Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Pharmacological, radioligand binding and molecular studies have subclassified the alpha-1 adrenoceptors into three subtypes, namely, alpha-1A, alpha-1B and alpha-1D for native receptors, and alpha-1a (historically, alpha-1c), alpha-1b and alpha-1d (historically, alpha-1a, alpha-1d or alpha-1a/d) for cloned receptors (Hieble et al., 1995a). These subtypes can be identified with selective and nonselective antagonists. For example, WB 4101 has high affinity for alpha-1A as well as for alpha-1D (Saussy et al., 1994; Buckner et al., 1996), 5-MU has high affinity for alpha-1A, BMY 7378 (Saussy et al., 1994; Piascik et al., 1995) and SK&F 105854 (Hieble et al., 1995a, b) have high affinity for alpha-1D and CEC selectively inactivates alpha-1B and, to a lesser extent, alpha-1D (see Hieble et al., 1995a) adrenoceptors. Subtypes of alpha-1 adrenoceptors also have been classified by binding and pharmacological studies according to high (pK_D > 9) and low affinity (pK_D < 9) for prazosin (Muramatsu et al., 1990; Oshita et al., 1992; Ford et al., 1994; Hieble et al., 1995a). All the cloned receptor subtypes correspond to those with high affinity.

Many studies have been conducted to identify alpha-1 adrenoceptor subtypes mediating vascular contraction in the rat aorta. Strong evidence, based on the effects of BMY 7375, WB 4101, 5-MU and CEC, has been obtained arguing the predominant presence of the alpha-1D subtype in the rat aorta (Saussy et al., 1996; Buckner et al., 1996; Kenny et al., 1995, 1996). Other investigators have suggested that alpha-1B as well as alpha-1D adrenoceptors also may be present (Van der Graaf et al., 1996). BMY administration in rats competitively antagonized the PE-induced pressor response, which suggests a role for alpha-1D adrenoceptors in the regulation of vascular resistance (Zhou and Vargas, 1996). Molecular studies in the rat aorta suggest that it transcribes the genes for alpha-1b, alpha-1c (alpha-1a) and alpha-1a/d adrenoceptor subtypes (Rokosh et al., 1994; Piascik et al., 1994) to make mRNA. Whether all are processed into expressed proteins is unclear.

In dog, density of plasma membrane [³H]prazosin binding sites was three times higher in the aorta than in the mesenteric artery, mesenteric vein and saphenous vein (Shi et al., 1989). Moreover, the K_D for [³H]prazosin binding in the dog aorta was significantly lower than in the other vessels (0.15 nM vs. 1-3 nM), which suggests the presence of high-affinity prazosin (alpha-1) binding sites (pK_D of 9.82) (Shi et al., 1989; Hoo et al., 1994). On the other hand, the densities of plasma membrane [³H]rauwolscine binding sites in the saphenous vein and the mesenteric vein were higher than in the aorta and mesenteric artery. Hoo et al. (1994) and Oriowa and Ruffolo (1992) have suggested that the alpha-1B adrenoceptor subtype is present in the dog aorta, but the presence of alpha-1D adrenoceptors remains to be assessed.

Recently, putative antagonists apparently having 50- to 100-fold selectivity for alpha-1D have been reported (Hieble et al., 1995a, b; Saussy et al., 1994, 1996). These are BMY 7378 (BMY) and MDL 73005EF (Saussy et al., 1994, 1996; Hieble et al., 1995a [IUPHAR nomenclature, 7th edition]) and SK&F 105854 (SKF) (Hieble et al., 1995b). Also a highly selective antagonist against alpha-1A adrenoceptors, RS 17053 (RS) (Ford et al., 1996), has been identified and an antagonist potent against this receptor but more potent against a presumptive new alpha-1 adrenoceptor, the 1L subtype (Testa et al., 1995, 1996, 1997; Leonardi, 1997) has become available. Because the alpha-2 adrenoceptor agonists were unable to produce contractions, and because of the low density of [³H]rauwolscine binding sites (Shi et al., 1989), we did not attempt to classify the alpha-2 adrenoceptors. In this study, we have characterized the pharmacological profile of alpha-1 adrenoceptor subtypes in the dog aorta with six antagonists against the various alpha-1 adrenoceptor subtypes. Our data indicate that two alpha-1 subtypes may be present in the dog aorta which resemble the alpha-1B and alpha-1D adrenoceptors in the rat aorta; however, the predominant receptor apparently is the alpha-1B adrenoceptor. Expression of alpha-1B adrenoceptors in this tissue was confirmed by immunocytochemical localization and Western blotting of microsomal membranes.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Animals and tissue preparation. Mongrel dogs (10-30 kg) were sacrificed by an overdose of sodium pentobarbital (100 mg·kg¹ i.v.), according to protocols approved by the Animal Care Committee at McMaster University and following the guidelines of the Canadian Council on Animal Care.

