Pediatr Nephrol(1990) 4:429-435 9 IPNA 1990

Pediatric Nephrology

Original article Molecular cloning and expression of the rat angiotensinogen gene John S. D. Chan, Albert H. H. Chan, Qin Jiang, Zeng-Rong Nie, Silvana LaChance, and Serge Carri~re Maisoneuve-RosemontHospitalResearchCenter, Universityof Montreal, Schoolof Medicine, 5415 L'AssomptionBoulevard,Montreal, Quebec, Canada, H1T 2M4

Abstract. To identify tissue- and hormonal-specific DNA control cis-elements in the rat gene, we have constructed fusion genes consisting of various lengths of the 5'-flanking region of the rat angiotensinogen gene linked to a human growth hormone (hGH) reporter gene and have introduced them into a subclone of rat pancreatic islet tumor cell line (1056A) which expresses the highest level of angiotensinogen mRNA. As a negative control, we have also introduced them into a human choriocarcinoma cell line (JEG-3), which does not express the endogenous angiotensinogen gene. The level of the expression of these fusion genes in these cells was determined by the level of immunoreactive hGH secreted into the culture medium. The expression of angiotensinogen-growth hormone (ANG-GH) fusion genes, pOGH (ANG N-1498/+18), pOGH (ANG N-688/+18), pOGH (ANG N-110/+18), pOGH (ANG N-53/+18), and pOGH (ANG N-35/+18) was 1.0, 1.8, 1.5, 12.0 and 3.0-fold higher, respectively, than the promoterless growth hormone expression vector (pOGH). The addition of dexamethasone (10.6 M), aldosterone (10. 5 M), and thyroid hormone, L-T3 (10. 7 M), stimulated the expression ofpOGH (ANG N-1498/+18) by 4.0-, 2.5-, and 2.0-fold above the control level, respectively. Combination of dexamethasone ( 104 M), L-T3 (107 M), and ethinyl-estradiol (10-6 M) stimulated the expression of the pOGH (ANG N-1498/+18) to greater than 10-fold over the control. Ethinyl-estradiol (10.6 M) or progesterone (10-6 M) alone had no effect on the expression of the pOGH (ANG N- 1498/+ 18). These studies demonstrate that the induction of expression of the angiotensinogen gene by dexamethasone and L-T3 in 1056A cells is due to a transcriptional mechanism and the 1056A cells could be useful for studying angiotensinogen gene regulation and for identifying the glucocorticoid and L-T3-responsive cisregulatory elements in the angiotensinogen gene. Key words: ANGiotensinogen - Gene expression Human growth hormone fusion gene - Steroids - Thyroid hormone Offprint requests to: J. S. D. Chan

Introduction Angiotensinogen, or renin substrate, is a member of the anti-trypsin, anti-thrombin III-ovalbumin family of proteins [1]. It is synthesized predominantly in the liver and secreted into the plasma, where it is converted to angiotensin I (a decapeptide) by renin (EC 3.4.23.15). It is subsequently cleaved by angiotensin-converting enzyme (dipeptidyl carboxypeptidase, peptidyl peptide hydroxylase, EC 3.4.15.1) into the biologically active octapeptide, angiotensin II. Biologically, angiotensin II causes an increase in blood pressure by direct vasoconstriction of arterioles [2]. The peptide has its effect within a few seconds and its duration of action is only a few minutes. It is one of the most potent stimulators of aldosterone production by the zona glomerulosa of the adrenal gland [3]. Furthermore, it has unique effects on the kidney, where it directly inhibits the release of renin by the juxtaglomerular cells [4] and decreases renal blood flow and glomerular filtration rate [5, 6]. In addition, angiotensin II has been shown to exert dose-dependent, biphasic effects on proximal tubular sodium reabsorption [7]. Thus, angiotensin II plays an important role in the control of blood pressure, water and electrolyte balance. The primary structure of angiotensinogen has recently been elucidated from the deduced amino acid sequence from cloned cDNA [8, 9]. These studies indicate that mature angiotensinogen consists of 452 amino acid residues with the angiotensin II sequence at its amino terminal portion. Human and rat angiotensinogen have been found to have 63.6% homology in their amino acid sequence [8, 9]. More recently, the structure of the angiotensinogen gene has been successfully identified and characterized [10]. Restriction mapping and nucleotide sequence analysis of the cloned DNA fragments indicate that the rat angiotensinogen gene is approximately 11.8 kb pairs long, consisting of five exons separated by four introns. Sequences at the exon-intron boundaries are consistent with the consensus splice junction sequence (GT/AG) observed for other genes [11].

