- Email: [email protected]
Identification of α1-adrenoceptor subtypes present in the human prostate
Life Sciences, VoL 54, No. 21, pp. 1595-1606, 1994 Copyright © 1994 Elsevier Science Ltd Printed in the USA. All rights r~ervtd 0024-3205/94 $6.00 + .00
Pergamon
IDENTIFICATION OF oq-ADRENOCEPTOR SUBTYPES PRESENT IN THE HUMAN PROSTATE C. Faure, C. Pimoule, G. Vallancien*, S.Z. Langer ~ and D. Graham Synth61abo Recherche (L.E.R.S.) 31, ave Paul Vaillant Couturier, BP 110, F-92225 Bagneux Cedex, France *Centre M6dico-Chirurgical de la Porte de Choisy 15, av. de la Porte de Choisy, 75013 Paris (Received in final form March 10, 1994) Summary (x~-Adrenoceptors (ARs) play an important role in mediating human prostatic smooth muscle conlraction. In the present study cDNA fragments covering different domains of 3 cq-AR subtypes (0qb, ~lc and ~d) were generated from human prostate by reverse transcription coupled to the polymerase chain reaction (RT-PCR). The reconstituted partial sequence (349 amino acids) of the human prostatic cqc-AR PCR products showed 94% identity at the amino acid level with that of the corresponding region of the cloned bovine brain oqc-AR. Using human a~-AR subtype selective cDNA probes in Northern blot analysis, the co-expression of mRNA transcripts corresponding to ~b-, cqc- and cqd-AR subtypes was detected in 4 different regions (apex, base, periurethra and lateral lobe) of the human prostate. Competitive inhibition experiments of [3H]-prazosin binding to membrane preparations of human prostate revealed that the non-selective ~rsubtype antagonist, alfuzosin, produced a monophasic inhibition curve, whereas oxymetazoline produced a 2-component inhibition curve with pK i values of 8.54 and 5.46. The high-affinity ~t-AR component of the oxymetazoline inhibition curve was predominant (57%-66%) and showed an affinity for oxymetazoline comparable to that of the oqc-AR subtype. As such our results illustrate the expression of different ~ - A R subtypes in human prostate and importantly that ~l~ represents the predominant (xt-AR subtype present in this tissue. Key Words: al-adrenoceptor subtypes, prostate, polymerase chain reaction, alfuzosin, oxymetazoline
The presence of both t~- and etz-adrenoceptors (ARs) in human prostate has been demonstrated by radioligand binding assays (1,2) although of these two AR classes functional assays suggest that the oq-AR subfamily plays the major role in controlling prostatic smooth muscle tone which possesses a noradrenergic innervation (3). The physiological importance of prostatic eq-ARs has been supported by recent clinical studies showing the efficacy of t~I-AR antagonists such as
1To whom all correspondence should be addressed
1596
Human Prostatic Ctl-adrenoeeptorSubtypes
Vol. 54, No. 21, 1994
alfuzosin (4) in the treatment of benign prostatic hyperplasia (BPH). In this context these tx~-AR antagonists reduce the dynamic component of prostatic smooth muscle tone and as such counteract bladder outlet obstruction that is a consequence of BPH pathophysiology. tx~-ARs have now been shown to be comprised of a subfamily of distinct subtypes (for review see 5). Notably, three pharmacologically- and structurally-distinct members of the tx~-AR subfamily have been identified by molecular cloning techniques. The ctm-AR subtype, originally described in tissues such as rat liver, has been cloned from hamster (6), rat (7) and man (8). Another subtype, isolated from bovine brain (9) and very recently from man (10), has been tentatively-assigned the nomenclature atc-AR, although the pharmacological properties of this subtype are very similar to those of the CttA-AR originally described in hippocampus and submaxillary gland of the rat (11,12). In addition, a third subtype, designated the cqd-AR subtype has been cloned from rat (13,14) and man (15). Although designated as the cqa-AR subtype (13,15), this subtype exhibits a pharmacological profile and tissue distribution pattern different to the CqA- and ct~B-subtypes (14), and, thus, it will be referred to here as the tx~d-AR. Given the structural and pharmacological diversity of the oq-AR subfamily we now report a study that we undertook to identify the txl-AR(s) that are expressed in human prostatic adenoma tissue taken from BPH patients. Our results suggesting that the eric-subtype is the major subtype expressed in this tissue might have important implications in the development of second generation cq-AR antagonists for the treatment of BPH.
Materials and Methods 1. Materials Phentolamine was obtained from Research Biochemicals Inc., oxymetazoline from Sigma and alfuzosin from Synth61abo. [3H]-(7-methoxy)prazosin (77 Ci/mmol) was obtained from Amersham. Permanent HeLa cell-lines stably transfected with either the hamster smooth muscle tx~b-AR (6), the bovine brain atc-AR (9) or the rat cerebral cortex tx~d-AR (13) were purchased from Tulco, Durham, N. Carolina, USA. Prostatic adenoma tissues obtained from patients undergoing open prostatectomy for symptomatic BPH were provided by Prof. G. Vallancien (Paris). Samples were frozen in liquid nitrogen immediately after surgery, and stored at -80°C until use. 2. RNA Preparation Total cellular RNA was isolated from tissue or cells by the method of Chomczymski and Sacchi (16) using acid guanidium thiocyanate-phenol-chloroform extraction (RNAzol, Bioprobe Systems). Poly(A)÷ RNAs were purified using two cycles of oligo(dT)-cellulose chromatography. Purity and quantitation of the RNAs were determined by measuring the absorbance at 260/280 nm. 3. Oligonucleotides The oligonucleotides used in the PCR were synthetized by Genset (Paris, France). For the selection of appropriate oligonucleotides that would amplify and clone oh-ARs, irrespective of subtypes, we aligned the amino acid sequences of all members of the ¢q-AR family cloned to date. Highly "degenerate" oligonucleotide primers a, b, c, d and e were designed from conserved
Vol. 54, No. 21, 1994
Human Prostatic al-adrenoceptor Subtypes
1597
amino acid residues in the transmembrane domains (TM). Sens primers: (a) 5' CTCGAATI'CGCITG(T/C)(A/C)A(T/C)(A/C)GICA(T/C)(T/C) 3' (I for inosine), corresponded to the coding strand for the AC(H/N)RH peptide at the beginning of the first intracellular loop (c) 5' C'I'CGAATTCGA(T/C)GTI(T/C)TNTG(T/C)TG(T/C)AC 3' (N for G, A, T or C) corresponded to the peptide DVLCC in TM3, and (e) 5' CTCGAATTCGTNATGTA(T/C) TG(T/C) (C/A)GNG 3', corresponded to the coding strand for VMYCR in TM5. Antisens primers: (b) 5' TT/'CTAGAACNC(G/T)(G/A)CA(G/A)TACATNAC 3', corresponded to the reverse sequence of the peptide VMYCR in TM5, and d) 5' TIq'CTAGAA(G/A)(G/A)TANCCNA (G/A)CCA(G/A)AA 3', corresponded to the reverse sequence of the peptide FWLGY in TM7. "Degenerate" oligonucleotide primers sens (h) 5' CTCGAATTCTI'(T/C)TGG(T/C)TNGGNTA (T/C)(T/C)T 3', derived from peptide sequence FWLGY in TM7, and antisens (i) 5' "Iqq'CTAGA(G/A)AA(G/A/T)AT(C/T)TI'CCA(C/T)TC(G/A)CA 3', corresponding to the peptide CEWKIF of the bovine c~c sequence (9), were also used. Human Cqd-AR specific oligonucleotides f and g were designed from the sequence reported by Bruno et al. (1991) (15); sens primer (f) 5' CTCGAA'ITCAGCGCTTCTGCGGTATCAC 3'; antisens primer (g) 5' TTTCTAGATCCGATGGCTTCAGCTG 3'. Restriction site sequences were added at the 5' end of each primer (underlined EcoRI and XbaI sequences, in sens and antisens primers, respectively) in order to allow subsequent cloning of PCR products.
