, the proteins on the gel were transferred in the cold onto a polyvinylidene difluoride membrane (Bio-Rad) in CAPS buffer (10 mm CAPS, 10% v/v methanol, pH 11) at 100 V/2 h. The membranes were then blocked for overnight at 4 °C with 5% nonfat dry milk in TBST (20 mm Tris, 136 mm NaCl, 0.15% Tween 20, pH 7.6), incubated with rabbit anti-HKII polyclonal antibody (Santa Cruz Biotechnology) at 22 °C for 1 h, followed by 1 h of incubation with a secondary antibody, horseradish peroxidase-conjugated anti-rabbit IgG (Amersham Biosciences). Finally, HKII protein was detected by an ECL system (Amersham Biosciences) according to the manufacturer's protocol. The DNA demethylase (dMTase) cDNA, a kind gift from Dr. M. Szyf (McGill University, Montreal, Canada), had been cloned previously in the mammalian expression vector pcDNA 3.1/His (Invitrogen) containing neomycin for selection of stable transfectants (24Bhattacharya S.K. Ramchandani S. Cervoni N. Szyf M. Nature. 1999; 397: 579-583Crossref PubMed Scopus (546) Google Scholar). Clone 9 cells were maintained in DMEM/Ham's F-12 (1:1) containing 10% fetal bovine serum, as described earlier. The cells were seeded in 6-well plates at a density of 2 × 105 cells/well. The dMTase expression construct (2 μg) was transfected per well using LipofectAMINETM 2000 reagent (Invitrogen) according to the manufacturer's protocol. After transfection for 48 h, the cells were split 1:10 in DMEM/Ham's F-12 (1:1) containing 10% fetal bovine serum and 400 μg/ml G418 (Geneticin®)-selective antibiotic (Invitrogen). The cells (CRLdM) were selected on G418 for 14 days, and viable colonies were expanded for further experiments. The GenBankTM accession numbers for the rat HKII promoter sequence from normal liver and hepatoma cells (AS-30D) are AY082375 andU19605, respectively. As an initial test of our hypothesis that methylation/demethylation events may be involved in regulating HKII gene expression in normal liver and hepatoma cells, we carried out a search for CpG dinucleotide rich “CpG islands” using computer algorithm “CpG Island Finder” (25Takai D. Jones P.A. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 3740-3745Crossref PubMed Scopus (1165) Google Scholar). Such islands frequently contain methylated cytosines in repressed genes (20Attwood J.T. Yung R.L. Richardson B.C. Cell. Mol. Life Sci. 2002; 59: 241-257Crossref PubMed Scopus (331) Google Scholar, 21Bird A. Genes Dev. 2002; 16: 6-21Crossref PubMed Scopus (5448) Google Scholar, 22Leonhardt H. Cardoso M.C. J. Cell. Biochem. 2000; 35 (suppl.): 78-83Crossref Google Scholar). Significantly, a high density of CpG dinucleotides was found in a response element-rich region straddling the transcription start site (Fig.1A). This region (−350 to +781 bp) shown in Fig. 1B contains 58.5% GC content with CGobs/CGexp ratio >0.8 and fits the criteria attributed to a classical CpG island (26Gardiner-Garden M. Frommer M. J. Mol. Biol. 1987; 196: 261-282Crossref PubMed Scopus (2661) Google Scholar). This finding applies to both normal liver and the AS-30D model hepatoma as they show >99% sequence identity (GenBankTM accession numbers AY082375 andU19605). The above analysis identifying a CpG island (−350 to +781 bp) in the HKII promoter raised the question as to whether this segment and perhaps other regions of the promoter are differentially methylated in hepatocytes and hepatoma cells. For this reason, we subjected genomic DNA obtained from freshly isolated hepatocytes and AS-30D cells to digestion with several methylation-sensitive restriction enzymes (BstUI,HhaI, HpaII, EagI, andClaI). The fully digested genomic DNA was subjected to Southern blot hybridization using a probe containing the 4.3-kb HKII promoter with first exon and intron (7Mathupala S.P. Rempel A. Pedersen P.L. J. Biol. Chem. 1995; 270: 16918-16925Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar). With only one exception (EagI), the results obtained (Fig.2) clearly showed more bands in the lanes containing restriction enzyme-digested hepatoma DNA than hepatocyte DNA (compare lanes 1 and 2; 3 and4; 5 and 6; and 9 and10). Moreover, the bands were in different positions in all cases. Thus, these findings strongly implicate hypermethylation of the HKII promoter in hepatocyte genomic DNA as compared with hepatoma DNA. Also, lanes containing digested hepatoma DNA showed much more intense bands than lanes containing the same amount of hepatocyte DNA consistent with our earlier work showing HKII gene amplification in the AS-30D hepatoma cell line (19Rempel A. Mathupala S.P. Griffin C.A. Hawkins A.L. Pedersen P.L. Cancer Res. 1996; 56: 2468-2471PubMed Google Scholar). The observations noted above provided the impetus for subjecting the HKII CpG island (−350 to +781 bp) to sodium bisulfite modification/sequence analyses. Sodium bisulfite converts cytosine to uracil in single-stranded DNA under conditions whereby 5-methylcytosine remains non-reactive. After PCR