Eur. J. Biochem. 213, 1117-1124 (1993) 0 FEBS 1993

The promoter and introdexon structure of the human and mouse /B-adrenergic-receptor genes Anke Van SPRONSEN, Clara NAHMIAS, StCphane KRIEF, Marie-Madeleine BRIEND-SUTREN, A. Donny STROSBERG and Laurent J. EMORINE! CNRS-UPR 0415 and Universitt Paris VII, Institut Cochin de GCnCtique MolCculaire, Paris, France

(Received January 13, 1993) - EJB 930052/1

Transcription-start sites for the mouse and human P3-adrenergic-receptor mRNA have been localized in a region comprised between 150 and 200 nucleotides 5’ from the ATG translation-start codon. Motifs potentially implicated in heterologous regulation of P3-adrenergic-receptor expression by glucocorticoids and by P-adrenergic agonists have been identified upstream from these cap sites. In mouse, a second mRNA initiation region is postulated to exist further upstream. Comparison of the nucleotide sequences of the 3’ end of the human and mouse P3-adrenergic-receptor genes to those of the corresponding cDNA revealed that in contrast to P1 and p2 adrenergic receptors, the p-adrenergic-receptor genes comprise several exons. A large exon (1.4 kb) encodes the first 402 and 388 amino-acid residues of the human and mouse p3 adrenergic receptor, respectively. In man, a second exon (700 bp) contains the sequence coding for the six carboxy-terminal residues of the receptor and the entire mRNA 3‘ untranslated region. In mouse, a second exon (68 bp) codes for the 12 carboxy-terminal residues of the receptor and a third exon contains the p-adrenergic-receptor mRNA 3’ untranslated region. The use of alternate acceptor splice sites generates two forms of exon 3 (600 bp and 700 bp), yielding two P3-adrenergic-receptor transcripts which are differentially expressed in white and brown adipose tissues. Human P3-adrenergic-receptor transcripts with different 3’ untranslated regions are produced by continuation of transcription beyond termination signals. Together, our results suggest that utilization of alternate promoters and/or 3’ untranslated regions may allow tissue-specific regulation of P3-adrenergic-receptors expression.

A third subtype of P-adrenergic receptor, the P3-adrenergic-receptor (P3-AR), has recently been characterized in human and in rodents (Emorine et al., 1989; Granneman et al., 1991; Muzzin et al., 1991; Nahmias et al., 1991). Thermogenesis and lipolysis in brown and white adipose tissues are under the control of this receptor subtype (Arch, 1989; Arch et al., 1991; Zaagsma and Nahorski, 1990). In addition to adipose tissues, expression of p - A R has also been predicted to occur in various other tissues (reviewed by Arch et al., 1991; Zaagsma and Nahorski, 1990) on the basis of pharmacological evidence and has been confirmed in the digestive tract and in gallbladder by the detection of p - A R mRNA (Granneman et al., 1991; Krief et al., 1993). This receptor subtype has thus been suggested to participate in the control by catecholamines of body-energy balance, possibly from assimilation in intestines to storage and mobilization in adipose tissues. Correspondence to L. J. Emorine, CNRS-UPR 0415, Institut Cochin de GCnCtique MolCculaire, 22 rue MCchain, F-75014, Paris, France Abbreviations. AR, adrenergic receptor; AP-1, activator protein1 ; CRE, CAMP-responsive element; GRE, glucocorticoid-responsive element; NF-1, nuclear-factor 1; PCR, polymerase chain reaction. Note. The novel nucleotide sequence data published here has been deposited with the EMBL sequence data banks and is available under accession numbers X72861 and X72862.

Several factors, such as temperature, feeding or fasting, and stress, influence body hormonal status and induce tissuespecific adaptive modifications of energy balance which may in part result from regulation of cellular P3-AR sensitivity. For example, glucocorticoids (Fkve et al., 1992) or P-adrenergic stimuli (Granneman and Lahners, 1992; Revelli et al., 1992; Thomas et al., 1992) modulate P3-AR density or responsiveness as well as P3-AR mRNA levels. Molecular determinants at the basis of such regulation may be found either on the receptor itself or in its gene or mRNA. For instance, post-translational modifications of the receptor in the third intra-cytoplasmic loop and in the carboxy-terminal tail (Bouvier et al., 1991 ; Bouvier et al., 1988; Hausdorff et al., 1990) modulate P-AR coupling to adenylate cyclase. Factors acting on transcription rate or on mRNA stability also affect cellular adrenergic responsiveness by modifying P-AR expression levels (Collins et al., 1989; Hadcock et al., 1989; Port et al., 1992). To gain more insight into the mechanisms forming the basis of the regulated expression of the P3-AR, it is thus of importance to determine the structure of the corresponding gene and mRNA. Previous studies have suggested the existence of several /33-AR transcripts, arising from the use of alternative promoters or polyadenylation signals (Emorine et al., 1989). Moreover, comparison of the p - A R amino-acid sequences predicted from the nucleotide sequences of the human and mouse genomic genes and rat cDNA has revealed

1118 unexpected differences in the length and sequence of the carboxy-terminal regions of the receptors. Although such differences could represent evolutionary species-related variations of the p3-AR, we reasoned that they could also be due to the existence of introns and alternate splice sites in the p3AR gene. In this study, we have characterized the promoters of the human and mouse P3-AR genes. Several regulatory elements have been identified in these regions and are potentially implicated in the regulated expression of the B-AR. We have also shown that the p3-AR gene is composed of several exons, a situation which in the adrenergic-receptor family is shared only with the alB-AR gene (Chodavarapu et al., 1992). p3-AR transcripts with alternate 3' untranslated regions may be generated by alternate splicing in mouse or by continuation of transcription beyond transcription-termination signals in human. The existence of introns in the p3AR gene and of transcripts with distinct 3' ends may allow further regulation of p3-AR transcription and/or mRNA stability.