4. RT-PCR Reverse transcription and PCR amplification were performed using the GeneAmp RNA PCR kit (Perkin-Elmer Cetus) according to the manufacturer's instructions. Poly(A)*-selected RNAs (250 ng) were reverse transcribed using random-hexamer priming and the resulting cDNAs were submitted to 30 cycles of PCR, using the different sets of primers (100 pmole each/reaction), under the following conditions: 1.5 min at 95°C, 2 min at 55°C and 2 min at 72°C, followed by a final extension period of 8 min at 72°C. The resulting products were fractionated by electrophoresis on a 2% agarose gel and visualized by staining with ethidium bromide (0.1 ~g/ml).
5. Analysis of PCR products The PCR reaction mixtures were phenol-chloroform extracted, ethanol precipitated, digested by EcoRI and XbaI, then separated on a 2% agarose gel containing ethidium bromide (0.1 pg/ml). PCR products of the expected size were cut from the gel, purified using QIAEX (Qiagen, CA), ligated into EcoRI/XbaI digested pBluescript II KS+ (Stratagene, CA) and transformed into Epicurian coli SURE competent cells (Stratagene, CA). Individual colonies were analysed on plasmid minipreparations by restriction mapping and sequencing using the dideoxynucleotide chain-termination method of Sanger et al. (17) (TAQuence, United States Biochemical Corp., Cleveland, OH). Sequence analysis was performed using the software program, Geneworks (Intelligenetics, CA).
6. Northern Blot Analysis Poly(A)÷-selected RNAs were fractionated by electrophoresis on 1% denaturing agarose gels (containing 0.66 M formaldehyde) and transferred on to nylon membranes (Hybond N, Amersham) by capillary blotting in 20xSSC (lxSSC = 0.15 M NaCI, 0.015 M sodium citrate, pH 7.0).
1598
Human Prostatic al-adrenoceptor Subtypes
Vol. 54, No. 21, 1994
cDNA fragments generated by RT-PCR were purified from recombinant pBluescript II KS ÷ plasmid preparations and labelled by random priming (Megaprime, Amersham) to a specific activity of ~ 1.109 cpm/pg DNA. The filters were prehybridized at 65°C for 1 h in 0.5 M sodium phosphate buffer (pH 6.5), 1.0 mM EDTA, 7% SDS and 100 pg/ml salmon sperm DNA. The prehybridization solution was replaced with fresh solution containing 3zp-labelled probe (2-4.10 6 cpm/ml), and incubation was performed for 15-20 h at 65°C. After hybridization, filters were washed in 2xSSC/0.1% SDS 3 times at room temperature and in 0.2xSSC/0.1% SDS twice 30 min at 55°C, then exposed to Hyperfilm MP (Amersham) at -80°C with two intensifying screens. Blots were stripped by incubating them twice in boiling 0.1% SDS and waiting until the temperature decreased to room temperature. Blots were reexposed to film and restripped if necessary before reprobing.
7. Radioligand Binding Assays Membranes from human prostatic adenoma tissue were prepared by homogenization and centrifugation twice at 37,000 x g for 10 min in 50 mM Tris-HCl buffer, pH 7.5. Membranes (120-180 ~g prot) were incubated with [3H]-prazosin for 30 min at 25°C in a total volume of 1 ml. Phentolamine (100 pM) was used to define non-specific binding. Incubations were terminated by rapid filtration using a Brandell cell harvester through Whatman GF/B glass fibre filters. The filters were washed with 3 x 5 ml of buffer, dried and the radioactivity measured by liquid scintillation spectrometry. Saturation and competitive inhibition data were analysed by nonlinear regression analysis programmes. Results
1. Selective Amplilqcation and Subcloning o( 3 Distinct o~t-AR Subtypes O'om Human Prostate Alignment of the primary structures of the 3 members of the cq-adrenoceptor subfamily cloned to date (cqb-, ¢xtc- and ~d-ARs) shows a high degree of sequence homology between subtypes, with notably 65 to 73% amino acid identity in the seven putative transmembrane spanning domains (TMs). Three different sets of highly "degenerate" oligonucleotide primers were designed from stretches of amino acid residues highly conserved between the 3 ¢x~-AR subtypes (Fig. 1). Sens primers a, c and e were designed respectively from consensus amino acid sequences AC(H/N)RH (in the first putative intracellular loop), DVLCC (in TM3) and VMYCR (in TM5). Antisens primers b and d were based on conserved peptide sequences VMYCR (in TM5) and FWLGY (in TM7), respectively. Each primer contained restriction site sequence at their 5' end in order to allow subsequent cloning of the PCR amplified products. Poly(A)÷ RNA (250 ng) prepared from the lateral lobe region of human prostatic adenoma tissue was reverse-transcribed and the resulting cDNA amplified by PCR using either a-b, c-d or e-d pairs of primers. After 30 cycles of amplification, PCR product bands of the expected sizes (~ 480, 650 and 380 bp, for a-b, c-d and e-d amplifications, respectively) were visualized upon agarose gel electrophoresis of the PCR reaction mixtures. No DNA band was detected when the reverse transcriptase was omitted before the PCR amplification reaction, showing that the PCR products resulted from cDNA and not from contaminating genomic DNA (Fig. 2). PCR amplified products of the predicted sizes were gel purified and subcloned into the plasmid pBluescript II KS÷. Minipreparations of plasmids with subcloned inserts were analysed by restriction mapping and sequencing.
Vol. 54, No. 21, 1994
Human Prostatic al-adrenoceptor Subtypes
1599
h~ flD,-
NH
2
TM 1 TM 2 TM3 ~
TM 4 TM5
TM6 TM7 ~
COOH
aiD--qlb clP,~
-,~d el~,-
~d FIG. 1 Schematic diagram of oligonucleotide primers chosen for RT-PCR amplification of cq-AR subtypes. The sequences of the different sets of primers (notated by letters) are described in Materials and Methods.
Primers • RT"
a/b ~
c/d e/d ~ - ~ - I ~+---~-1
FIG. 2 RT-PCR analysis of ~x~-AR subtypes expressed in human prostate. Poly(A+)-selected RNA (250 ng) was reverse-transcribed in the presence (+) or absence (-) of Reverse Transcriptase (RT) and the resulting cDNA was amplified by PCR with the primer sets a-b, c-d and e-d shown in Fig. 1 under conditions described in Materials and Methods. PCR products were electrophoresed on 2% agarose gel and visualized by ethidium bromide staining. The size markers are shown on the left.
