624

Pituitary-Specific Transcription Factor (Pit-1)Binding Site in the Human Renin Gene

5'-Flanking DNA Stimulates Promoter Activityin Placental Cell Primary Cultures and Pituitary

Lactosomatotropic Cell Lines

Jidong Sun, Carole Oddoux, Matthew T. Gilbert, Yan Yan, Amy Lazarus,Wallace G. Campbell, Jr, Daniel F. Catanzaro

Abstract Renin gene expression is limited to a number ofspecific tissues, including the kidney, adrenal glands, repro-ductive organs (of particular relevance to this study, theplacenta), and the pituitary gland. In the present study, weinvestigated the human renin (hRen) S'-flanking DNA se-quences required to drive the expression of a luciferasereporter gene in placental and pituitary cells and in two celllines, 293 and JEG-3, which have been proposed as modelsystems with which to study transcriptional regulation of reningenes. The activities of specific sequences in the hRen 5'-flanking DNA sequences in human placental cell primarycultures were very similar to those that we previously reportedin pituitary cells, suggesting the involvement of commonpromoter elements and related transcription factors. Accord-ingly, the binding site for the pituitary-specific transcriptionfactor (Pit-1) was the major determinant of renin promoteractivity in both pituitary and placental cells. Gel mobility shiftanalysis showed a placental nuclear factor with a gel mobilitydifferent from that of Pit-i. However, Northern blot analysis

A the rate-limiting enzyme in the angiotensin-aldosterone cascade, renin, released from renaljuxtaglomerular (JG) cells, plays a central role in

determining blood pressure and electrolyte homeostasis.1Although elevated blood pressure should suppress reninrelease from JG cells, a spectrum of renin levels is foundin hypertensive patients. Those with high renin levelshave been shown to have a higher incidence of myocar-dial infarction and stroke.23 Because renin release mustbe accompanied by de novo synthesis to replenish de-pleted JG storage granules, the mechanisms that directrenin gene expression are likely to play an important rolein the pathogenesis and sequelae of hypertension. Un-derstanding these mechanisms may provide better ap-proaches to the treatment of hypertension.

Studies in transgenic mice have broadly defined therenin genomic sequences required for JG cell-specific

Received March 1, 1994; accepted June 20, 1994.From the Cardiovascular Center and Departments of Medicine

(J.S., M.T.G., A.L., D.F.C.), Pathology (W.G.C.), and Physiology(C.O., Y.Y., D.F.C.), Cornell University Medical College, NewYork, NY.

Reprint requests to Daniel F. Catanzaro, PhD, CardiovascularCenter, New York Hospital-Cornell Medical Center, 525 East68th St, New York, NY 10021.© 1994 American Heart Association, Inc.

failed to demonstrate abundant Pit-i-related mRNAs inrenin-expressing cultures of chorionic and decidual cells, sug-gesting that the placental factor is not closely related to Pit-i.Although a factor from 293 cells also bound to the Pit-1 site, ithad gel mobility shift characteristics different from Pit-1 andthe placental factor. Moreover, the low promoter activity in293 cells was independent of this site or, indeed, of sequencesupstream from the TATA box. In JEG-3 cells, renin 5-flanking DNA sequences showed virtually no transcriptionalactivity. Taken together, these data suggest that commoncis-acting sequences direct renin gene expression to pituitaryand placental cells and that the activity of these sequences isdependent on the presence of transcription factors specific tothese cells. Our studies also indicate that 293 and JEG-3 celllines may not be appropriate for the study of renin geneexpression. (Circ Res. 1994;75:624-629.)Key Words * renin gene expression * pituitary * placenta

*transcription factor

expression.4-7 However, in the absence of JG cell lines,it has been difficult to identify the specific cis-actingsequences involved and the trans-acting factors withwhich they interact. Since renin is also expressed in thepituitary gland8'9 and the placenta,10 cell culture systemsfrom these tissues have been used to study the transcrip-tional mechanisms that direct renin gene expression."1-14Using renin-luciferase hybrid genes transfected into

the pituitary tumor cell line GC, Sun et al14 demon-strated that human renin gene (hRen) 5'-flanking DNAsequences contain both positive and negative regulatoryelements. Sequences contained between -148 and + 18(numbered relative to the transcription start site at + 1)were found to be sufficient for maximal expression.Within this region, binding of the pituitary-specifictranscription factor (Pit-1) was shown to be the majordeterminant of promoter activity.14 In studies usingprimary cultures of placental decidual cells, sequencescontained between -100 and + 18 were also found to besufficient for the expression of renin-CAT hybridgenes.12 However, the elements involved were not fur-ther delineated.

