Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Alpha₁-adrenoceptors mediate many of the physiological functions of the sympathoadrenal transmitters adrenaline and noradrenaline (Ruffolo and Hieble, 1994). Therefore, they are important drug targets, e.g., in the cardiovascular and urogenital systems (Ruffolo et al., 1995). In recent years it has become clear that at least three subtypes of alpha₁-adrenoceptors exist, which are designated alpha_1A, alpha_1B and alpha_1D (Hieble et al., 1995; Michel et al., 1995). The genes and/or cDNAs encoding these three subtypes have been cloned in rats and humans (Hieble et al., 1995), and species homologs for some subtypes have additionally been cloned in hamsters, cows, rabbits and mice (Cotecchia et al., 1988; Schwinn et al., 1990; Alonso-Llamazares et al., 1995; Miyamoto et al., 1997; Suzuki et al., 1997). The pharmacological characterization of tissue alpha₁-adrenoceptor subtypes has mainly been performed in rats (Michel et al., 1995). Some data on the characterization of tissue alpha₁-adrenoceptor subtypes are available in humans (Gross et al., 1989; Lepor et al., 1993; Michel et al., 1996), guinea pigs (Garcia-Sainz et al., 1992; Haynes and Pennefather, 1993, Büscher et al., 1996) and cows (Büscher et al., 1996) but only very few in mice (Garcia-Sainz et al., 1994).

Transgenic techniques are a powerful tool of molecular pharmacology because they allow the study of phenotypes that are caused by well-defined genotypes (Chien, 1996). The most frequently used species in the generation of transgenic animals is the mouse. In the alpha₁-adrenoceptor field, transgenic mice have been generated that overexpress constitutively active alpha_1B-adrenoceptors in their myocardium (Milano et al., 1994). More recently, knockout mice have been generated that lack endogenous alpha_1B-adrenoceptors (Cavalli et al., 1997). Studies of these mice have yielded intriguing observations, but their interpretations are limited by the scarcity of knowledge regarding the physiological characteristics of murine alpha₁-adrenoceptors relative to those of other species. Therefore, we have characterized alpha₁-adrenoceptors in a variety of murine tissues by using competition binding studies with several subtype-selective drugs and receptor inactivation studies with chloroethylclonidine.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Tissue preparation. Mice of either sex (strain HLG, 25-35 g) were obtained from the central animal breeding facility at the University of Essen. In some experiments alpha_1B-adrenoceptor knockout mice (C57 × 129 hybrids) were used, which have been described in detail previously (Cavalli et al., 1997). Mice were sacrificed by decapitation, and the cerebral cortex, cerebellum, heart, kidney, liver, lung and spleen were removed rapidly. The tissues were macroscopically cleared from adhering connective tissue, rapidly frozen in liquid nitrogen and stored at 80°C for up to 3 months until analysis.

Radioligand binding. Radioligand binding with the use of [³H]prazosin as the ligand was performed as previously described (Michel et al., 1993a). Briefly, aliquots of the membrane suspensions were incubated in a total volume of 1000 µl of binding buffer (50 mM tris [hydroxymethyl]aminomethane, 0.5 mM ethylenediamine tetraacetate at pH 7.5) for 45 min at 25°C. In competition binding experiments with agonists, 100 µM guanosine triphosphate was always added to prevent guanosine diphosphate-dependent formation of agonist high-affinity states. The incubation was terminated by rapid vacuum filtration over Whatman GF/C filters, and each filter was washed twice with 10 ml of binding buffer. After drying of the filters for 1 h at 60°C, 4 ml of scintillator (Quickszint 1, Zinsser, Frankfurt, Germany) was added to each filter, and after vigorous shaking of each sample, the radioactivity on the filters was quantified in a scintillation counter at 42% efficiency. Nonspecific binding was defined as binding in the presence of 10 µM phentolamine. In some experiments, membrane preparations were treated with the indicated concentrations of chloroethylclonidine or vehicle for 30 min at 37°C, followed by two washout centrifugations before incubation with the radioligand.

