cis-regulatory elements and trans-acting factors directing basal and camp-stimulated human renin...

P Borensztein, S Germain, S Fuchs, J Philippe, P Corvol and F PinetcAMP-stimulated human renin gene expression in chorionic cells

cis-regulatory elements and trans-acting factors directing basal and

ISSN: 1524-4571 Copyright © 1994 American Heart Association. All rights reserved. Print ISSN: 0009-7330. Online

TX 72514Circulation Research is published by the American Heart Association. 7272 Greenville Avenue, Dallas,

1994, 74:764-773Circulation Research

http://circres.ahajournals.org/content/74/5/764located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://www.lww.com/reprintsReprints: Information about reprints can be found online at [email protected]. E-mail:

Fax:Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters http://circres.ahajournals.org//subscriptions/Subscriptions: Information about subscribing to Circulation Research is online at

by guest on July 14, 2011http://circres.ahajournals.org/Downloaded from

http://circres.ahajournals.org/

764

Original Contributions

cis-Regulatory Elements and trans-Acting FactorsDirecting Basal and cAMP-Stimulated HumanRenin Gene Expression in Chorionic Cells

Pascale Borensztein, Stephane Germain, Sebastien Fuchs, Josette Philippe,Pierre Corvol, Florence Pinet

Abstract Much knowledge was accumulated in the regula-tion of plasma renin activity and renin secretion during recentyears. However, the mechanisms of renin gene transcription,especially for the human gene, have been poorly studiedbecause of the lack of cell lines expressing renin. Cells derivedfrom chorion tissue were used to study renin gene transcrip-tion because these cells express renin and regulate reninsecretion in a similar way to JG cells. The present study wasperformed to determine the cis-regulatory elements and thetrans-acting factors involved in human renin gene expressionusing chorionic cells. Transient DNA transfections were per-formed with various constructs containing the 5'-flankingregion of the human renin gene. 5'-Deletion analysis of thehuman renin promoter (from -2616 to -67 bp) revealed thepresence of two proximal negative cis-regulatory elementsbetween -374 and -273 bp and between -273 and -137 bp.These elements were not present in a non-renin-producingcell line, JEG-3 cells. DNase I footprinting revealed that twosequences located within these regions bind trans-factorspresent in chorionic cellular nuclear extract: AGE3-like se-quence (-293/-273) and apolipoprotein Al regulatory pro-

tein-1-like sequence (-259/-245). The first 110 bp of therenin promoter were sufficient to direct specific expression inchorionic cells and contained two footprints sharing homologywith ets (-29/-6) and pituitary-specific factor (Pit-1) (-70/-62)sequences. Furthermore, one footprint (-234/-214) containedthe sequence TAGCGTCA, which shares strong homology to thecAMP-responsive element (CRE) binding site. Gel shift analysisshowed specific DNA/protein complexes within this region,which were displaced by the somatostatin consensus CRE. Fi-nally, luciferase analysis of 5'-deletion mutant revealed that -273to +16 bp of the renin promoter was sufficient to confercomplete forskolin stimulation, whereas deletion to -130 (dele-tion of the CRE) decreased cAMP responsiveness by 50% andthose to -67 bp (deletion of the CRE and Pit-1-like sequences)suppressed it. Thus, these latter two sequences probably acttogether to confer complete cAMP responsiveness. (Circ Res.1994;74:764-773.)Key Words * renin gene * chorionic cells * luciferase -

cis-acting elements * trans-acting factors * cAMP-responsiveelement * pituitary-specific factor * ets

R enin (E.C.3.4.23.15), an aspartyl protease thatcleaves angiotensin I from angiotensinogen inthe rate-limiting step of the renin-angiotensin

system, plays a major role in blood pressure regulationand fluid and electrolytic homeostasis.During recent years, much knowledge has accumu-

lated concerning the numerous factors influencing reninsecretion, such as f3-adrenergic agonists, peptidic hor-mones, and drugs.' Several intracellular mediators in-fluencing renin release, such as cAMP, cGMP, andintracellular calcium, have also been identified.2 Most ofthese studies were performed in isolated perfused kid-ney, kidney slices, or juxtaglomerular cell cultures be-cause kidney-derived renin represents most of the cir-culating enzyme. Another important step in the controlof renin secretion is the rate of transcription of the reningene. However, studies examining this step are difficultto perform, in particular because of the lack of estab-lished renin-producing cell lines.The transcriptional regulatory elements of the mouse

renin genes have been largely characterized. One of theinteresting features of this species is the presence of aduplicated renin gene in some strains, eg, DBA/2J. The

Received October 8, 1993; accepted January 10, 1994.From INSERM Unit 36, College de France, Paris.Correspondence to Dr Florence Pinet, INSERM Unit 36,

College de France, 3 rue d'Ulm. 75005 Paris. France.

RenJ gene is expressed mainly in the kidney, whereasthe Ren2 gene is expressed in both the kidney and thesubmandibular gland. Nakamura et a13 have shown thepresence of a negative regulatory element in the 5'-flanking region of both mouse renin genes; this elementis functional in the Renl but not the Ren2 gene, prob-ably because of the interference by an adjacent 150-bpinsertion in this latter gene.4 This difference in function-ality may play a role in the differential expression of therenin genes in the submandibular gland. First, thisnegative regulatory element could bind a specific nu-clear protein, present in the submandibular gland, re-sulting in the inhibition of Reni expression in thistissue.4 Second, the high expression of submandibularRen2 could be due to the nonfunctionality of thenegative regulatory element.4 Recently, Tamura et al'identified, by transient DNA transfection, two positivecis-acting sequences located within a large insertion inthe mouse gene and not present in the human gene,6suggesting that transcriptional regulation of the reningene could differ between species. In addition, theyshowed that renin gene expression depends strongly onthe cell model used and on the presence or absence oftrans-acting factors in these cells.Very little is known about the mechanisms controlling

human renin gene expression, mainly because of thelack of established renin-producing cell lines. Human



Borensztein et al cis-Acting Elements in the Human Renin Gene 765

renin is expressed in many cell types,7 but its expressionis particularly high in juxtaglomerularl and chorioniccells.8-'0 Juxtaglomerular cells are difficult to isolate andto culture, and they lose their ability to produce reninafter a single passage.1',12 Attempts to immortalizethese cells by the use of different SV40 mutants resultedin limited expression of the human renin gene.'3 Al-though these cells allowed the study of some of themechanisms involved in renin secretion, they cannot beused for renin gene expression studies. Chorion tissuerepresents the second major site of production, afterkidney, in human tissues.10 Moreover, chorionic cells inprimary and secondary cultures produce large quanti-ties of renin, predominantly in the prorenin form, andare easily accessible.14 Renin regulation in this modelwas studied and showed a role of cAMP and proteinkinase C.15 This is similar to the regulation demon-strated in primary cultures of JG cells.2 Therefore,these cells may provide a useful model for studying theregulation of renin gene expression.