MATERIALS AND METHODS Determination of mRNA-transcription-start sites The mRNA-transcription-start sites were determined using both nuclease-S1 mapping and primer-extension procedures. For nuclease-S1 mapping, single-stranded DNA from a M13 subclone of the murine p-AR-gene promoter region (fragment BglII-NarI, positions 1-560, Fig. 1D) was used as template for probe synthesis. After polymerization in the presence of [a-"PIdATP, DNA was digested with AlwNI (position 249, Fig. 1D) and the single-stranded probe (380 nucleotides including 310 bases from the p3-AR gene and 70 bases from M13) was isolated on denaturing 6% polyacrylamide gels containing 7 M urea. Probe (105cpm) and total RNA (50 pg) were mixed in 30 pl 40mM Pipes, pH 6.4, containing 0.4 M NaCl, 1 mM EDTA and 70% formamide and denatured for 10 rnin at 85°C. After hybridization (1416 h at 30"C), samples were digested for 30 rnin at 37°C with 100-200U S1 nuclease in 400pl 50mM sodium acetate, pH 4.8, 280 mM NaC1, 4.5 mM ZnSO,. Following incubation for 10 rnin at 65"C, samples were phenol extracted, ethanol precipitated and analyzed on 6 % acrylamide, 7 M urea sequencing gels. For primer extension, antisense oligonucleotide A-74 (36 residues, bp 520-485, Fig. 1D) was radiolabelled using T4 polynucleotide kinase in the presence of [y'2P]ATP and hybridized (5 X lo5cpm) to 50 pg total RNA, in 30 p1 of 20 mM TrisMCl, pH 8, 50 mM NaCI, 10 mM MgCl,, 1 mM dithiotreitol, and 0.1 mM EDTA, for 14-16 h at 37°C. Nucleic acids were ethanol precipitated and incubated for 90 rnin at 42°C after resuspension in 25 pl 50 mM Tris/HCl, pH 8.3,s mM MgCl,, 30 mM KC1 containing 0.5 mM each dNTP, 50 U placenta RNase inhibitor and 40 U avian-myeloblastosis-virus reverse transcriptase. Reactions were terminated by addition of 1 p1 of both 0.5 M EDTA and pancreatic RNase A (1 mg/ml) and incubation for 30 min at 37 "C. After phenol extraction and ethanol precipitation, samples were analyzed on 6% acrylamide, 7 M urea sequencing gels.

Nucleotide-sequence determination Cloning and initial sequencing of the human and mouse D-AR genomic genes have already been described (Emorine

et al., 1989; Nahmias et al., 1991). A 2.5-kb SphI-BglII fragment and a 2.2-kb XhoI-BglII fragment spanning the entire 3' untranslated regions of the human and mouse p3AR genomic genes, respectively, were inserted in both orientations into M13 and sequenced by the standard chain-termination method (Taquence DNA sequencing kit, USB). Primers were nested oligonucleotides synthesized step by step during the course of sequence determination. Polymerase-chain-reaction (PCR)-generated products were purified on agarose or acrylamide gels and directly sequenced (Sequenase DNA sequencing kit, USB) using the PCR oligonucleotides as well as other internal sense and antisense oligonucleotides as primers.

RNA analyses Total RNA was prepared from frozen powdered tissues and digested for 15 rnin at 37°C with 0.3 U RNase-free DNase I (Promega)/pg nucleic acid in 100 mM Tris/HCI, pH 7.5; 50 mM MgC12, in the presence of 2 U/pl of placenta RNase inhibitor. RNA (0.25 pg) was then treated with 400 U of Maloney-murine-leukaemia-virusreverse transcriptase (Gibco BRL) in 20 pl PCR buffer (67 mM Tris/HCI, pH 8.4, 6.7 mM MgC12, 6.7 mM EDTA, 10 mM 2-mercapto-ethanol, 16 mM (NH4),S04, 0.1 mg/ml gelatine) containing 0.4 mM each dNTP, 1OmM random hexanucleotides and 2U/p1 RNase inhibitor. A control without reverse transcriptase was performed to verify that amplification did not proceed from residual genomic DNA. Reaction volumes were brought to 100 pl by adding PCR buffer, 2.5 U Thermophylus aquaticus polymerase (Perkin Elmer-Cetus), each sense and antisense oligonucleotide primers, each dNTP and dimethylsulfoxide at final concentrations of 125 nM, 125 mM and 10% (by vol.), respectively. Samples were then submitted to 29 temperature cycles (92"C, 1 min; 57"C, 1.5 min; 72"C, 1.5 min) followed by 7 min of extension at 72°C in a temperature cycler (PREMTM111, LEP Scientific).

RESULTS The human and mouse /?3-AR-genes proximal promoters Using RNA from human perirenal and omental adipose tissues which contain high levels of p3-AR mRNA (Krief et al., 1993), extension from primer A-74 allowed localization of mRNA cap sites in a region between 150 and 200 nucleotides upstream from the translation-initiation codon (Fig. 1). In murine 3T3-F442A adipocyte RNA, both primer extension and S1-nuclease-mapping assays gave concordant results and also allowed localization of mRNA cap sites at 150-180 nucleotides from the translation-initiation codon (Fig. 1). Two TATA box approximations exist within a canonical distance (25-30 nucleotides) 5' from the cap sites in the human gene and one in the mouse gene (Fig. 1). In both genes, a direct and a reverse CCAAT box are also present upstream from the cap sites and TATA-like sequences. The scores of these TATA motifs, calculated using a weight matrix for TATA box (Bucher, 1990), was low (2396 and 2257 in human and 2872 in mouse) compared to the maximal theoretical score (3380). This situation is certainly responsible for the observed inaccuracy of mRNA initiation. Upstream from the mRNA cap sites, close analogues of the hexanucleotide WGTYCT (W is A or T and Y is T or C) representing glucocorticoid-response elements (GRE) are present in the immediate vicinity of the heptanucleotide

1119

I

0 L

_ _CRE _--_. AGATCTCACCAAGCTGA~TCPTGGGAGAGGAGATACTGGCTGAGCCCTATTACTTAATTT~TACCTTAGGGGAGGCCACCCAAG~ATGC~C 100

------

! CRE TCCTGTGAATCCTTTGCTTGACTCCAGCGGGTTACCTTACCTTTGCCTCTGATACAT~GGGTTACCTGGGGATGGGAGCGCTCTCCTCTCTCCTTCCCCTGCCTTGCT 200

.

