molecular cloning of the promoter for rat hepatic neutral cholesterol ester hydrolase: evidence for...

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 243, 349–355 (1998)ARTICLE NO. RC978030

Molecular Cloning of the Promoter for Rat HepaticNeutral Cholesterol Ester Hydrolase: Evidencefor Transcriptional Regulation by Sterols

Ramesh Natarajan, Shobha Ghosh, and W. McLean GroganDepartment of Biochemistry and Molecular Biophysics, Medical College of Virginia,Virginia Commonwealth University, Richmond, Virginia 23298

Received December 19, 1997

expressed (4,5), characterization of the gene has notNeutral cholesterol ester hydrolase is a key enzyme yet been reported.

in regulating hepatic free cholesterol. Using the CEH Perturbation of cellular cholesterol metabolism hasspecific cDNA sequence in the 5*-untranslated region a marked effect on CEH mRNA, protein and activity.as a primer, we amplified and cloned 1.3 Kb of pro- Cholesterol feeding suppresses both CEH mRNA andmoter sequence upstream of the ATG initiation codon. activity in rats (2,6). Whereas intravenous infusion ofAnalysis of the sequence revealed the presence of a

the cholesterol precursor, mevalonate, decreases CEHconsensus GC-box, which can bind the positive tran-mRNA, protein and activity, infusion of lovastatin, anscription factor Sp1, 35 bases upstream from the tran-HMGCoAR inhibitor, increases both CEH mRNA andscription start site. Transcriptional regulation byactivity (6). Feeding of cholestyramine, a bile acid se-agents perturbing cholesterol metabolism was studiedquesterant, increases hepatic CEH mRNA and activityin HepG2 cells by transient transfection assays of the2-fold (6). Interruption of enterohepatic circulation inpromoter activity in deletion constructs linked to therats by chronic biliary diversion also causes a 2-foldluciferase reporter gene. Three functional sterol re-increase in CEH mRNA and activity (6).sponse sequences were identified at positions -92, -160,

While most of the data reported to date suggest thatand -280 of the CEH promoter. The sterol response se-quence at position -92 was shown to bind SREBP-2. regulation of CEH by sterols is under transcriptionalTherefore, the CEH gene is similar to other genes in- control, there has been no direct evidence to supportvolved in regulation of cholesterol homeostasis, in that this mode of regulation. In the present study, we haveit appears to be transcriptionally regulated by sterols. used hepatic CEH cDNA sequences to isolate the 5*-q 1998 Academic Press flanking region of the CEH gene. In addition, transcrip-

Key Words: cholesterol ester hydrolase; promoter; tional regulation of the CEH gene by sterols was stud-HepG2 cells; gene regulation; sterol response element. ied in human hepatoblastoma HepG2 cells, a cell line

commonly used to study regulation of cholesterol ho-meostasis, by transient transfection assays of the pro-moter activity of deletion constructs linked to the lucif-Hepatic cholesterol ester hydrolase (CEH) catalyzes erase reporter gene. Our results indicate that agentsthe hydrolysis of stored cholesteryl esters to cholesterol that perturb cholesterol metabolism affect the tran-and free fatty acid, thereby regulating the free choles- scription of the CEH gene, mainly through sterol re-terol pool. CEH plays an essential role in the regulation sponse sequences in the proximal 226 bases upstreamof cholesterol homeostasis (1-3), in concert with 3-hy- of the ATG initiation codon.droxy-3-methyl-glutaryl CoA reductase (HMGCoAR),

cholesterol 7a-hydroxylase and acyl-CoA:cholesterolEXPERIMENTAL PROCEDURESacyl transferase. Whereas this enzyme has been puri-

fied from rat liver and its cDNA has been cloned andMaterials. Mevalonolactone and o-nitrophenyl-b-D-galactopyra-

noside were purchased from Sigma Chemical Co. (St. Louis, MO);Promoter Finder DNA Walking Kit, from Clontech (Palo Alto, CA);Abbreviations used: CEH, cholesterol ester hydrolase; HMGCoAR,

3-hydroxy-3-methyl-glutaryl CoA reductase; Luc, luciferase gene; DNA modifying enzymes, tissue culture media and other media sup-plies, from GIBCO-BRL (Grand Island, NY); reporter lysis buffer,NF-Y, nuclear factor-Y; nt, nucleotide; RLU, relative light units;

SRE, sterol regulatory element; SREBP-2, sterol regulatory element luciferase assay system, TA-cloning vector pGEMT-Easy and the lu-ciferase reporter vector pGL3-Basic, from Promega (Madison, WI);binding protein-2.