Using transient DNA transfections, Duncan et a116reported that human chorionic cells are able to use thefirst 600 bp of the renin promoter to direct chloramphen-icol acetyl transferase (CAT) expression without needingan enhancer. Other studies have been performed usingthe JEG-3 cell line, which is derived from a humanchoriocarcinoma.17'18 However, this model has severallimitations: these cells do not express the renin gene,18'19and analysis of the 5'-flanking region of the gene requiredthe use of the herpes simplex virus thymidine kinase (tk)promoter to direct CAT expression.'7'18

There is a single recent study on a trans-acting factorcontrolling human renin gene expression showing thatnuclear extracts isolated from a rat lactotrope precursorcell line (GC) interact with a pituitary-specific factor(Pit-1) binding site consensus sequence.20 Since Pit-1 isexpressed only in pituitary cells,21 the functionality ofthis cis-acting sequence remains to be shown in renin-expressing cells.The present study was performed to determine the

cis-regulatory elements and the trans-acting factorsinvolved in human renin gene expression, using renin-producing cells derived from chorionic tissue. The pro-moter elements involved in the transcription of thehuman renin gene in chorionic cells were localized bydifferent approaches. Luciferase assays in human chori-onic cells transiently transfected with constructs madeby sequential deletions of the promoter region allowedthe mapping of basal and cAMP-induced renin geneexpression. Footprinting analysis of the human reninpromoter, reported here for the first time, revealed thepresence of several cis-regulatory elements within theproximal renin promoter. These consisted of sequenceswith varying degrees of homology with, respectively, thePit-i-like binding site consensus sequence, the cAMP-responsive element (CRE), the apolipoprotein Ai reg-ulatory protein-1 (ARP-1) binding site, an ets bindingsite, and the AGE3 sequence on the angiotensinogengene. Finally, gel mobility shift analysis showed that theCRE sequence specifically binds nuclear extracts fromchorionic cells.

Materials and MethodsMaterials

All DNA-modifying enzymes were purchased from NewEngland Biolabs. Oligonucleotides were synthesized with an

Applied Biosystems synthesizer. Reverse transcriptase, TaqDNA polymerase, ATP, and culture media were obtainedfrom Boehringer Mannheim. [a-'2P]dCTP and [y-`P]ATPwere purchased from Amersham. Purified luciferase fromPhotinus pyralis, synthetic D-luciferin, and forskolin were ob-tained from Sigma.

Cell Culture and TransfectionsPrimary and secondary cultures of human chorionic cells

were prepared essentially as previously described.'4 Briefly,the chorion was separated from other membrane layers anddigested by successive incubation periods in 0.05% collagenaseand 0.05% trypsin. After filtration (45-,um nylon mesh) andcentrifugation, the cell pellets were resuspended in Dulbecco'smodified Eagle's medium (DMEM) supplemented with strep-tomycin (10 gg/mL), penicillin (10 UI/mL), 4 mmol/L gluta-mine, and 10% fetal calf serum (FCS) and then plated onto the25-cm2 flasks containing 4 mL DMEM. When the cells wereconfluent, the first subculture was carried out: chorionic cellswere treated with 0.05% collagenase and then with 5 mmol/LEDTA. The cell suspension was centrifuged, and the resultantcell pellets were resuspended in DMEM supplemented asabove and then plated at a 1:4 dilution into similar flaskscontaining DMEM.The human choriocarcinoma cell line (JEG-3) was obtained

from the American Type Culture Collection and maintained inDMEM supplemented as above.

Transient DNA transfections into chorionic and JEG-3 cellswere performed by calcium phosphate precipitation22 using 25 ,ugDNA per plate consisting of 5 gg RSV-CAT (as an internalcontrol for plate-to-plate transfection efficiency) and 20 jig reninluciferase reporter plasmid. After overnight incubation, the cellswere treated for 2 minutes with a 15% glycerol shock. Themedium was then replaced with serum-free defined mediumcontaining 50% Dulbecco's medium and 50% Ham F12 supple-mented with streptomycin (10 gg/mL), penicillin (10 UI/mL),selenium (3x 10`8 mol/L), palmitic acid (1 gg/mL), oleic acid (5gg/mL), linoleic acid (5 gg/mL), bovine serum albumin free fattyacid (1 mg/mL), transferrin (5 ,ug/mL), and glutamine (4 mmol/L). The cells were incubated for 24 hours with or withoutforskolin 10` mol/L, and then the luciferase activity was deter-mined and the CAT immunoassay performed.

Direct renin immunoassay was performed on culture me-dium by enzyme-linked immunosorbent assay (ELISA) with asensitivity limit of 20 pg/mL.23

Reporter AssaysTransfected cells were washed three times with phosphate-

buffered saline (PBS), lysed by addition of 0.1 mL 1% Triton, 10mmol/L MgCl2, 1 mmol/L EDTA, 25 mmol/L Tris-phosphate(pH 7.8), 15% glycerol, 1 mmol/L dithiothreitol (DTT), and 0.2mmol/L phenylmethylsulfonyl fluoride, and harvested by scrap-ing. The lysates were transferred to Eppendorf tubes, and aftercentrifugation, the supernatant was saved. Luciferase activity wasmeasured by a liquid scintillation counter (model 1211 Rackbeta,LKB). Luminescence was integrated for 1 minute after additionof 150 ,umol/L luciferin and 400 gmol/L ATP.24

Quantitative determination of CAT was performed by asandwich enzyme immunoassay (Boehringer Mannheim).During the first step, CAT contained in cell extracts bindsspecifically to modules coated with anti-CAT antibodies.Then, a digoxigenin-labeled anti-CAT antibody is bound to thefixed CAT, and during the last step, the digoxigenin-labeledanti-CAT antibody is detected by a peroxidase-labeled anti-digoxigenin antibody.

Plasmid ConstructionHuman renin gene fragments were isolated from the two

plasmids phrnCAT30 and phrnCAT06, kindly provided by DrFukamizu (University of Tsukuba, Japan) and derived fromthe plasmid subclone AHRn88.25 The plasmid pRSVL, desig-



766 Circulation Research Vol 74, No 5 May 1994

nated here as pRSV-luciferase and in which luciferase tran-scription is driven by the Rous sarcoma virus long-terminalpromoter, was provided by Dr Swesh Subramani (versionLpJD201).26 The promoterless plasmid (BS luci) was con-structed by subcloning a HindIII/BamHI-digested fragment ofLpJD201 (corresponding to the intronless luciferase constructwith the SV40 small t antigen intron and an SV40 polyadenyl-ation signal) into the polylinker region of the plasmid Blue-Script SK (Stratagene). The plasmids p582+ and p582- wereconstructed as follows: a 598-bp fragment of the human reningene (-582 to + 16) was isolated by Bgl II/HindIll digestion ofphrnCATO6 and subcloned in either orientation into theHindlIl site of BS luci. The resulting p582+ plasmid containednucleotides -582 to + 16 of the renin gene followed by theluciferase coding sequence. The plasmid containing the an-tisense insert (+16 to -582) was designated as p582-. Theplasmid p892+ was constructed as follows: a 908-bp fragmentof the human renin gene (-892 to +16) was isolated byHindIII digestion of phrnCAT30 and inserted into the HindlIlsite of BS luci. The plasmid p2616+ was constructed byisolating a 2632-bp fragment of the human renin gene (-2616to +16) by Bgl II/BamHI/Pst I digestion of phrnCAT30 andinserted into the HindIII site of BS luci.