! GZe - ! -AP;!-! GTGGGAACTTCTGGGAARGGAGGTGCAGGGCA~CTCCAGG~GCCAGTGCCCAGGGAGTGCTATGCTGAGTCCAGGAGCCTGGCCACGGCAG~TGGACAGA 300

NF-1 ! b d CCAAT ! CCAAT TOO~;GGCAGAGGMCCACGGTGTCCCTTCCTCCAGATTTAGCTAAAGGAARCGTGGAGCATCCCATTGGCCATCCTCCCCACTCTCC~TT~CTCCAG400 ! TATA! TAIA ! ! ! F N , -

-

--__

AGGCCCCTCCAGACTATAGGCAGCTGCCCCTTTAAGCGTCGCTACTCCTCCCCCAAGAGCGGTGGC~CCGAGGGAGTT~GGTG~A~CTGA~GC 500 ! - ! !* I T C T G G C T G G G T T A C C A C A G C T A G A G A n G R T G G C C C A ~ C T G G G G ~ G T C ~ - C ~ ~ ~ ~ A ~ ~ ~ ~ T ~ ~ ~ ~ ~ ~ ~ C ~ ~ ~ ~ C ~ - ~ A G C C A ~ T G600 ATTTGGGAGACCCCCTCC

!la;lsl

TTCCTTCTTTCCCTACCGCCCCACGCGCGACCCGGGG ATG GCT CCG TGG CCT CAC GAG AAC Met A l a Pro TTP PI0 His G l r i A s n

D A G A T C T G T A A T C C C A G C A C T G G G G A G G T T G A G G C A G A A G G A T C ~ A G G T C C A G A C C A A T C T ~ C ~ C A T A T A G ~ G A C T A T C T C ~ C A A T A A100 GAT~

b!

!

CCTTAGGGAGAGCATCCAAGCAGAAGAGGGGCTATCTTGGAT~TTT~TTGTTCGGTTTTGTTTT~TTTG~TCTGGATGGTTGCCTTCCTTGTTGG 200

. . 1 _GRE_NF-I ! -!- -CRE ---. GTAAAGGATAGGGTGCGGGGGTTTCTCTTCTTTGCAGGGTTGCCTCAGGTTCTGCCAGGAAGGAGC~C~AGCTCCAGG~CCGGTGCTGAGGGAGTG 300

!-%I-

!

CCAAT

CCAAT

TATA

!

!

TC~GACAGGACGCCCCTCTCCACCCTCCAATTCCCCACCAGA~CCTCTCTTGTGACTATT~ACGCTGTTCCTTT~GCAGCCACTCCTCCCGGC~ 400

i

-

!

-

.\,

!

!U--

!

-

CTe~T~TACATGGW3GGTGAGATGGAGGGARGCTGACAGACTTACCCCAGC~TTAGGGAAGATGGCCCAGGCTffiAAGAGT~CTCCCAAGCCCTAC 500 T G T C C C C T T C C C T A A G C C A G C G G G T C T G G G G A G G A G G G G G A T 2 GCT CCG TGG CCT CAC AGA AAC

IA-741

s1

4

Met A l a Pro T r p Pro His A r g A m

Fig. 1. Promoter characterization of the human and mouse 83-AR genes. Human m-AR mRNA cap sites were identified by primer extension (A) using perirenal (lane 1 ) and omental (lane 3) adipose tissues RNA. Mouse PJ-AR mRNA cap sites were determined by primer extension (B; left panel) and S1 nuclease protection assays (B; right panel) using 3T3-F442A adipocytes RNA (lanes 4).RNA from Chinese hamster ovary cells expressing the human pl-AR gene (lanes 2) was used as control. Lane p shows the undigested probe used in S1-nuclease-mapping assay. Fragment lengths were determined using sequencing-reaction products as molecular-size markers (M) and are indicated (in bp) on the sides. The nucleotide sequences of the human (C) and mouse (D) promoters show the positions of mRNA cap sites determined by primer extension ( 0 )and S1-nuclease mapping (V).CCAAT and TATA-like boxes, GRE and CRE, and AP-1-recognition and NF-1-recognition sequences are highlighted. Arrows with boxes indicate positions and identification numbers of sense (e.g. S-78) and antisense (e.g. A-74) primers used in this study. The sequence of sense primers is read directly on the figure, whereas that of antisense primers is the reverse-complement of the sequence presented. Oligonucleotide A-74 used for primer extension was synthesized according to the mouse sequence; the homologous region is indicated (---) in the human sequence. The most 3’ nucleotide of the probe used for S1nuclease mapping is indicated S1 (J) in the mouse sequence.

TGASTMA (S is G or C andM is A or C) recognized by the transcription factor activator protein 1 (AP-1; Fig. 1). Two direct TGGCA recognition sequences for nuclear-factor 1 (NF-1) exist in the human, and a single one in reverse orientation, in the mouse promoters (Fig. 1). Motifs similar to the consensus sequence GGCWCTGGTCAKG (K is G or T) defined as fat-specific elements (Hunt et al., 1986) are found in direct (+) and reverse (t) orientations in the pro-

moter region of the human (position 237t248, 279+290, 499+509 and 565t577) and mouse (position 16+27, 39+49,248+259,334+346 and 522+534) P3-AR genes. In the human gene, CAMP-responsive elements (CRE ; Fig. 1C) have been characterized by their ability to confer CAMP responsiveness to a viral promoter (Thomas et al., 1992).Another CRE was identified upstream from the region presented here. In mouse, we found a single potential CRE

1120 (TGAGCTCC, nucleotides 270- 277) in the promoter region so far sequenced (Fig. 1D). The human and mouse P3-AR genes display several ATG codons between the mRNA cap sites and the receptors’ translation-initiation codons. In the human gene (Fig. lC), there is a 16-codon open reading frame containing two ATG codons (positions 524 and 551). In mouse (Fig. lD), there is an 8-codon open reading frame containing 2 ATG (positions 412 and 424) and a second open reading frame (beginning at nucleotide 465) of 16 codons which is highly similar to that found in the human gene. The translation of these reading frames is M A Q A G E(R) V A L(P) M(K) P C(Y) C P L P (the mouse residues are indicated in parentheses at positions which differ between the two species).