430 A l t h o u g h the liver is the p r i n c i p a l site o f a n g i o t e n s i n o g e n synthesis, recent studies h a v e s h o w n that angiotensino g e n m e s s e n g e r r i b o n u c l e i c acid ( m R N A ) is also exp r e s s e d at l o w e r levels in m a n y e x t r a - h e p a t i c tissues, inc l u d i n g k i d n e y , brain, m e s e n t e r y , atria, lung, adrenal, large intestine, stomach, spleen and the b r o w n a d i p o s e tissues surrounding the aorta [12, 13]. Studies b y I n g e l f i n g e r et al. [ 14] using in situ h y b r i d i z a t i o n techniques s h o w e d that the a n g i o t e n s i n o g e n m R N A is p r e s e n t p r i m a r i l y in renal p r o x i m a l tubules, w i t h a lesser a m o u n t in renal vessels and in the renal m e d u l l a , w h i l e r e n i n m R N A is present p r i m a r i l y in the j u x t a g l o m e r n l a r apparatus and g l o m e r u l a r tuft. O u r r e c e n t studies using N o r t h e r n blot also s h o w e d that the a n g i o t e n s i n o g e n m R N A is e x p r e s s e d in i s o l a t e d rat renal p r o x i m a l tubules ( u n p u b l i s h e d results). A n g i o t e n s i n o g e n m R N A is not d e t e c t a b l e in n o r m a l rat p a n c r e a s [12]. H o w ever, a n g i o t e n s i n o g e n m R N A is f o u n d in all t u m o r cell lines d e r i v e d f r o m a r a d i a t i o n - i n d u c e d rat t u m o r islet cell line, R I N - r [15, 16]. A n g i o t e n s i n o g e n m R N A isolated f r o m the rat islet t u m o r cell line 1 0 5 6 A is a p p r o x i m a t e l y 200 nucleofides l o n g e r than that o f liver, w h i c h has b e e n s h o w n to b e an e x t e n s i o n o f the 3 ' - u n t r a n s l a t e d r e g i o n [16]. F u r t h e r m o r e , studies b y B r a s i e r et al. [16] have d e m o n st-rated that the a d d i t i o n o f d e x a m e t h a s o n e increases angiot e n s i n o g e n m R N A levels a p p r o x i m a t e l y 9 - f o l d a b o v e control. T h e s e studies d e m o n s t r a t e that d e x a m e t h a s o n e induction o f a n g i o t e n s i n o g e n m R N A levels in 1056A cells m i g h t be due, at least in part, to a transcriptional r e s p o n s e and that 1056A cells c o u l d be useful for studying angiotensinogen gene r e g u l a t i o n and for identifying the steroid-res p o n s i v e r e g u l a t o r y sequences in the a n g i o t e n s i n o g e n gene. O u r p r e s e n t studies r e p o r t the D N A sequence o f the 5 ' - f l a n k i n g (-1498 bp) sequence of the rat a n g i o t e n s i n o g e n gene and attempt to l o c a l i z e the cis-regulatory e l e m e n t s w h i c h are i m p o r t a n t for the b a s a l expression, as w e l l as for m e d i a t i n g the effect o f steroid h o r m o n e s (glucocorticoid, estradiol, p r o g e s t e r o n e , aldosterone) and t h y r o i d h o r m o n e (L-T3) in the e x p r e s s i o n o f a n g i o t e n s i n o g e n fusion genes in 1056A cells.