Chemicals. Dye reagent for the protein assay was purchased from Bio-Rad (Munich, Germany); methoxamine HCl and (-)-noradrenaline bitartrate from Sigma (Deisenhofen, Germany); chloroethylclonidine HCl, 5-methylurapidil, (+)-niguldipine HCl and BMY 7378 from Research Biochemicals Inc. (Natick, MA); and [³H]prazosin (specific activity, 80 Ci/mmol) from New England Nuclear (Dreieich, Germany). The following drugs were gifts of the respective companies: tamsulosin HCl [(-)-isomer, formerly known as YM 617; Yamanouchi Pharmaceutical Co., Tokyo, Japan], SB 216469 (formerly known as Rec 15/2739; Recordati, Milan, Italy) and phentolamine HCl (Ciba-Geigy, Basel, Switzerland).

Data analysis. Data are shown as means ± S.E.M. of n experiments. Saturation binding experiments were analyzed by fitting rectangular hyperbolic functions to the experimental data. Competition binding experiments were analyzed by fitting monophasic and biphasic sigmoidal functions to the experimental data; a two-site fit was accepted only when it resulted in a significant improvement over the one-site fit as assessed by an F test. K_i values were calculated from the IC₅₀ values in the binding and functional experiments according to the equation K_i = IC₅₀/[(L/K_d)], where L is the concentration of radioligand and K_d is its affinity. All curve-fitting procedures were performed by using the InPlot program (GraphPAD Software, San Diego, CA). Statistical significance of differences was assessed by two-tailed t tests by using the InStat program (GraphPAD Software), and a P < .05 was considered significant.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

[³H]Prazosin identified specific, saturable, high-affinity binding sites in the murine cerebral cortex, cerebellum, liver, lung, kidney and heart (table 1), but no quantifiable specific binding was detected in murine spleen (data not shown). K_d values for [³H]prazosin were similar in all tissues and amounted to 54 ± 14 pM in cerebral cortex, 87 ± 15 pM in cerebellum, 78 ± 11 pM in liver, 67 ± 10 pM in lung, 73 ± 11 pM in kidney and 66 ± 12 pM in heart (n = 4-5 each).

View this table: [in this window] [in a new window]   TABLE 1 1-Adrenoceptor density in control and chloroethylclonidine-treated murine tissues

The murine alpha₁-adrenoceptors were characterized pharmacologically by competition binding experiments by using the subtype-selective agonists noradrenaline and methoxamine and the antagonists 5-methylurapidil, BMY 7378, (+)-niguldipine, SB 216469 and tamsulosin. In murine liver, all agonists and antagonists competed for [³H]prazosin binding with steep and monophasic curves of low affinity (table 2, figs. 1-7), indicating the presence of a homogeneous population of alpha_1B-adrenoceptors.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   5-Methylurapidil competition for [3H]prazosin binding to murine cerebral cortex (open squares), kidney (filled squares), lung (open circles) and liver (filled circles). Data are means of the total number of experiments indicated in tables 2, 3, 5 and 6, which also contain the quantitative analysis of the data.

In murine cerebral cortex, competition curves for the alpha_1D-selective BMY 7378 were steep, monophasic and of low affinity (table 3, fig. 2). Competition curves for 5-methylurapidil, methoxamine, (+)-niguldipine and tamsulosin were significantly better explained by a two-site model than a one-site model in most cases; the percentage of high-affinity sites for these compounds ranged between 14% and 43% (table 3, figs. 1, 3, 4 and 7). Although competition curves for noradrenaline and SB 216469 also were slightly shallow, they were not significantly better explained by a two-site than a one-site model in at least half of all experiments; the calculated affinities of both compounds were low (table 3, Figs. 5 and 6). In cerebral cortex of alpha_1B-adrenoceptor knockout mice, the competition curves for (+)-niguldipine and tamsulosin were considerably steeper than in control mice; they were no longer significantly better explained by a two-site model in most cases, and the calculated affinity of both compounds was high [log K_i (+)-niguldipine, 9.40 ± 0.18 and tamsulosin, 10.29 ± 0.27; n = 4 each; fig. 8).