Introdexon structure of the human and mouse 83-AR genes Determination of the nucleotide sequences of the 3’ regions of the human and mouse P3-AR genomic genes revealed that polyadenylation signals are found 3 kb and 2.8 kb 3’ from the mRNA-initiation regions. In these two species however, the size of P3-AR transcripts is in the range 2.1 2.4 kb (Fkve et al., 1991; Krief et al., 1993; Nahmias et al., 1991), suggesting the presence of introns in the P3-AR genes. This was further supported by PCR experiments showing that cDNA fragments amplified using primers in the 3‘ regions of the genes, were shorter than expected from the structure of the genomic genes (Fig. 2). In cDNA from human gallbladder and adipose tissues, the length of the fragment generated using primers S-102/A-82 was about 320 bp whereas it was, as expected, about 1350 bp in genomic DNA (data not shown). With primers S-86/A-103, a fragment of the expected size (588 bp) was obtained both from cDNA and genomic DNA. In cDNA from mouse white and brown adipose tissues, two fragments (744 bp and 831 bp) were amplified with primers S-102/A-101 whereas genomic DNA yielded a single 1550-bp fragment (data not shown). The nucleotide sequences of the amplified cDNA fragments were determined and compared to those of the corresponding regions of the human and mouse /33-AR genomic genes. This revealed the presence in both the human and mouse D-AR genes, of introns interrupting the coding and 3’ untranslated sequences of the cognate cDNA (Fig. 3). The human p - A R gene is composed of two exons and a single intron (Fig. 3A). Exon 1 (about 1.4 kb), spans the 5’ untranslated region and the major part of the coding block (402 amino-acid residues). It is separated by a 1025-bp intron from exon 2 (660 bp) which contains 19 bp of the coding region and the entire 3‘ untranslated region. In mouse, there are three exons and two introns (Fig. 3B). Exon 1 (about 1.3 kb) contains the 5‘ untranslated region and most of the coding block (388 amino-acid residues). It is separated by a 463-bp intron from exon 2 which contains 37 coding nucleotides and 31 bp of 3‘ untranslated region, the remainder of which is carried by exon 3. Determination of the nucleotide sequence of the two fragments amplified by PCR with primers S-102/A-101 in mouse RNA (Fig. 2) demonstrated that they represented two forms of exon 3 (680 bp and 593 bp) depending on whether excision of intron 2 occurred between donor-splice-site 2 and either acceptor-splice-site 2 or 2* (Fig. 3B). In the human gene, a consensus donor-splice sequence was found 4 bp downstream from the one determined above (Fig. 3A). Alternate splicing at this site would generate

Fig. 2. PCR analysis of fl-AR-mRNA 3‘ untranslated regions. RNA samples were treated (+) or not treated (-) with reverse transcriptase and submitted to PCR using the sense and antisense primers (Fig. 3) indicated under the panels. PCR-amplified fragments were visualized on 2% agarose gels. Human RNA was prepared from gallbladder (lanes 1) and omental (lanes 2) and perirenal (lanes 3) adipose tissues. Mouse RNA was prepared from white (lanes 4, 6 and 8) and brown (lanes 5, 7 and 9) adipose tissues from three animals. Human RNA was analyzed with two overlapping pairs of primers because poor amplification was obtained with primers S-102/A-103 corresponding to those used for mouse. Migration of molecular-size markers (123-bp ladder, GIBCO-BRL) is shown on the right.

a 402-amino-acid instead of 408-amino-acid P - A R . To investigate this possibility, two 21-residue sense PCR primers with identical 5‘ sequences (the last 17 nucleotides of exon 1) but with 3‘ ends being the first four nucleotides either of exon 2 or of the intron (S-TGCCAACGGCTCGACGGGGCT-3’ or 5’-TGCCAACGGCTCGACGGGTAG-3’, respectively) were synthesized. Using each of these sense primers in combination with the antisense primer A-82, amplification of a cDNA fragment by PCR was obtained only with the sense primer corresponding to the sequence spliced at the first donor site (Fig. 4). Do several &3-AR transcripts with alternative 5’ or 3’ ends exist?

Previous studies have revealed that in addition to the major P3-AR mRNA species (2.2-2.4 kb), additional P3-AR transcripts of higher sizes are present in mouse and rat adipose tissues (Emorine et al., 1989; Fkve et al., 1991 ; Granneman et al., 1991 ; Muzzin et al., 1991; Nahmias et al., 1991). These could result from utilization of either alternate promoters or polyadenylation signals. The possibility that transcription of the mouse P3-AR gene could proceed from a second distal promoter was suggested here by the observation that a 3 10-nucleotide frag-

1121

A ... (

Leu I l e Tyr cys Arg Ser P r o 1 6 6 8 ) - C E TPIC,--p2-C>s-&GC-ssG s-10 /

Pro Arg L e u cys Gln Arg Leu Asp G1

....... / I .......CCC

ADO CTT TGC CAA CGG CTC GAC GG gtaggtaaccggggcagagggaccggcggctc

Donor

A- A-

402 1874

agggtcgggaagcatgcgatgtgtccgtgggtcaactttttgagtgtggagtttattaagagaaggtgggatggctttgcttggagagaaaagggaacgaggagtagcgaaccaaaatgg gacccagggtccttttctttccggatccagtcactagggtagaagcaaaggagggcgagcgggccgtcgttcctcacccaaggacccaaggtgcgccaccggaaagcgctgcggtgtccc gaggactctcgcctcgcctggtcggctttagggattttttttttttttaaatagagacagggtttcgtctctgtcgcccacgcgggaatgcagtggtgcgatctcagctcactgcagtct tgaactcctggctcctgggctcaagcgatcctcccacctcagcctcctgagtatctgggactacaggcgagccccaccaatcccagctatttttaaaatttcttgtagagatggggtctt gctatgttgcccaggcttgtcttgaacttctg~cctcaagtgatccttctgcctcagccttccaaagcattaggattacaggccggagccagggcgccgggtcggctctagttttggttt tccagctcagttctttgcccccctcccccgatttcttgccatcactagacctggctcggacttgaaggcagggctagtgcccccccacccgccccccaagccctcggcct~gttctggg ttttctcaaaggtttgacagctgtg~aggtgagaatccacttccggtatgaagtacagttgtgagtgaggagcctgtgagtgcagaCgtg~gccctcccgctccctgggctgggttggag tagggatggggtggggcgtgtgtggctgggtggtgccctggcgtttttgtgtaactaaatatgcgttccagggtctctgatctctgtcaCtcccctcagtgcacctgttgctcctttcac