Materials and methods Materials. Both restriction and modified enzymes were purchased either from Bethesda Research Laboratories (Burlington, Ontario, Canada), Boehringer-Mannheim (Dorval, Quebec, Canada), or Pharmacia (Bale d'Urfe, Quebec, Canada). The pGEM-3 vector was purchased from Promega-Biotech (Madison, Wis., USA). Both pOGH and pTKGH, vectors containing the entire genomic sequence of the human growth hormone gene (hGH) with or without thymidine kinase promoter sequence fused in the 5'-end of the hGH gene, respectively, were purchased from Nichols Institute of Diagnostics (La JoUa, Calif., USA). Alpha-[35S] ATP (>1000 Ci/mmol) and gamma-[32P] ATP (3000 Ci/mmol) were purchased from New England Nuclear, Dupont (Boston, Mass., USA). Na-125I was purchased from ICN Biomedicals (Montreal, Quebec, Canada). The radioimmunoassay (RIA) kit for hGH was a gift from NIADDK (NIH, USA). The double-antibody RIA procedure was similar to that previously described for ovine placental lactogen [17]. 12sI hGH was prepared by a slight modification of the lactoperoxidase method of Thorell and Johanson [18]. NIAMDD-hGH-I-1 (AFP-4793B) was used for iodination and as a hormone standard.

Other reagents were molecular biology grade and obtained either from Sigma St. Louis, USA, Boehringer-Mannheim or Pharmacia Isolation of angiotensinogen genomic clones. Recombinant genomic DNA encoding the rat angiotensinogen gene was isolated from a fetal rat liver genomic library in lambda vector EMBL-3 by hybridization of synthetic oligonucleotides complementary to the rat angiotensinogen 5'-flanking sequence and a cRNA probe to the rat angiotensinogen cDNA. The library was constructed by partial (SAU3A) digestion of rat fetal genomic liver DNA cloned into the BamHI site of EMBL-3 [19]. Two positive clones were identified. One of these (clone I) has a molecular weight of 7.5 kb pairs and contains approximately 1.5 kb of the 5'-untranslated sequence, exon I, intron I and part of exon II (position + 380 bp). The angiotensinogen DNA of this clone was transferred into a plasmid vector pGEM-3, via the polycloning site EcoRI. This clone was tentatively designated as pGEM-ANG I. In the present study, we have used pGEM-ANG I exclusively as the starting material. Subcloning and sequencing. Digestion products were separated by electrophoresis in 1.0% low-melting agarose (Boehringer-Mannheim, Canada) and the DNA fragment was extracted from the agarose according to the method described by Maniatis et al. [20]. Ligations for subcloning were performed by combining all products of restriction digests of insert and vector fragments purified as above, using T4-DNA ligase (Bethesda Research Laboratories). The ligation mixture was used to transform HB 101 ofEscherichia coli K-12 by the standard calcium chloride precipitation procedure [21]. Plasmid DNA was prepared from transformed cells by the method of Birnboim [22], including cesium chloride centrifugation [23]. Sequences of DNA fragments subcloned into pGEM-3 were determined using the dideoxynucleotide chain-termination reaction of Sanger et al. [24] or by the modified chain-termination reaction using T7 DNA polymerase with the reagents purchased in kit form from Pharmacia. Construction of fusion genes. A DNA fragment (28 bp) (XhoII-Xho[I, N-53 to N+75) was removed from the pGEM (ANG I) and inserted into pGEM-3 via the polylinkersite, BamHI. This plasmid was termed pGEM (ANG N-53/+75). The orientation and the sequence of the DNA insert were determined by DNA sequencing using both T7 and SP6 primers supplied by Promega-Biotech (Madison, Wis., USA). Deletion at the T-border of the pGEM (ANG N-53/+75) was created by using restriction site, EcoRI, at the polylinker region of pGEM-3 and then sequentially removing by digestion with endonuclease BAL-31. Following repair with DNA polymerase I large fragment (Klenow) and digestion with HindlII, the DNA fragment was isolated from low-melting-pointagarose and inserted into pOGH via polylinker sites (HindlII/HinclI), The plasmid containing angiotensinogen (N-53 to N+18) insert was isolated and designated as pOGH (ANG N-53/+18). This angiotensinogengrowth hormone (ANG-GH) fusion gene has a T-boundary of n+18 relative to the transcriptional start site. The ANG-GH fusion gene, pOGFI (ANG N-1498/+18) was created by isolating the DNA fragment (HindIII-XhoI, N-1498 to N-35) from pGEM (angiotensin I) and inserting into pOGH (ANG N-53/+18) which had been previously digested with restriction enzymes HindIII-XhoI to release DNA fragments (N-53 to N-36). The pOGH (ANG N-688/+18) was constructed by isolating the DNA fragment (BamHI-BamHI, N-688 to N+18) from pOGH (ANG-1498/+18) and re-inserting to pOGH via the polylinker site, BamHI. The pOGH (ANG N-110/+18) was created by removing the DNA fragment of ANG N- 1498 to N-111 from pOGH (ANG N-1498/+ 18) by HindIII/NcoI, repairing with Klenow, religating and transferring into bacteria HB101. The sequences for all fusion genes were confirmed by dideoxy sequencing [24]. Cell culture and DNA transfection. The rat islet cell line RIN 1056A is derived from a radiation-inducedrat islet tumor and is described in detail elsewhere [25]. The human placental choriocarcinoma (JEG-3) cell line was obtained from the American Type Tissue Culture Collection. (The JEG cells were used as a negative control because these cells express no angiotensinogen mRNA as observed by us (unpublished work) and Brasier et al. [26]. These cell lines were grown in 100 x 20 mm plastic Petri