5-Deletion mutants were generated by linearization of thep582+ plasmid by Xho I and Apa I, followed by exonucleaseIII digestion, S1 nuclease digestion, repair, and self-ligation.The 5-extent of the exonuclease III digestion is given innucleotide pairs corresponding to the published sequence.25The plasmid p110- was constructed as follows: a 127-bpfragment (-110 to + 16) of the renin gene was isolated by KpnI and Hindlll digestion of the plasmid p110+ (previouslyobtained by the exonuclease III digestion above) and insertedinto the Hindlll site of BS luci. All constructions were verifiedby the dideoxy sequencing method (Sequenase Version 2.0DNA Sequencing Kit, USB).

Reverse Transcriptase Reaction and PolymeraseChain Reaction

Total RNA was extracted from chorionic and JEG-3 cells bythe method of Chomczynski and Sacchi.27 Total RNA (500 and50 ng) was used to prepare cDNA with the M-MLV reversetranscriptase (200 U/,uL, GIBCO-BRL) in the appropriatebuffer (Boehringer Mannheim) in the presence of 7.5 mmol/LoligodT, 0.5 U/,L of the RNase inhibitor rRNasin (40 000U/mL, Promega), 50 mmol/L DTT, 1 ,ug yeast tRNA, and 2.5mmol/L of each deoxyribonucleotide. After 1 hour of incuba-tion at 37°C, the reaction was stopped by heating the samplesfor 2 minutes at 95°C.The polymerase chain reaction (PCR) was carried out as

described by Caroff et al.28 Briefly, 3 ,uL of the final cDNAsolution was mixed with 1 ,uL (10 pmol) of both primers, 4.54L of 25 mmol/L MgCl2 solution, 1 ,L of a 25 mmol/L solutionof each deoxyribonucleotide, 5 ,uCi of [a-32P]dCTP, 14 ,uLH20, and 2.5 ,L of 10x PCR buffer (supplied with 1.25 U ofTaq polymerase). Thirty cycles of PCR were performed,consisting of denaturation at 94°C, annealing at 48°C, andextension at 72°C. After PCR, polyacrylamide gel electropho-resis was performed with the amplification products originat-ing from the renin mRNA, followed by autoradiography. Toavoid amplification of genomic DNA coding for renin, the twoprimers 5'(GTGTCTGTGGGGTCATCCACCTTG)3' and5'(GGATTCCTGAAATACATAGTCCGT)3' were chosen,the first sequence being present in exon 7 of the renin gene andthe second spanning the exon 8/exon 9 border.

Quantitative reverse transcriptase PCR was also carried outas described previously28 by use of an internal standard. Serialdilutions of total RNA (from 500 to 30 ng) from chorionic andJEG-3 cells were used with a fixed amount of internal standard(10 pg) to prepare cDNA before PCR. Bands corresponding toPCR products were excised and counted in a 8-counter.

Primer Extension AnalysisTotal RNA was isolated from human chorionic cells, after

transfection as described above with the plasmids BS luci,pS82+, and p110+, according to the method of Chomczynskiand Sacchi.27 Primer extension reaction was carried out ac-cording to the protocol of the AMV reverse transcriptaseprimer extension system (Promega). Briefly, RNA was resus-pended in 5 ,uL H20 and incubated with 1 uL (100 fmol) of32P-labeled luciferase primer (corresponding to the sequence+37 to +63 of the luciferase gene) and 5 pL of AMV primerextension 2X buffer. Primer and RNA were annealed byheating at 85°C for 5 minutes, then at 60°C for 2 hours. AMVreverse transcriptase was added to the same buffer withsodium pyrophosphate (6 mmol/L) and incubated at 42°C for1 hour. The reaction was stopped by addition of loading dye,and the primer extension products were analyzed on denatur-ing 8% polyacrylamide/7 mol/L urea gel.

Preparation of Nuclear Extracts and DNase IFootprinting AssaysHuman chorionic cell nuclear extracts were prepared as

previously described.29 The -336 to + 16 fragment of the reningene promoter was isolated by Avr II/Hindll digestion of thep582+ plasmid and inserted into the EcoRV site of Blue-Script SK. The synthetic DNA fragment corresponding to-336 to + 16 was obtained by PCR amplification using theBlue-Script SK and KS universal primers. The SK or KSprimer was labeled with [y-32PJATP and T4 polynucleotidekinase before PCR amplification. Footprint analysis was per-formed in a 10 ,L reaction mixture containing 4 mmol/LMgC12, 4 mmol/L spermidine, 10 mmol/L HEPES (pH 7.9), 50mmol/L KCl, 0.1 mmol/L EDTA, 0.1 mmol/L EGTA. 2.5%glycerol, 0.5 ,ug of double-stranded poly(dI-dC), 30 ,ug ofnuclear extract, and 15 000 cpm of end-labeled fragment.After 20 minutes of incubation at 4°C, 2 ,L of DNase I(Boehringer Mannheim, 10 IU/,uL) at various dilutions rang-ing from 1150 to 1/400 was added, and digestion was allowed toproceed for 1 minute at 20°C. The reaction was stopped byaddition of 30 pL of a solution containing 50 mmol/L EDTA,0.1% sodium dodecyl sulfate, 0.2 mg/mL yeast tRNA, and 10mg/mL of proteinase K. The reaction mixture was incubatedfor 45 minutes at 42°C. The DNA was extracted once withphenol/chloroform, precipitated with 2 volumes of ethanol,resuspended in 98% formamide dye, and electrophoresed on a6% acrylamide/7 mol/L urea sequencing gel.