0006-291X/98 $25.00Copyright q 1998 by Academic PressAll rights of reproduction in any form reserved.

349

AID BBRC 8030 / 6948$$$$81 01-16-98 07:24:27 bbrcg AP: BBRC

Vol. 243, No. 2, 1998 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

reporter vector pCMVb and the MBS Mammalian Transfection Kit, TA-cloned into pGEMT-Easy vector and sequenced by the ABI auto-mated DNA sequencing system. The sequence has been submittedfrom Stratagene Cloning Systems (La Jolla, CA); squalestatin, by

Glaxo Research Group, (Middlesex, United Kingdom, UB6 OHE); to the GENBANK (accession number AF013599)human hepatoblastoma cell line, by American Type Culture Collec-

Primer extension. Radiolabelled primer PAS3 was annealed to rattion (Rockville, MD).liver total RNA (15 mg) and extended with AMV reverse transcriptase

Isolation of rat hepatic CEH promoter. The rat hepatic CEH pro- (1.3U) for 30 minutes at 427C. Primer extension products were ana-moter was isolated by PCR with the Promoter Finder DNA Walking lyzed on an 8% denaturing polyacrylamide gel, followed by autoradi-Kit. Oligonucleotides SEQP8 5*-TACCCCCAAGCTGTGC-ACGCAG- ography . Radiolabelled standards were used to determine the sizeCAAG-3 *, corresponding to positions 56 to 31 of the cDNA for rat of the primer extension product.hepatic CEH, and anchor primer AP1 5*-GTAATACGACTCACT-ATAGGGC-3 * of the promoter finder kit were used in primary PCR Construction of CEH promoter/luciferase reporter genes. In order

to clone the full length promoter into the luciferase reporter vectorwith 5 different libraries of rat genomic DNA as template. PrimaryPCR products were diluted and used in secondary PCR with nested pGL3-Basic, PCR was performed with the primer PAS5BglII 5*- gca-

agatctGATGACAGAAAAGCTCTC-3 * (capitals indicate gene se-primers PAS3 5*-GCGCATTGTG-GAAGGAACAAATAGCCC-3 *, cor-responding to CEH cDNA specific regions 6 to -21 in the 5*-untrans- quence), bearing a Bgl II restriction site (underlined) and correspond-

ing to position -37 upstream of the first ATG codon, and anchorlated portion, and anchor primer AP2 5*-ACTATAGGGCACGCG-TGGT-3 *. PCR and thermal cycling conditions are in manufacturers primerAP2, which has a Mlu I restriction site. The PCR product was

then cloned into the Bgl II-Mlu I sites of pGL3-Basic to give p-instructions. Secondary PCR products from libraries 3 and 4 were

FIG. 1. Nucleotide sequence of the rat CEH promoter. The sequence shown contains 1317 nucleotides of 5*-flanking DNA upstream ofthe ATG initiation codon. Bases in lower case are identical to the reported CEH cDNA sequence. Pertinent cis-acting elements are in boldor in a box. The transcription start site identified by primer extension analysis of total liver RNA is indicated by an asterisk (*). Sequenceidentities of the SRE-1 elements to the consensus SRE-1 sequence are indicated.

350



FIG. 2. Construction of rat CEH/luciferase chimeric genes. Deletion clones were generated by PCR or by Exonuclease III digestion asdescribed in Experimental Procedures. Numbers are relative to the ATG initiation codon of the CEH cDNA.

1317Luc. Other constructs were obtained by generation of unidirec- the initial 20 seconds of reaction. b-galactosidase was assayedwith o-nitrophenyl-b-D galactopyranoside substrate, measuringtional nested deletion breakpoints in p-1317Luc with exonuclease III

(7). All plasmids were verified by restriction digestion analysis and absorbance at 420 nm. Luciferase activity was normalized fortransfection efficiency and extract collection, by ratioing emissionsequencing and purified by double banding in CsCl gradients or with

Qiagen columns per manufacturers instructions. to b-galactosidase activity.