View larger version (14K): [in this window] [in a new window]   Fig. 2.   BMY 7378 competition for [3H]prazosin binding to murine cerebral cortex (open squares), kidney (filled squares), lung (open circles) and liver (filled circles). Data are means of the total number of experiments indicated in tables 2, 3, 5 and 6, which also contain the quantitative analysis of the data.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Methoxamine competition for [3H]prazosin binding to murine cerebral cortex (open squares), cerebellum (filled triangles), kidney (filled squares), lung (open circles) and liver (filled circles). Data are means of the total number of experiments indicated in tables 2 through 6, which also contain the quantitative analysis of the data.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   (+)-Niguldipine competition for [3H]prazosin binding to murine cerebral cortex (open squares), cerebellum (filled triangles), kidney (filled squares), lung (open circles) and liver (filled circles). Data are means of the total number of experiments indicated in tables 2 through 6, which also contain the quantitative analysis of the data.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Noradrenaline competition for [3H]prazosin binding to murine cerebral cortex (open squares), cerebellum (filled triangles), kidney (filled squares), lung (open circles) and liver (filled circles). Data are means of the total number of experiments indicated in tables 2 through 6, which also contain the quantitative analysis of the data.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   SB 216469 competition for [3H]prazosin binding to murine cerebral cortex (open squares), cerebellum (filled triangles), kidney (filled squares), lung (open circles) and liver (filled circles). Data are means of the total number of experiments indicated in tables 2 through 6, which also contain the quantitative analysis of the data.

In murine cerebellum, competition curves for methoxamine, (+)-niguldipine, noradrenaline and SB 216469 were not significantly better explained by a two-site than a one-site model in at least half of all experiments, and the calculated affinities were low (table 4, figs. 3-6). In contrast, the competition curves for tamsulosin were more shallow and significantly better explained by a two-site model in all cases (table 4, fig. 7).

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Tamsulosin competition for [3H]prazosin binding to murine cerebral cortex (open squares), cerebellum (filled triangles), kidney (filled squares), lung (open circles) and liver (filled circles). Data are means of the total number of experiments indicated in tables 2 through 6, which also contain the quantitative analysis of the data.

In murine kidney, BMY 7378 and noradrenaline had steep and monophasic competition curves in almost all cases, and the calculated affinities were low (table 5, figs. 2 and 5). In contrast, 5-methylurapidil, methoxamine, (+)-niguldipine and SB 216469 had shallow and biphasic competition curves; the percentage of high-affinity sites for all compounds was similar and amounted to 48% to 55% (table 5, figs. 1, 3, 4 and 6). Although the competition curves for tamsulosin also were shallow, they were significantly better fitted to a two-site than a one-site model in only two of four experiments (table 5, fig. 7).

In murine lung, BMY 7378, methoxamine, (+)-niguldipine, noradrenaline and SB 216469 had steep and monophasic competition curves in the majority of cases; the calculated affinities for all compounds was low (table 6, figs. 2-6). In contrast, 5-methylurapidil and tamsulosin had shallow and biphasic competition curves, which detected 54% to 59% high-affinity sites (table 6, figs. 1 and 7).

Treatment with chloroethylclonidine (10 µM, 30 min, 37°C) significantly reduced the density of detectable alpha₁-adrenoceptors in all tissues (table 1). Thus, almost complete inactivation was observed in the liver and kidney, an ~90% inactivation in the cerebral cortex and heart and an ~75% to 80% inactivation in the cerebellum and lung. Although K_d values could not be reliably calculated in experiments with very extensive inactivation and/or small control receptor densities, no major change of apparent [³H]prazosin K_d values was seen in the remaining experiments (data not shown).

To resolve the discrepancy between alpha_1B-adrenoceptor estimates in the competition binding studies and the chloroethylclonidine experiments, the concentration-dependent inactivation of murine tissue alpha₁-adrenoceptor subtypes by chloroethylclonidine was investigated. For these experiments, liver tissue from control mice served as a source of alpha_1B-adrenoceptors, whereas cerebral cortex from alpha_1B-adrenoceptor knockout mice served as a source of alpha_1A-adrenoceptors. Chloroethylclonidine (0.1-10 µM) concentration-dependently activated alpha₁-adrenoceptors in both preparations (fig. 9). Although the percentage of inactivation was significantly smaller for alpha_1A- than for alpha_1B-adrenoceptors at each tested concentration of chloroethylclonidine, the differences were only small, and 10 µM chloroethylclonidine inactivated ~90% of alpha_1A-adrenoceptors.