In other studies, transfection of JEG-3 choriocarci-noma cells with CAT constructs containing hRen 5'-flanking sequences linked to the herpes simplex virusthymidine kinase (HSV-TK) promoter indicated the

by guest on February 21, 2018http://circres.ahajournals.org/

Dow

nloaded from

Sun et al Renin Expression in Placenta and Pituitary 625

presence of both positive and negative regulatory ele-ments.11' 3 More recently, mouse Ren-lc 5'-flankingDNA sequences were shown to be transcriptionallyactive in the human embryonal kidney cell line 293.15Although a 293 cell nuclear factor bound to the con-served Pit-1 site, activation by this element was weakunless a sequence contained within a 476-bp insertionspecific to mouse renin genes was also present.Although these studies identified a number of posi-

tive and negative regulatory elements, the level ofresolution at which regulatory regions in the renin5'-flanking DNA were mapped was not sufficient todetermine whether common mechanisms direct reningene expression to these different tissues. The identifi-cation of common mechanisms would simplify studies ofthe molecular mechanisms and provide evidence forcommon pathways through which extracellular signalsmight regulate renin gene expression.

In the present study, we compared the promoteractivity of hRen 5'-flanking DNA sequences in humanplacental cell primary cultures and in JEG-3 and 293cell lines. We also compared binding of transcriptionfactors from each of these cell preparations to the Pit-1site. These studies show that although common cis-acting sequences appear to direct transcription in bothplacental and pituitary cells, activity in these two celltypes appears to be mediated by different trans-actingfactors. In contrast, renin promoter activity in JEG-3and 293 cells appears to be due to basal transcriptionfrom the TATA box, independent of 5'-flanking DNAsequences further upstream.

Materials and MethodsCell Cultures of Chorion Laeve and Decidua

Chorionic or decidual cells were prepared by using modifi-cations of established protocols.16"17 Term placentas wereobtained from consenting patients by following a protocolapproved by the New York Hospital-Cornell Medical CenterHuman Ethical Review Committee. Within 30 minutes afterdelivery, placental membranes were rinsed free of blood with0.9% saline at room temperature. After removing the amnioticlayer, the decidual layer was gently scraped off with a straight-edged razor blade. Chorionic and decidual tissues (10 to 20 geach) were incubated separately in 50 to 100 mL of Hank'sbalanced salt solution (calcium and magnesium free, CMF-H)containing 0.05% trypsin (Sigma type I) and 0.05% collage-nase (Worthington type CLS II). After a 20-minute digestion,10 mg DNase I (Sigma type IV) was added, and the incubationcontinued for 10 minutes. Tissue pieces were allowed to settle,the medium was decanted and mixed with an equal volume ofcalf serum, and the cells were pelleted for 10 minutes at 50g.This procedure was repeated twice. Medium from the firstdigestion contained mainly erythrocytes and cell debris, whichwere discarded. Cells from the second two incubations werecombined in 5 mL CMF-H and applied to a Percoll stepgradient (15% and 40% steps in CMF-H) and centrifuged at1200g for 20 minutes. Cells banding at the interface werecollected, washed three times in CMF-H, and plated at adensity of :106 cells/mL in HEPES-buffered Dulbecco's mod-ified Eagle's medium (DMEM) containing 10% fetal bovineserum (FBS). JEG-3 and 293 cells were obtained from theAmerican Type Culture Collection and grown in HEPES-DMEM containing 10% FBS. The use of these cells in studiesof renin promoter activities has been described by otherresearchers."",15,18

Immunocytochemistry and Renin AssayAn antiserum to purified recombinant human renin was

prepared in rabbits. Serum obtained after the fourth booster

was used in these studies. This antiserum stained intenselythe juxtaglomerular apparatus (JGA) of normal mice and ofhuman kidneys with advanced arterial nephrosclerosis. Italso stained a renal JGA tumor. Preincubation of theantiserum with highly purified recombinant human proreninabolished staining to sections of placental tissue and tochorionic cell monolayers (authors' unpublished data). Ananti-a-actin monoclonal antibody was obtained from Boeh-ringer Mannheim.