Contractility experiments. For muscle bath experiments, the dog aorta was removed and cleaned of surrounding tissue. A piece of the aorta was dissected open, and strips of the aorta were prepared by cutting along the perimeter of the aortic opened ring. These procedures usually resulted in tissues of 1 cm in length and about 2 mm in thickness. Tissue was dissected in a Petri dish filled with oxygenated Krebs' solution at 25°C. The composition of Krebs' solution in mM was: NaCl, 115; KCl, 4.9; MgCl₂, 1.16; NaH₂PO₄, 1.10; NaHCO₃, 21.9; CaCl₂, 2.5; glucose, 11.0. The removal of endothelium by rubbing the luminal surface of the vessels was confirmed by the absence of carbachol (1 µM)-induced relaxation of precontractions produced by 60 mM K⁺.

Data analysis. All contractile responses are expressed as a percentage of the contractile response to 100 mM K⁺ unless otherwise stated. Apparent K_B values were calculated comparing the mean EC₅₀ values of PE concentration-response curves (estimated for each curve using the logistic function in MicroCal Origin Software, Northampton, MA) to those from simultaneously studied controls. This was done because relaxation after PE-induced contraction in the dog aorta required prolonged washing of more than 60 min, precluding repetitive concentration-response curves. EC₅₀ values were evaluated for the significance of differences with treatment with one-way analysis of variance. Data from determinations of K_B values were expressed as pK_B and standard errors and significance of differences between values at different antagonist concentrations determined. Statistical significance was accepted at P values of less than .05. When pK_B values across a range of several antagonist concentrations were not significantly different, the values were pooled. Data are expressed as means ± S.E.M.

Immunocytochemical studies. Four healthy dogs of either sex were sacrificed and blood vessels collected from aorta and mesenteric arteries as described above. Blood vessels were opened, rinsed free of blood and pinned out on Sylgard silicon rubber-coated dishes and fixed with 4% paraformaldehyde with 0.1 M phosphate buffer, pH 7.4. The tissues to be used for cryostat sectioning were cut into small pieces and then stored in 15% sucrose containing phosphate-buffered saline for cryoprotection at 4°C for 24 hr and sectioned in 16-µm thicknesses in a cryostat (Leitz 1720 digital). The sections were collected on the slides coated with gelatin. Cryostat sections were incubated overnight at 4°C in 1:300 dilutions of rabbit antisera raised against residues 506-515 at the carboxyl terminus of the hamster alpha-1B adrenoreceptor which had been coupled to keyhole limpet hemocyanin (Fonseca et al., 1995, 1997).The antibody was visualized with CY3 labeled goat anti-rabbit goat anti-mouse antibodies (Jackson ImmunoResearch, West Grove, PA). Specificity of staining was ascertained with preimmune serum and by saturation of the antibody with the peptide epitope against which it was raised (5 µg/ml) during exposure of cryostat sections. Background and autofluorescence was evaluated by omission of the primary antibody. After washing with phosphate-buffered saline, the sections were then mounted in 80% glycerol in phosphate-buffered saline (pH 10) and viewed on a Leitz microscope equipped with fluorescence epiluminator and I₂ filter. Kodak T-MAX 400 film was used for black and white photography.