1994 2114 2234 2354 2474 2594 2714 2834

v Ala ier E Trp Gly Val Ser cccagggtctattatctccactttt.tttcccagp GCT TCT 'rGG GGA GTT TCT tag gcctgaaggacaagaagcaacaactctgttgatcagaacctgtggaaaacctctgg A Accaptor / cctctgttcagaatgagtcccatgggattccccggctgtgacactctaccctccagaacctgacgactgggccatgtgacccaaggagggatccttaccaagtgggttttcaccatcctc

408 2945

IA-821

\

3065

ttgctccctgtctgagagatgttttctaaaccccagccttgaacttcactcctccctcagtggtagtgtccaggtgccgtggagcagcaggctggctttggtaggggcacc~atcacccg 3185 gcttgcctgtgcagtcagtgagtgcttagggcaaagagagctcccctggttccattccttctgccacc~aaccctgatgagaccttagtgttctccaggctctgtggcccaggctg~ga 3305 gcagcagggtagaaaagaccaagatttggg9ttCtatctctggttcccttattactgctctcaagcagtggcctctctcactttagccatggaatggctccgatctacctca~agcagtg 3425 Poly A tcagaaggacttcgccagggttttgggagctccagggttcataagaaggtgaaccattagaacagatcccttctttt~cttttgcaatcagata~tat~~ctgaatgcagttcat 3545

IA-1031

\

--.....-.

I

cctcggccccctttccctccgtttgttttcttttcataatccacttactcccttcccttctactctgctggcttttgacagaggcgtaaattaggcctaatcctcactctttt~ttccta 3665 atgttcatcaaagaaaaa

\

B

Val Ile Tyr cys Arg S e r P r o (1588)-GTC ATC TAC TCC CGC AGC CCG

IA-1001

387 1771 Donor-llicgtctctgtctctatttttgagtttggagatt9ggg~a~~ggaagatgta~at9999~t9c~9t9tgtgtgt99~t9g9gggtggcctttgtcttgagaggacagaaaagaggtaggaac 1891 taaaacgggccctttctcttcttggatccaatccctgggtctgaagcaaaagggaggaaggggataattgcgcaccttaggaccaggtgacccccacaggcagttgctgctcttccggca 2011 ggtttctgacctctctggtcgcctctagtttg999tttgtttgttttt9ttt9tttgtttgtttgttttgtttttttagttcccttcttcgggaacccaggcatctctatacctgt~tgg 2131 Ala Arg Gln S e r Pro P r o Leu Asn Ar AGC CAD AGT CCA CCG CTC AAC AG gtaggggacacgagcgggggaccggagtctctgggtgggga

... I / ... GCC

g Phe Asp Gly Tyr Glu Gly Ala Arg Pro P h e P r o T h r * gatatccatagacagcaatggacttccctagtcctcggcctcagtcccgctctctctcaaag G Tl"r GAT GGC TAT GAA GGT GCG CGT CCG TIT CCC ACG tgaaggg

400 2237

-AACC*ptOr-l

ccgtgaagatccagcaaggaagct tgagttggcttggagttgctttcctccctcagggactggattagaactatagggtgggacttgggggggagggagggtgcaggatggaccct~tg Donor-1 ggatttgggggtggagtagagggatgcgggaatggtccctatatcttt9aaaagtgaatatgcttttcagggttcctgaat~cttccctcttccttccagtgcttgatccccatcttct tgactggttgccccaagaaatattgtttccgtttttgcaggacttctg9g9attttttttttcctccagaaagacaagcaacggctatggatgcaacatttttataatgcctttgatttc Accaptor-2 tactcagagtgagtcccctggaacctcaactctccaacgctccagaaccgatgactagaccacgaggtgtaagggaaatcttaccaaatgggtttcaccgtcctctctctctttccgaga

&

2357 2477 2597

-A

-A

.

2717

ACCOptOS-J*

gaagttgtctaagacccaccttgaacttcactactacctcagcagctgggacggcaggccacctgtgcttgacggccctgggaggagccctatggccttggaggcctgccagtccctgcc tatgtttgtgctgtatgcttagggaaaagagagcacccctccctccctttcttcctactgctttcctaaccct9atgatcgacatgttcctccacaaatcactctgtctccaggctctgt

2837 2957

lA-91/S-921 gtctctggttagtttgagagcaggaatccaggaaaaaaaaaaagtttgaggtttcatccctgtctcctcactatggctctctaagcaccatcttggaccatctctcacaataggcacaaa 3077 Poly A acagctctaatctacctcacagttaggacttcaaggtttgggggggaaattccagggttcataggaagaagtcaaactattggaatgggtcctttttccacttaaaatcaaatt~ 3197

IA-1011

\

-

w

.

tattattgaa~~t~~~~tgtcccctgctcgccttttctc~g9gtttgttttcttttcgtggcctgcttgctggcttccttgctccgagctgcgttttgacaggggcagtaaattaggagt 3317

==>>:-:-

\

aatccttgcctctttcttcctaatcctcatcagacacaaccagaaagtctgtctgtgtaagtgaggcagtcgagtctttgcctcgccttcctccccaccttttctgaaacttttgagat~ 3437

Fig. 3. Sequence analysis of jD-AR-mRNA 3' untranslated regions. Comparison of the nucleotide sequences of the 3' extremity of the human (A) and mouse (B) p - A R genomic genes to that of the corresponding PCR-amplified cDNA fragments (Fig. 2) allowed to determine the intron/exon stmcture of the genes. Numbering of the sequences is the continuation of that in Fig. 1, taking into account the coding regions of the genes (accessible in GENBANK and EMBL databases). In human, a single intron is spliced out at positions indicated by arrowheads between the underlined donor and acceptor sites; further 3' from the donor splice site, a second potential donor site is also underlined. In mouse, excision of introns occurs between splice sites donor-I and acceptor-I and between splice sites donor-2 and, alternatively, acceptor-2 or acceptor-2". In exons, codons are in capital letters and their amino-acid translation indicated above. In the intron of the human gene, the region homologous to mouse exon 2 is underlined. In the 3' untranslated region of both genes, polyadenylation signals are boxed and transcription-termination signals are highlighted by dotted lines. Primers are indicated as in Fig. 1. Primer S-102 corresponds to a sequence within the coding region of the mouse B3-AR gene; the homologous region in the human sequence is indicated (---).