Cell cultures at or near confluence were fixed briefly inBouin's solution (Polyscientific) and then dehydrated througha series of graded ethanol incubations. Anti-renin and anti-a-actin antisera were applied at dilutions of 1:1000 and 1:20,respectively. After incubation for 16 hours at 4°C, the mono-layers were washed in phosphate-buffered saline. Bound anti-body was localized by using an Immunogold kit (Amersham)with a 1-nm gold particle size. Renin activity in cell culturemedium was assayed as previously described.'9

Constructions and TransfectionsRenin-luciferase hybrid genes used in the present study are

described in detail elsewhere.'4 These constructions containhRen sequences extending from + 18 to various positions in the5'-flanking DNA up to -2748. Two of the constructionscontain mutations in the Pit-1 site that reduce or prevent Pit-1binding (-148,A1 and -98/A2).'4 Transfections of placentalcells and 293 and JEG-3 cell lines were carried out by thecalcium phosphate method20 on 60-mm dishes of each cellpreparation at z50% confluence. Eight micrograms of DNAwas applied to each dish for 12 to 16 hours, after which themedium was replaced. Cells were harvested 48 hours aftertransfection and lysed in 200 j,uL luciferase assay buffer, ofwhich a 100-,uL aliquot was assayed for activity. pRSV.luc,2which contains the Rous sarcoma virus long terminal repeat,was transfected in parallel, and the data were normalized tothe activity of this construct.Gel mobility shift assays were carried out with nuclear

extracts prepared from cell cultures as described previously'4by using a 32P end-labeled oligonucleotide corresponding topositions -80/-58 in the hRen 5'-flanking DNA.

Northern Blot AnalysisRNA was prepared by using chaotropic agents.22 Total RNA

(10 jig per lane) was fractionated on a 1.2% agarose gelcontaining 20 mmol/L MOPS (pH 7.0) and 2.4% formalde-hyde. Size markers run in parallel were from GIBCO-BRL.After electrophoresis, RNA was transferred by blotting in 10xstandard saline citrate (SSC) to a nylon membrane, which wasthen dried at 60°C for 1 hour and UV-crosslinked for 30minutes. Membranes were prehybridized and hybridized insolutions containing 5x SSPE, 50% formamide, 0.2% sodiumdodecyl sulfate (SDS), 100 ,ug/mL denatured salmon spermDNA, and 5 x Denhardt's solution. Probe was labeled byrandom priming of a 369-bp fragment spanning the DNAbinding domain of Pit-1 (amino acids 141 through 263).23Hybridization was at 37°C for 24 hours, and the membrane wasthen washed two times for 20 minutes in 1 x SSC and 1% SDSat 20°C, 47°C below the calculated melting temperature (Tm).

ResultsRenin Secretion and Synthesis by Primary Culturesof Chorionic and Decidual Cells

Cell preparations from both chorion laeve and de-cidua grew rapidly and achieved confluence after 5 to 7days in culture. At confluence, both chorionic anddecidual cell cultures secreted prorenin activity of10000 to 20000 ng angiotensin I * mL-1*h` in 24hours. These cultures remained viable for up to 7 weeksand continued to secrete prorenin through two to threepassages.

by guest on February 21, 2018http://circres.ahajournals.org/

Dow

nloaded from

626 Circulation Research Vol 75, No 4 October 1994

)-t

0

_'i4

4.

B

1.-tIr

FIG 1. Photomicrographs showing immunocytochemical detec-tion of prorenin (A) and a-actin (B) in primary cell cultures ofhuman placental chorion laeve cells. A diversity of cell typesshows finely granular cytoplasmic staining for prorenin. Largepolygonal cells stain especially intensely. Cytoplasm of manyelongated cells showed a filamentous pattern for a-actin consis-tent with contractile elements. Original magnification x400.

Decidual cell cultures contained mostly elongatedcells, whereas chorion laeve cultures exhibited a nuumberof different cell morphologies, including irregular polyg-onal and elongated cells (Fig 1). In chorion laevecultures, overall in excess of 70% of cells stained withthe anti-renin antibody (Fig 1A). Of these, a smallerpopulation (5% to 15%) stained intensely. The broad,flattened morphology of these strongly staining cellssuggests that they might be more mature cells, whichexpress high levels of prorenin. Most of the cells con-

tained in these preparations also bound to antibodiesspecific for the smooth muscle-specific marker cr-actin(Fig 1B). Actin staining in larger cells often exhibitedfilamentous structures typical of the organization ofactin into contractile bundles (Fig 1B), suggesting thatrenin-producing cells in these cultures are derived fromsmooth muscle cells.

hRen Promoter Activity in Placental Cell PrimaryCultures and JEG-3 and 293 Cell LinesTo compare the activity of hRen 5'-flanking DNA

sequences in different cell types, renin-luciferase hybridgenes were transfected into primary cultures of chorionlaeve and decidua and JEG-3 and 293 cell lines (Fig 2).