Western blotting. Plasmalemma-enriched microsomal membrane fractions used for Western blotting studies were isolated from dog aortic smooth muscle layers according to fractionation procedures previously developed and characterized in this laboratory (Kwan et al., 1984). Mic I fraction represents the postmitochondrial fraction 4- to 6-fold enriched in plasmalemma content over the postnuclear supernatent, and Mic II fraction represents further a refined fraction from Mic I with plasmalemma content twice that in Mic I. Postnuclear supernatent membranes were prepared from rat spleen, as a tissue source of alpha-1B adrenoceptors (Hoo et al ., 1994).

Drugs. Unless otherwise stated, the drugs were dissolved in double distilled, deionized water. L-phenylephrine (Sigma, ST. Louis, MO), WB 4101 (2-(2,6-dimethoxyphenoxyethyl)aminomethyl-1,4-benzodioxane hydrochloride (Research Biochemicals Inc. [RBI], Natick, MA), BMY 7378 (8-(2-[4-(2-methoxyphenyl)-1-piperazinyl]-ethyl)-8-azaspiro[4,5]decane-7,9-dione dihydrochloride, RBI), MDL 72832, {8-[4-(1,4-benzodioxan-2-ylmethylamino)butyl]-8-azaspirol[4,5]decane-7,9-dione HCl and MDL 73005EF, {8-[2-(1,4-benzodioxan-2-ylmethylamino)ethyl]-8-azaspirol[4,5]decane-7,9-dione HCl (Tocris Cookson Chemicals, Bristol, UK), prazosin (dissolved in dimethyl sulfoxide to a stock of 10 mM and protected from light, Sigma) and chloroethylclonidine dihydrochloride (CEC, RBI). SK&F 105854 was a generous gift from Dr. J.P. Hieble (SmithKline Beecham Pharmaceuticals, King of Prussia, PA).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effects of prazosin on PE concentration-response curves. The aortic strips were treated with prazosin, a poorly selective alpha-1 antagonist, at 10, 30 and 100 nM before cumulative additions of PE to construct a concentration-response curve (fig. 1, table 1). In tissues treated with prazosin, rightward shift of PE concentration-response curves was indicated by the EC₅₀ values of PE. The pK_B values for concentrations of prazosin of 10⁸, 3 × 10⁸ and 10⁷ M were not significantly different (8.98 ± 0.22, n = 7; 9.02 ± 0.16, n = 7; and 8.95 ± 0.23, n = 6), equivalent to a Schild plot slope not different from 1, so these values were pooled and a mean pK_B of 8.99 ± 0.11, n = 20 (K_B = 1.03 nM) for prazosin was obtained.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   The effects of prazosin at 0 (time controls, filled squares, n = 8), 10 nM (open squares, n = 7), 30 nM (filled triangles, n = 7) and 100 nM (open triangles, n = 7) on PE concentration-response curves were examined. A rightward shift in the EC50 values of PE was observed with each concentration of prazosin used compared with time controls. Values are means ± S.E.M. (number of experiments).

Effects of WB 4101 on PE concentration-response curves. WB 4101, which has high affinity for alpha-1A and alpha-1D, shifted PE concentration-response curves in dog aorta as shown in figure 2. At 10, 30 and 100 nM, WB 4101 affected EC₅₀ values of PE concentration-response curves (table 1) with pK_B values of 8.80 ± 0.10, n = 8; 8.81 ± 0.20, n = 8; and 8.65 ± 0.10, n = 7, respectively. These were not significantly different, and a Schild slope was not different from 1, so the data were pooled with a resultant pK_B value for WB 4101 of 8.75 ± 0.08 (n = 23) and K_B = 1.76 nM. The value suggests the dog aorta alpha receptors have an affinity for WB 4101 that is intermediate between alpha-1A or alpha-1D adrenoceptors (pK_B > 9) and alpha-1B adrenoceptors (pK_B   8) [see Saussy et al. (1996) and Ford et al . (1994)].

View larger version (18K): [in this window] [in a new window]   Fig. 2.   The effects of WB 4101 (WB) at 0 (time controls, filled squares, n = 8), 10 nM (open squares, n = 8), 30 nM (filled triangles, n = 8) and 100 nM (open triangles, n = 7) on PE concentration-response curves were examined. A rightward shift in the EC50 values of PE was observed with each concentration of WB used compared with time controls. Values are means ± S.E.M. (number of experiments).