Gel Mobility Shift AssaysThe following double-stranded synthetic oligonucleotide, cor-

responding to the -234 to -200 renin promoter (containing theCRE-like sequence, underlined), was used for gel shift analysis(only the + strand is shown): REN, 5'-GAGGGCT-GCTAGCGTCACTGGACACAAGATTGCTTT-3'. The fol-lowing double-stranded oligonucleotide of the rat somatostatinpromoter containing the consensus CRE30 was used as a com-petitor: SMS, 5'-CTGGGGGCGCCTCCTTGGCTGCT-GACGTCAGAGAGAGAG-3'. The REN oligonucleotide wasend-labeled with [y-32PJATP and T4 polynucleotide kinase. Nu-clear extract (7 ,mg) was incubated for 15 minutes at 4°C in 18 4Lof a reaction mixture containing 10 mmol/L HEPES (pH 7.8), 1mmol/L Na2HPO4 (pH 7.2), 0.1 mmol/L EDTA, 50 mmol/L KCI,4 mmol/L MgC12, 4 mmol/L spermidine, 2.5% glycerol, 2 ,ug ofdouble-stranded poly(dI-dC), 1 ,ug of salmon sperm DNA, and20 000 cpm of labeled double-stranded renin oligonucleotide inthe presence or absence of 2- to 50-fold excesses of competitoroligonucleotides. The protein-DNA complexes were analyzed bynondenaturing electrophoresis through 6% polyacrylamide gelsrun in 0.25 x Tris borate EDTA.

Statistical AnalysisAll results are given as mean±+SEM. Levels of significance

were calculated by Student's t test; P<.05 was consideredsignificant.




% relative luciferase activityFIG 1. Bar graph showing transient

0 2 4 6 8 10 transfection analysis of renin/luciferase+16 L * * (luci) constructs in chorionic and JEG-3

-2676 =W p2616. cells. Transient cotransfections of renin/-2616 p2616

luciferase and RSV-chloramphenicolacetyl transferase (CAT) reporter gene

+16 were assayed in chorionic (stippled-892 -*p98~p892+ bars) and JEG-3 cells (solid bars). Lu-

ciferase activity was normalized to co-+16 transfected RSV-CAT activity. Results

-582 - Wuci p582* are given as mean±SEM of the percent-age of RSV luciferase activity to adjust

-582 for differences in transfection efficiency.___6______lu~ci~ p582Each bar represents two to five indepen-16 luciS p582- *dent transfections in triplicate flasks.

ResultsHuman Renin Gene Expression inTransfected CellsHuman renin gene expression was evaluated by tran-

sient DNA transfections into two cell types derivedfrom human chorionic tissue: chorionic cells in second-ary culture, which expressed the renin gene to a highextent (renin content of the culture medium measuredby ELISA reached 20 ng - mL- 1 24 hours-1), andJEG-3 cells, which did not express it (no renin detectedin the culture medium). Plasmids containing the 5'-flanking region of the human renin gene (up to + 16relative to the transcription start site) were fused to theluciferase reporter gene, which was preferred to theCAT gene because of its higher sensitivity.31 To normal-ize for plate-to-plate differences in transfection effi-ciency, a reporter plasmid RSV-CAT was cotransfectedand assayed independently. Luciferase activity was nor-malized to the cotransfected internal control RSV-CATand plotted as a percentage of the RSV luci signal foreach transfection.

Fig 1 shows the basal activity of the renin promoter. Inchorionic cells, there was a moderate but not significantdecrease in luciferase activity with shorter plasmids fromp2616+ to p582+. The basal luciferase activities ofp582+, p892+, and p2616+ were 4.65%, 7.26%, and8.43% of RSV luci, respectively. This suggests that majorbasal cis-regulatory elements are located downstream-582 bp. Two controls were performed: (1) The antisenseplasmid p582- did not express luciferase activity(0.1+0.1%, n=2) compared with the sense plasmid p582+(4.65±2.08%, n=6). (2) In JEG-3 cells, the renin lu-

-582 ` 16

164o(.-582-501 >

-455 >

-273 - *>-137 >-110 >+16 -110

-67

p582ip582 -

p501+

p455+

p374+p273.

pl 37+

p110+

p110 -

p67+

ciferase fusion reporter gene displayed very low activity(0.23%, 0.17%, and 0.04% of RSV luci for the sameplasmids, respectively) compared with that in chorioniccells.To locate cis-regulatory elements, sequential dele-

tions of the 5'-flanking region of the p582+ plasmidwere performed, and the resulting renin-luciferase con-structs were transfected into chorionic cells. Resultswere expressed as percentage of the activity of thep582+ plasmid to normalize the results of differenttransfections. Fig 2 shows that progressive 5'-deletionsextending to nucleotide - 374 had no effect on promoteractivity. In contrast, deletions extending to nucleotide-273 gave a twofold stimulation above that of the renin5'-flanking region of 582 bp. Further deletion of se-quences between -273 and -67 greatly increased pro-moter activity: the plasmids pl37+, p110+, and p67+demonstrated threefold to fivefold increases in lu-ciferase activity compared with p582+ plasmid. Thespecificity of the renin promoter activity was confirmedby the very low luciferase activity observed with thep110- plasmid, where the renin DNA was inserted inthe antisense orientation. The 5'-sequences withinp110+ were sufficient to direct specific expression ofthis renin gene construct in the chorionic cells, sinceJEG-3 cells transfected with p110+ plasmid did notexhibit the same high level of expression as identicallytransfected chorionic cells (Table). The p67+ plasmiddemonstrated a 1.6-fold increase in luciferase activity,suggesting that the 67 bp of renin promoter are notsufficient to direct specific expression of renin gene.Smith and Morris18 described that JEG-3 cells ex-pressed no detectable renin mRNA by Northern blot

FIG 2. Bar graph showing 5-deletion% relative luciferase activity analysis of the human renin gene pro-

moter. 5-Deletion of the p582+ plas-0 100 200 300 400 500 600 mid was performed by exonuclease Ill

as described in "Materials and Meth-ods." Plasmid constructs were trans-fected into human chorionic cells. Lu-ciferase activity was normalized tocotransfected RSV-chloramphenicolacetyl transferase activity, and the re-sults (mean+SEM) are given as per-centage of p582+ lucferase activity.

_ t~~~~~~~~~~achbox represents a minimum ofthree independent transfections in trip-licate flasks, except for plasmidsp582- and p110- (two independenttransfections).



768 Circulation Research Vol 74, No S May 1994

Relative Luciferase Activity of Renin PromoterDeletions in Chorionic and JEG-3 Cells

Chorionic Cells JEG-3

Plasmid %±SEM n %±SEM n

p582+ 100 7 100 3

p110+ 346+58 6 123+4 3

p67+ 476+58 3 167 22 3

Luciferase activity was normalized to cotransfected RSV-chloramphenicol acetyl transferase activity, and the results(mean+SEM) are given as percentage of p582+ luciferaseactivity. n represents the number of independent transfections intriplicate flasks.

analysis, and little renin mRNA detected by PCRcompared with chorionic cells (Fig 3) confirmed thesedata. Moreover, by quantitative reverse transcriptasePCR, we showed that renin mRNA from JEG-3 cellsrepresented 0.3% of renin mRNA from chorionic cells(data not shown). Finally, the exact site of transcriptiondriven by the renin/luciferase fusion gene constructswas determined by primer extension. The p582+ andp110+ plasmids were transfected into chorionic cells,the RNA was extracted, and the transcriptional initia-tion site was mapped by primer extension experiments.One major fragment was obtained with both plasmids,confirming the correct initiation of transcription (datanot shown).