431 5'

(

~.Sk~

E.Col R11

l Sac1

-1498 -1143

11+8kb

s.6kb

:~

XbaI Sail Barn H1 -959 -821

-688

Nci-I -265

I I Xho II Xho II -52 +75

Fig. 1. Schematic representation of the rat angiotensinogen gene. The genomic map of the rat angiotensinogen gene is shown to span 11.8 kb pairs and contain five exons and four introns. The plasmid, pGEM (ANG-1), contains 1498 bp of the Y-flanking sequence, exon I, intron I and part of exon II (N+380, cDNA coordinates). The restriction enzymedigested fragments of the 5'-flanking sequence were subcloned in pGEM-3 vectors, and the nucleotide sequence was analyzed in both directions by the dideoxynucleotide chain termination method

dishes (Gibco, Burlington, Ontario, Canada) in Dulbecco's modified Eagle's medium (DMEM), pH 7.45, supplemented with 10% fetal bovine serum (FBS), 100 units/ml penicillin, and 100 ~g/ml streptomycin. The cells were grown in a humidified atmosphere in 95% air, 5% CO2 at 37 ~C. For subculturing, cells were trypsinized (0.05% trypsin and EDTA) and plated at 2.5 • 104 cells/cm2.

-1489

Plasmids and]or ANG-GH fusion genes were transfected into JEG-3 and RIN 1056A cells utilizing calcium phosphate~mediated endocytosis or the DEAE dextran approach [27, 28]. Cells were transfected with a total of 10 gg supercoiled DNA containing 5 ~tg or less of ANG-GH fusion genes and 5 p.g or more of carrier plasmid pGEM-3. After transfection, cells were grown overnight in DMEM containing 10% FBS. Every 48 h thereafter, media were collected and replaced with fresh media containing depleted 10% FBS. The media were kept at-20 ~C until assayed for immunoreactive hGH activity by RIA-hGH. The depleted FBS was prepared by incubation with 1% activated charcoal and 1% resin (AG 1 x 8) (Bio-RAD, Richmond, Calif., USA) for 16 h or more at room temperature as described by Samuels et al. [29]. To study the effect(s) of steroids and L-T3 on the expression of ANG-GH fusion genes, the cells were incubated in DMEM supplemented with depleted 10% FBS, and various hormones were added as indicated.

Results

Isolation and sequencing analysis of the 5'-flanking sequence of rat angiotensinogen gene Figure 1 shows the schematic representation of the rat angiotensinogen gene and the strategy for sequencing the 5'-untranslated flanking sequence. Figure 2 shows the nucleotide sequence of the 5'-untranslated region of the rat

5'ag tct ctc tct got cac tac cca tot cct gag atq acc atg ago cct tea tot coo

-1433 aga ctt cto aca ggg qac aac aag atg ttc agt gag ctt aag oat cca qcc ate tta qtg SACI cgq t t q oga qct cco aaa aaa ago aaa oaa aaa