In secondary cultures of chorion laeve, the most activehRen construct contained sequences -148/+18. Thisconstruct exhibited :1% of the activity of pRSVluc,which yielded 3 x 10' to 7 x 106 light units in the luciferaseassay in the experiments shown. Generally, we found thatcotransfection of an expression vector to normalize trans-

04 293

0,2ii .. iiio.2-JE

00O ND ND~

FIG 2. Bar graphs showing activity of hRenluc constructstransfected into cultures of chorion laeve (HC9), GC, 293, andJEG-3 cells. Dotted overlay on HC9 shows the mean+SEMvalues for selected constructs from four separate experiments(see text). Activities of hRen /uc constructs in GC cells wererecalculated from previously published data14 by normalizing theactivity of each construct to the activity of RSV.Iuc transfected ina parallel dish (n=4). Data for 293 and JEG-3 cells are forduplicate transfections.

fection efficiencies suppressed the activity of the hRenpromoter. Therefore, activities were normalized topRSVluc transfected in parallel dishes. Data are shownfor preparation 9 (HC9). However, similar results wereobtained in four separate experiments by using cells fromdifferent placentas and at least two different prepara-tions of a number of key constructs. The mean±SEMvalues for these replicate experiments are shown super-imposed on the results obtained with HC9.The relative activities of the various hRen 5'-flanking

DNA sequences were similar in chorion laeve and GCcells (Fig 2), suggesting that common promoter ele-ments are involved. Deletion of sequences between-2748 and -148 led to increased activity, whereasmutagenesis or deletion of the Pit-1 binding site mark-edly reduced activity. As in GC cells, loss of the Pit-1site caused the greatest reduction in activity of any

cc

u20

0)cc

9Wk'.1L.

1. A.

by guest on February 21, 2018http://circres.ahajournals.org/

Dow

nloaded from

Sun et al Renin Expression in Placenta and Pituitary 627

(L)

x:HC9 HC5

(D \i I I 2 4 6_ _ t

P::

FIG 3. Gel mobility shift analysis of nuclear factor binding to thehuman renin pituitary-specific transcription factor (Pit-1) site.Nuclear extracts were from GC cells, human chorionic (HC)cells, and human decidual (HD) cell cultures. Numbers indicateseparate preparations. P2 and P3 denote the passage numberof the HC culture, whereas HD was secondary culture (P2). Mand D indicate monomer and dimer bands, respectively. Lanes8 through 10 show a titration of the amount of extract added (seetext).

mutation. Although deletion of sequence between -148and -98 also resulted in losses in activity, mutagenesisof the Pit-1 site resulted in the same loss of activity as

deletion of the entire region -148 to -65, suggestingthat any positive effect of sequences contained between-148 and -98 is dependent on the Pit-1 site. This isalmost identical to the pattern of activities observed in

GC cells (Fig 2).hRen luciferase constructs were also tested for pro-

moter activity in 293 and JEG-3 cells. Whereas inchorionic cell primary cultures hRen -148/+18 5'-flanking DNA stimulated luciferase expression 70-foldover that of the promoterless construct pZluc, in 293cells this stimulation was only 17-fold, and in JEG-3cells it was 1.4-fold. Indeed, in JEG-3 cells, none of thehRen 5'-flanking sequences stimulated luciferase ex-

pression more than 2-fold. The modest promoter activ-ity of hRen 5'-flanking sequences observed in 293 cellswas near maximal with the -34 sequence. Since thissegment of 5'-flanking DNA contains little more thanthe TATA box, the activity observed in these cells ismost likely due mainly to basal transcription by RNApolymerase independent of the effect of factors boundto sequences further upstream. Similar patterns of basalexpression were also found in other choriocarcinomacell lines, BeWo and JAR (not shown).