Footprint Analysis of the Human Renin PromoterChanges in the expression of the serial deleted renin/

luciferase genes (p582+ to p67+) suggest that factorswithin chorionic cells interact with renin cis-actingsequences. To demonstrate that these regulatory re-gions involved in the expression of the renin gene aresites for contact with DNA binding proteins from chori-onic cells, the promoter sequence spanning -336 to+ 16 was subjected to DNase I footprint analyses in thepresence and absence of nuclear extracts from humanchorionic cells. This study showed the presence of sixfootprints, designated A through F (Fig 4). Footprint Acomprised sequences from -29 to -6 and contained acentrally located, purine-rich sequence (GGAA) shownto be a conserved recognition site for ets-domain pro-teins.32 Footprint B, which extended from -79 to -62,was homologous with the Pit-I binding site consensussequence A(A/T)TTANCAT.33 Footprint C (from-107 to -83) did not show any sequence homology withpreviously described regulatory elements. Footprint D,from -234 to -214, showed strong homology with theCRE, TGACGTCA.3(1 Footprint E extended from -259

1 2 3 4

-._^ 243bp

FIG 3. Northern blot analysis of renin mRNA by reverse tran-scriptase polymerase chain reaction (PCR) in chorionic cells(lanes 1 and 3) and JEG-3 cells (lanes 2 and 4). Total RNA 50 ng(lanes 1 and 2) and 500 ng (lanes 3 and 4) were subiected toreverse transcription and PCR amplification, electrophoresed on8% acrylamide gel, and autoradiographed for 3 hours.

to -245, and its sequence was similar to the binding sitefor the ARP-1.34 Footprint F, from 293 to 272,showed strong homology with the sequence AGE3described recently by Tamura et aP35 on the humanangiotensinogen gene.

Functionality of the CRE ElementThe functionality of the CRE in the renin promoter in

chorionic cells was further assessed by two methods: (1)transient DNA transfections performed in the presence offorskolin and (2) gel mobility shift analysis using the reninCRE-like sequence and chorionic cell extracts. TransientDNA transfections were performed with the differentconstructs described above, and the cells were treated ornot treated with 10` mol/L forskolin for 24 hours. Alltransfections were performed in the absence of FCS toavoid nonspecific interference in renin gene expression.Fig 5 shows a clear twofold to threefold increase ofpromoter activity within the region -2616 to -273. Withthe p137+ and p110+ plasmids, which lack element D(-234 to -214), there was only modest stimulation ofpromoter activity (1.81±0.2 P<.05, and 1.62±0.28,P=NS, respectively). No significant stimulation was ob-served with the p67+ plasmid. The promoterless plasmid,BS luci, was not stimulated by forskolin (0.96 of unstimu-lated luciferase activity), demonstrating that the inductionof luciferase activity in cells transfected with the renin/luciferase constructs was mediated only via the reninpromoter sequence.

Finally, the specific interaction of chorionic cell nu-clear factors with element D (containing the CRE-likesequence) was investigated using gel mobility shiftanalysis. Double-stranded oligonucleotide containingelement D was used as a labeled probe. Two protein/DNA complexes (Fig 6, arrows) were observed thatwere specifically competed by an excess of either ho-mologous unlabeled DNA (REN) or somatostatin(SMS) oligonucleotide containing the consensus CRE.Both REN and SMS probe were able to completelycompete the chorionic cell nuclear binding protein. Thissuggested that a CRE binding protein (CREB)-likesubstance interacted with element D of renin promoter.

DiscussionSince the structure of the human renin gene was first

determined,625 several studies have been carried out toidentify the cis-regulatory elements that control itstranscriptional activity. Because human renin gene ex-pression very likely depends on transcriptional regula-tory factors present only in renin-expressing cells, it isimportant to perform such studies in cell lines thatexpress renin. Cells derived from chorionic tissue are amajor extrarenal site of renin production after kidneytissue.8"^" Human chorionic cell cultures have beencharacterized by Pinet et al,14 who used electron micros-copy to show that these cultures contained a single typeof elongated cell. Immunofluorescence studies usingspecific renin antibodies have shown that all cells inculture were stained and therefore contained both reninand prorenin. Chorionic cells in culture produce pre-dominantly prorenin, and they seem to have only theconstitutive pathway. Nevertheless, even though thechorionic cell in culture does not process prorenin, thismodel could be used to study renin gene transcription.




G T 1 23

0.

ateZr:

Pt? ta S . -.U

-# ,.a VAZ.. .4 .I* .

5C

'4

*4'

4- .A

fl.p.)

U

[c]

[B]

K1

GT 1

n

1*f

*t

la f

AW-

FIG 4. DNase footprinting of the 5-proxi-mal region of the human renin gene C 336to + 16). Footprint analysis was performedwith human chorionic cell nuclear extracts.Binding reactions and DNase treatmentwere carried out as described in "Materialsand Methods." G and T represent the se-quencing ladder. In panels through Ill, lane1 shows the reaction performed withoutnuclear extract; lanes 2 and 3, the reactionsperformed with 30 gig of nuclear extract anddecreasing amounts of DNase 1. Panel 1,DNase footprinting analysis of the noncod-ing strand of the -336 to +16 fragment.Panels II and Ill, DNase footprinting analy-sis ofthe coding strand ofthe -336 to ±16fragment. Panel IV, sequences (A through F,in boxes) of the DNase I-protected regionsof the human renin promoter in the pres-ence of human chorionic cell nuclearextract.

Iv-364 GGGGTTGGGTCTGGGGIAGGGAGCTGGAAACGIAGGITTTTACGCGTTGTCCGAGITTITGAT

F E-3Q4 GTTAGCCCTG GCAGIGGIGITIGITCATGAGG TCTGCGIGCTC AGGGGTGAGAGGG~C-

D-244 AAGCCAGAIA AGGGCTGCIAGCGIGACIGG[AACAAGAITGGITIGGGCACAGGTGTGGI-184 ICCIGGAGGGCCTCTGCTGGGCAIGGGGAAACGTGGGTACGGIIGACCCACCTAGTCI'GG

c B-1 24 TCGCGCAGIGAGTTTTATIGGITGACTGCCCITGCGCATCITAC CC AG~GIAAIAAATGAGGG

643GCAGAATTGCAATGACCCCATGCATGGAGT IATAAAAGGGGAAGGGGCTAAGGG G

-4 CCACAGAACCICAGIGGAIC+1

Human chorionic cells are a suitable model to study thetranscriptional regulatory elements involved in renin geneexpression: (1) These cells synthesized large amounts ofprorenin. Their renin mRNA was identical in size (1.6 kb)to that of kidney renin mRNA14,16 and was easily detect-able by Northern blot analysis. In contrast, in non-renin-producing cells, JEG-3 cells, renin mRNA could be de-tected only slightly by PCR. (2) A highest luciferaseactivity (between 20- and 200-fold) of renin promoter!luciferase constructs was obtained with chorionic cellscompared with JEG-3 cells by transient DNA transfection.This finding in JEG-3 cells is in agreement with the results

of Smith and Morris,'8 who showed that, in JEG-3 cells,the human renin promoter was not able to direct CATtranscription in the absence of the herpes simplex virus tkpromoter. The CAT activity detected when the reninpromoter region was fused to the tk promoter17'18mighthave been driven by the tk promoter, which is more potentthan the renin promoter. In contrast, Duncan et al'6reported that human chorionic cells are capable of usingthe first 600 bp of the renin promoter to direct CATexpression. (3) A correct initiation of transcription wasfound with renin/luciferase fusion genes transfectedwithin chorionic cells.

