Promoter paper

Basal level transcriptional regulation of the human angiotensin IItype 1 receptor gene

Xin Zhao, Mickey M. Martin, Terry S. Elton *Department of Chemistry and Biochemistry, Brigham Young University, C206 Benson Building, P.O. Box 25700, Provo, UT 84602, USA

Received 14 July 2000; received in revised form 1 September 2000; accepted 5 September 2000

Abstract

The peptide hormone angiotensin II regulates a variety of physiological responses which are mediated by its interaction with high affinityG protein-coupled receptors localized on the surface of target cells. To gain insights into the transcriptional regulation of the humanangiotensin II type 1 receptor (hAT1R) gene, we have isolated 1 kb of the 5P-flanking sequence of this gene. Expression constructscontaining various 5P-deletions of the hAT1R promoter region, fused upstream to the luciferase reporter gene, were transiently transfectedinto H295-R, HEC-1B and A549 cells. It was demonstrated that a 145 bp sequence within the promoter region was required for basal levelexpression of the hAT1R gene in all of the three cell lines investigated. Computer analysis indicated the existence of numerous putativetranscription factor binding sites in this region. Further detailed deletion data suggested essential transcription factor binding sites between398 and 379 bp. Electrophoretic mobility shift assays revealed that four protein^DNA complexes were formed within the 398 to 379 bpregion of the hAT1R gene when incubated with H295-R cell nuclear extract. Site-directed mutagenesis experiments showed that a putativeSp1 binding site was critical for the basal level expression of the hAT1R gene. ß 2000 Elsevier Science B.V. All rights reserved.

Keywords: Angiotensin II receptor; Transcriptional regulation; Sp1

The peptide hormone angiotensin II (Ang II) is believedto play a pivotal pathogenic role in the development ofhypertension and atherosclerosis [1,2]. Ang II evokes di-verse physiological responses including arterial vasocon-striction, stimulation of aldosterone secretion and renalsodium reabsorption [3,4]. Additionally, Ang II has beendemonstrated to be a growth-promoting factor for cul-tured rat vascular smooth muscle cells [5,6], renal mesan-gial cells [7], cardiomyocytes [8] and cardiac ¢broblasts [9].

The biological responses to Ang II are mediated by itsinteraction with high a¤nity G protein-coupled receptorslocalized on the surface of target cells [10]. Two main AngII receptor subtypes, AT1R and AT2R, have now beenpharmacologically identi¢ed [4]. AT1R activation by AngII stimulates phosphatidylinositol-speci¢c phospholipaseC, leading to the generation of inositol trisphosphateand diacylglycerol, which are involved in intracellularCa2� mobilization [11,12] and protein kinase C activation[13].

To begin to study AT1R gene regulation in humans and

to begin to investigate its potential role in cardiovasculardisease, our laboratory [14,15] and others [16,17] havecloned the hAT1R gene and demonstrated that it is com-prised of at least four exons and spans greater than 60 kb.To examine the promoter of the hAT1R gene, variousdeletions of the 5P-£anking region (31002 to +39 bp)were generated by PCR and the resulting products weresubsequently subcloned upstream (i.e., blunt end ligatedinto the SmaI site) of the luciferase reporter gene con-tained in the promoterless, enhancerless expression vector(pGL3-Basic, Promega) (Fig. 1). These fusion constructswere designated phAT1R(31002/+39)Luc, phAT1R(3657/+39)Luc, phAT1R(3324/+39)Luc, phAT1R(3207/+39)-Luc, phAT1R(3145/+39)Luc, phAT1R(398/+39)Luc,phAT1R(385+39)Luc, and phAT1R(359/+39)Luc, re-spectively. The authenticity and the appropriate 5P to 3Porientation of the promoter fragments, relative to the lu-ciferase gene, was con¢rmed in all the constructs by di-deoxy chain termination sequencing. These reporter con-structs, or the control construct, pGL3-Basic, were testedfor promoter activity in cell culture (Fig. 2). The cell linesinvestigated were chosen for their ability to express theAT1R (i.e., H295-R human adrenocortical carcinoma[18]) or cells that were derived from human tissues that

0167-4781 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved.PII: S 0 1 6 7 - 4 7 8 1 ( 0 0 ) 0 0 2 2 8 - 1