Binding of Placental Nuclear Factor(s) to the hRenPit-1 Site

Gel mobility shift assays were carried out by using thehRen Pit-1 binding site (-80/-58)'4 as the probe, withnuclear extracts from chorion laeve and decidual cells atvarious passage numbers, and GC and 293 cells (Fig 3).Nuclear extracts from both chorion laeve and decidualcells resulted in a single band (lanes 4 through 10),which ran between the monomer and dimer forms of the

Pit-1 complex formed with GC cell nuclear extract (lane2). Increasing amounts of chorionic cell nuclear extractled to proportional increases in the intensity of thiscomplex (lanes 8 through 10), suggesting high-affinitybinding. The specificity of interactions involved in for-mation of this complex was further demonstrated by theability of an oligonueleotide containing a strong Pit-ibinding site from the human growth hormone gene'4 tocompete for complex formation (not shown). Nuclearextracts from later passages, where the rate of proreninsecretion into the media was reduced (suggesting lowerlevels of renin gene expression) exhibited proportionallylower formation of this complex (compare lanes 4 and5). Nuclear extract from 293 cells formed a complex(lane 3) with different mobility to GC, chorionic, anddecidual cell extracts, whereas no binding was evidentwith extract from JEG-3 cells (not shown).Northern Blot Analysis of POU Gene Expression inRenin-Expressing Tissues

Further studies were carried out to determine whetherPit-i-related sequences are expressed in placental cellcultures. RNA was prepared from chorion laeve anddecidual cell primary cultures and subjected to Northernblot analysis with a Pit-1 probe. RNA from GC cells thatexpress the endogenous Pit-1 gene was included as apositive control. Since sequences from the DNA bindingdomain of the POU family are highly conserved, thisregion of the Pit-1 eDNA was used as the probe. North-ern blot analysis was carried out at low-stringency con-ditions under which sequences with as little as 50%identity to the POU DNA binding domain would beexpected to bind the probe. Although RNA from GCcells (which express high levels of Pit-1) showed a strongband of the expected size (2.4 kb)24 and a weaker bandcorresponding to the primary transcript (4.5 kb),24 RNAfrom both chorion laeve and decidual cell culturesshowed only very weak hybridization with the probe (notshown). Although this might indicate that a POU familyfactor is expressed at low levels, the intensity of the bandobserved in gel mobility shift assays with chorion laevenuclear extracts (described above) suggested that thefactor involved is highly abundant. Therefore, it seemsmore likely that the placental cell factor that binds thehRen Pit-1 site is only distantly related to Pit-i.

DiscussionIn the present study, we compared the activity of

renin-luciferase hybrid genes containing various hRen5'-flanking DNA sequences in placental cell culturesand a number of cell lines. Further, we compared thebinding of nuclear factors from these different cell typesto a sequence we had previously shown to bind thepituitary-specific factor Pit-1, the major transcriptionalactivator of renin gene expression in pituitary cells. Theresults of these experiments show that (1) hRen 5'-flanking DNA sequences are transcriptionally active inprimary cultures of chorion laeve and decidual cells, (2)common 5'-flanking DNA sequences appear to directhRen promoter activity in placental and pituitary cells,(3) low-level promoter activity in 293 human embryonalkidney cells appears to be largely independent of se-quences contained upstream of the TATA box, (4) hRen5'-flanking DNA sequences are virtually inactive intheir ability to drive transcription in human choriocar-cinoma cell lines JEG-3, JAR, and BeWo, and

by guest on February 21, 2018http://circres.ahajournals.org/

Dow

nloaded from

628 Circulation Research Vol 75, No 4 October 1994

(5) binding of a nuclear factor from placental cells tothe Pit-1 site is correlated with promoter activity. Al-though the placental Pit-i-like factor appeared to berelatively abundant, Northern analysis of placental cellRNA indicated that at the primary sequence level, thisfactor is at best only distantly related to Pit-i. Takentogether, these observations suggest that common pro-moter elements direct renin gene expression in pituitaryand placental cells but do so through interactions withcell-specific transcription factors.

Renin has been detected both enzymatically and im-munocytochemically in chorion laeve10,25 and is alsofound in cell cultures of chorion laeve.10'16'7 Althoughthe source of placental renin remains the subject ofconjecture,26 placental cells in culture are the only un-transformed cell culture systems that express the humanrenin gene. Since, in term placentas, chorion laeve isgenerally more abundant than attached decidua, wecharacterized renin expression in chorion laeve culturesmore extensively. Using a polyclonal antibody raisedagainst highly purified recombinant human renin, wewere able to demonstrate renin immunoreactivity inprimary cultures of chorion laeve. A large number ofcells in these preparations also stained for the smoothmuscle-specific marker a-actin. Since >70% of cellsstained for each marker, it seems likely that renin-expressing cells in this preparation were either derivedfrom smooth muscle cells or, at least, express commongenes. This would imply that common mechanisms mayalso be involved in directing renin expression to smoothmuscle-derived JG cells. Indeed, our preliminary studieswith the mouse renal renin-expressing cell line As 4.127suggest that the cis-acting elements in hRen 5'-flankingDNA that control renin transcription in placental andpituitary cells also function in these renal cells.Renin and a number of other proteins, such as