-1373 ctt gcc tag caa ggg caa ggc t g q q t t

-1313 aaa aaa aaa aaa gaa t c c agc cat ctq tog t c a gaa aca qaq aoq aqa qaa cct atc t a t -1253 cao t a t

act tog qaa otc tac tct tot qtc cco aqc aca qqq cct cot oct oaa coc caa

-1193 ago tao t a a ggc atc cac aca ctc ctc too qaa ate oat oca coo ctq ccc ttt oat tot -1133 aca too ott got cot cot qqc ttc ctt agt cot gag toc aoa tcc otc ato qct oca oct -1073 tot ~gt goa gac cot cat oct gta Oct tto ato oct coo aea act OOO act tcc cac too

-1013 oat ago tot tct got aat tat ttt aca tot otq cqc ago goa ogt oac tot eta oaq ooo SACI -953 agc t,ca gao aoc arc oat tcc tqc atg qtc cat cto Got tct tct gag cca ago act oct -893

-833 -773

ctg cca ate etc oct ttc eqc ctt eac ttt cta qag cca ctt tag got aaa qqc aqg ttq SALI t g c ctg t q t cga c t c eta qgc caa qag a t t act tga cat act gga too aag qgc aaq t t a

-713

gtc gee a c e cat ggt qga eta cag ggq gaa ace gtc toe cog acc toe too tqc tot eft BamHI ctc agc tcc cac cct tcc ttg gga fcc ace cgt otc art otc ~ta ~ao ere cat oac tat

-653

tcc a~c act caa ggg cog tgc tct gaa acc tcc aqc ctc caa aae a q c cgc ctt eta tcc

-593

aca age cca gaa cat tit qtt tca ata tog eft tac cac agt toq got ttc cca acc toa

-533

~ca got gig cta gaa gtt tcc caa ago tqc ago aag gga c~q qaa qcc cct oto cog aac

-473

ago caa cta acg taa tta caa eca tic aaq Qfc qte cat cct qaq aOc eOc too tqa oft

-413

atq aga qcr~ tom t c t ccc gee ogt ago tec e f t cot ago atq too too oao OCt t a t aaq

-353 -293

cag atc acg aqg qgc ~tt acg tcc c a a qqt etq cat Got Oqc ttc aqt tot cac aac cca NciI *** cog agt cct gtg act otq ate coo oct ecq qoq aac too qqt aaa tat ota act caa coc

-233

cog ccc ttt tit ttt ttt ttt ttt ttt ttt ttt too oto aag ctc arc too cac tag eqc

-173

ttc ctg gaa ggg acc cog cgg aac ate gcc ctg goc tto otc cat ctt qqc taa qcc tog

-113

art cec atg gte ccc oga oct gg~ ice tee oct agc ctc tgt aca qaq tag ~ct oqq aat XhoII XhoI ago tcc arc tto acc coo tcg agt ata oat aag oct oct tag ttc aoe aog qqA TAG CTG

-53

GCT TGT CTG GGC TGG AGC TAA AGO taa gca qaq ccc tct qqg qtc ctt tct oot cct tcq XhoII tg a 9at cc -3'

Fig. 2. Nucleotide sequence of the 5'-flanking region of the rat angiotensinogen gene. The 5'-flanking region of the rat angiotensinogen gene is displayed with relevant restriction endonuclease sites. The capital letters represent the bases in exon I. Bases which deviate from the published sequence are identified by 9 and 9 for Ohkubo et al. [8] and Brasier et al. [26], respectively

432 E.Col R1 Hind III I

l.Col RI

'" i

I

:7

PGEM (ANG-I)

Cul with Xho II

. . . . ..

,.

+.3

...1,

....

(._

~

-~

pGEM (ANG N -53/N +75)

m l / | l l 311KLINOW/HiI'tlli nl ligile with POGH (HIn4 IIl~linc II) H i n d IH

, Xh~ I

I a m I11

_, ( ' . 1lind

pOGH(ANG N-53/+18)

III

| E l m Ht +q

';'

X11d I ~/

NC:O I

"'""; U l l m N'l I

/m~= m

9611d Barn H1

)

"

HiMI [illXltll l i r a 111

I

"" 11,,,,...o,,0,,,,,.,.,+

pOGII(ANG N-35/+18)

t~Ll llgltlefl I Xfld I lllm H~I , _, !