* Corresponding author. Fax: +1-801-378-5474;E-mail : terry_elton@byu.edu

BBAEXP 91478 31-10-00

Biochimica et Biophysica Acta 1494 (2000) 181^184

www.elsevier.com/locate/bba

express the AT1R (HEC-1B human endometrial adenocar-cinoma and A549 human lung carcinoma). Plasmids forall transfection experiments were prepared by CsCl equi-librium gradient centrifugation. The luciferase plasmidconstructs (2 Wg) were cotransfected with the pCMV L-galactosidase (L-gal) plasmid (Promega) (2 Wg) utilizingthe transfection reagent LipofectAMINE Plus (GibcoBRL, Gaithersburg, MD, USA), following the manufac-turer's recommended procedure. Cells were harvested 48 hafter transfection and luciferase activities were determinedby standard procedures utilizing an Optocomp I luminom-eter (MGM Instruments Inc., Hamden, CT, USA). L-Galactivity was used to normalize for variations in transfec-tion e¤ciency and was determined using standard assays.All experiments were performed three times and normal-ized for transfection e¤ciency. The relative luciferase ac-tivity was calculated by dividing the luciferase light unitsof the reporter deletion constructs by the luciferase lightunits of the pGL3-Basic vector. The luciferase assay re-sults in H295-R cells showed that deleting regions of thehAT1R promoter from 31002 to 3324 bp did not signi¢-cantly a¡ect the luciferase activity (Fig. 2A) while the de-letion from 3324 to 3207 bp dramatically decreased lu-ciferase activity. A deletion from 3207 to 3145 bpresulted in a return of luciferase activity to a level almostthe same as before the deletion of 3324 to 3207 bp,suggesting that a positive response element resides between3324 and 3207 bp and a negative response element re-sides between 3207 and 3145 bp. Further deletion of thepromoter from 398 to 385 bp resulted in a signi¢cantdecrease in luciferase activity in H295-R cells.

To examine potential cell line speci¢city of the hAT1R

promoter, the luciferase activities of the various phAT1R/Luc constructs were also assayed in HEC-1B (Fig. 2B) andA549 (Fig. 2C) cells which do not express hAT1R. Theluciferase activity pro¢le of these various constructs wassimilar to that observed in H295-R cells demonstratingthat the 398 to 385 bp region is important for basalexpression of the hAT1R gene. However, in HEC-1Band A549 cells there was also a signi¢cant decrease inactivity from 31002 to 3657 bp, indicating the presenceof a positive response element within this region. Takentogether, these results suggest that the promoter regionregulating the basal expression of the hAT1R gene (i.e.,398 to 385 bp) is similar in all of the three cell lines

Fig. 1. Schematic representation of the reporter constructs used for tran-sient transfections. A 1041 bp fragment (31002 to +39 bp, relative tothe transcription initiation site) of the hAT1R 5P-£anking sequence wascloned upstream of the luciferase reporter gene in the pGL3-Basic vec-tor to produce the phAT1R(31002/+39)Luc construct. hAT1R promoterdeletion constructs containing progressively shorter sequences at the 5P-end were generated using PCR.

Fig. 2. Transient transfection of 5P-£anking deletion phAT1R/Luc re-porter constructs in (A) H295-R, (B) HEC-1B and (C) A549 cells. Plas-mids for transfection experiments were prepared by CsCl equilibriumgradient centrifugation. The luciferase activity levels, normalized with L-gal activities, were expressed as fold increase over the control (i.e.,pGL3-Basic plasmid). The mean activities þ S.E.M. from three inde-pendent transfections are shown.

BBAEXP 91478 31-10-00

X. Zhao et al. / Biochimica et Biophysica Acta 1494 (2000) 181^184182

investigated. Therefore, this region was selected for furtherstudies.

DNase I footprint analysis was performed according tothe manufacturer's protocol (Core Footprinting System,Promega). The hAT1R promoter region DNA fragmentwas isolated from phAT1R (3145/+39)Luc by digestionwith KpnI and HindIII and [Q-32P]ATP end-labeled. Nu-clear protein extracts were isolated from H295-R andHEC-1B cell lines as previously described [19] and proteinconcentrations were determined by the Coomassie PlusProtein Assay Reagent (Pierce). The DNase I digestionproducts were resuspended in formamide loading bu¡erand separated by electrophoresis through an 8% denatur-ing polyacrylamide gel. Dried gels were subjected to auto-radiography. As illustrated in Fig. 3, a clear protection ofthe 399 to 369 bp region of the hAT1R promoter wasobserved when this fragment was incubated with nuclearextract from H295-R and HEC-1B cells, suggesting thatputative transcription factor(s) can indeed bind to thisregion.