Cell culture and transfections. Human hepatoblastoma HepG2 Gel mobility shift assay. The probe used for the gel mobility shiftcells were grown in 75 cm2 flasks as described by Pandak et al (8). assay was a double stranded oligonucleotide corresponding to posi-Cells were seeded in 35 mm dishes with 2 ml of medium and grown tions -87 to -107 of the CEH promoter. This probe included the puta-to confluence. Cells were then transiently transfected by CaPO4 DNA tive SRE-1 sequence at position -92 (Fig. 1). A 20 ml reaction volume,coprecipitation (9) with the MBS mammalian transfection kit (Stra- containing 25 mM Hepes, pH 7.9, 50 mM KCl, 1 mM EDTA, 5% (w/tagene). Specifically, 2.0 mg of test plasmid and 0.5 mg of pCMVb, v) Ficoll, 1 mM DTT, 0.5 mg poly [d(I.C)], 10% glycerol, 10 fmol (4 1were incubated at room temperature for 10-20 minutes with CaCl2 104 cpm) of the 32P-labelled double stranded oligonucleotide probe(0.125 mM) and BBS (N,N-bis(2-hydroxyethyl)-2-aminoethanesul- and the indicated concentration of SREBP-2 protein, was incubatedfonic acid and buffered saline, pH6.95). Culture medium was re- for 30 minutes on ice and electrophoresed on a 6% polyacrylamideplaced with 2 ml fresh medium containing 6% modified bovine serum gel in 25 mM Tris-borate/ 0.5 mM EDTA buffer (47C, 2-3 hr, 150V)instead of 10% fetal bovine serum. Following addition of the DNA and exposed to Kodak XAR film overnight at 0707C.suspension, cells were incubated for 3 hr at 357C in 3% CO2. Cells

Statistical analysis. Data was analyzed for statistically signifi-were washed 31 with phosphate-buffered saline (PBS) and refedcant differences by Students t-test.with indicated concentration of agent or vehicle in serum-free me-

dium. Transfected cells were incubated at 377C in 5% CO2 for timesstated in figure legends. RESULTS

Reporter enzyme assays. Cells were washed 21 with PBS, lysedand harvested in 300 ml of reporter lysis buffer per manufacturers Cloning of the rat CEH promoter. To isolate the pro-instructions. Luciferase activity was assayed with luciferase re-

moter for the rat CEH gene, we used a PCR basedagent (100 ml) in cell extracts (5 - 20 ml) at 257C in a luminometer(Lumat LB9501, Berthold), measuring total light emission during gene walking method. The primary PCR reaction with

351



FIG. 3. Basal promoter activity of rat CEH/luciferase chimeric genes transfected into HepG2 cells. Confluent HepG2 cultures weretransfected as described under Experimental Procedures and incubated for 40 hrs. in serum-free medium. Cells were harvested and reporterenzyme activities determined as described. Normalized promoter activities are expressed as % control (pGL3-Basic) and represent the mean{ S.E.M. of three independent determinations; * indicates difference at põ0.001.

primers SEQP8 and AP1 yielded products ranging from gene, confluent HepG2 cells were transfected with CEHpromoter/luciferase chimeric genes shown in Fig. 2. As250 bp to 4 Kb in all five ‘‘libraries’’. Primary PCR

products were diluted and used as templates for sec- shown in Fig. 3, full-length and deletion clones weremore active than the promoter-less pGL3-Basic alone.ondary PCR reactions with nested primers PAS3 and

AP2. Libraries 3 and 4 gave approximately 1.3 Kb and CEH/luciferase chimeric genes downstream of, and in-cluding p-599Luc, increased luciferase reporter activity400 bp products, respectively, which were then TA-

cloned into pGEMT-Easy vector and sequenced. The 1.6 to 2-fold as compared to the longest construct, p-1317Luc. This suggests that positive cis-acting ele-1.3 Kb product from library 3 overlapped the 400 bp

product from library 4 and was identical to it in se- ments are located downstream of nt-599 and repressorsequences, upstream of nt-599.quence. Analysis of the 1.3 Kb sequence revealed the

presence of several putative cis-acting elements in theEffect of agents that perturb cholesterol metabolism on5*-flanking region including two sterol responsive ele-

CEH promoter activity. Perturbations of cellular choles-ments (SRE) in the proximal 226 bases upstream fromterol metabolism by agents that increase or decrease lev-the first ATG codon. Fig. 1 shows the nucleotide se-els of intracellular cholesterol have a marked effect onquence of the cloned 1.3 Kb of the promoter for the ratCEH mRNA, protein and activity. Intravenous infusionCEH gene as well as the position of the putative SRE’s.of mevalonate, an agent known to increase intracellularPrimer extension analysis was performed to identifycholesterol levels, causes a compensatory decrease inthe transcription start site (data not shown). Compari-CEH activity and protein mass (6). Moreover, lovastatin,son of the length of the primer extension product to thea potent competitive inhibitor of HMGCoAR, increasesDNA sequence of the 5*-flanking region of the CEHCEH mRNA 2-fold in cultured rat primary hepatocytesgene localized the transcription start site to a ‘G’ nucle-(6). Therefore, to determine the sterol responsiveness ofotide, 60 bases upstream from the initiation ATG codonthe CEH gene, we tested the effect of mevalonate, a pre-(Fig. 1). Although no canonical TATA-box sequencescursor for cholesterol biosynthesis, and squalestatin, anwere found around the start site, a consensus GC-box,inhibitor of squalene synthase and cholesterol biosynthe-which is a putative binding site for Sp1, was found 35sis, on the activity of the CEH promoter. Mevalonate (2bases upstream from the start site.mM) treatment for 24 hrs. repressed reporter gene activ-ity of p-1317Luc and p-1190Luc by 74% and 84% respec-Determination of basal CEH promoter activity. To

determine the basal promoter activity of the rat CEH tively (Fig. 4A). Luciferase reporter activity was restored