769




Forskolin stimulation (fold increase)

0 1.

-2616 /

-892

TagCGTCA BS Luci

--~--> +16 p2616+>6p2+

mi---- p892+

2 3 4I

-582 p582+

-273 t-- p273+

-137 p137+

-110 p110+

-67* p67+

FIG 5. Bar graph showing effects of 10-5 mol/L forskolin on the expression of the human renin promoter. The different renin/luciferaseconstructs were transfected into chorionic cells and assayed for luciferase activity after 24 hours of 10-5 mol/L forskolin stimulation.Luciferase activity was normalized to cotransfected RSV-chloramphenicol acetyl transferase activity. Results (mean -+-SEM) are given asforskolin-induced fold increases in basal unstimulated luciferase activity. The luciferase activity of the control plasmid, BS luci, was notstimulated by forskolin (0.96 of unstimulated luciferase activity). Each box represents two to five independent transfection experimentsin triplicate flasks. The presence of the putative cAMP-responsive element (-226/-219) is indicated by the black marker on the variousconstructs.

In the present study, no major basal regulatory ele-ments were detected upstream of -582 up to -2616. Incontrast, deletion analysis of the 5'-flanking region ofthe p582+ plasmid revealed that the first 110 bp of therenin promoter were sufficient for a specific and highexpression in human chorionic cells (more than three-

competitor

b I

a)0

REN SMS

0 0 0nc tin c' N" - Ce in

a oh 01% 04

W #,

wsst.qmmu.mm

FIG 6. Gel mobility shift analysis of chorionic cell nuclear factorinteractions with the human renin promoter (-234/-200). Adouble-stranded synthetic oligonucleotide including the foot-print D was used as the probe. Competitions were performedwith homologous DNA (REN) or with the somatostatin (SMS)oligonucleotide.30 Arrows indicate the specific DNA/protein com-plexes. The fold molar excess of competitor is indicated for eachcompetitor sequence.

fold higher than p582+) and could therefore corre-spond to the "basal promoter." Indeed, this highestactivity is not due to plasmid "read-through" artifact,since the activity of the shortest 5'-flanking DNA frag-ment (p110+ and p67+) was not increased when trans-fected into a non-renin-producing cell line, the JEG-3cells. A negative regulatory element was found betweenthe -374 and -273 bp by deletion analysis of the 582 bpof renin promoter. DNase I footprinting showed thatthis region of DNA binds trans-activating factors pre-sent in chorionic cellular nuclear extract. This footprinthas strong homology with a sequence of the mouseangiotensinogen gene that binds the constitutive factornamed AGF3 by the authors.35 They suggested thatAGF3 could play important roles in the differentiation-dependent promoter activation of the mouse angioten-sinogen gene in adipocytes. In the case of the humanrenin promoter, the present results suggested that an"AGF3-like" protein from chorionic cells could represshuman renin expression. Further deletions of the 5'-region until -137 bp revealed the presence of anothernegative regulatory element. Interestingly, the DNase Iprotection assay revealed that this region contains asequence, footprint E (-259 to -245), that bindsnuclear extract from chorionic cells. This putative cis-element has strong homology with a sequence of thehuman apolipoprotein AI gene that binds the ubiqui-tous ARP-1,34 a member of the orphan steroid receptorsuperfamily that decreases apoAl gene expression.These results suggest that an ARP-1-like protein mightdecrease renin gene expression in chorionic cells. Bothnegative elements found in chorionic cells are notoperative in JEG-3 cells, since the p10O+ plasmid didnot exhibit a highest activity, suggesting that thesetrans-acting factors, AGF3 and ARP-1, may not bepresent in these latter cells.No difference in luciferase activity was found with the

plasmids pl37+ and p110+ in relation to the fact thatno footprint was found between -137 and -110 bp ofthe human renin promoter. Further deletions until -67bp showed an increase in luciferase activity. This DNA

t ._ __j

EEEH

-lt:




c'

41-214 -107/-83 -791-62 -291-6 +1 +16

D C B A

ARen TATAAAAGGGGAAGGGCTAAGGGAA Ets Cons. GGAW GGAW

B Ren GGTAATAAATCAGGGCAGPit-1 Cons. ATGNATAAWT

D RenCRE Cons.

Ii

GAGGGCTGCTAGCGTCACTGGTGACGTCA

FIG 7. Schematic representation of the hu-man renin promoter. The stippled boxes are

the protein-binding sites identified by DNasefootprinting analysis. For each footprint(A--F), renin sequence was aligned to possi-ble factor binding site consensus sequence.Boldface letters in renin sequence representhomology with the consensus sequence.W=A+T, N=G+A+T+C. LUCI indicates lu-ciferase; ARP-1, apolipoprotein Al regulatoryprotein-i; CRE, cAMP-responsive element;and Pit-1, pituitary-specific factor.

Ren AGGGGTCACAGGGCCE ARP-1 Cons. AGGGGTCA-AGGGNTCA

F Ren GCAGTGCT-GTTTCTCATCAGCCAGE3 Cons. AGCTGTGCTTGT

region binds two factors: one named element C, which didnot share any homology with previously described regu-

latory elements; the other, element B, is an adenine/thymine rich region very similar to the Pit-1 binding siteconsensus sequence.33 Recently, Sun et a120 showed thatthis region of the human renin promoter binds nuclearextracts isolated from GC cells, a pituitary lactotropeprecursor cell line. Furthermore, they showed that theactivity of the renin promoter/luciferase constructstransfected into HeLa cells was greatly increased bycotransfection with a Pit-1 expression vector. Pit-1, a

specific pituitary factor, is a member of the POU familyof transcription factors, which plays a critical role in theproliferation of specific cells and in their expression ofspecific genes.36 In contrast to the results of Sun et a120in GC cells, deletion of this Pit-i-like binding site didnot affect luciferase gene expression in chorionic cells.This suggests cell-specific use of this regulatory ele-ment. Such an observation, concerning the differentialexpression of a promoter in different cell lines, has beenmade by Paulweber et a137 for the apolipoprotein Bpromoter in HepG2 (hepatic) and CaCo-2 (intestinal)cell lines. They showed that this difference was due tothe relative amounts of nuclear factors that bind tospecific sequences located within the promoter. An-other explanation for this discrepancy could be thepresence, in chorionic cells, of a nuclear factor bindingthe downstream element A (-29 to -6 bp), whichcontains the consensus GGAA ets motif.32 Many of theets-domain proteins have been shown to be transcriptionactivators in various tissues.32 Thus, the high luciferaseactivity of plasmid p67+ might result from the presence,in chorionic cells, of an ets-domain protein involved inthe expression of the human renin gene. This domainmight be not operative in GC cells.The functionality of the renin promoter was further