NCd

-II0

pOGH(ANG N-14891+18)

-~5 e l i

~

/

reliaatlt WltPt pOG11 { a i m t.I1

a/

11inll

IIIINCo I

g|ll

XHO I IIIIm )4 I

~0

!

[ I -35

p()GIIIAN(, N-6881+18)

angiotensinogen gene. The sequence on nucleotide base position N-688 to N+I is almost identical to that reported by Ohkubo et al. [8] and Brasier et al. [26] with few base substitutions, as indicated. Construction of ANG-GH fusion genes

The 5'-untranslated region of the rat angiotensinogen gene was fused to the structural gene of hGH reporter gene [30]. Figure 3 shows the schematic representation of the procedure for constructing the plasmids containing the ANGGH fusion gene. Both DNA fragments (ANG N-1498 to N-35, ANG N-53 to N+75) were initially isolated from pGEM (ANG-1) for the construction of fusion genes. Transient expression of fusion-gene constructs in the rat pancreatic islet tumor cell line RIN 1056A and a nonislet human choriocarcinoma cell line JEG-3

The rat pancreatic islet tumor cell line, RIN 1056A, is known to produce somatostatin, glucagon, insulin, and angiotensinogen genes [15, 16], whereas the JEG-3 cell line does not express the angiotensinogen gene [26]. Plasmid containing thymidine kinase enhancer/promoter (TK) fused to the hGH structural gene (pTKGH) was used as positive control, whereas the plasmid containing no promoter (pOGH) was used as negative control. It is apparent that pTKGH is expressed in both 1056A (3.0_+ 0.7 ng/ml, n=9) and JEG-3 (3.5_+1.0 ng/ml, n--9) cell lines, and pOGH is not expressed in either cell line (0.5 _+0.2 ng/ml, n=9, 0.35 +_0.2 ng/ml, n=9, respectively) (Figs. 4, 5).

L 111"INL~'IG~OW"r11 HC,'-.MCPIE

.18

pOGIi(AN(; N-110/+18i

Fig. 3. Schematicrepresentationof the constructionof angiotensinogengrowthhormone(ANG-GH)fusion genes

The full-length ANG-GH fusion gene, pOGH (ANG N-1498/+18), was not expressed in any of the cell lines (0.4_+0.2 ng/ml, n=9, 0.6_+0.2 ng/ml, n--9, respectively, Figs. 4, 5). Other ANG-GH fusion genes, however, pOGH (ANG-688/+18), pOGH (ANG N-110/+18), pOGH (ANG N-53/+18), and pOGH (ANG N-35+18) were 1.8(0.9_+0.2 ng/ml, n=9), 1.5- (0.7_+0.1 ng/ml, n=9), 12.0(6.5-+0.2 ng/ml, n=9) and 3.0- (1.8-+0.4 ng/ml, n=9) fold higher than pOGH (0.5_+0.2 ng/ml) in expressing hGH activity, respectively (Fig. 4). No significant expression of any of these fusion genes was observed in JEG-3 cells (Fig. 5). Effect of steroids and L-T3 on the expression of ANG-GH fusion gene in 1056A cell lines

It has been previously observed that dexamethasone (10-6-10 .9 M) and L-T3 (10-8 M) enhance the level of angiotensinogen mRNA in RIN 1056A cells in vitro [16]. Therefore, we sought to determine whether these hormones increased the transcription of ANG-GH fusion genes in this cell line. Figure 6 shows that the addition of either dexarnethasone (10 -6 M), aldosterone (10-5 M), or L-T3 (10-7 M) increased the expression of ANG-GH fusion gene, pOGH (ANG N-1498/+18) by 4.0- (1.7 __0.2 ng/ml, n=9), 2.5-(1.0+0.4 ng/ml, n=9), and 2.0 (0.8 +0.2 ng/ml, n=9) -fold, respectively, as compared with the control (0.4 +- 0.1 ng/ml, n = 9, without addition of hormones), whereas neither progesterone (I0-6 M) nor ethinylestradiol (E2) (10.6 M) had any effect on the expression of pOGH (ANG N- 1498/+ 18). Furthermore, combined treatment with dexamethasone (10-6 M), E2 (1@6 M), and

433

L-T3 (10--7 M) increased the expression of ANG-GH fusion gene, pOGH (ANG N-1498/+18), to greater than 10fold (4.9 • 0.7 ng/ml, n =9) compared with controls.