Three distinct types of PCR products, related to the eq-AR subfamily, were generated from human prostatic tissue. The amino acid sequences deduced from 2 of them were 100% identical to those reported for the corresponding regions of the human t~ab- (8) and human (tld- (15) ARs (data not shown). The deduced amino acid sequences of the third type of PCR products were closely related to those of the corresponding regions of the cqc-AR subtype cloned previously from bovine brain (9). Two "degenerate" primers were designed in order to get more sequence information for the
1600
Human Prostatic al-adrenoceptor Subtypes
Vol. 54, No. 21, 1994
human Ct~c-AR subtype. Using sens primer h, designed from consensus peptide FWLGY in TM7, and antisens primer i, designed from the sequence CEWKIF in the C terminal region of bovine cq~-AR sequence (9), RT-PCR amplification of human prostatic RNA allowed us to subclone an additional cDNA fragment covering part of the C terminal region of the human (x~-AR. Alignment of the reconstituted human tx~-AR sequence, which spans the region from the first intracellular loop domain to the middle of the C terminal region, with that of the corresponding region of the cloned bovine cqc-AR shows 94% identity at the amino acid level (Fig. 3). TM2 Human ctlc-AR
Bovine czlc-AR
TM3
ACHRHLRSVTHYYIVNLAVADLLLTS~AIF~VlLGYWAFGRVFCI~I [WAAVDVLCCTASIMGLCRSIDRYI I 75
ACnSnLmVTnVV~VADLLLTSrVLPFSAIF~ I IL~VWAmRWC~I~IWAAVOVLCCrASIM~LCnSmRW I 75 TM4
Humanalc-AR Bovine ctlc-AR
TM5
GVSYPLRYPTIVTQ]RtRGLMAI1CVWALSLVISIGPLFGWRQPAPEDETICQINEEI~YVLFS ALGSF~I~ 150 GVSYPLRYPTIVTQIK]RGLMALLCVWALSLVISIGPLFGWRQPAPEDETIOqlNEEPGYVLFSALGSF~Xl_..~I1[~.~ 150
Bovine ctlc-AR
~
Human ctIc-AR
KTLGIVV~CWLPFFLVMPIGSFFPDI~K~-'~------~ FWLGYLNSCINPI l YPCSSQEFKKAFQNVLRIQCI4
Human =lc-AR
CRVYVVAKRESRGLKSGLKTDKSDSEQVTLRIHRKN~4PA1 " ~ - - ~ S ~ 1]KTHFSVRLLKFSREKKAAI 220 CRVYVVAKRESRGLKSGLKTDKSDSEQVTLRIHRKNAIQVIGGSGI VTISAKINIKTHFSVI~LKFSREKKAAI 220 TM6
TM7
Bovine ctlc-AR
KTLGIWGCFVLCWLPFFLVMPIGSFFPDF~ RIPSEWFKI IAIFWLGYLNSCINP I I YPCSSQEFKKAFqNVLRIQCLI
Humanczlc-AR Bovine c~lc-AR
~
295 295
p ~-'] QAV~ - - - - ~ MI--~---~ ~ R ISKTDGVCEWI~F['~ 349 A[PSI HVL~GQHKI~L~rRIPVGSIAIETFY]KISKTDGVCEWIqIIF I 349
FIG. 3 Alignment of the deduced amino acid partial sequence of the human ~k-AR with that of the corresponding region of the bovine ~k-AR. Putative transmembrane-spanning
domains (TMs) are indicated. 2. Northern Blot Analysis Human O~I-ARsubtype selective cDNA probes, covering the region which includes the third intracellular loop of ~lb-, t~lc- and ~d-ARs were obtained by RT-PCR from prostatic mRNA. The ~b-AR and cttc-AR probes spanned the region from the fifth to seventh transmembrane domains and resulted from RT-PCR amplification using "degenerate" e and d primers. The t~ld-ARcDNA probe used for the present report was generated using human t~d-AR specific oligonucleotides f and g (Fig. 1). The subtype-selectivity of these 3 probes towards their corresponding t~-AR subtype was confirmed by Northern blot hybridization of poly(A)÷-selected RNAs extracted from permanently-established t~t-AR cell-lines, expressing either the hamster cqb-AR, the bovine ~]c-AR or the rat t~ld-AR (Fig. 4). These 3 probes were then used to examine the pattern of expression of ~I-AR subtypes in 4 different regions (notably, periurethra, base, apex and lateral lobe) of human prostatic adenoma tissue (Fig. 5). Similar ~ - A R subtype expression patterns were noted in each region (Fig. 5). Using the eqb-AR cDNA probe, a single mRNA species of 2.9 Kb was detected. The ~c-AR cDNA probe hybridized to 2 major mRNA species of 5.7 and 3.0 Kb, as well as a minor mRNA of 3.9 Kb. In addition, 2 major mRNA species of 3.0 and 1.7 Kb were visualized in the human prostate using the (X~d-AR cDNA probe.
Vol. 54, No. 21, 1994
Human Prostatic al-adrenoceptor Subtypes
(x~-AR cell line :
1601
O~lb (Xlc (Xld
(Xlb (Xlc O~ld
(7"1b ~1c O~ld
(Zlb-AR
alc-AR
~ld-AR
4.4 _
2.4 _
1.4
i
(Kb) Probe :
FIG. 4 Subtype-selectivity of the different RT-PCR cloned human ~I-AR eDNA probes. Northern blot analysis of poly(A+)-selected RNAs (5 pg per lane) from HeLa cells stably transfected with hamster ~lb-, bovine eric- and rat ctm-AR cDNAs was performed using 32p-labelled human ct~-AR cDNA probes cloned by RT-PCR. After hybridization, final washing conditions of the filters were 0.2 x SSC, 0.1% SDS at 55°C. Filters were exposed to Amersham MP films at -80°C for 24 h.
1
2
3
4
1
2
3
4
1
2
3
4
7.5i
4,-
!i
iil;ill i!ii~
2.4 --
+iiii+i + .........
?++~iii~~ ! 1.4--
(~)
Probe :
O~lb-AR
cqc-AR
O~ld-AR
FIG. 5 Northern blot analysis of a~-AR subtype mRNAs in human prostatic tissue. The RNA blot was prepared with poly(A÷) RNA extracted from different prostatic regions (1, periurethra; 2, base; 3, apex; 4, lateral lobe) and probed with 32P-labelled human al-AR subtype-selective cDNA probes (rehybridization of the same blot with the 3 cq-AR probes). After hybridization, final washing conditions of the filters were 0.2xSSC, 0.1% SDS at 55°C. Filters were exposed to Amersham MP films with two intensifying screens at -80°C for 24 h. The quality of RNA blotted was controlled by reprobing with 32P-labelled human Glyceraldehyde 3-phosphate dehydrogenase (G3PDH) probe.
1602
Human Prostatic al-adrenoceptor Subtypes
Vol. 54, No. 21, 1994
3. [3H]-Prazosin Binding to Human Prostate Equilibrium saturation analysis of [3H]-prazosin binding to membrane preparations of human prostatic adenoma tissue revealed the binding of this radioligand to an apparently homogenous class of binding sites with a K d of 0.16 nM and a B,,~x of 130 fmol/mg prot. Competitive inhibition experiments of [3H]-prazosin binding to membrane preparations of human prostate were performed using varying concentrations of unlabelled oxymetazoline and alfuzosin (Fig. 6). [3H]-Prazosin binding was inhibited by alfuzosin in a monophasic fashion with a pK~ of 8.29. The oxymetazoline inhibition curve showed a significantly better fit for a 2 site binding model compared to a single site model (p < 0.002) with pK~ values of 8.54 and 5.46 at the high and low affinity binding sites, respectively. The high-affinity component corresponded to 57%-66% of the [3H]-prazosin binding sites present in these tissue homogenates (n = 3 experiments).