Effects of CEC on PE concentration-response curves. Chloroethylclonidine irreversibly inactivates alpha-1B and a large fraction of alpha-1D adrenoceptors. It also can inactivate some alpha-2 adrenoceptors and can interact competitively with other alpha subtypes (Michel et al., 1993; Low et al., 1994; Nunes and Guimaraes, 1993). In the dog aorta, low concentrations of CEC (fig. 3; table 1) (100, 300 and 1000 nM) had no functional effect on PE concentration-response curves. However, higher concentrations (10 and 100 µM) of CEC noncompetitively, irreversibly and almost completely abolished PE responses.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   The effects of CEC at 0 (time controls, filled squares, n = 13), 100 nM (open squares, n = 9), 300 nM (filled triangles, n = 8), 1 µM (open triangles, n = 10), 10 µM (filled diamond, n = 3) and 100 µM (open diamond, n = 4) on PE concentration-response curves were examined. EC50 values of PE concentration-response curves were not affected by CEC concentrations between 100 nM and 1 µM. Close to complete inhibition of Emax was observed when tissues were pretreated with CEC at 10 and 100 µM. Values are means ± S.E.M. (number of experiments).

Effects of 5-MU on PE concentration-response curves. The antagonist 5-MU discriminates between alpha-1A and alpha-1D adrenoceptor subtypes with high affinity only for the alpha-1A subtype. In the dog aorta, 5-MU at concentrations of 10, 30, 100 and 300 nM did not significantly alter the concentration-response curves to PE (fig. 4; table 1).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   The effects of 5-MU at 0 (time controls, filled squares, n = 4), 10 nM (open squares, n = 3), 30 nM (filled triangles, n = 3), 100 nM (open triangles, n = 4) and 300 nM (filled diamonds, n = 4) on PE concentration-response curves were examined. No consistent effect of 5-MU was observed. Values are means ± S.E.M. (number of experiments).

Effects of RS 17053 on PE concentration-response curves. A recently described antagonist (Ford et al., 1996; Lachnit et al., 1997), RS 17053, also was reported to be highly selective for alpha-1A adrenoceptors but to have lower affinity for the putative alpha-1L subtype. It, like 5-MU, had no effect on concentration-response curves of the dog aorta to PE (fig. 5). For concentrations of RS 17053 of 3, 30 and 300 nM, the dose ratios (treated/control) from EC₅₀ values were not significantly different from one: 1.13 ± 0.41, 1.26 ± 0.31 and 1.63 ± 0.27, respectively (mean ± S.E.M., n = 4).

View larger version (18K): [in this window] [in a new window]   Fig. 5.   The lack of significant effects of RS 17053, a highly selective alpha-1A adrenoceptor antagonist, is shown. At 3 (filled circles), 30 (open triangles) or 300 (closed triangles) nM, this antagonist did not significantly shift or depress the PE concentration-response curve compared with controls (open circles). Each curve contains data from four independent experiments.

Effects of MDL 72832 on PE concentration-response curves. MDL 72832 also is considered a somewhat selective alpha-1A adrenoceptor antagonist, based on binding affinities for expressed receptors (Saussy et al., 1996). It shifted the PE concentration-response curve rightward at 10, 100 or 1000 nM with pK_B values of 7.83 ± 0.26, 7.93 ± 0.07 and 8.10 ± 0.41 (mean ± S.E.M., n = 4), respectively, yielding a pooled value of 7.95 ± 0.15 (mean ± S.E.M., n = 12) with K_B of 11.2 nM. These pK_B values were too low to be consistent with the presence of alpha-1A adrenoceptors (expected values, 8.4-8.6) and did not distinguish alpha-1B from alpha-1D subtypes.