1400

z~

O O

~00o 8OO ~0o

9I -

400

Discussion

-r

e ~o O

g

g

.7

.7

~:

Y

.~

Fig. 4. Expression of ANG-GH fusion genes in rat pancreatic islet tumor calls (RIN 1056A). Plasmids, pOHG and pTKGH were used as negative and positive controls, respectively. Amounts of immunoreactive human growth hormone (hGH) released into media were quantified by specific radioimmunoassay for hGH. The lower limit of sensitivity of the assay was 0.2 ng/ml. The inter- and intra-assay coefficients of variation were 10% (n = 10) and 12% (n = 10), respectively. Results are expressed as mean _+SD as percentage of control (pOGH, 100%). Each point represents a minimum of three independent transfections in triplicate plates

A600

E oo~, 19

n

60

~

~

~

Fig. 5. Expression of ANG-GH fusion genes in human choriocarcinoma cells. Results are expressed as mean +_SD as percentage of control (pOGH, 100%)

p ~

J

J

pOGH(AN6 N-1498/+1~) +

L-T3 ~

§

§

E2

9

§

§

/~IW~EI~q~GESlI~/~E L-13 D~EIH/~ONE E2

Fig. 6. Effect of steroids and thyroid hormone (L-T3) on the expression of pOGH, (ANG N-1498/+18 ) in RIN 1056A cells. The plasmid, pOGH (ANG N-1498/+18 ) was transiently transfected into RIN 1056A cells. Dexamethasone, aldosterone, progesterone, ethinyl-estradiol (E2) and L-T3 were added to the culture medium to examine their effect on the expression of pOGH (ANG-N 1498/+18 ). Results are expressed as mean +_SD as percentage of control (without addition of hormones, 100%). The concentrations of steroids and L-T3 added were dexamethasone (10~5 M), aldosterone (10 -s M), L-T, (10-7 M), E2 (10~6 M), and progesterone (104 M)

In the present study, we have used the hGH reporter gene to analyze the regulation of angiotensinogen gene expression, based on the secretion of hGH from transiently transfected cells in culture. This approach offers two unique advantages over the generally utilized transient expression system (i. e., CAT reporter gene system): (1) it is not necessary to destroy cells to assay for expression, and (2) relatively large numbers of cells can be transfected in batches divided into equal aliquots which allows experiments that require many assay points to be done. As a result, the effect of a variety of factors on identically transfected cells can be tested, a task which would be difficult if not impossible with assays requiring a separate plate or flask for each assay point. Moreover, the sensitivity of the hGH reporter gene system is at least 10-fold greater than that of the CAT reporter gene system [30]. By application of this system, we have identified and mapped several locations of cis-acting elements in the angiotensinogen 5"-flanking region which contribute to the basal expression of a hGH reporter gene in a rat pancreatic islet tumor cell line, 1056A. In comparison with plasmids containing pTKGH, it is evident that pOGH (ANG N-1498/+18) fusion genes are not active in both 1056A and JEG-3 cell lines (Figs. 4, 5). The low basal transcription rate of the angiotensinogen gene is consistent with the observation that this gene is under suppressive regulation by the negatively acting ciselements which bind to putative suppressor transacting factor(s). Indeed, the studies of Bouhnik et al. [31] demonstrated that somatic-hybrid cells formed by fusion: of hepatoma cell lines which produce low levels of anglotensinogen with hepatoma cell lines which produce large amounts of angiotensinogen failed to express the angiotensinogen gene. Thus, these studies suggest that dominant putative suppressor transacting factor(s) is present in cells which inhibit the expression of the angiotensinogen gene. As shown in Fig. 4, the expression of the ANG-GH fusion gene is indeed under both positive and negative regulation, The basal activity of pOGH (ANG N-688/+18) was 1.8-f01d greater than that of the promoterless pOGH (IR-hGH, 0.9+0.2 ng/ml vs 0.5_+0.2 ng/ml P ~