ment, corresponding to the distance between the NarI and AZwNI extremities of the probe, was protected from nucleaseS1 digestion (Fig. 1A). PCR experiments using primers located upstream from the actually determined cap sites (S-78/ A-79 for human and S-76/A-77 for mouse) confirmed the existence of alternate promoter(s) in the mouse, but not in the human p - A R gene (Fig. 5 ) . The existence of alternate 3' ends in the human and mouse m-AR mRNA was investigated by PCR using a sense primer within the 3' untranslated regions and an antisense primer located 3' from the AATAAA polyadenylation signal (Fig. 5). In human gallbladder and adipose tissue cDNA, primers S-86/A-100 yielded a fragment of 750 bp, suggesting that B - A R transcription may proceed beyond the AATAAA

polyadenylation signal characterized in this study. In contrast, using primers S-92/A-105, no transcript extending beyond the AATAAA polyadenylation signal could be detected in P3-AR cDNA from mouse adipose tissues (Fig. 5 ) . The sequences downstream from the polyadenylation signals were screened against a weight matrix for transcription-termination signal (Gribskov and Devereux, 1991). Two overlapping terminator motifs were found in the human gene whereas four are present in the mouse gene, at a canonical distance (20- 50 bp) downstream from the polyadenylation motif. The scores of these motifs where 340 and 303 in human and 346, 315, 340 and 303 in mouse, for a maximum theoretical score of 391. Transcription termination of the AR gene may thus be more efficient in mouse than in human.

m-

1122 AR gene and of transcripts with distinct 3’ ends may allow further regulation of p3-AR transcription andor mRNA stability.

The promoters of the human and mouse m - A R genes

Fig.4. PCR analysis of the potential utilization of an alternate donor splice sequence in the human m-AR gene. Utilization of the second donor splice sequence identified in the human P3-AR gene (Fig. 3) was analyzed by PCR in RNA from human gallbladder (lanes 1) and adipose tissues from perirenal (lanes 2), omental (lanes 3) and subcutaneous (lanes 4) deposits. Antisense primer A-82 was used in combination with any of two sense primers sharing identical 5‘ sequences (the last 17 nucleotides of exon 1) but with 3’ ends being the first four nucleotides either of exon 2 (lanes a) or of the intron (lanes b). Amplification of a cDNA fragment was obtained only when the first sense primer was used although the second primer allowed amplification of a fragment of the expected size (1188 bp) when genomic DNA (lanes 5 ) was used as template.

Fig. 5. Alternate 3‘ and 5’ untranslated sequences in human and mouse jY3-AR mRNA. Existence of P3-AR transcripts extending either 3’ from the transcription-termination signals or 5‘ from the cap sites characterized here was analyzed in RNA from human gallbladder (lanes 1) and omental adipose tissue (lane 2 ) and from mouse white adipose tissue (lanes 4). Positive controls were human (lanes 3) and mouse (lanes 5 ) genomic DNA. PCR primers used for these studies were S-86lA-100 and S-92lA-105 for human and mouse /?3-AR mRNA 3’ ends, respectively (Fig. 3), and S-781 A-79 and S-76/A-77 for human for mouse P3-AR mRNA 5’ ends, respectively (Fig. 1).

DISCUSSION In this study, we have characterized the minimal promoters of the human and mouse p3-AR genes. Several potential regulatory elements have been structurally identified and their proposed biological role is now amenable to further functional analysis. We have also shown that the p3-AR gene is composed of several exons, a situation which in the adrenergic-receptor family is shared only with the alB-AR gene (Chodavarapu et al., 1992). P3-AR transcripts with alternate 3’ untranslated regions may be generated, either by alternate splicing or by continuation of transcription beyond transcription-termination signals. The existence of introns in the p3-

Several transcription-start sites have been mapped 150200 bp 5’ from the translation-initiation codon of the human and mouse p3-AR genes. Utilization of multiple mRNA cap sites has often been observed for genes, such as those for the p2-AR and alB-AR (Chodavarapu et al., 1992; Emorine et al., 1987; Kobilka et al., 1987), the promoters of which contain CCAAT boxes or Spl-binding sites, but lack canonical TATA boxes. This is particularly true for the human P3-AR gene where the TATA-like region largely departs from the consensus TATA-box sequence (Bucher, 1990). The cap sites, and the TATA-like and CCAAT boxes found in their vicinity represent the RNA-polymerase-I1 promoters of the P3-AR genes and allow tentative ascription of regulatory roles to upstream sequence motifs. Elements homologous to sequences conserved among several fat-cell-specific genes (Hunt et a]., 1986) are present in the promoter regions of the human and mouse P3-AR genes. Moreover, both genes display the TGGCA sequence recognized by NF-1-like transcription factors which have recently been suggested to be involved in tissue-specific expression of adipocyte P2 gene (Graves et al., 1991). These findings are in agreement with the predominant expression of P3-AR in adipose tissues of all species studied (Arch, 1989; Fbve et al., 1991; Granneman et al., 1991; Krief et al., 1993; Langin et al., 1991; Muzzin et al., 1991; Nahmias et al., 1991 ; Zaagsma and Nahorski, 1990). In the promoter regions of the human and mouse p3-AR genes, close approximations of GRE are found in the vicinity of binding sites for transcription factor AP-1. As for other genes negatively regulated by glucocorticoids, inhibition of P3-AR mRNA transcription by dexamethasone (Fkve et al., 1992) may thus result from negative interactions of the glucocorticoid receptor with transcription factor AP-1 (Diamond et al., 1990; Yang-Yen et al., 1990). Alternatively, since several fat-specific elements occur in this region, it is possible that negative interferences of the glucocorticoid receptor with adipose-specific factors are responsible for this phenomenon. A recent study has suggested that treatment of mouse 3T3-F442A adipocytes with isoproterenol increased cellular P3-AR mRNA, possibly by a transcriptional mechanism (Thomas et al., 1992). This effect could tentatively be attributed to CAMP-induced stimulation of p3-AR mRNA transcription via the CRE described here. The results of Thomas et al. (1992), however, differ from those of others (Granneman and Lahners, 1992; Revelli et al., 1992) who reported decreased p3-AR-mRNA steady-state levels in white and brown adipose tissues of rats exposed to P-adrenergic stimuli. Since it is suggested here that mouse P3-AR-mRNA transcription may proceed from several promoters, it is possible that differences between the results of these in-vitro and invivo studies are due to differential actions of regulatory effector(s) acting on one or the other promoter.