loo
H
Alfuzosin
E N
50
E
oo o~
0
I
t
I
i - log [drug]
FIG. 6 Inhibition of [3H]-prazosin binding to human prostatic adenoma tissue. Membranes were incubated with [3H]-prazosin in the absence and presence of different concentrations of oxymetazoline and alfuzosin as described in Materials and Methods. Discussion The concept of cq-AR subtypes, originally defined on the basis of pharmacological studies (18,19), has been confirmed and extended by the cloning of 3 pharmacologically- and structurallydistinct members of this subfamily (5-15). For the present report we have undertaken the identification of oq-AR subtype(s) expressed in human prostatic adenoma tissue by looking for the presence of mRNA transcripts corresponding to the ~lb-' [~lc- and oqa-AR subtypes. In addition, a complementary study on the pharmacological profile of [3H]-prazosin binding to human prostatic tissue has been performed. Based upon the reported sequences of cDNAs encoding ~ r A R subtypes that have been cloned to date various sets of degenerate oligonucleotide primers were designed for use in a RT-PCR strategy to permit identification of the expression of oqb-, ~c- and cqa-AR subtypes in human tissues. Application of this RT-PCR strategy to human prostatic tissue and subsequent characterization of subcloned PCR products by restriction mapping and sequencing revealed the presence in this tissue of 3 distinct oq-AR subtypes. The deduced amino acid sequences of two
Vol. 54, No. 21, 1994
Human Prostatic al-adrenoceptor Subtypes
1603
of these PCR products were identical to those reported in the literature for the corresponding regions of the human ~X~b-(8) and human ~d- (15) ARs. In addition, the deduced amino acid sequences of the other PCR products revealed that they were closely related to the (xac-AR subtype previously cloned from bovine brain (9). In fact, the reconstituted sequence of these human PCR products showed 94% identity at the amino acid level with the cloned bovine tXl~-AR. Also, the extensive amino acid sequence information that we present on this human cqc-AR subtype matches exactly that reported very recently for the human cq¢-AR subtype (10). Thus, this RT-PCR technique permitted the identification of the expression of 3 oq-AR subtypes (Ohb, 0~1¢ and 0qd) in human prostate. It should be noted, however, that the highly-degenerate oligonucleotide probes chosen for this RT-PCR study do not permit interpretations to be made as to the relative expression levels of these different tx~-AR subtypes in this tissue. Using human cq-AR subtype selective cDNA probes the pattern of expression of cq-AR subtypes in different regions of human prostatic adenoma tissue was examined by Northern blot hybridization studies. This technique revealed the co-expression of all 3 cloned to date members of the cq-AR subfamily (~b, cite and Cqd) in 4 different regions of the prostate namely, apex, base, periurethra and lateral lobe. This observation showing the expression in the human prostate of mRNA transcripts corresponding to the Cttb-, ~t~- and (t~d-AR subtypes thus concords with the results of a recent study (20). Furthermore, the quantitative solution hybridization protocol adopted in that study permitted those authors to conclude that the predominant (60-75%) mRNA txI-AR subtype present in human prostate was in fact cq~. For that reason, we have undertaken a complementary study using in vitro [3H]-prazosin radioligand binding assays to investigate at the protein level the degree of ctI-AR subtype heterogeneity in the human prostate. Equilibrium saturation binding of [3H]-prazosin to human prostatic adenoma tissue revealed that this radioligand bound to a single class of high-affinity binding sites in this tissue (Ka = 0.16 riM). Moreover, in competition inhibition experiments, the cq-AR antagonist alfuzosin potently inhibited [3H]-prazosin binding to this tissue with a K~value in the low nanomolar range. In agreement with previous studies (1,3,21), therefore, our [3H]-prazosin binding data illustrate the presence of ¢q-ARs in human prostate. The (x~-AR antagonist, alfuzosin (22), however, displays similar potencies at inhibiting [3H]-prazosin binding both to the 3 cloned to date members (Cqb, Cqc and tx~a) of the cq-AR subfamily as well as to the classical pharmacologically-defined tX~A-AR of rat salivary gland (C. Pimoule and D. Graham, unpublished data). As such, we chose in addition to examine the profile of inhibition of [3H]-prazosin binding by oxymetazoline, a compound that shows a very marked selectivity for the ~A- and cqc-AR subtypes compared to (~lb- and Cqd-ARs (12-14). Interestingly, oxymetazoline inhibition of [3H]-prazosin binding to human prostatic adenomas showed the presence of two cq-AR subtypes in this tissue. The high-affinity ct~-AR component of [3H]-prazosin binding displayed an affinity for oxymetazoline comparable to that exhibited by the rat salivary gland (t~n- (12) and cloned bovine brain t~c-AR (9) subtypes for this compound. Moreover, this high-affinity cq-AR subtype represented the predominant (57%-66%) oq-AR subtype present in this tissue. As such, our data on [3H]-prazosin binding are concordant with the results of a recent study on 125I-HEAT binding to human prostate indicating on the basis of biphasic inhibition curves for WB4101 and 5-methylurapidil that the dominant eq-AR subtype in this tissue showed the pharmacological profile of the O~IA-ARsubtype (23). Since the pharmacological profiles of the classical pharmacologically-defined t~A-AR and the cloned t~c-AR are very similar, the finding that the tx~c-AR is the predominant oq-AR mRNA subtype present in the human prostate (20) probably infers that the majority cq-AR subtype at the protein level is in fact the (x~c-AR. In conclusion, both [3H]-prazosin binding and Northem blot analysis indicate the presence of several Ctl-AR subtypes in human prostatic adenoma tissue and importantly that ctlc represents the
1604
Human Prostatic al-adrenoeeptor Subtypes
Vol. 54, No. 21, 1994
predominant cq-AR subtype present in this tissue. It will therefore be particularly pertinent to verify in functional studies the precise importance of the cqc-AR subtype in mediating human prostatic smooth muscle contraction. In this context, the possibility of developing a selective cqc-AR antagonist potentially devoid of vascular side-effects mediated by other cq-AR subtypes could lead to second generation ~ - A R blockers of improved safety and efficacy for the treatment of BPH.
Acknowledgements: The authors thank Caroline Gouhier and Christiane Sellier for their technical assistance and Fran~oise P6choux for secretarial assistance in the preparation of this manuscript. References 1. 2. 3.
4. 5. 6. 7. 8. 9.
10. 11. 12. 13.
14. 15. 16. 17. 18. 19. 20. 21.
22.
H. LEPOR and E. SHAPIRO, J. Urol. 1321226-1229 (1984). E. SHAPIRO and H. LEPOR, J. Urol. 1351038-1042 (1986). C.R. CHAPPLE, M.L. AUBRY, S. JAMES, P.M. GREENGRASS, G. BURNSTOCK, R.T. TURNER-WARWICK, E.J.G. MILROY and M.J. DAVEY, Brit. J. Urol. 6___33487-496 (1989). A. JARDIN, H. BENSADOUN, M.C. DELAUCHE-CAVALLOR and P. ATFALl, Lancet 3371457-1461 (1991). D.B. BYLUND, D.C. EIKENBERG, J.P. HIEBLE, S.Z. LANGER, R.J. LEFKOWlTZ, K.P. MINNEMAN, P.B. MOLINOFF and R.R. RUFFOLO, Jnr., Pharmacol. Revs. (in press). S. COTECCHIA, D.A. SCHWINN, R.R. RANDALL, R.J. LEFKOWITZ, M.G. CARON and B.K. KOBILKA, Proc. Natl. Acad. Sci. 8...557159-7163 (1988). M.M. VOIGT, J. KISPERT and H. CHIN, Nucl. Acids Res. 1__.881053(1990). C.S. RAMARAO, J.M.K. DENKER, D.M. PEREZ, R.J. GAIVIN, R.P. RIEK and R.M. GRAHAM, J. Biol. Chem. 26721936-21945 (1992). D.A. SCHWlNN, J.W. LOMASNEY, W. LORENZ, P.J. SZKLUT, R.T. FREMEAU, JR., T.L. YANG-FENG, M.G. CARON, R.J. LEFKOWITZ and S. COTECCHIA, J. Biol. Chem. 265, 14, 8183-8189 (1990). A. HIRASAWA, K. HORIE, T. TANAKA, K. TAKAGAKI, M. MURAl, J. YANO and G. TSUJIMATO, Biochem. Biophys. Res. Commun. 195902-909 (1993). K.P. MINNEMAN, C. HAN and P.W. ABEL, Mol. Pharmacol. 3__3509-514 (1988). A.D. MICHEL, D.N. LOURY and R.L. WHITING, Br. J. Pharmacol. 98883-889 (1989). J.W. LOMASNEY, S. COTECCHIA, W. LORENZ, W-Y LEUNG, D.A. SCHWINN, T.L. YANG-FENG, M. BROWNSTEIN, R.J. LEFKOWITZ and M. CARON, J. Biol. Chem. 266 6365-6369 (1991). D.M. PEREZ, M.T. PIASCIK and R.M. GRAHAM, Mol. Pharmacol. 4..._0876-883 (1991). J.F. BRUNO, J. WHIqTAKER, J. SONG and M. BERELOWITZ, Biochem. Biophys. Res. Comm. 1791485-1490 (1991). P. CHOMCZYMSKI and N. SACCHI, Anal. Biochem. 162156-159 (1987). F. SANGER, S. NICKLEN and A.R. CARLSON, Proc. Natl. Acad. Sci. 7__.445463-5467 (1977). K.P. MINNEMAN, C. HAN and P.W. ABEL, Mol. Pharmacol. 3...33509-514 (1988). K.P. MINNEMAN, Pharmacol. Rev. 4._0_0,2, 87-119 (1988). D.J. PRICE, D.A. SCHWINN, J.W. LOMASNEY, L.F. ALLEN, M.C. CARON and R.J. LEFKOWITZ R.J., J.Urol. 150546-551 (1993). H. SCHOEMAKER, H. BLANCHARD, C. PIMOULE, F. LEFI~VRE-BORG, P. MANOURY, A. JARDIN and S.Z. LANGER, in: Prostate and Alpha Blockers, Excerpta Medica pp. 328-343 (1989). F. LEFI~VRE-BORG, S.E. O'CONNOR, H. SCHOEMAKER, P.E. HICKS, J. LECHAIRE, E. GAUTIER, F. PIERRE, C. PIMOULE, P. MANOURY and S.Z. LANGER, Br. J.
Vol. 54, No. 21, 1994
23.