Effects of BMY 7378 on PE concentration-response curves. BMY 7378, a selective alpha-1D antagonist, at concentrations of 100, 300 and 1000 nM, gave pK_B values of 7.13 ± 0.22 (n = 3), 7.07 ± 0.15 (n = 6) and 7.45 ± 0.28 (n = 4), respectively. The three mean pK_B values were not significantly different, so these data were pooled, resulting in a pooled pK_B of 7.21 ± 0.13 (n = 13) with K_B = 62 nM (fig. 6; table 1). This pK_B value was much less than that reported in rat aorta, 8.88, or the pK_i value for the human alpha-1D adrenoceptor, 9.39, and close to the pK_i for the human alpha-1B adrenoceptor, 7.25 (Saussy et al., 1996)

View larger version (19K): [in this window] [in a new window]   Fig. 6.   The effects of BMY 7378 (BMY) at 0 (time controls, filled squares, n = 6), 100 nM (open squares, n = 6), 300 nM (filled triangles, n = 6) and 10 µM (open triangles, n = 5) on PE concentration-response curves were examined. Competitive rightward shift in the EC50 values of PE was observed with each concentration of BMY used compared with time controls. Values are means ± S.E.M. (number of experiments).

Effects of MDL 73005EF on PE concentration-response curves. MDL 73005EF also was reported to be a alpha-1D adrenoceptor selective antagonist (Saussy et al., 1996). At concentrations of 10, 100 and 1000 nM, it caused rightward shifts of concentration-response curves. However, the calculated pK_B values decreased significantly with antagonist concentrations. At 10, 100 and 1000 nM the respective values were: 7.88 ± 0.06, 7.56 ± 0.28 and 6.92 ± 0.18 (corresponding to K_B values of 13.2, 27.5 and 120.2 nM). In rat aorta, the pK_B value for this compound was 8.00, corresponding to a pK_i value for the expressed human alpha-1D adrenoceptor of 8.16 (Saussyet al., 1996). Thus for all but the lowest concentration, the pK_B values were too low for alpha-1D adrenoceptors, but appropriate at higher concentrations for alpha-1B adrenoceptors (pK_i = 6.88 for the expressed human receptor).

Effects of SK&F 105854 on PE concentration-response curves. Concentration-response curves to PE were constructed in the absence and in the presence of SK&F 105854, a putative, selective alpha-1D antagonist. The pK_B values calculated for the four SKF concentrations used were not significantly different from each other [5.95 ± 0.33 (n = 3), 5.68 ± 0.10 (n = 4), 5.90 ± 0.12 (n = 4) and 5.78 ± 0.16 (n = 4) for SKF concentrations of 1, 3, 10 and 30 µM, respectively], so these pK_B values were pooled to give a value of 5.81 ± 0.08 with K_B = 1.52 µM (n = 15) (see fig. 7; table 1).

View larger version (21K): [in this window] [in a new window]   Fig. 7.   The effects of SK&F 105854 (SKF) at 0 (time controls, filled squares, n = 5), 1 µM (open squares, n = 3), 3 µM (filled triangles, n = 4), 10 µM (open triangles, n = 4) and 30 µM (filled diamond, n = 4) on PE concentration-response curves were examined. A rightward shift of the EC50 values of PE was observed with increasing concentrations of SKF used. Values are means ± S.E.M. (number of experiments).

Immunocytochemistry of alpha-1B adrenoceptors in dog aorta. Studies in dog mesenteric artery strongly suggest that it contains predominantly alpha-1A or alpha-1L adrenoceptors and few or no alpha-1B adrenoceptors (Daniel EE, Lu-Chao H, Low AM, Brown RD and Kwan CY, unpublished). Therefore, staining with an antibody for alpha-1B adrenoceptors was compared in mesenteric artery and aorta. As reported elsewhere, mesenteric artery smooth muscle did not recognize an antibody against an epitope exclusive to the alpha-1B adrenoceptor (Fonseca et al., 1995). However, as shown in figure 8, the aorta stained strongly and all cells were stained. In 15-µm-thick sections the staining appeared to encompass the entire cell except in cells cross-sectioned through the nucleus (fig. 8A). However, when cell edges were cut tangentially the staining appeared to be particulate. This figure shows that all cells within bundles were immunostained, a uniform finding whether tissues were cut in cross-sections (fig. 8A) or in longitudinal sections (fig. 8B). The preimmune serum did not immunostain aorta smooth muscle (fig. 8, C and D), and saturation of the antibody with 5 µg/ml of the peptide antigen used to raise the antibody nearly abolished staining (fig. 8E), leaving only nonspecific staining such as produced after omission of the primary antibody (fig. 8F),