prolactin and apolipoprotein B (apo B), are unusual inthat they are expressed specifically but in a number ofdifferent cell types. In the prolactin gene, a downstreampromoter interacts with an upstream enhancer to drivelactotrope-specific expression.28 Early in development,the prolactin promoter region appears to be sufficient todirect expression to lactosomatotropic precursor cells,which synthesize both prolactin and growth hormone.29Prolactin expression in the placenta, on the other hand,is directed by a distal promoter that is entirely separatefrom the pituitary-specific control elements.3031 Apo Bexpression is also controlled by promoter and enhancerelements.32,33 However, in this case, different sequencesin the proximal promoter region in combination withcell-specific enhancer and reducer sequences situatedfurther upstream direct expression to liver and colon.The findings of the present study implicate yet anothermechanism in which common promoter elements inter-act with different cell-specific transcription factors.Gel mobility shift assays demonstrated the high-

affinity binding of a factor from placental cell nucleithat correlated closely with expression of both endoge-nous and transfected renin genes. However, low-strin-gency Northern blot analysis of RNA from chorionicand decidual cells with a probe corresponding to thehighly conserved DNA binding domain of the POUfamily of transcription factors failed to detect anyabundant cross-hybridizing mRNAs. Although we can-not rule out a POU family member, the high abundanceof the placental factor in relation to the absence of any

high abundance POU mRNA would suggest the involve-ment of a more distantly related factor. Such a relationoccurs with the homeobox proteins, which are homolo-gous to the POU family in the DNA binding domain.Although individual POU and homeobox proteins mayshare as little as 25% identity in their DNA bindingdomains, these factors all appear to bind the same, orsimilar, AT-rich sequences.34Whereas a number of different transcription factors

may stimulate renin gene expression, not all factorscapable of binding the Pit-1 site are either able orsufficient to activate transcription. Although a nuclearfactor from 293 cells was found to bind the Pit-1 site, theweak promoter activity of renin S'-flanking DNA se-quences in these cells was independent of this site.Previously, mouse renin promoter activity in 293 cellshad been shown to require both the Pit-1 site and a sitecontained in an insertion element specific to mouserenin genes.15 Our findings further show that in theabsence of sequences resembling the mouse-specificinsertion element, there are no other sequences in thehuman renin 5'-flanking DNA up to position -2748,which can activate basal transcription from the hRenTATA box more than twofold. Thus, hRen promoteractivity in 293 cells appears to be due mainly to modestlevels of basal transcription originating from the TATAbox independent of the influence of promoter elementsfurther upstream. A similar pattern of activities hasbeen described for the apo B gene.33 Although differentcell-specific sequences upstream from position -85 arerequired for high-level transcription in apo B-express-ing hepatoma and colon carcinoma cells, proximal se-quences contained downstream from -85 give rise tothe highest levels of activity in Chinese hamster ovarycells, which do not express the endogenous apo B gene.

Burt et al,18 using the JEG-3 choriocarcinoma cellline, demonstrated a twofold stimulation of CAT activ-ity by hRen sequences -892/+13. Since no stimulationwas found in a number of other cell lines derived fromrenin-expressing tissues, the authors proposed thatJEG-3 cells might serve as a useful model system inwhich to study transcriptional control elements in thehuman renin gene. To create a system in which basalexpression was elevated, Burt et al"l subsequentlylinked hRen 5'-flanking DNA sequences to an HSV-TKpromoter. These studies suggested that sequences-892/-149 contain positive regulatory elements whoseactivity is suppressed by sequences -148/+13. In thepresent study, we found no evidence for negative regu-latory elements in the more proximal region. Indeed, inboth placental and pituitary cells this region containedthe strongest positive control elements, whereas nega-tive elements were contained further upstream. Inagreement with the earlier report of Burt et al,18 wefound that hRen 5'-flanking DNA sequences stimulatedexpression of the reporter gene in JEG-3 cells one- totwofold. However, this activity was independent of theextent of 5'-flanking DNA tested and 50- to 100-foldlower than the stimulation afforded in placental orpituitary cells. Taken together, these observations dem-onstrate that the effects of renin 5 '-flanking DNAsequences on the HSV-TK promoter do not reflect thenet effects of these sequences in cells in which the nativehRen promoter is active.