Human Prostatic al-adrenoceptor Subtypes
Pharmacol. 109 1282-1289 (1993). H. LEPOR, R. TANG, S. MERETYK and E. SHAPIRO, J. Urol. 149 640-642 (1993).
1605
Pergamon
IDENTIFICATION OF oq-ADRENOCEPTOR SUBTYPES PRESENT IN THE HUMAN PROSTATE C. Faure, C. Pimoule, G. Vallancien*, S.Z. Langer ~ and D. Graham Synth61abo Recherche (L.E.R.S.) 31, ave Paul Vaillant Couturier, BP 110, F-92225 Bagneux Cedex, France *Centre M6dico-Chirurgical de la Porte de Choisy 15, av. de la Porte de Choisy, 75013 Paris (Received in final form March 10, 1994) Summary (x~-Adrenoceptors (ARs) play an important role in mediating human prostatic smooth muscle conlraction. In the present study cDNA fragments covering different domains of 3 cq-AR subtypes (0qb, ~lc and ~d) were generated from human prostate by reverse transcription coupled to the polymerase chain reaction (RT-PCR). The reconstituted partial sequence (349 amino acids) of the human prostatic cqc-AR PCR products showed 94% identity at the amino acid level with that of the corresponding region of the cloned bovine brain oqc-AR. Using human a~-AR subtype selective cDNA probes in Northern blot analysis, the co-expression of mRNA transcripts corresponding to ~b-, cqc- and cqd-AR subtypes was detected in 4 different regions (apex, base, periurethra and lateral lobe) of the human prostate. Competitive inhibition experiments of [3H]-prazosin binding to membrane preparations of human prostate revealed that the non-selective ~rsubtype antagonist, alfuzosin, produced a monophasic inhibition curve, whereas oxymetazoline produced a 2-component inhibition curve with pK i values of 8.54 and 5.46. The high-affinity ~t-AR component of the oxymetazoline inhibition curve was predominant (57%-66%) and showed an affinity for oxymetazoline comparable to that of the oqc-AR subtype. As such our results illustrate the expression of different ~ - A R subtypes in human prostate and importantly that ~l~ represents the predominant (xt-AR subtype present in this tissue. Key Words: al-adrenoceptor subtypes, prostate, polymerase chain reaction, alfuzosin, oxymetazoline
The presence of both t~- and etz-adrenoceptors (ARs) in human prostate has been demonstrated by radioligand binding assays (1,2) although of these two AR classes functional assays suggest that the oq-AR subfamily plays the major role in controlling prostatic smooth muscle tone which possesses a noradrenergic innervation (3). The physiological importance of prostatic eq-ARs has been supported by recent clinical studies showing the efficacy of t~I-AR antagonists such as
1To whom all correspondence should be addressed
1596
Human Prostatic Ctl-adrenoeeptorSubtypes
Vol. 54, No. 21, 1994
alfuzosin (4) in the treatment of benign prostatic hyperplasia (BPH). In this context these tx~-AR antagonists reduce the dynamic component of prostatic smooth muscle tone and as such counteract bladder outlet obstruction that is a consequence of BPH pathophysiology. tx~-ARs have now been shown to be comprised of a subfamily of distinct subtypes (for review see 5). Notably, three pharmacologically- and structurally-distinct members of the tx~-AR subfamily have been identified by molecular cloning techniques. The ctm-AR subtype, originally described in tissues such as rat liver, has been cloned from hamster (6), rat (7) and man (8). Another subtype, isolated from bovine brain (9) and very recently from man (10), has been tentatively-assigned the nomenclature atc-AR, although the pharmacological properties of this subtype are very similar to those of the CttA-AR originally described in hippocampus and submaxillary gland of the rat (11,12). In addition, a third subtype, designated the cqd-AR subtype has been cloned from rat (13,14) and man (15). Although designated as the cqa-AR subtype (13,15), this subtype exhibits a pharmacological profile and tissue distribution pattern different to the CqA- and ct~B-subtypes (14), and, thus, it will be referred to here as the tx~d-AR. Given the structural and pharmacological diversity of the oq-AR subfamily we now report a study that we undertook to identify the txl-AR(s) that are expressed in human prostatic adenoma tissue taken from BPH patients. Our results suggesting that the eric-subtype is the major subtype expressed in this tissue might have important implications in the development of second generation cq-AR antagonists for the treatment of BPH.
Materials and Methods 1. Materials Phentolamine was obtained from Research Biochemicals Inc., oxymetazoline from Sigma and alfuzosin from Synth61abo. [3H]-(7-methoxy)prazosin (77 Ci/mmol) was obtained from Amersham. Permanent HeLa cell-lines stably transfected with either the hamster smooth muscle tx~b-AR (6), the bovine brain atc-AR (9) or the rat cerebral cortex tx~d-AR (13) were purchased from Tulco, Durham, N. Carolina, USA. Prostatic adenoma tissues obtained from patients undergoing open prostatectomy for symptomatic BPH were provided by Prof. G. Vallancien (Paris). Samples were frozen in liquid nitrogen immediately after surgery, and stored at -80°C until use. 2. RNA Preparation Total cellular RNA was isolated from tissue or cells by the method of Chomczymski and Sacchi (16) using acid guanidium thiocyanate-phenol-chloroform extraction (RNAzol, Bioprobe Systems). Poly(A)÷ RNAs were purified using two cycles of oligo(dT)-cellulose chromatography. Purity and quantitation of the RNAs were determined by measuring the absorbance at 260/280 nm. 3. Oligonucleotides The oligonucleotides used in the PCR were synthetized by Genset (Paris, France). For the selection of appropriate oligonucleotides that would amplify and clone oh-ARs, irrespective of subtypes, we aligned the amino acid sequences of all members of the ¢q-AR family cloned to date. Highly "degenerate" oligonucleotide primers a, b, c, d and e were designed from conserved
Vol. 54, No. 21, 1994
Human Prostatic al-adrenoceptor Subtypes
1597
amino acid residues in the transmembrane domains (TM). Sens primers: (a) 5' CTCGAATI'CGCITG(T/C)(A/C)A(T/C)(A/C)GICA(T/C)(T/C) 3' (I for inosine), corresponded to the coding strand for the AC(H/N)RH peptide at the beginning of the first intracellular loop (c) 5' C'I'CGAATTCGA(T/C)GTI(T/C)TNTG(T/C)TG(T/C)AC 3' (N for G, A, T or C) corresponded to the peptide DVLCC in TM3, and (e) 5' CTCGAATTCGTNATGTA(T/C) TG(T/C) (C/A)GNG 3', corresponded to the coding strand for VMYCR in TM5. Antisens primers: (b) 5' TT/'CTAGAACNC(G/T)(G/A)CA(G/A)TACATNAC 3', corresponded to the reverse sequence of the peptide VMYCR in TM5, and d) 5' TIq'CTAGAA(G/A)(G/A)TANCCNA (G/A)CCA(G/A)AA 3', corresponded to the reverse sequence of the peptide FWLGY in TM7. "Degenerate" oligonucleotide primers sens (h) 5' CTCGAATTCTI'(T/C)TGG(T/C)TNGGNTA (T/C)(T/C)T 3', derived from peptide sequence FWLGY in TM7, and antisens (i) 5' "Iqq'CTAGA(G/A)AA(G/A/T)AT(C/T)TI'CCA(C/T)TC(G/A)CA 3', corresponding to the peptide CEWKIF of the bovine c~c sequence (9), were also used. Human Cqd-AR specific oligonucleotides f and g were designed from the sequence reported by Bruno et al. (1991) (15); sens primer (f) 5' CTCGAA'ITCAGCGCTTCTGCGGTATCAC 3'; antisens primer (g) 5' TTTCTAGATCCGATGGCTTCAGCTG 3'. Restriction site sequences were added at the 5' end of each primer (underlined EcoRI and XbaI sequences, in sens and antisens primers, respectively) in order to allow subsequent cloning of PCR products.
4. RT-PCR Reverse transcription and PCR amplification were performed using the GeneAmp RNA PCR kit (Perkin-Elmer Cetus) according to the manufacturer's instructions. Poly(A)*-selected RNAs (250 ng) were reverse transcribed using random-hexamer priming and the resulting cDNAs were submitted to 30 cycles of PCR, using the different sets of primers (100 pmole each/reaction), under the following conditions: 1.5 min at 95°C, 2 min at 55°C and 2 min at 72°C, followed by a final extension period of 8 min at 72°C. The resulting products were fractionated by electrophoresis on a 2% agarose gel and visualized by staining with ethidium bromide (0.1 ~g/ml).