Since transcription factors can speci¢cally interact withthe 398 to 385 bp region of the hAT1R promoter, thissequence was computer analyzed using the Genomatixsoftware GmbH `MatInspector V2.2' software [20]. Thisanalysis revealed the presence of a number of putativetranscription factor binding sites that may play a role in regulating the hAT1R gene. Among these, the putative

Sp1 binding site is very similar to the consensus Sp1DNA binding site. To further de¢ne which speci¢c ele-ment(s) within the 398 to 385 bp region of the hAT1Rpromoter contributed to constitutive hAT1R expression,the putative Sp1 binding site harbored in thephAT1R(398/+39)Luc construct was mutated by site-di-rected mutagenesis (Fig. 4A). Subsequently, the wild-typeand mutant phAT1R(398/+39)Luc reporter constructs aswell as the deletion phAT1R(385/+39)Luc constructs wereindependently transfected into H295-R, HEC-1B andA549 cells (Fig. 4B). The luciferase activity of cells trans-fected with the mutant construct decreased to approxi-mately the same level as the deletion constructphAT1R(385/+39)Luc. These results suggest that thetwo mutated base pairs are critical for the binding of pu-tative transcription factors which play an important rolein regulating the basal expression of the hAT1R gene.

The loss of expression observed upon deletion of the 13bp region located between 398 and 385 bp upstream ofthe transcription start site suggested the presence of basalregulatory element(s). To determine if speci¢c DNA^pro-tein complexes could be detected within this region, elec-trophoretic mobility shift assays (EMSAs) were performed(Fig. 5). Double-stranded probes were prepared by hybrid-ization of complementary oligonucleotides correspondingto 398 to 379 bp of the hAT1R promoter region. Gel-puri¢ed double-stranded probes were end-labeled with[Q-32P]ATP using T4 polynucleotide kinase. Mobility shiftbinding reactions typically contained 15 Wg of H295-R,HEC-1B or A549 cell nuclear extract and 40 000 cpm of

Fig. 3. DNase I footprinting analysis of the hAT1R promoter region. Asize marker (lane 1) and a parallel sequencing reaction (data not shown)were run to precisely identify the protected regions. The protected re-gions are indicated with brackets and nucleotide numbers. The nuclearprotein extracts used for each reaction are as indicated.

Fig. 4. (A) Nucleotide sequence of a portion of the hAT1R promoter re-gion which harbors a putative Sp1 binding site shown in bold letters.The point mutations introduced into the binding site are denoted by as-terisks. (B) Transient transfection of wild type or mutant phAT1R(398/+39)Luc constructs and the phAT1R(386+39)Luc constructs in H295-R, HEC-1B and A549 cells. The luciferase activity levels, normalizedwith L-gal activities, were expressed as fold increase over the control(i.e., pGL3-Basic plasmid). The mean activities þ S.E.M. from three in-dependent transfections are shown.

BBAEXP 91478 31-10-00

X. Zhao et al. / Biochimica et Biophysica Acta 1494 (2000) 181^184 183

end-labeled double-stranded probe in a ¢nal volume of 25Wl of EMSA bu¡er (TE bu¡er, pH 7.5, with 50 mM NaCl,1 mM dithiothreitol and 5% glycerol). Poly(dI-dC)bpoly-(dI-dC) (2 Wg) was included as a non-speci¢c competitor inall reactions. When competitive binding studies were beingperformed, 100-fold molar excess of unlabeled speci¢c ormutated oligonucleotides was pre-mixed with the nuclearextract prior to addition of the probe. The reactions wereincubated for 20 min at room temperature and the result-ing DNA^protein complexes were resolved by electropho-resis using a 6% non-denaturing polyacrylamide gel and0.5U TBE running bu¡er. When nuclear extract fromH295-R cells was used, EMSA experimental results dem-onstrated that four DNA^protein complexes were formedwith the 398 to 379 bp region of hAT1R promoter (Fig.5, lane 2). All four complexes were competed when anexcess of unlabeled speci¢c competitor DNA was used(Fig. 5, lane 3). Inclusion of unlabeled mutated double-stranded oligonucleotides as a competitor did not a¡ectformation of the complexes I, II and IV (Fig. 5, lane 4)demonstrating that these protein^DNA complexes are se-quence-speci¢c. When nuclear extracts from HEC-1B orA549 cells were used, the protein^DNA complexes formedwere distinct from those formed with H295-R cell extract(Fig. 5, lanes 5 and 8), suggesting that distinct transcrip-tion factors can interact with this region. These resultsindicate the possibility that basal level expression of thehAT1R gene may be di¡erentially regulated in a cell-spe-ci¢c manner.