352



FIG. 4. Effect of cholesterol perturbing agents on promoter activity of rat CEH/luciferase chimeric genes. Confluent HepG2 cultureswere transfected with rat CEH/luciferase promoter constructs and incubated for 24 hrs. in serum-free medium in the presence or absenceof 2 mM mevalonolactone alone (A) or 2 mM mevalonolactone and 1 mM squalestatin (B). Normalized promoter activities are expressed as% control and represent the mean { S.E.M. of three separate observations; * indicates difference at põ0.001; ** indicates difference atpõ0.005; *** indicates difference at põ0.01.

to basal levels in p-859Luc and p-540Luc. While activity Inasmuch as squalestatin did not reverse the inhibitionassociated with p-1190Luc, it is possible that the regionof p-226Luc was inhibited 57% by mevalonate, activity ofbetween nt-1190 and nt-860 is uniquely responsive top-418Luc was only repressed 30%, suggesting that sterolnon-sterols.response elements (SRE) are present between nt-1190

and nt-860, nt-418 and nt-227, and between nt-226 and Gel mobility shift assay. Binding of SREBP-2 tont-37. Simultaneous treatment for 24 hrs. with 2 mM CEH promoter sequences was verified by a mobilitymevalonate and 1 mM squalestatin restored the activity shift assay. As seen in Fig. 5, electrophoretic mobility ofof p-226Luc and p-418Luc (Fig. 4B), but failed to bring the probe (nt-87 to -107) containing the putative SRE-1the activity of p-1190Luc back to basal levels. These data site was retarded by binding of SREBP-2. The retarda-

tion pattern was similar to that obtained by the bindingsuggest the presence of SRE’s between nt-226 and nt-37.

353



HepG2 cells. Mevalonate was used to stimulate synthe-sis of both sterols and non-sterol intermediates in thecholesterol biosynthetic pathway and squalestatin wasused to selectively lower sterol endproducts. Mevalo-nate has been shown to decrease HMGCoAR transcrip-tion (12) and both HMGCoAR and CEH steady-statemRNA levels (6,12). In the present study, mevalonatedecreased CEH promoter activity of p-226Luc, p-418Luc and p-1190Luc by 57%, 30% and 89% respec-tively. In contrast, simultaneous treatment withsqualestatin and mevalonate, restored the promoter ac-tivity of p-226Luc and p-418Luc, indicating transcrip-tional regulation of the CEH gene by sterols and func-tional sterol responsive sequences located between nt-226 and nt-37 and between nt-418 and nt-227. TwoSRE-1 like sequences are present between nt-226 andnt-37: CACCGAAC at position -92, which has a 7/8 basematch, and CACCGATC at position -160 (Fig.1), whichhas a 6/8 base match with the consensus SRE-1 se-quence CACC(C/G)(C/T)AC, in promoters for HMGCoAsynthase, HMGCoAR and the LDL receptor (13). Simi-larly, the region between nt-418 and nt-227 has oneSRE-1 sequence, CACCGTTC at position -280 (Fig.1),which also has a 7/8 base match with the consensusSRE-1 sequence. Thus, these putative SRE’s probablymediate the observed regulation of the CEH gene bysterols. The SRE-1 sequences of the LDL receptor andHMGCoA synthase promoters, which bind multipleforms of SREBP, are responsible for regulation of these