amplification and sequencing, all cytosines that remain are the ones that were originally methylated. Results presented in Fig. 3, Aand B, provide examples of how these analyses were conducted for the HKII CpG island in hepatocyte and hepatoma DNA, whereas Fig.4 provides a complete accounting of methylated and unmethylated sites in these two cases. Specifically, data presented in Fig. 3A verify that the efficiency of sodium bisulfite treatment of hepatocyte and hepatoma DNA is nearly 100% (“Experimental Procedures”). Thus, examination of lanes 1–6 show that HKII DNA when untreated with bisulfite exhibits a single band (lane 1), which is unaffected by digestion with restrictions enzyme, Tsp509I (↓AATT↑) andApoI (↓AATT↑Y) (lanes 2 and 3). However, when the DNA is modified with bisulfite, Tsp509I completely cuts the DNA to give a smaller fragment (lane 6, arrow), showing nearly 100% efficiency of conversion to this site in bisulfite-modified DNA. Lanes 7–12 present an identical type of control experiment for the hepatoma HKII DNA, comparelanes 10 and 12 (arrow). In experiments not presented here, where a different set of PCR products were generated after bisulfite treatment, the other selected restriction enzyme (ApoI) also completely cleaved the modified but not the unmodified DNA. In all experiments, prior to performing any DNA sequence analysis, the efficiency of bisulfite conversion was assessed, with sequencing being performed only when conversion was complete or very near completion.Figure 4Summary of methylation analysis of the CpG island located in the rat HKII gene in the region encompassing the transcription initiation site. Each strand of sodium bisulfite-treated genomic DNA from rat hepatocyte and hepatoma (AS-30D) cells was PCR-amplified using appropriate sets of modified PCR primers as detailed under “Experimental Procedures.” Subsequently, the PCR products were cloned in pCR® 2.1 vector and sequenced using a M13R primer from the vector backbone. Upper panel, schematic representation of hypermethylated CpG dinucleotides in the rat HKII CpG island. The primary sequence of the HKII CpG island is shown with all CpG dinucleotides marked as vertical lines. The putative motifs for binding of different transcription factors in this CpG island are also shown. Information about the primers used for the amplification of bisulfite-treated DNA and the region amplified are given in Table I. The black circles represent the CpG dinucleotides that show different methylation patterns in the hepatocyte and hepatoma HKII CpG island. The position of CpG dinucleotide with respect to the transcription initiation site has been marked below each black circle. The bent arrow denotes the transcription initiation site. Middleand lower panels, the methylation profile of the HKII gene within the CpG island found in rat hepatocytes and hepatoma cells. The methylation profile of 15 individual, bisulfite-treated clones from hepatocytes (middle panel) and hepatomas (lower panel) is shown. Only the CpG dinucleotides that show differences in methylation between hepatocyte and hepatoma cells have been depicted in the figure. The remaining CpG sites do not show any differences in their methylation pattern between rat hepatocyte and hepatoma cells (data not shown). The open circles represent unmethylated CpGs, and the black circles represent methylated CpGs. The degree of methylation is indicated by a + number (−, 0%; +, 1–25%; ++, 26–50%; +++, 51–75%; ++++, 76–100%). Each row represents a single clone. The methylated CpG dinucleotide mapping of the rat HKII CpG island: −350 to +1 bp (A) and +1 to +781 bp (B).View Large Image Figure ViewerDownload Hi-res image Download (PPT) Results presented in Fig. 3B provide examples of the sequencing data following bisulfite treatment. Here certain cytosines in the hepatocyte HKII CpG island remained unmodified by bisulfite (Fig. 3B, upper panel), implicating methylation, whereas the corresponding regions in the hepatoma CpG island were modified (G/C →A) implicating the absence of methylation (Fig.3B, lower panel). In a more detailed analysis of the CpG island containing 90 potential methylation sites (Fig. 4, A and B), 18 sites (CpG−294, CpG−291, CpG−266, CpG−226, CpG−202, CpG−164, CpG−70, CpG−55, CpG+101, CpG+142, CpG+155, CpG+167, CpG+172, CpG+387, CpG+540, CpG+560, CpG+572, and CpG+717) were found to be methylated to varying degrees in hepatocytes. In sharp contrast, no methylation was observed in the entire CpG island of hepatoma HKII. The remaining 72 CpG sites in the CpG island of the HKII gene showed no methylation in hepatocytes or hepatoma cells. In addition to the above, two other observations of potential interest emerged from these analyses. First, in the hepatocyte CpG island, the CpG sites downstream to the transcription initiation site (CpG+387, CpG+540, CpG+560, CpG+572, and CpG+717) showed higher degrees of DNA methylation (26–75%) as compared with the
Referência(s)