Introdexon structure of the P - A R genes The human and mouse P3-AR genomic genes were initially characterized by our group (Emorine et al., 1989; Nah-

1123 mias et al., 1991). By homology with those for the pl-AR and p - A R , it was assumed that the P3-AR gene was also intronless. However, the present study demonstrates that in contrast to the pl-AR and m-AR genes, the P3-AR gene contains one or several introns interrupting the mRNA coding and 3’ untranslated region. A large exon encodes the major part of the human and mouse p3-AR (402 and 388 amino-acid residues, respectively). In man, a second exon encodes six additional carboxy-terminal residues and the 3’ untranslated region. In mouse, two further exons encode the 12 carboxy-terminal residues of the receptor and the 3’ untranslated region, respectively. Although a sequence similar to that of mouse exon 2 exists in the human gene, splicing of exon 1 to this potential exon does not seem to occur in man in spite of the presence of a potential acceptor splice site at its 5’ extremity. The size of the spliced human and mouse p3-AR transcripts (2.1 kb) calculated by subtracting the introns from the distance between the transcription initiation and termination sites identified here, is in good agreement with that experimentally determined for the major p3-AR mRNA species (2.2-2.4kb; Fkve et al., 1991; Krief et al., 1993; Nahmias et al., 1991). This suggests that neither genes contain any intron other than those described here. In addition, when mouse exon 1 is spliced to exon 2, the resulting coding region presents uninterrupted sequence similarities with that of the rat P - A R cDNA (Granneman et al., 1991;Muzzin et al., 1991). Finally, the sizes of PCR-amplified fragments obtained with several pairs of oligonucleotides distributed over the whole of exon 1 were identical whether genomic DNA or cDNA matrices were used. Characterization of Chinese hamster CHW-1102 fibroblasts transfected with the entire human P3-AR gene (Nantel et al., 1993) showed that the full-length receptor (408 aminoacid residues), has the same pharmacological profile as the shorter receptor (402 amino-acid residues) initially expressed in Chinese hamster ovary cells (Emorine et al., 1989). This is in agreement with experiments showing that truncation of most of the carboxy-terminal tail of the p - A R had no influence on ligand binding and receptor coupling (Dixon et al., 1987; Johnson et al., 1990). In mouse adipose tissues, alternate splicing generates two P - A R transcripts differing in their 3’ untranslated regions. In three animals examined, the shorter transcript was more represented in white than in brown adipose tissue whereas the longer was slightly predominant in brown adipose tissue. These different expression levels of the two P3-AR transcripts may be due to different, possibly tissue-specific, modulation either of mRNA splicing or stability. The additional sequence present in the longer transcripts contains a 10-nucleotide poly(U) region and a 11-nucleotide AU-rich motif. This is of particular interest in view of the potential involvement of such motifs in the agonist-induced destabilization of p2-AR mRNA (Port et al., 1992), and may be more generally of mRNA of receptors positively coupled to adenylate cyclase. Similar mechanisms could also occur in man since we have shown here that P3-AR transcripts with different 3’ ends may be generated by continuation of transcription further from the polyadenylation and transcriptiontermination signals. We wish to thank Dr D. Caput (Sanofi Elf BioRecherches, Labkge, France) for helpfull discussions and Dr A. Szabo (INSERM Unit6 152, Paris, France) for critical reading of the manuscript. This work was supported by grants from the Centre National de la Re-

cherche Scientifque, the Universite‘ Paris VII, the Institut National de In Sante‘ et de la Recherche Mkdicale, the Minist&-e de la Recherche et de la Technologie and the Bristol-Myers-Squibb Company (Princeton, USA).

REFERENCES Arch, J. R. S. (1989) The brown adipocyte P-adrenoceptor, Proc. Nutr. SOC.48, 215-223. Arch, J. R. S., Cawthorne, M. A., Coney, K. A., Gusterson, B. A,, Piercy, V., Sennit, M. V., Smith, S. A., Wallace, J. & Wilson, S. (1991) P-adrenoceptor-mediated control of thermogenesis, body composistion and glucose homeostasis, in Obesity and cachexia (Rothwell, N. J. & Stock, M. J., eds) pp. 241 -268, John Wiley & Sons Ltd., London. Bouvier, M., Hausdorff, P., De Blasi, A., O’Dowd, B. F., Kobilka, B. K., Caron, M. G. & Lefkowitz, R. J. (1988) Removal of phosphorylation sites from the fl-adrenergic receptor delays onset of agonist-promoted desensitisation, Mature 333, 370- 373. Bouvier, M., Guilbault, N. & Bonin, H. (1991) Phorbol-ester-induced phosphorylation of the fl-adrenergic receptor decreases its coupling to Gs, FEBS Lett. 279, 243-248. Bucher, P. (1990) Weight matrix description of four eukaryotic RNA polymerase I1 promoter elements derived from 502 unrelated promoter sequences, J. Mol. Biol. 212, 563-578. Choddvarapu, S. R., Kincade Denker, J. M., Perez, D. M., Gaivin, R. J., Riek, R. P. & Graham, R. M. (1992) Genomic organization and expression of the human a,,-adrenergic receptor, J. Biol. Chem. 30, 21 936-21 945. Collins, S., Bouvier, M., Bolanowski, M., Caron, M. & Lefkowitz, R. (1989) CAMP stimulates transcription of the fl-adrenergic receptor gene in response to short-term agonist exposure, Proc. Natl Acad. Sci. USA 86,4853-4857. Diamond, M. I., Miner, J. N., Yoshinaga, S. K. & Yamamoto, K. R. (1990) Transcription factor interactions: selectors of positive and negative regulation from a single DNA element, Science 249, 1266- 1272. Dixon, R. A. F., Sigal, I., Candelore, M. R., Register, R. B., Scattergood, W., Rands, E. & Strader, C. D. (1987) Structural features required for ligand binding to the P-adrenergic receptor, EMBO J. 6 , 3269-3275. Emorine, L. J., Marullo, S., Delavier-klutcho, C., Kaveri, S. V., Durieu-Trautman, 0. & Strosberg, A. D. (1987) Structure of the gene for the human fl-adrenergic receptor: expression and promoter characterisation, Proc. Natl Acad. Sci. USA 84, 69956999. Emorine, L. J., Marullo, S., Briend-Sutren, M.-M., Patey, G., Tate, K., Delavier-Klutchko, C. & Strosberg, A. D. (1989) Molecular characterization of the human P3-adrenergic receptor, Science 24.5, 1118-1121. Fkve, B., Emorine, L. J., Lasnier, F., Blin, N., Baude, B., Nahmias, C., Strosberg, A. D. & Pairault, J. (1991) Atypical P-adrenergic receptor in 3T3-F442A adipocytes. Pharmacological and molecular relationship with the human 83-adrenergic receptor, J. Biol. Chem. 266, 20329-20336. Fkve, B., Baude, B., Krief, S., Strosberg, A. D., Pairault, J. & Emorine, L. J. (1992) Inhibition by dexamethasone of p3-adrenergic receptor responsiveness in 3T3-F442A adipocytes: evidence for a transcriptional mechanism, J. Bid. Chem. 267, 15909-25915. Granneman, J. G., Lahners, K. N. & Chaudhry, A. (1991) Molecular cloning and expression of the rat 83-adrenergic receptor, Mol. Phamacol. 40, 895- 899. Granneman, J. G. & Lahners, K. N. (1992) Differential adrenergic regulation of /?l- and p3- adrenoreceptor messenger ribonucleic acids in adipose tissues, Endocrinology 130, 109-114. Graves, R. A., Tontonoz, P., Ross, S. R. & Spiegelman,B. M. (1991) Identification of a potent adipocyte-specific enhancer: involvement of an NF-1-like factor, Genes & Dev. 5, 428-437. Gribskov, M. & Devereux, J. (1991) Sequence analysis primer, University of Wisconsin Biotechnology Center Biotechnical Resource Series (Burgess, R. R., ed) Stockton Press, Madison.