5. Analysis of PCR products The PCR reaction mixtures were phenol-chloroform extracted, ethanol precipitated, digested by EcoRI and XbaI, then separated on a 2% agarose gel containing ethidium bromide (0.1 pg/ml). PCR products of the expected size were cut from the gel, purified using QIAEX (Qiagen, CA), ligated into EcoRI/XbaI digested pBluescript II KS+ (Stratagene, CA) and transformed into Epicurian coli SURE competent cells (Stratagene, CA). Individual colonies were analysed on plasmid minipreparations by restriction mapping and sequencing using the dideoxynucleotide chain-termination method of Sanger et al. (17) (TAQuence, United States Biochemical Corp., Cleveland, OH). Sequence analysis was performed using the software program, Geneworks (Intelligenetics, CA).
6. Northern Blot Analysis Poly(A)÷-selected RNAs were fractionated by electrophoresis on 1% denaturing agarose gels (containing 0.66 M formaldehyde) and transferred on to nylon membranes (Hybond N, Amersham) by capillary blotting in 20xSSC (lxSSC = 0.15 M NaCI, 0.015 M sodium citrate, pH 7.0).
1598
Human Prostatic al-adrenoceptor Subtypes
Vol. 54, No. 21, 1994
cDNA fragments generated by RT-PCR were purified from recombinant pBluescript II KS ÷ plasmid preparations and labelled by random priming (Megaprime, Amersham) to a specific activity of ~ 1.109 cpm/pg DNA. The filters were prehybridized at 65°C for 1 h in 0.5 M sodium phosphate buffer (pH 6.5), 1.0 mM EDTA, 7% SDS and 100 pg/ml salmon sperm DNA. The prehybridization solution was replaced with fresh solution containing 3zp-labelled probe (2-4.10 6 cpm/ml), and incubation was performed for 15-20 h at 65°C. After hybridization, filters were washed in 2xSSC/0.1% SDS 3 times at room temperature and in 0.2xSSC/0.1% SDS twice 30 min at 55°C, then exposed to Hyperfilm MP (Amersham) at -80°C with two intensifying screens. Blots were stripped by incubating them twice in boiling 0.1% SDS and waiting until the temperature decreased to room temperature. Blots were reexposed to film and restripped if necessary before reprobing.
7. Radioligand Binding Assays Membranes from human prostatic adenoma tissue were prepared by homogenization and centrifugation twice at 37,000 x g for 10 min in 50 mM Tris-HCl buffer, pH 7.5. Membranes (120-180 ~g prot) were incubated with [3H]-prazosin for 30 min at 25°C in a total volume of 1 ml. Phentolamine (100 pM) was used to define non-specific binding. Incubations were terminated by rapid filtration using a Brandell cell harvester through Whatman GF/B glass fibre filters. The filters were washed with 3 x 5 ml of buffer, dried and the radioactivity measured by liquid scintillation spectrometry. Saturation and competitive inhibition data were analysed by nonlinear regression analysis programmes. Results
1. Selective Amplilqcation and Subcloning o( 3 Distinct o~t-AR Subtypes O'om Human Prostate Alignment of the primary structures of the 3 members of the cq-adrenoceptor subfamily cloned to date (cqb-, ¢xtc- and ~d-ARs) shows a high degree of sequence homology between subtypes, with notably 65 to 73% amino acid identity in the seven putative transmembrane spanning domains (TMs). Three different sets of highly "degenerate" oligonucleotide primers were designed from stretches of amino acid residues highly conserved between the 3 ¢x~-AR subtypes (Fig. 1). Sens primers a, c and e were designed respectively from consensus amino acid sequences AC(H/N)RH (in the first putative intracellular loop), DVLCC (in TM3) and VMYCR (in TM5). Antisens primers b and d were based on conserved peptide sequences VMYCR (in TM5) and FWLGY (in TM7), respectively. Each primer contained restriction site sequence at their 5' end in order to allow subsequent cloning of the PCR amplified products. Poly(A)÷ RNA (250 ng) prepared from the lateral lobe region of human prostatic adenoma tissue was reverse-transcribed and the resulting cDNA amplified by PCR using either a-b, c-d or e-d pairs of primers. After 30 cycles of amplification, PCR product bands of the expected sizes (~ 480, 650 and 380 bp, for a-b, c-d and e-d amplifications, respectively) were visualized upon agarose gel electrophoresis of the PCR reaction mixtures. No DNA band was detected when the reverse transcriptase was omitted before the PCR amplification reaction, showing that the PCR products resulted from cDNA and not from contaminating genomic DNA (Fig. 2). PCR amplified products of the predicted sizes were gel purified and subcloned into the plasmid pBluescript II KS÷. Minipreparations of plasmids with subcloned inserts were analysed by restriction mapping and sequencing.
Vol. 54, No. 21, 1994
Human Prostatic al-adrenoceptor Subtypes
1599
h~ flD,-
NH
2
TM 1 TM 2 TM3 ~
TM 4 TM5
TM6 TM7 ~
COOH
aiD--qlb clP,~
-,~d el~,-
~d FIG. 1 Schematic diagram of oligonucleotide primers chosen for RT-PCR amplification of cq-AR subtypes. The sequences of the different sets of primers (notated by letters) are described in Materials and Methods.
Primers • RT"
a/b ~
c/d e/d ~ - ~ - I ~+---~-1
FIG. 2 RT-PCR analysis of ~x~-AR subtypes expressed in human prostate. Poly(A+)-selected RNA (250 ng) was reverse-transcribed in the presence (+) or absence (-) of Reverse Transcriptase (RT) and the resulting cDNA was amplified by PCR with the primer sets a-b, c-d and e-d shown in Fig. 1 under conditions described in Materials and Methods. PCR products were electrophoresed on 2% agarose gel and visualized by ethidium bromide staining. The size markers are shown on the left.