Studies using transgenic animals have shown that aslittle as 0.9 kb 5'-flankingDNA is sufficient for cell-specific

by guest on February 21, 2018http://circres.ahajournals.org/

Dow

nloaded from

Sun et al Renin Expression in Placenta and Pituitary 629

expression of the human renin gene, but only when codingand intervening sequences are also present.6,35 Althoughthese findings show that 5'-flanking DNA sequences prox-imal to hRen are sufficient for gene expression, sequencesdownstream from the transcription start site may containcell-specific transcriptional enhancer elements or contrib-ute to mRNA stability. The low promoter activities of thehRen S'-flanking DNA sequences that we describe hereare compatible with both models. It will be important todetermine whether the proximal sequences we haveshown to drive maximal renin transcription in placentaland pituitary cells are also necessary and sufficient forrenal renin gene expression in transgenic models in vivoand to determine the role of genomic sequences in direct-ing renin gene expression both in transfection experimentsand transgenic animals. We are currently investigating thenature of the factor(s) that interacts with the hRen Pit-1site in placental and renal cells and the role of othersequences that contribute to activity such as the region-148/-99.

AcknowledgmentsThis study was supported by a Grant-in-Aid from the

American Heart Association, New York City Affiliate, Inc (DrCatanzaro); the Michael Wolk Heart Foundation; the MellonFoundation; and Hypertension SCOR grant HL-18323. Wethank Drs John Laragh and Jean Sealey for their ongoingencouragement and support.

References1. Sealey JE, Laragh JH. The renin-angiotensin-aldosterone system

for normal regulation of blood pressure and sodium and potassiumhomeostasis. In: Laragh JH, Brenner BM, eds. Hypertension:Pathophysiology, Diagnosis and Management. New York, NY:Raven Press Publishers; 1990:1287-1317.

2. Alderman MH, Madhavan S, OOi WI, Cohen H, Sealey JE,Laragh JH. Association of the renin-sodium profile with the risk ofmyocardial infarction in patients with hypertension. N Engl J Med.1991;324:1098-1104.

3. Brunner HR, Laragh JH, Baer L, Newton MA, Goodwin FT,Krakoff LR. Essential hypertension: renin and aldosterone, heartattack and stroke. N Engl J Med. 1972;286:441-449.

4. Fukamizu A, Seo MS, Hatae T, Yokoyama M, Nomura T, KatsukiM, Murakami K. Tissue-specific expression of the human reningene in transgenic mice. Biochem Biophys Res Commun. 1989;165:826-832.

5. Sigmund CD, Gross KW. Structure, expression, and regulation ofthe murine renin genes. Hypertension. 1991;18:446-457.

6. Sigmund CD, Jones CA, Kane CM, Wu C, Lang JA, Gross KW.Regulated tissue- and cell-specific expression of the human reningene in transgenic mice. Circ Res. 1992;70:1070-1079.

7. Tronik D, Dreyfus M, Babinet C, Rougeon F. Regulated expressionof the Ren-2 gene in transgenic mice derived from parental strainscarrying only the Ren-1 gene. EMBO J. 1987;6:983-987.

8. Saint AJ, Rohmer V, Pinet F, Rousselet MC, Bigorgne JC, CorvolP. Renin and cathepsin B in human pituitary lactotroph cells: anultrastructural study. Histochemistry. 1989;91:291-297.

9. Deschepper CF, Mellon SH, Cumin F, Baxter JD, Ganong WF.Analysis by immunocytochemistry and in situ hybridization ofrenin and its mRNA in kidney, testis, adrenal, and pituitary of therat. Proc NatlAcad Sci USA. 1986;83:7552-7556.

10. Poisner AM, Wood GW, Poisner R, Inagami T. Localization ofrenin in trophoblasts in human chorion laeve at term pregnancy.Endocrinology. 1981;109:1150-1156.

11. 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.

12. Duncan KG, Haidar MA, Baxter JD, Reudelhuber TL. Regulationof human renin expression in chorion cell primary cultures. ProcNatlAcad Sci USA. 1990;87:7588-7592.

13. 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 USA. 1989;86:56-59.

14. Sun J, Oddoux C, Lazarus A, Gilbert MT, Catanzaro DF. Promoteractivity of human renin 5-flanking DNA sequences is activated by thepituitary-specific factor Pit-1. J Biol Chem. 1993;268:1505-1508.

15. 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;20:3617-3623.

16. 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.

17. 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.