1124 and down-regulation in obesity., J. Biol. Chem. 266, 24053Hadcock, J., Ros, M. & Malbon, C. (1989)Agonist regulation of 8-adrenergic receptor mRNA. Analysis in 549 mouse lymphoma 24058. mutants, J . BioE. Chem. 264,13956-13961. Nahmias, C., Blin, N., Elalouf, J.-M., Mattei, M. G., Strosberg, A. Hausdorff, W. P., Caron, M. G. & Lefkowitz, R. J. (1990)Turning D. & Emorine, L. J. (1991)Molecular characterization of the off the signal: desensitization of P-adrenergic receptor function, mouse P3-adrenergic receptor: relationship with the atypical reFASEB J. 4 , 2881 -2889. ceptor of adipocytes, EMBO J. 10, 3721-3727. Hunt, C., Ro, J., Dobson, D., Min, H. & Spiegelman, B. (1986) Nantel, F., Bonin, H., Emorine, L. J., Zilberfarb, V., Strosberg, A. Adipocyte P2 gene: developmental expression and homology of D., Bouvier, M. & Marullo, S. (1993)The human p-adrenergic 5’-flanking sequences among fat cell-specific genes., Proc. Nut2 receptor is resistant to short-term agonist promoted desensitizAcad. Sci. USA 83,3786-3790. ation, Mol. Pharmacol., in the press. Johnson, J. A., Clark, R. B., Friedman, J., Dixon, R. A. F. & Strader, Port, J. D., Huang, L.-Y. & Malbon, C. C. (1992)/?-adrenergicagonC. D. (1990)Identification of a specific domain in the /?-adrenists that down-regulate receptor mRNA up-regulate a M,35000 ergic receptor required for phorbol ester-induced inhibition of protein(s) that selectively binds to /?-adrenergic receptor mRNAs, catecholamine-stimulatedadenylyl cyclase, Mol. Pharmacol. 38, J. Biol. Chem. 267,24103-24108. 289-293. Revelli, J.-P., Muzzin, P. & Giacobino, J.-P. (1992)Modulation in Kobilka, B., Frielle, T., Dohlman, H., Bolanowski, M., Dixon, R., vivo of P-adrenergic receptor subtypes in rat brown adipose tisKeller, P., Caron, M. & Letkowitz, R. (1987)Delineation of the sue by the thermogenic agonist Ro 16-8714, Biochem. J. 286, intronless nature of the genes for the human and hamster p2743-746. adrenergic receptor and their putative promoter region, J. Biol. Thomas, R. F., Holt, B. D., Schwinn, D. A. & Liggett, S. B. (1992) Chem. 262,7321 -7327. Long-term agonist exposure induces upregulation of P3-adrenKrief, S.,Lonnqvist, F., Raimbault, S., Baude, B., Van Spronsen, A., ergic receptor expression via multiple CAMPresponse elements, Amer, P., Strosberg, A. D., Ricquier, D. & Emorine, L. J. (1993) Proc. Natl Acad. Sci. USA 89, 4490-4494. Tissue distribution of p-adrenergic receptors mRNA in man, J. Yang-Yen, H.-F., Chambard, J.-C., Sun, Y. L., Smeal, T., Schmidt, Clin. Invest. 91, 344-349. T. J., Drouin, J. & Karin, M. (1990)Transcriptional interference Langin, D., Portillo, M. P., Saulnier-Blache, J.-S. & Lafontan, M. between c-Jun and the glucocorticoid receptor: mutual inhibition (1991)Coexistence of three P-adrenoceptor subtypes in white fat of DNA binding due to direct protein-protein interaction, Cell cells of various mammalian species, Eul: J. Physiol. 199, 291 62, 1205-1215. 301. Muzzin, P., Revelli, J.-P., Kuhne, F., Gocayne, J. D., McCombie, W. Zaagsma, J. & Nahorski, S. R. (1990)Is the adipocyte P-adrenoceptor a prototype for the recently cloned atypical ‘fl-adrenocepR., Venter, J. C., Giacobino, J.-P. & Fraser, C. M. (1991)An tor’, Trends Pharmacol. Sci. 11, 3-7. adipose tissue-specific a-adrenergic receptor. Molecular cloning