assessed by forskolin stimulation. Previous studies byDuncan et a138 had shown that, in chorionic cells,plasmid constructs containing either the first 600 or 100bp of the human renin promoter fused to CAT were

markedly stimulated by 8 hours of incubation with 10-6mol/L forskolin. In contrast, plasmids containing the'-flanking region of the renin gene (from - 584 to -146

and from -146 to +11) fused to the tk promoter were

not induced by forskolin.38 However, these experimentswere performed in the presence of FCS, which iscapable of stimulating the transcription of several genes

via different serum response elements.39 cAMP mightalso activate serum response elements,40 and therefore,interactions between serum- and cAMP-induced stimu-lation could not be excluded. In JEG-3 cells, Burt et al17had also used plasmids containing the 5'-flanking regionof the renin gene fused to the tk promoter and reporteda modest 60% cAMP-induced stimulation. It was forthese reasons that further studies on the effects ofcAMP on renin gene expression were required.

In the present study, all stimulations were performed inthe absence of FCS to avoid nonspecific interference. Theresults show that element D (-234 to -214) was requiredfor forskolin to stimulate transcription twofold to three-fold. This element contains a motif, TAGCGTCA, thatshares a six-nucleotide homology with the consensus CREoctamer TGACGTCA.30 It contains the short motifCGTCA, which has been shown to be a binding site forCREB.41 The present results show that nuclear extractsisolated from chorionic cells bound to this sequence. Twospecific DNA/protein complexes were found and couldresult from the formation of monomeric and dimericDNA/protein complexes as previously described.42 In ad-dition, this DNA/protein binding was displaced by theSMS oligonucleotide containing the consensus binding sitefor CREB.30 Taken together, these results are in favor ofelement D being a functional CRE site. However, themodest forskolin-stimulated increase in luciferase activityobserved with constructs p137+ and pl10+ (50%) cannotbe explained by this CRE-like sequence. Interestingly,Peers et a143 have shown that Pit-1 is involved in the cAMPstimulation of the human prolactin gene and that severalPit-1 binding sites fused to the tk promoter confer cAMPresponsiveness to GC cells. It is therefore possible that thePit-i-like binding site present on the renin gene is alsoinvolved in regulation by cAMP and that stimulation bycAMP is the result of multiple responses acting in combi-nation to yield a measurable degree of stimulation. Fur-ther studies will be necessary to determine whether thePit-i-like sequence is necessary to confer cAMP respon-siveness on the renin gene.

-2931-272 -2591-245F E

=

A.4) 40 A..

-23d




Overall, these results clearly identify several func-tional cis-regulatory regions in the renin gene andpotential trans-activating factors governing renin geneexpression in human chorionic cells as representedschematically in Fig 7. No studies have yet been re-ported in human JG cell regulation of renin transcrip-tion because of the lack of suitable cell lines or cellculture models. However, like chorionic cells, JG cellsrespond to forskolin by an increase in renin mRNA andrenin release,44 and factors similar or identical to theCREB identified in the present study could be involved.Moreover, chorionic cells in culture respond to angio-tensin II by an elevation of intracellular calcium and adecrease in renin production (unpublished data) like JGcells.2However, characterization of transcription factors

that bind human renin promoter from kidney cortex orjuxtaglomerular cells would permit determination ofrenin-producing cell-specific factors. We can speculatethat some of the transcription factors characterizedcould be implied in the renin tissue-specific expressionand in renin regulation in vivo.

AcknowledgmentsWe wish to thank Dr A. Fukamizu and Dr S. Subramani

for having kindly provided the renin and RSV luciferaseplasmids, respectively. The authors are grateful to Dr M.Day for editorial help and to Nicole Braure for secretarialassistance. We wish to thank G. Masquelier and A. Boisquil-Ion for artwork.

References1. Hackenthal E, Paul DG, Taugner R. Morphology, physiology, and

molecular biology of renin secretion. Physiol Rev. 1990;70:1067-1116.

2. Kurtz A. Cellular control of renin secretion. Rev Physiol BiochemPharmacol. 1989;113:1-40.

3. Nakamura N, Burt DW, Paul M, Dzau VJ. Negative controlelements and cAMP responsive sequences in the tissue-specificexpression of mouse renin genes. Proc NatlAcad Sci US A. 1989;86:56-59.

4. Barret G, Horiuchi M, Paul M, Pratt RE, Nakamura N, Dzau VJ.Identification of a negative regulatory element involved in tissue-specific expression of mouse renin genes. Proc NatlAcad Sci U SA.1992;89:885-889.

5. Tamura K, Tanimoto K, Murakami K, Fukamizu A. A combi-nation of upstream and proximal elements is required for efficientexpression of the mouse renin promoter in cultured cells. NucleicAcids Res. 1992;14:3617-3623.

6. Soubrier F, Panthier JJ, Houot AM, Rougeon F, Corvol P. Seg-mental homology between the promoter region of the human reningene and the mouse Renl and Ren2 promoter regions. Gene.1986;41:85-92.

7. Deboben A, Inagami T, Ganten D. Tissue renin. In: Hypertension:Physiopathology and Treatment. New York, NY: McGraw-Hill;1983:194-209.

8. Acker GM, Galen FX, Devaux C, Foote S, Papernik E, Pesty A,Menard J, Corvol P. Human chorionic cells in primary culture: amodel for renin biosynthesis. J Clin Endocrinol Metab. 1982;55:902-909.

9. Poisner AM, Wood GM, Poisner R, Inagami T. Localization ofrenin in trophoblasts in human chorion leave at term pregnancy.Endocrinology. 1981;109:1150-1155.

10. Skinner SL, Lumbers ER, Symonds EM. Renin concentration inhuman fetal and maternal tissues. Am J Obstet Gynecol. 1968;101:529-533.

11. Conn JW, Cohen EL, Lucas CP, McDonald WJ, Mayor GH,Blough WM, Eveland WC, Bookstein JJ, Lapides J. Primaryreninism. Arch Intern Med. 1972;130:682- 696.

12. Galen FX, Corvol MT, Devaux C, Gubler MC, Mounier F, Cam-illeri JP, Houot AM, Menard J, Corvol P. Renin biosynthesis byhuman tumoral juxtaglomerular cells: evidences for a renin pre-cursor. J Clin Invest. 1984;73:1144-1155.