18. Burt DW, Nakamura N, Kelley P, Dzau VJ. Functional humanrenin promoter in transfected cells: evidence for cell-specificexpression. J Hypertens. 1988;6(suppl 4):S429-S431.

19. Itskovitz J, Rubattu S, Rosenwaks Z, Liu HC, Sealey JE. Rela-tionship of follicular fluid prorenin to oocyte maturation, steroidlevels, and outcome of in vitro fertilization. J Clin EndocrinolMetab. 1991;72:165-171.

20. Kingston RE. Calcium phosphate transfection. In: Ausubel FM,Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, StruhlK, eds. Current Protocols in Molecular Biology. New York, NY:John Wiley & Sons Inc; 1991:9.1.1-9.1.9.

21. Schadlow VC, Barzilai N, Deutsch PJ. Regulation of geneexpression in PC12 cells via an activator of dual second mes-sengers: pituitary adenylate cyclase activating polypeptide. MolBiol Cell. 1992;3:941-951.

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

23. He X, Treacy MN, Simmons DM, Ingraham HA, Swanson LW,Rosenfeld MG. Expression of a large family of POU-domain reg-ulatory genes in mammalian brain development. Nature. 1989;340:35-41.

24. 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 a pituitaryphenotype. Cell. 1988;55:519-529.

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

26. Shaw KJ, Do YS, Kjos S, Anderson PW, Shinagawa T, Dubeau L,Hsueh WA. Human decidua is a major source of renin. J ClinInvest. 1989;83:2085-2092.

27. Sigmund CD, Okuyama K, Ingelfinger J, Jones CA, Mullins JJ,Kane C, Kim U, Wu CZ, Kenny L, Rustum Y, Dzau V, Gross KW.Isolation and characterization of renin-expressing cell lines fromtransgenic mice containing a renin-promoter viral oncogene fusionconstruct. J Biol Chem. 1990;265:19916-19922.

28. Crenshaw EB, Kalla K, Simmons DM, Swanson LW, RosenfeldMG. Cell-specific expression of the prolactin gene in transgenicmice is controlled by synergistic interactions between promoterand enhancer elements. Genes Dev. 1989;3:959-972.

29. Jones GJ, Catanzaro DF. Interactions between rat prolactin genepromoter and enhancer regions in mammosomatotrope and lac-totrope cell lines. Mol Endocrinol. 1991;5:1836-1844.

30. Hiroaka Y, Tatsumi K, Shiozawa M, Aiso S, Fukasawa T, YasudaK, Miyai K. A placenta-specific 5' non-coding exon of humanprolactin. Mol Cell Endocrinol. 1991;75:71-80.

31. DiMattia GE, Gellersen B, Duckworth ML, Friesen HG. Humanprolactin gene expression: the use of an alternative non-codingexon in decidua and the IM-9-P3 lymphoblast cell line. J BiolChem. 1990;265:16412-16421.

32. Paulweber B, Levy WB. The mechanisms by which a human apo-lipoprotein B gene enhancer and reducer interact with thepromoter are different in cultured cells of hepatic and intestinalorigin. J Biol Chem. 1991;266:24161-24168.

33. Paulweber B, Onasch MA, Nagy BP, Levy WB. Similarities anddifferences in the function of regulatory elements at the 5' end ofthe human apolipoprotein B gene in cultured hepatoma (HepG2)and colon carcinoma (CaCo-2) cells. J Biol Chem. 1991;266:24149-24160.

34. Levine M, Hoey T. Homeobox proteins as sequence-specific tran-scription factors. Cell. 1988;55:537-540.

35. Fukamizu A, Hatae T, Kon Y, Sugimura M, Hasegawa T,Yokoyama M, Nomura T, Katsuki M, Murakami K. Human reninin transgenic mouse kidney is localized to juxtaglomerular cells.Biochem J. 1991;278:601-603.

by guest on February 21, 2018http://circres.ahajournals.org/

Dow

nloaded from

J Sun, C Oddoux, M T Gilbert, Y Yan, A Lazarus, W G Campbell, Jr and D F Catanzaropituitary lactosomatotropic cell lines.

5'-flanking DNA stimulates promoter activity in placental cell primary cultures and Pituitary-specific transcription factor (Pit-1) binding site in the human renin gene

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1994 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.75.4.6241994;75:624-629Circ Res. 

http://circres.ahajournals.org/content/75/4/624World Wide Web at:

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

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on February 21, 2018http://circres.ahajournals.org/

Dow

nloaded from