Three distinct types of PCR products, related to the eq-AR subfamily, were generated from human prostatic tissue. The amino acid sequences deduced from 2 of them were 100% identical to those reported for the corresponding regions of the human t~ab- (8) and human (tld- (15) ARs (data not shown). The deduced amino acid sequences of the third type of PCR products were closely related to those of the corresponding regions of the cqc-AR subtype cloned previously from bovine brain (9). Two "degenerate" primers were designed in order to get more sequence information for the
1600
Human Prostatic al-adrenoceptor Subtypes
Vol. 54, No. 21, 1994
human Ct~c-AR subtype. Using sens primer h, designed from consensus peptide FWLGY in TM7, and antisens primer i, designed from the sequence CEWKIF in the C terminal region of bovine cq~-AR sequence (9), RT-PCR amplification of human prostatic RNA allowed us to subclone an additional cDNA fragment covering part of the C terminal region of the human (x~-AR. Alignment of the reconstituted human tx~-AR sequence, which spans the region from the first intracellular loop domain to the middle of the C terminal region, with that of the corresponding region of the cloned bovine cqc-AR shows 94% identity at the amino acid level (Fig. 3). TM2 Human ctlc-AR
Bovine czlc-AR
TM3
ACHRHLRSVTHYYIVNLAVADLLLTS~AIF~VlLGYWAFGRVFCI~I [WAAVDVLCCTASIMGLCRSIDRYI I 75
ACnSnLmVTnVV~VADLLLTSrVLPFSAIF~ I IL~VWAmRWC~I~IWAAVOVLCCrASIM~LCnSmRW I 75 TM4
Humanalc-AR Bovine ctlc-AR
TM5
GVSYPLRYPTIVTQ]RtRGLMAI1CVWALSLVISIGPLFGWRQPAPEDETICQINEEI~YVLFS ALGSF~I~ 150 GVSYPLRYPTIVTQIK]RGLMALLCVWALSLVISIGPLFGWRQPAPEDETIOqlNEEPGYVLFSALGSF~Xl_..~I1[~.~ 150
Bovine ctlc-AR
~
Human ctIc-AR
KTLGIVV~CWLPFFLVMPIGSFFPDI~K~-'~------~ FWLGYLNSCINPI l YPCSSQEFKKAFQNVLRIQCI4
Human =lc-AR
CRVYVVAKRESRGLKSGLKTDKSDSEQVTLRIHRKN~4PA1 " ~ - - ~ S ~ 1]KTHFSVRLLKFSREKKAAI 220 CRVYVVAKRESRGLKSGLKTDKSDSEQVTLRIHRKNAIQVIGGSGI VTISAKINIKTHFSVI~LKFSREKKAAI 220 TM6
TM7
Bovine ctlc-AR
KTLGIWGCFVLCWLPFFLVMPIGSFFPDF~ RIPSEWFKI IAIFWLGYLNSCINP I I YPCSSQEFKKAFqNVLRIQCLI
Humanczlc-AR Bovine c~lc-AR
~
295 295
p ~-'] QAV~ - - - - ~ MI--~---~ ~ R ISKTDGVCEWI~F['~ 349 A[PSI HVL~GQHKI~L~rRIPVGSIAIETFY]KISKTDGVCEWIqIIF I 349
FIG. 3 Alignment of the deduced amino acid partial sequence of the human ~k-AR with that of the corresponding region of the bovine ~k-AR. Putative transmembrane-spanning
domains (TMs) are indicated. 2. Northern Blot Analysis Human O~I-ARsubtype selective cDNA probes, covering the region which includes the third intracellular loop of ~lb-, t~lc- and ~d-ARs were obtained by RT-PCR from prostatic mRNA. The ~b-AR and cttc-AR probes spanned the region from the fifth to seventh transmembrane domains and resulted from RT-PCR amplification using "degenerate" e and d primers. The t~ld-ARcDNA probe used for the present report was generated using human t~d-AR specific oligonucleotides f and g (Fig. 1). The subtype-selectivity of these 3 probes towards their corresponding t~-AR subtype was confirmed by Northern blot hybridization of poly(A)÷-selected RNAs extracted from permanently-established t~t-AR cell-lines, expressing either the hamster cqb-AR, the bovine ~]c-AR or the rat t~ld-AR (Fig. 4). These 3 probes were then used to examine the pattern of expression of ~I-AR subtypes in 4 different regions (notably, periurethra, base, apex and lateral lobe) of human prostatic adenoma tissue (Fig. 5). Similar ~ - A R subtype expression patterns were noted in each region (Fig. 5). Using the eqb-AR cDNA probe, a single mRNA species of 2.9 Kb was detected. The ~c-AR cDNA probe hybridized to 2 major mRNA species of 5.7 and 3.0 Kb, as well as a minor mRNA of 3.9 Kb. In addition, 2 major mRNA species of 3.0 and 1.7 Kb were visualized in the human prostate using the (X~d-AR cDNA probe.
Vol. 54, No. 21, 1994
Human Prostatic al-adrenoceptor Subtypes
(x~-AR cell line :
1601
O~lb (Xlc (Xld
(Xlb (Xlc O~ld
(7"1b ~1c O~ld
(Zlb-AR
alc-AR
~ld-AR
4.4 _
2.4 _
1.4
i
(Kb) Probe :
FIG. 4 Subtype-selectivity of the different RT-PCR cloned human ~I-AR eDNA probes. Northern blot analysis of poly(A+)-selected RNAs (5 pg per lane) from HeLa cells stably transfected with hamster ~lb-, bovine eric- and rat ctm-AR cDNAs was performed using 32p-labelled human ct~-AR cDNA probes cloned by RT-PCR. After hybridization, final washing conditions of the filters were 0.2 x SSC, 0.1% SDS at 55°C. Filters were exposed to Amersham MP films at -80°C for 24 h.
1
2
3
4
1
2
3
4
1
2
3
4
7.5i
4,-
!i
iil;ill i!ii~
2.4 --
+iiii+i + .........
?++~iii~~ ! 1.4--
(~)
Probe :
O~lb-AR
cqc-AR
O~ld-AR
FIG. 5 Northern blot analysis of a~-AR subtype mRNAs in human prostatic tissue. The RNA blot was prepared with poly(A÷) RNA extracted from different prostatic regions (1, periurethra; 2, base; 3, apex; 4, lateral lobe) and probed with 32P-labelled human al-AR subtype-selective cDNA probes (rehybridization of the same blot with the 3 cq-AR probes). After hybridization, final washing conditions of the filters were 0.2xSSC, 0.1% SDS at 55°C. Filters were exposed to Amersham MP films with two intensifying screens at -80°C for 24 h. The quality of RNA blotted was controlled by reprobing with 32P-labelled human Glyceraldehyde 3-phosphate dehydrogenase (G3PDH) probe.
1602
Human Prostatic al-adrenoceptor Subtypes
Vol. 54, No. 21, 1994
3. [3H]-Prazosin Binding to Human Prostate Equilibrium saturation analysis of [3H]-prazosin binding to membrane preparations of human prostatic adenoma tissue revealed the binding of this radioligand to an apparently homogenous class of binding sites with a K d of 0.16 nM and a B,,~x of 130 fmol/mg prot. Competitive inhibition experiments of [3H]-prazosin binding to membrane preparations of human prostate were performed using varying concentrations of unlabelled oxymetazoline and alfuzosin (Fig. 6). [3H]-Prazosin binding was inhibited by alfuzosin in a monophasic fashion with a pK~ of 8.29. The oxymetazoline inhibition curve showed a significantly better fit for a 2 site binding model compared to a single site model (p < 0.002) with pK~ values of 8.54 and 5.46 at the high and low affinity binding sites, respectively. The high-affinity component corresponded to 57%-66% of the [3H]-prazosin binding sites present in these tissue homogenates (n = 3 experiments).
loo
H
Alfuzosin
E N
50
E
oo o~
0
I
t
I
i - log [drug]
FIG. 6 Inhibition of [3H]-prazosin binding to human prostatic adenoma tissue. Membranes were incubated with [3H]-prazosin in the absence and presence of different concentrations of oxymetazoline and alfuzosin as described in Materials and Methods. Discussion The concept of cq-AR subtypes, originally defined on the basis of pharmacological studies (18,19), has been confirmed and extended by the cloning of 3 pharmacologically- and structurallydistinct members of this subfamily (5-15). For the present report we have undertaken the identification of oq-AR subtype(s) expressed in human prostatic adenoma tissue by looking for the presence of mRNA transcripts corresponding to the ~lb-' [~lc- and oqa-AR subtypes. In addition, a complementary study on the pharmacological profile of [3H]-prazosin binding to human prostatic tissue has been performed. Based upon the reported sequences of cDNAs encoding ~ r A R subtypes that have been cloned to date various sets of degenerate oligonucleotide primers were designed for use in a RT-PCR strategy to permit identification of the expression of oqb-, ~c- and cqa-AR subtypes in human tissues. Application of this RT-PCR strategy to human prostatic tissue and subsequent characterization of subcloned PCR products by restriction mapping and sequencing revealed the presence in this tissue of 3 distinct oq-AR subtypes. The deduced amino acid sequences of two
Vol. 54, No. 21, 1994
Human Prostatic al-adrenoceptor Subtypes