View larger version (19K): [in this window] [in a new window]   Fig. 8.   (+)-Niguldipine (filled squares) and tamsulosin (open circles) competition for [3H]prazosin binding to murine cerebral cortex of alpha1B-adrenoceptor knockout mice. Data are means ± S.E.M. of four experiments.

View larger version (14K): [in this window] [in a new window]   Fig. 9.   Concentration-dependent inactivation of murine alpha1A- and alpha1B-adrenoceptors by chloroethylclonidine. Membrane preparations from murine liver from normal mice were used as a homogeneous source of alpha1B-adrenoceptors (filled circles) and from cerebral cortex from alpha1B-adrenoceptor knockout mice as a source of alpha1A-adrenoceptors (open squares). They were incubated in the absence (control) or presence of the indicated concentrations of chloroethylclonidine for 30 min at 37°C, followed by two washout centrifugations. Thereafter, alpha1-adrenoceptor density was determined by [3H]prazosin saturation binding experiments. Data are mean ± S.E.M. of four experiments. Alpha1-adrenoceptor density under control conditions in liver and cerebral cortex was 92 ± 17 and 59 ± 4 fmol/mg protein, respectively. * and **, P < .05 and .01, respectively, in a two-tailed t test compared with liver.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The tissue- and cell type-specific distribution of alpha₁-adrenoceptor subtypes can be studied at the mRNA, protein and functional levels. In the mouse, studies at the mRNA level have been performed by reverse transcription polymerase chain reaction (Alonso-Llamazares et al., 1995; Cavalli et al., 1997) and Northern blotting (Garcia-Sainz et al., 1994). They have detected alpha_1A-adrenoceptor mRNA in the heart, lung, liver, spleen, kidney, aorta, adipose tissues and several brain regions, including the cortex and cerebellum. Alpha_1B-adrenoceptor mRNA was also detected in all of these tissues, although at somewhat lower abundance in the spleen and adipose tissue. Alpha_1D-adrenoceptor mRNA was also detected in all of these tissues, although its presence in liver was seen in one (Alonso-Llamazares et al., 1995) but not another (Cavalli et al., 1997) study. Our data demonstrate that alpha₁-adrenoceptors are detectable at the protein level in murine tissues by [³H]prazosin binding, with the rank order of cerebral cortex > cerebellum > liver > lung > kidney > heart > spleen, with the spleen not exhibiting detectable expression.

Further experiments were designed to investigate possible qualitative species heterogeneity, i.e., to define the subtypes of alpha₁-adrenoceptors in the various murine tissues by using competition binding with various subtype-selective agonists and antagonists and inactivation by chloroethylclonidine. Our results demonstrate that murine liver expresses a homogeneous population of alpha_1B-adrenoceptors. Similar results have previously been obtained by other investigators (Garcia-Sainz et al., 1994). Although the situation in the other tissues was somewhat more complex, the competition binding data with most subtype-selective drugs indicate that murine cerebellum and lung also express predominantly alpha_1B-adrenoceptors, whereas murine cerebral cortex and kidney express mixed alpha_1A- and alpha_1B-adrenoceptors in approximate ratios of 30% to 70% and of 50% to 50%, respectively.

Detection of tissue alpha_1D-adrenoceptors at the protein level has long been hampered by the lack of selective antagonists. Recently, BMY 7378 has been introduced as an antagonist that is ~100-fold selective for alpha_1D- over alpha_1A- and alpha_1B-adrenoceptors (Goetz et al., 1995). In our study, BMY 7378 competition curves in murine liver, cerebral cortex, kidney and lung were steep and monophasic, with affinity estimates 100 times lower than those reported for cloned rat alpha_1D-adrenoceptors (Goetz et al., 1995; Yang et al., 1997b). Although a low affinity of BMY 7378 for murine alpha_1D-adrenoceptors cannot be excluded, these data indicate that the alpha_1D-adrenoceptor protein is absent in the respective murine tissues despite the presence of corresponding mRNA (Alonso-Llamazares et al., 1995; Cavalli et al., 1997). Although alpha_1D-adrenoceptors have been detected at the protein level in rat aorta (Deng et al., 1996), our present data are similar to those found in several rat tissues (Deng et al., 1996; Yang et al., 1997b) and the human prostate (Michel et al., 1996), where no alpha_1D-adrenoceptor protein was detected in radioligand binding studies despite the presence of corresponding mRNA. The consistency of this finding across three different species indicates that alpha_1D-adrenoceptor mRNA may not be efficiently translated into a functional protein and/or that the functional protein is rapidly degraded. Studies with cloned human alpha_1D-adrenoceptors in our laboratory have indicated that agonist exposure causes less alpha_1D-adrenoceptor down-regulation, if any, compared with alpha_1A- or alpha_1B-adrenoceptors (Yang et al., 1997a). This makes inefficient translation a more likely explanation of low protein abundance than does rapid degradation of alpha_1D-adrenoceptors. Further investigations into this question were beyond the scope of the present paper. The remaining part of the discussion will therefore focus on the alpha_1A- and alpha_1B-adrenoceptor distribution at the protein level in murine tissues.