View larger version (142K): [in this window] [in a new window]   Fig. 8.   This figure shows staining of cryostat sections from canine aorta with an antibody against the alpha-1B adrenoceptor and appropriate controls. (A) Staining of muscle cells cut in cross-section. Note particulate nature of stain when cell boundaries are cut tangentially. Unstained central regions of cells reflect cells sectioned through the nucleus. (B) Staining of muscle cells cut in longitudinal section. Note that all cells were stained. (C, D) Very weak or absent staining of similarly cut sections (C, cross-sectioned; D, longitudinally sectioned) by preimmune serum from the same rabbit. (E) Abolition of staining when the antibody was saturated with 5 µg/ml of the peptide epitope used to raise the antibody. (F) Nonspecific staining when the primary antibody was omitted. Length bars: 25 µm in B, C, D and F; 12.5 µm in A and E.

Western blot of aortic microsomal membranes. Mic I and Mic II membranes were studied and immunoblotted with immune serum, preimmune serum and immune serum after preabsorption with the peptide used to raise the antibody. Other microsomal membranes from mesenteric artery (DMA) and rat spleen (RSP) also were used for comparison. As expected, a protein with a molecular weight about 80 kdaltons was found in aortic membranes, more densely present in Mic II than Mic I membranes (fig. 9, lanes 1 and 2). Exposure of the blots to the preimmune serum (lanes1a and 2a) or the immune serum after saturation with 0.2 µg/ml of peptide antigen (not shown) did not lead to staining. Staining of a protein of the same molecular weight was found in membranes from rat spleen in which the alpha-1B receptor was expressed (lane 4), but no expression was found in mesenteric artery membranes (lanes 3), which we have found to contain mostly alpha-1A adrenoceptors (Daniel et al., unpublished). In no case was there staining by preimmune serum.

View larger version (66K): [in this window] [in a new window]   Fig. 9.   This figure contains a representative Western blot (n = 4) of microsomal membranes from canine aorta, DMA and from the membranes of rat spleen (RSP) containing alpha-1B adrenoceptors. Molecular weight standards are shown by dots in the left hand lane. Staining of MicII, Mic I membranes from dog aorta (DAO), membranes from dog mesenteric artery (DMA) and from the rat spleen (RSP) cells are shown in sequence in lanes labeled 1, 2, 3 and 4. In the same sequence, lack of staining of membranes by preimmune serum are shown to the right (1a, 2a, 3a and 4a). As expected, a band at ~ 80 Kd was stained deeply in DAO membranes, more deeply in MicII than in MicI. Minimal staining was present in DMA membranes, and clear but lesser staining than in DAO membranes was present in RSP membranes at the same molecular weight. PNS, postnuclear supernatant.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Molecular pharmacology studies in rats and humans have demonstrated the expression of three subtypes of alpha-1 adrenoceptors (see Hieble et al., 1995a) but have not provided any information about the relative contribution of each receptor subtype to the behavior of whole tissue, such as blood vessels. Pharmacological classification of alpha adrenoceptor subtypes evolved along with the discovery of subtypes and the availability of pharmacological antagonists. In this study we used a panel of nine different alpha-1 adrenoceptor antagonists to classify the subtypes of alpha-1 adrenoceptor that are present in the endothelium-denuded dog aorta. Our evidence suggests that the smooth muscle of dog aorta contains mainly alpha-1B adrenoceptors but some alpha-1D adrenoceptors also may be present.