1603
of these PCR products were identical to those reported in the literature for the corresponding regions of the human ~X~b-(8) and human ~d- (15) ARs. In addition, the deduced amino acid sequences of the other PCR products revealed that they were closely related to the (xac-AR subtype previously cloned from bovine brain (9). In fact, the reconstituted sequence of these human PCR products showed 94% identity at the amino acid level with the cloned bovine tXl~-AR. Also, the extensive amino acid sequence information that we present on this human cqc-AR subtype matches exactly that reported very recently for the human cq¢-AR subtype (10). Thus, this RT-PCR technique permitted the identification of the expression of 3 oq-AR subtypes (Ohb, 0~1¢ and 0qd) in human prostate. It should be noted, however, that the highly-degenerate oligonucleotide probes chosen for this RT-PCR study do not permit interpretations to be made as to the relative expression levels of these different tx~-AR subtypes in this tissue. Using human cq-AR subtype selective cDNA probes the pattern of expression of cq-AR subtypes in different regions of human prostatic adenoma tissue was examined by Northern blot hybridization studies. This technique revealed the co-expression of all 3 cloned to date members of the cq-AR subfamily (~b, cite and Cqd) in 4 different regions of the prostate namely, apex, base, periurethra and lateral lobe. This observation showing the expression in the human prostate of mRNA transcripts corresponding to the Cttb-, ~t~- and (t~d-AR subtypes thus concords with the results of a recent study (20). Furthermore, the quantitative solution hybridization protocol adopted in that study permitted those authors to conclude that the predominant (60-75%) mRNA txI-AR subtype present in human prostate was in fact cq~. For that reason, we have undertaken a complementary study using in vitro [3H]-prazosin radioligand binding assays to investigate at the protein level the degree of ctI-AR subtype heterogeneity in the human prostate. Equilibrium saturation binding of [3H]-prazosin to human prostatic adenoma tissue revealed that this radioligand bound to a single class of high-affinity binding sites in this tissue (Ka = 0.16 riM). Moreover, in competition inhibition experiments, the cq-AR antagonist alfuzosin potently inhibited [3H]-prazosin binding to this tissue with a K~value in the low nanomolar range. In agreement with previous studies (1,3,21), therefore, our [3H]-prazosin binding data illustrate the presence of ¢q-ARs in human prostate. The (x~-AR antagonist, alfuzosin (22), however, displays similar potencies at inhibiting [3H]-prazosin binding both to the 3 cloned to date members (Cqb, Cqc and tx~a) of the cq-AR subfamily as well as to the classical pharmacologically-defined tX~A-AR of rat salivary gland (C. Pimoule and D. Graham, unpublished data). As such, we chose in addition to examine the profile of inhibition of [3H]-prazosin binding by oxymetazoline, a compound that shows a very marked selectivity for the ~A- and cqc-AR subtypes compared to (~lb- and Cqd-ARs (12-14). Interestingly, oxymetazoline inhibition of [3H]-prazosin binding to human prostatic adenomas showed the presence of two cq-AR subtypes in this tissue. The high-affinity ct~-AR component of [3H]-prazosin binding displayed an affinity for oxymetazoline comparable to that exhibited by the rat salivary gland (t~n- (12) and cloned bovine brain t~c-AR (9) subtypes for this compound. Moreover, this high-affinity cq-AR subtype represented the predominant (57%-66%) oq-AR subtype present in this tissue. As such, our data on [3H]-prazosin binding are concordant with the results of a recent study on 125I-HEAT binding to human prostate indicating on the basis of biphasic inhibition curves for WB4101 and 5-methylurapidil that the dominant eq-AR subtype in this tissue showed the pharmacological profile of the O~IA-ARsubtype (23). Since the pharmacological profiles of the classical pharmacologically-defined t~A-AR and the cloned t~c-AR are very similar, the finding that the tx~c-AR is the predominant oq-AR mRNA subtype present in the human prostate (20) probably infers that the majority cq-AR subtype at the protein level is in fact the (x~c-AR. In conclusion, both [3H]-prazosin binding and Northem blot analysis indicate the presence of several Ctl-AR subtypes in human prostatic adenoma tissue and importantly that ctlc represents the
1604
Human Prostatic al-adrenoeeptor Subtypes
Vol. 54, No. 21, 1994
predominant cq-AR subtype present in this tissue. It will therefore be particularly pertinent to verify in functional studies the precise importance of the cqc-AR subtype in mediating human prostatic smooth muscle contraction. In this context, the possibility of developing a selective cqc-AR antagonist potentially devoid of vascular side-effects mediated by other cq-AR subtypes could lead to second generation ~ - A R blockers of improved safety and efficacy for the treatment of BPH.
Acknowledgements: The authors thank Caroline Gouhier and Christiane Sellier for their technical assistance and Fran~oise P6choux for secretarial assistance in the preparation of this manuscript. References 1. 2. 3.
4. 5. 6. 7. 8. 9.
10. 11. 12. 13.
14. 15. 16. 17. 18. 19. 20. 21.
22.
H. LEPOR and E. SHAPIRO, J. Urol. 1321226-1229 (1984). E. SHAPIRO and H. LEPOR, J. Urol. 1351038-1042 (1986). C.R. CHAPPLE, M.L. AUBRY, S. JAMES, P.M. GREENGRASS, G. BURNSTOCK, R.T. TURNER-WARWICK, E.J.G. MILROY and M.J. DAVEY, Brit. J. Urol. 6___33487-496 (1989). A. JARDIN, H. BENSADOUN, M.C. DELAUCHE-CAVALLOR and P. ATFALl, Lancet 3371457-1461 (1991). D.B. BYLUND, D.C. EIKENBERG, J.P. HIEBLE, S.Z. LANGER, R.J. LEFKOWlTZ, K.P. MINNEMAN, P.B. MOLINOFF and R.R. RUFFOLO, Jnr., Pharmacol. Revs. (in press). S. COTECCHIA, D.A. SCHWINN, R.R. RANDALL, R.J. LEFKOWITZ, M.G. CARON and B.K. KOBILKA, Proc. Natl. Acad. Sci. 8...557159-7163 (1988). M.M. VOIGT, J. KISPERT and H. CHIN, Nucl. Acids Res. 1__.881053(1990). C.S. RAMARAO, J.M.K. DENKER, D.M. PEREZ, R.J. GAIVIN, R.P. RIEK and R.M. GRAHAM, J. Biol. Chem. 26721936-21945 (1992). D.A. SCHWlNN, J.W. LOMASNEY, W. LORENZ, P.J. SZKLUT, R.T. FREMEAU, JR., T.L. YANG-FENG, M.G. CARON, R.J. LEFKOWITZ and S. COTECCHIA, J. Biol. Chem. 265, 14, 8183-8189 (1990). A. HIRASAWA, K. HORIE, T. TANAKA, K. TAKAGAKI, M. MURAl, J. YANO and G. TSUJIMATO, Biochem. Biophys. Res. Commun. 195902-909 (1993). K.P. MINNEMAN, C. HAN and P.W. ABEL, Mol. Pharmacol. 3__3509-514 (1988). A.D. MICHEL, D.N. LOURY and R.L. WHITING, Br. J. Pharmacol. 98883-889 (1989). J.W. LOMASNEY, S. COTECCHIA, W. LORENZ, W-Y LEUNG, D.A. SCHWINN, T.L. YANG-FENG, M. BROWNSTEIN, R.J. LEFKOWITZ and M. CARON, J. Biol. Chem. 266 6365-6369 (1991). D.M. PEREZ, M.T. PIASCIK and R.M. GRAHAM, Mol. Pharmacol. 4..._0876-883 (1991). J.F. BRUNO, J. WHIqTAKER, J. SONG and M. BERELOWITZ, Biochem. Biophys. Res. Comm. 1791485-1490 (1991). P. CHOMCZYMSKI and N. SACCHI, Anal. Biochem. 162156-159 (1987). F. SANGER, S. NICKLEN and A.R. CARLSON, Proc. Natl. Acad. Sci. 7__.445463-5467 (1977). K.P. MINNEMAN, C. HAN and P.W. ABEL, Mol. Pharmacol. 3...33509-514 (1988). K.P. MINNEMAN, Pharmacol. Rev. 4._0_0,2, 87-119 (1988). D.J. PRICE, D.A. SCHWINN, J.W. LOMASNEY, L.F. ALLEN, M.C. CARON and R.J. LEFKOWITZ R.J., J.Urol. 150546-551 (1993). H. SCHOEMAKER, H. BLANCHARD, C. PIMOULE, F. LEFI~VRE-BORG, P. MANOURY, A. JARDIN and S.Z. LANGER, in: Prostate and Alpha Blockers, Excerpta Medica pp. 328-343 (1989). F. LEFI~VRE-BORG, S.E. O'CONNOR, H. SCHOEMAKER, P.E. HICKS, J. LECHAIRE, E. GAUTIER, F. PIERRE, C. PIMOULE, P. MANOURY and S.Z. LANGER, Br. J.
Vol. 54, No. 21, 1994
23.
Human Prostatic al-adrenoceptor Subtypes
Pharmacol. 109 1282-1289 (1993). H. LEPOR, R. TANG, S. MERETYK and E. SHAPIRO, J. Urol. 149 640-642 (1993).
1605