13. Pinet F, Corvol MT, Dench F, Bourguignon J, Feunteun J, MenardJ, Corvol P. Isolation of renin-producing human cells by trans-fection with three simian virus 40 mutants. Proc Natl Acad SciUSA. 1985;82:8503-8507.

14. Pinet F, Corvol MT, Bourguignon J, Corvol P. Isolation and char-acterization of renin-producing human chorionic cells in culture.J Clin Endocrinol Metab. 1988;67:1211-1220.

15. Poisner AM, Agrawal P, Poisner R. Renin release from humanchorionic trophoblasts in vitro. Trophoblast Res. 1987;2:45-60.

16. Duncan KG, Haidar MA, Baxter JD, Reudelhuber TL. Controlelements in the human renin gene. Trans Am Assoc Phys. 1987;100:1-9.

17. Burt DW, Nakamura N, Kelley P, Dzau VJ. Identification ofnegative and positive regulatory elements in the human renin gene.J Biol Chem. 1989;264:7357-7362.

18. Smith DL, Morris BJ. Transient expression analyses of DNAextending 2.4 kb upstream of the human renin gene. Mol CellEndocrinol. 1991;80:139-146.

19. Ekker M, Sola C, Rougeon F. The activity of the mouse reninpromoter in cells that do not normally produce renin is dependentupon the presence of a functional enhancer. FEBS Lett. 1989;255:241-247.

20. Sun J, Oddoux C, Lazarus A, Gilbert MT, Catanzaro DF.Promoter activity of human renin 5'-flanking DNA sequences isactivated by the pituitary-specific transcription factor Pit-1. J BiolChem. 1993;268:1505-1508.

21. Ingraham HA, Chen R, Mangalam HJ, Elsholtz HP, Flynn SE, LinCR, Simmons DM, Swanson L, Rosenfeld MG. A tissue-specifictranscription factor containing a homeodomain specifies apituitary phenotype. Cell. 1988;55:519-529.

22. Graham FL, Van Der Eb AJ. A new technique for the assayof infectivity of human adenovirus 5 DNA. Virology. 1975;52:456-476.

23. Menard J, Bews J, Heusser C. A multirange ELISA for the mea-surement of plasma renin in humans and primates. J Hypertens.1984;2(suppl 3):275-279.

24. N'Guyen VT, Morange M, Bensaude 0. Firefly luminescenceassays using scintillation counters for quantitation in transfectedmammalian cells. Anal Biochem. 1988;171:404-408.

25. Fukamizu A, Nishi K, Nishimatsu S, Miyazaki H, Hirose S,Murakami K. Human renin gene of renin-secreting tumor. Gene.1986;49:139-145.

26. De Wet JR, Wood KV, Deluca M, Helinski DR, Subramani S.Firefly luciferase gene: structure and expression in mammaliancells. Mol Cell Biol. 1987;7:725-737.

27. Chomczynski P, Sacchi N. Single-step method of RNA isolation byacid guanidinium thiocyanate-phenol-chloroform extraction. AnalBiochem. 1987;162:156-159.

28. Caroff N, Della Bruna R, Philippe J, Corvol P, Pinet F. Regulationof human renin secretion and renin transcription by quantitativePCR in cultured chorionic cells: synergistic effect of cyclic AMPand protein kinase C. Biochem Biophys Res Commun. 1993;193:1332-1338.

29. Shapiro DJ, Sharp PA, Wahli WW, Keller M. A high efficiencyHela cell nuclear transcription extract. DNA. 1988;7:47-55.

30. Montminy MR, Sevarino KA, Wagner JA, Mandel G, GoodmanRH. Identification of a cyclic-AMP-responsive element withinthe somatostatin gene. Proc Natl Acad Sci U SA. 1986;83:6682-6686.

31. Alam J, Cook JL. Reporter genes: application to the study ofmammalian gene transcription. Anal Biochem. 1990;188:245-254.

32. Wasylyk B, Hahn SL, Giovane A. The Ets family of transcriptionfactors. EurJBiochem. 1993;211:7-18.

33. Nelson C, Albert VR, Elsholtz HP, Lu LIW, Rosenfeld MG.Activation of cell-specific expression of rat growth hormone andprolactin genes by a common transcription factor. Science. 1988;239:1400-1405.

34. Ladias JAA, Karathaniasis SK. Regulation of the apolipoproteinAI gene by ARP-1, a novel member of the steroid receptor super-family. Science. 1991;251:561-565.

35. Tamura K, Tanimoto K, Ishii M, Murakami K, Fukamizu A.Proximal and core DNA elements are required for efficient angio-tensinogen promoter activation during adipogenic differentiation.J Biol Chem. 1993;268:15024-15032.

36. He X, Treacy MN, Simmons DM, Ingraham HA, Swanson BW,Rosenfeld MG. Expression of a large family of POU-domain reg-ulatory genes in mammalian brain development. Nature. 1989;340:35-42.

37. Paulweber B, Onasch MA, Nagy BP, Levy-Wilson B. Similaritiesand differences in the function of regulatory elements at the 5' end




of the human apolipoprotein B gene in cultured hepatoma(HepG2) and colon carcinoma (CaCo-2) cells. J Biol Chem. 1991;266:24149-24160.

38. Duncan KG, Haidar MA, Baxter JB, Reudelhuber TL. Regulationof human renin expression in chorion cell primary cultures. ProcNatl Acad Sci USA. 1990;87:7588-7592.

39. Faisst S, Meyer S. Compilation of vertebrate-encoded transcriptionfactors. Nucleic Acids Res. 1992;20:3-26.

40. Fukumoto Y, Kaibuchi K, Oku N, Hori Y, Takai Y. Activation ofthe c-fos serum-response element by the activated c-Ha-ras proteinin a manner independent of protein kinase C and cAMP-dependent protein kinase. J Biol Chem. 1990;265:774-780.

41. Nichols M, Weih F, Schmid W, De Vack C, Kowenz-Leutz E,Luckow B, Boshart M, Schuitz G. Phosphorylation of CREB affects

its binding to high and low affinity sites: implications for cAMPinduced gene transcription. EMBO J. 1992;11:3337-3346.

42. Yamamoto KK, Gonzales GA, Biggs WH, Montminy MR.Phosphorylation-induced binding and transcriptional efficacy ofnuclear factor CREB. Nature. 1988;334:494-498.

43. Peers B, Monget P, Nalda MA, Voz ML, Berwaer M, Belayew A,Martial JA. Transcriptional induction of the human prolactingene by cAMP requires two cis-acting elements and at leastthe pituitary-specific factor Pit-i. J Biol Chem. 1991;266:18127-18134.

44. Della Bruna R, Kurtz A, Corvol P, Pinet F. Renin mRNA quan-tification using polymerase chain reaction in cultured juxtaglo-merular cells. Circ Res. 1993;73:639-648.



cis-regulatory elements and trans-acting factors directing basal and camp-stimulated human renin...

Documents