It has been postulated that all three subtypes, which have high affinity (pK_D > 9) in binding studies for both prazosin and HV-723, are members of the alpha-1H adrenoceptor family (Muramatsu et al., 1990; Oshita et al., 1992; Ford et al., 1994). Our previous radioligand studies showed high-affinity [³H]prazosin binding in the dog aorta (Shi et al., 1989; Hoo et al., 1994) with a pK_D of 9.8. This value is somewhat higher than the pK_B value, 9.0, but is consistent with the demonstration of a high correlation between binding interactions of alpha-1 adrenoceptors in canine aorta and binding to recombinant alpha-1b adrenoceptors recently reported by Leonardi et al. (1997). The mechanism of the approximate one log unit discrepancy between prazosin affinities in ligand binding and in functional studies is unclear but possibly may be related to the different experimental conditions (Grever et al., 1997; Jasper et al., 1997; Michel et al., 1997). However, the lower affinity for prazosin from functional studies cannot be attributed to uptake of agonist into nerve terminals because synthetic PE was used instead of noradrenaline. Furthermore, the dog aorta is sparsely innervated by adrenergic nerves if innervated at all (Kwan et al., 1984). Moreover, phenylephrine has minimal low-affinity interactions with alpha-2 or -adrenoceptors. There are few functional studies in which pK_B values near 10 for prazosin are found; e.g., in the recent study of Leonardi et al. (1997) interactions between prazosin and rabbit urethra, prostate, aorta and ear artery all yielded pK_B values between 7.85 and 8.82; the last value was from the aorta. Therefore, it is possible that the higher affinity values reflect the conditions of binding studies which differ from conditions of functional studies.

Our studies revealed no evidence for the presence of alpha-1A adrenoceptors. Two highly selective antagonists, 5-MU and RS 17053, had no significant antagonistic effects in concentrations which should have markedly inhibited responses. Although less selective, another antagonist, MDL 72832, also failed to suggest the presence of appreciable numbers of these receptors, because the pK_B values were equally consistent with the presence of alpha-1B adrenoceptors. In studies to date antagonists effective against alpha-1A adrenoceptors also have had some potency against the putative alpha-1L subtype (Leonardi et al., 1997). Thus they are also unlikely to be functional in this blood vessel. If alpha-1A and alpha-1L adrenoceptors are functionally absent from the dog aorta, then only alpha-1B or alpha-1D adrenoceptors or both must mediate contraction.

With use of the alkylating agent CEC to irreversibly inactivate the alpha-1B adrenoceptors, Hoo et al. (1994) reported that CEC pretreatment (30 min, 4°C, 100 µM) reduced [³H]prazosin binding sites by about 75% and suggested that the dog aorta contains mainly alpha-1B adrenoceptors. Subsequently, the selectivity of CEC for alpha-1B adrenoceptors has been questioned (see Hieble et al., 1995a; Michel et al., 1993). Our present functional study supports the presence of the alpha-1B adrenoceptors in the dog aorta, which are sensitive to CEC and relatively insensitive to antagonists, BMY 7378, MDL 73005EF and SK&F 105854, alpha-1D selective in rodents and humans. In this study, the pK_B for BMY was 6.95, which is a lower value than reported by Saussy et al. (1996) for human recombinant alpha-1D expressed in rat fibroblasts (9.39) and in rat aorta (8.88), a tissue which demonstrates predominance of the alpha-1D subtype. The pK_i value for BMY with human alpha-1B adrenoceptors expressed in fibroblasts was 7.25, close to the value we observed, but clearly not different from the value for the expressed human alpha-1A adrenoceptor of 6.8. The pK_B values for MDL 73005EF, a structural analog of BMY, decreased with concentration from 7.88 to 6.92. This compound had a pK_B in rat aorta of 8.00 and a pK_i of 8.16 for the human alpha-1D adrenoceptor (Saussy et al., 1996). Thus our values were also inconsistent with the predominant receptor in canine aorta. At low concentrations, the value was consistent with the pK_i of the human alpha-1D adrenoceptor, but at higher concentrations the values were similar to those of the human alpha-1B adrenoceptor (6.88). Although our value for the pK_B of SK&F 105854 (5.82) is similar to the pK_B (5.77) reported by Hieble et al. (1995b) in rat aorta, both values are lower than that for recombinant receptors at alpha-1D adrenoceptors (pK_i = 7.14) and closer to the values for alpha-1B or alpha-1A adrenoceptors (pK_i values of 6.11 and 5.48, respectively). Thus, the inappropriately low affinities which we observe for alpha-1D selective antagonists argue that the CEC-sensitive contractions in dog aorta are subserved mainly by alpha-1B, rather than the alpha-1D, subtype. However, our results allow the presence of a population of alpha-1D adrenoceptors as well.