FIG. 5. Binding of SREBP-2 to rat CEH promoter. A double genes by sterols. In the present study, we have shownstranded probe corresponding to nucleotides -87 to -107 of the ratbinding of one of these proteins (SREBP-2) to the SRE-CEH promoter was end labelled with [g-32P]ATP and incubated with1 sequence at position -92 of the CEH promoter, provid-various amounts of SREBP-2, as indicated under conditions de-

scribed in ‘‘Experimental Procedures.’’ Arrows indicate the shifted ing further evidence for a functional regulatory ele-probe, bound to SREBP-2, and the free probe. ment and suggesting that SREBP-2 may mediate regu-

lation of CEH by sterols.Sterol response elements are often located in close

proximity to other positive elements that are requiredof SREBP-2 to a probe containing one copy of the SRE-for SRE function. SRE’s of LDL receptor (14) and fatty1 element of the LDL receptor promoter (data notacid synthase (15) promoters require Sp1 sequencesshown). It therefore appears that the SRE-1 sequencefor expression of sterol responsiveness. In contrast, theat position -92 of the CEH promoter confers sterol re-HMGCoAR promoter has multiple nuclear factor-1sponsiveness mediated by binding of SREBP-2.(NF-1) sites, which are necessary for SRE activity (16).By analogy, the CEH promoter has potential Sp1 andDISCUSSIONNF-Y sites near the SRE-1 sites (Fig. 1), which may berequired for SRE function. Thus, the CEH promoterThis is the first reported evidence for a unique CEH

gene. The CEH promoter is similar to that of has structural and sequence characteristics commonto other known sterol-responsive promoters, consistentHMGCoAR (10) and human squalene synthase (11)

promoters in that it has no consensus TATA-box like with transcriptional regulation of CEH by sterols.Although the sequence between nt-1190 and nt-859sequences immediately upstream of the transcription

start site . Like the HMGCoAR and the human squa- also inhibited promoter activity in the presence of me-valonate, activity was only partially restored bylene synthase promoters, the CEH promoter has a GC-

box sequence, which can bind the positive transcription squalestatin. Thus it is possible that this region has acis-acting element that responds to non-sterol interme-factor Sp1 to drive transcription, and an inverted

CCAAT box, which can bind another positive transcrip- diates in cholesterol biosynthesis. The existence of sucha ‘‘non-sterol’’ response element would be contrary totion factor, NF-Y (Fig.1).

In this study, known cholesterol perturbing agents the widely held view that non-sterol intermediates actonly at the post-translational level. Identification of thewere used to alter intracellular cholesterol levels in

354



5. Ghosh, S., Mallonee, D. M., Hylemon, P. B., and Grogan, W. M.non-sterol response element and its putative DNA(1995) Biochim. Biophys. Acta 1259, 305–312.binding proteins will enable us to elucidate this novel

6. Ghosh, S., Natarajan, R., Pandak, W. M., Hylemon, P. B., andmechanism of regulation. Grogan, W. M. (1998) Am. J. Physiol., in press.7. Henikoff, S. (1987) Meth. Enzymol. 155, 156–165.8. Pandak, W. M., Stravitz, R. T., Lucas, V., Heuman, D. M., andACKNOWLEDGMENTS Chiang, J. Y. L. (1996) Am. J. Physiol. 270, G401–G410.9. Graham, F. L., and Van der Eb, A. J. (1973) Virology 52, 456–

This work was supported by a grant from the National Institutes 467.of Health (DK44613). SREBP-2 was graciously provided by Dr. Gre- 10. Reynolds, G. A., Basu, S. K., Osborne, T. F., Chin, D. J., Gil, G.,gorio Gil. Brown, M. S., and Goldstein, J. L. (1984) Cell 38, 275–285.

11. Guan, G., Jiang, G., Koch, R. L., and Schecter, I. (1995) J. Biol.Chem. 270, 21958–21965.

REFERENCES 12. Goldstein, J. L., and Brown, M. S. (1990) Nature 343, 425–430.13. Smith, J. R., Osborne, T. F., Brown, M. S., Goldstein, J. L., and

Gil, G. (1988) J. Biol. Chem. 263, 18480–18487.1. Ghosh, S., Kounnas, M. Z., and Grogan, W. M. (1990) Lipids 25,14. Dawson, P. A., Hofmann, S. L., Van der Westhuyzen, D. R., Sud-221–225.

hof, T. C., Brown, M. S., and Goldstein, J. L. (1988) J. Biol.2. Grogan, W. M., Bailey, M. L., Heuman, D. M., and Vlahcevic, Chem. 263, 3372–3379.Z. R. (1991) Lipids 26, 907–914.15. Bennett, M. K., Lopez, J. H., Sanchez, H. B., and Osborne, T. F.

3. McGovern, R. F., and Quackenbush, F. W. (1973) Lipids 8, 473– (1991) J. Biol. Chem. 270, 25578–25583.478. 16. Osborne, T. F., Gil, G., Goldsteil, J. L., and Brown, M. S. (1988)

J. Biol. Chem. 263, 3380–3387.4. Ghosh, S., and Grogan, W. M. (1991) Lipids 26, 793–798.

355


molecular cloning of the promoter for rat hepatic neutral cholesterol ester hydrolase: evidence for...

Documents