In order to identify and characterize the transcriptionfactors necessary for basal expression of this promoter, weare currently performing DNA/protein UV-crosslinkingand supershift experiments with the important hAT1Rpromoter region, 398 to 379 bp.

In conclusion, we have shown that the 398 to 379 bpregion of hAT1R promoter is necessary for basal levelexpression of the hAT1R gene in all of the cell lines inves-tigated. We have also shown that, in di¡erent cell lines,there are distinct transcription factors or complexes whichconfer cell-speci¢c basal level expression of the hAT1Rgene. It is important to gain an understanding of hAT1Rgene regulation since aberrant regulation of this gene maylead to cardiovascular disease.

This project was supported by NIH/NHLBI ResearchGrant HL48848.

References

[1] G.K. Owens, Hypertension 9 (1987) 178^187.[2] J.S. Powell, J.P. Clozel, R.K.M. Muller, H. Kuhn, F. Hefti, M.

Hosang, H.R. Baumgartner, Science 245 (1989) 186^188.[3] M.J. Peach, Physiol. Rev. 57 (1977) 313^370.[4] P.B.M.W.M. Timmermans, P.C. Wong, A.T. Chin, W.F. Herblin, P.

Ben¢eld, D.J. Carini, R.J. Lee, R.R. Weber, J.A.M. Saye, R.D.Smith, Pharmacol. Rev. 45 (1993) 205^251.

[5] G.H. Gibbons, R.E. Pratt, V.J. Dzau, J. Clin. Invest. 90 (1992) 456^461.

[6] H. Weber, D.S. Taylor, C.J. Molloy, J. Clin. Invest. 93 (1994) 788^798.

[7] T. Homma, Y. Akai, K.D. Burns, R.C. Harris, J. Biol. Chem. 267(1992) 23129^23135.

[8] J. Sadoshima, S. Izumo, Circ. Res. 73 (1993) 413^423.[9] W. Schorb, G.W. Booz, D.E. Dostal, K.M. Conrad, K.C. Chang,

K.M. Baker, Circ. Res. 72 (1993) 1245^1254.[10] F.A. Mendelsohn, J. Hypertens. 3 (1985) 307^316.[11] M.J. Berridge, Nature 361 (1993) 315^325.[12] A.R. Marks, Circulation 86 (1992) III61^III67.[13] Y. Nishizuka, Nature 334 (1988) 661^665.[14] B. Su, M.M. Martin, K.B. Beason, P.J. Miller, T.S. Elton, Biochem.

Biophys. Res. Commun. 204 (1994) 1039^1046.[15] B. Su, M.M. Martin, T.S. Elton, in: M.K. Raizada et al. (Eds.),

Recent Advances in Cellular, Molecular Aspects of Angiotensin Re-ceptor, Plenum Press, NY, 1996, pp. 11^21.

[16] D.F. Guo, H. Furuta, M. Mizukoshi, T. Inagami, Biochem. Biophys.Res. Commun. 200 (1994) 313^319.

[17] K.M. Curnow, L. Pascoe, E. Davies, P.C. White, P. Corvol, E.Clauser, Mol. Endocrinol. 9 (1995) 1250^1262.

[18] I.M. Bird, J.I. Mason, W.E. Rainey, Endocrinology 134 (1994) 2468^2474.

[19] S.M. Abmayr, J.L. Workman, in: F.M. Ausubel et al. (Eds.), CurrentProtocols in Molecular Biology, Green Publishing, New York, 1991,pp. 12.1.1^12.1.9.

[20] K. Quandt, K. Frech, H. Karas, E. Wingender, T. Werner, NucleicAcids Res. 23 (1995) 4878^4884.

Fig. 5. EMSAs for the binding of nuclear extracts from H295-R, HEC-1B and A549 cells to the critical hAT1R proximal promoter region(398 to 379 bp). Addition of 100-fold molar speci¢c or mutated coldcompetitor DNA is indicated at the top of the appropriate lanes. Pro-tein^DNA complexes are labeled I, II, III, IV or N.S. (for non-speci¢cbinding).

BBAEXP 91478 31-10-00

X. Zhao et al. / Biochimica et Biophysica Acta 1494 (2000) 181^184184