The reported affinities of WB 4101 for alpha-1 adrenoceptor, pK_i values of 8.71 and 9.76 for rat alpha-1D and alpha-1A adrenoceptors, respectively, and pK_i values of 9.33, 9.54 and 8.62 for human 1D_, 1A and 1B adrenoceptors, respectively (Saussy et al., 1996), compared with the pK_B of 8.6 observed for dog aorta alpha-1 adrenoceptors in this study, are consistent with the possibility that alpha-1B or alpha-1D subtype receptors are indeed present in the dog aorta. Moreover, WB 4101 was reported to inhibit competitively [³H]prazosin binding in both control and CEC-treated membranes (Hoo et al., 1994) as expected if the residual alpha adrenoceptors sensitive to WB 4101 after CEC alkylation are the alpha-1D subtype. The proposal that WB 4101 acts on alpha-1B/D adrenoceptors rather than on alpha-1A subtypes would explain the lack of effect of 5-MU or RS17053 and the low potency (pK_B, 7.15, compared with pK_i values of 8.6 or 8.4 for rat and human alpha-1A adrenoceptors) of MDL 72832 observed in this study.

Thus all the functional evidence is consistent with the premise that alpha-1B adrenoceptors are present and predominantly determine the functional responses of this tissue. This evidence was strongly supported by the Western blot studies which showed a protein of the appropriate molecular weight in aortic plasmalemma as well as in membranes from a cell line in which the receptor was expressed, which was recognized by an antibody against an epitope of the alpha-1B adrenoceptor (Fonseca et al., 1995). The specificity of this recognition was shown by the absence of recognition by preimmune serum, by the ability of the epitope peptide to abolish staining and by the absence of staining in membranes of mesenteric artery which lacks functional evidence of the alpha-1B adrenoceptor. The questions remain, is there an additional receptor in the aorta and how is the alpha-1B adrenoceptor distributed?

If there are both alpha-1B and alpha-1D adrenoceptors on the dog aorta, then an important question is whether they are distributed as a cellular mosaic with some cells exclusively containing each subtype, or as a subcellular mosaic with each cell having some of each subtype? The antibody to the alpha-1B adrenoceptor subtype which recognized the alpha-1B adrenoceptor in Western blots of plasmalemma was also active in immunocytochemical studies (Fonseca et al., 1995). This enabled us to approach this question and to provide further substantiation for the predominance of alpha-1B adrenoceptors. Our studies showed that all smooth muscle cells were immunoreactive to the antibody for alpha-1B subtype adrenoceptors, which suggests that, if a mosaic exists, it is subcellular in distribution. The strong staining obtained suggests that the major receptor subtype expressed in canine aorta is alpha-1B. In this regard, we note that alpha-1B adrenoceptor immunostaining presents a nonuniform or punctate appearance in some cellular profiles, which suggests that receptors may localize to distinct subcellular structures. An antibody specific for alpha-1D adrenoceptors and functional in immunocytochemistry is not available, so we could not test the presence and distribution of alpha-1D adrenoceptors with this technique. The antibody that recognized alpha-1B adrenoceptor also recognized a protein in Western blot studies of ~80 kdaltons molecular weight, near the value reported for alpha-1B adrenoceptors (Fonseca et al., 1995).

In conclusion, the dog aorta smooth muscle exhibits predominantly functional alpha-1B-like adrenoceptor activities. Immunocytochemical data support strong prominent expression of alpha-1B adrenoceptors throughout this vessel. The functional activities of the subtype found in canine aorta correspond to previously reported high-affinity [³H]prazosin binding sites in the same tissue (Shi et al., 1989; Hoo et al., 1994) with a reported pK_D of 9.82 for prazosin. Our conclusions are similar to those of Leonardi et al. (1997) in which binding to canine aortic adrenoceptors correlated highly with binding to recombinant alpha-1b adrenoceptors. There may be additional receptors of the alpha-1D subtype, but so far the uncloned alpha-1L adrenoceptors, as suggested by Leonardi et al. (1997), are unlikely to be present.