The present study has obtained consistent affinity estimates for all compounds across all investigated tissues. These estimates are in good agreement with those we have previously reported after using identical methods for the human prostate (Michel et al., 1996), bovine cerebral cortex (Büscher et al., 1996), a variety of rat (Büscher et al., 1996; Michel et al., 1993a) and guinea pig (Büscher et al., 1996) tissues and for cloned rat and bovine alpha₁-adrenoceptors (Michel and Insel, 1994; Chess-Williams et al., 1996; Michel et al., 1996). They are also in good agreement with those previously reported for murine hepatic alpha_1B-adrenoceptors by other investigators (Garcia-Sainz et al., 1994). Taking into consideration the large interlaboratory variation in reported drug affinities of alpha₁-adrenoceptor subtypes, they are also in the mainstream of affinities reported by other investigators for cloned subtypes (Michel et al., 1995). Thus, at least for the presently investigated drugs, murine alpha₁-adrenoceptors appear to have a qualitatively similar pharmacological profile as those in rats, guinea pigs, cows and humans.

The utility of mice as model systems depends not only on the similarity of pharmacological receptor profiles but also on the quantitative and qualitative receptor subtype expression in comparison to other species. Our data indicate that the interspecies variation in tissue-specific alpha₁-adrenoceptor subtype expression appears to be considerably greater than that of drug affinities at a given subtype. For example, on a quantitative level, we were unable to detect quantifiable alpha₁-adrenoceptor binding in murine spleen, whereas splenic expression of alpha₁-adrenoceptors (mostly of the alpha_1B-subtype) is well detected in rats (Michel et al., 1993a) and guinea pigs (Büscher et al., 1996). On a qualitative level, the murine liver exhibited a homogeneous population of alpha_1B-adrenoceptors in the present and a previous study (Garcia-Sainz et al., 1994). Although homogeneous populations of alpha_1B-adrenoceptors are also found in rat (Garcia-Sainz et al., 1994; Büscher et al., 1996) and hamster (Garcia-Sainz et al., 1994) liver, homogeneous populations of alpha_1A-adrenoceptors were reported for human (Garcia-Sainz et al., 1995) and guinea pig (Garcia-Sainz and Romer-Avila, 1993) liver and mixed alpha_1A/alpha_1B-adrenoceptor populations for rabbit liver (Torres-Marquez et al., 1991; Garcia-Sainz et al., 1992; Taddei et al., 1993). In the cerebral cortex, the contribution of alpha_1A-adrenoceptors in mice (vide infra) appears to be similar to that in guinea pigs (Büscher et al., 1996), lower than that in rats (Han and Minneman, 1991; Michel et al., 1993a) and humans (Gross et al., 1989), and much lower than that in cows (Büscher et al., 1996). In the kidney, similar densities of alpha_1A- and alpha_1B-adrenoceptors were seen in mice (vide infra) and rats (Han et al., 1990; Michel et al., 1993a), but alpha_1B-adrenoceptors were predominantly reported in guinea pigs (Büscher et al., 1996). Thus, considerable interspecies heterogeneity exists with regard to the quantitative and qualitative expression of alpha₁-adrenoceptor subtypes in several tissues. Therefore, the identification of a given alpha₁-adrenoceptor subtype as a mediator of a specific functional response in mice may not always correctly predict which subtype mediates this response in other species. This caveat may also limit the utility of knockout mice as model systems.

In some murine tissues, tamsulosin behaved differently from other alpha_1A-selective drugs. Specifically, tamsulosin differentiated two types of sites in the murine lung and cerebellum that were not discriminated by a variety of other drugs, e.g., (+)-niguldipine, despite their greater subtype selectivity. We have previously reported a similar aberrant behavior of tamsulosin in guinea pigs (Büscher et al., 1996). Thus, in guinea pig kidney, a panel of seven subtype-selective drugs indicated a homogeneous population of alpha_1B-adrenoceptors, whereas tamsulosin detected 39% high-affinity sites. In the guinea pig cerebral cortex, these seven subtype-selective drugs detected alpha_1A- and alpha_1B-adrenoceptors in approximately a 25% to 75% ratio, whereas tamsulosin detected a 51% to 49% ratio. This raises the possibility that the high- and low-affinity sites for tamsulosin in some murine and guinea pig tissues may not correspond to alpha_1A- and alpha_1B-adrenoceptors, respectively. However, it should be noted that noradrenaline behaved similarly to tamsulosin in the guinea pig tissues (Büscher et al., 1996), and 5-methylurapidil behaved similarly to tamsulosin in murine lung (vide infra). To clarify the nature of tamsulosin low-affinity sites in some murine tissues, we have performed competition binding experiments for tamsulosin and, as a reference compound, for (+)-niguldipine in the cerebral cortex of alpha_1B-adrenoceptor knockout mice. According to our previous studies, this tissue is a homogeneous source of murine alpha_1A-adrenoceptors (Cavalli et al., 1997). Because tamsulosin and (+)-niguldipine detected only high-affinity sites in the cerebral cortex of alpha_1B-adrenoceptor knockout mice, these data suggest that high- and low-affinity sites for tamsulosin indeed correspond to alpha_1A- and alpha_1B-adrenoceptors, respectively. The reason for quantitatively different ratios of the two subtypes as determined with tamsulosin versus other alpha_1A-selective drugs is not fully clear, but it should be noted that [³H]tamsulosin may label different numbers of alpha₁-adrenoceptors than [³H]prazosin in some tissues and even with cloned subtypes (Yazawa et al., 1992; Michel and Goepel, 1998). Thus, it is possible that alpha₁-adrenoceptor ligands from distinct chemical classes label alpha₁-adrenoceptor subtypes in a quantitatively different manner. Although elucidation of the reasons for such differences was beyond the scope of this paper, these data indicate that percentages of high- and low-affinity sites for a single alpha_1A-selective compound should be interpreted only with great care with regard to the quantitative presence of these subtypes.

The alkylating agent chloroethylclonidine has historically been useful in establishing alpha₁-adrenoceptor heterogeneity (Han et al., 1987). In some studies on rat tissues, the proportion of chloroethylclonidine-sensitive and -insensitive sites has corresponded well with the proportion of alpha_1B- and alpha_1A-adrenoceptors, respectively, as determined by competition binding with subtype-selective antagonists (Minneman et al., 1988; Wilson and Minneman, 1989; Feng et al., 1991). In contrast, chloroethylclonidine readily alkylates cloned alpha_1A-adrenoceptors (Han et al., 1995; Schwinn et al., 1995; Büscher et al., 1996; Hirasawa et al., 1997), and this has contributed to the initial failure to correctly classify the cloned bovine alpha_1A-adrenoceptor (Schwinn et al., 1990). Meanwhile, it has become clear that the ability of chloroethylclonidine to alkylate alpha₁-adrenoceptor subtypes depends on the incubation temperature and duration and on the osmolarity of the buffer system. When all of these conditions are kept constant, some subtype selectivity can indeed be seen in the alkylating effects of chloroethylclonidine (Schwinn et al., 1995). We have previously demonstrated that incubation with 10 µM chloroethylclonidine for 30 min at 37°C in a low-osmolarity buffer is necessary and sufficient for full inactivation of rat tissue alpha_1B-adrenoceptor but has little effect on rat tissue alpha_1A-adrenoceptors (Michel et al., 1993b). Under these conditions, chloroethylclonidine powerfully inactivated alpha₁-adrenoceptors in all murine tissues of the present study. In this respect, no major differences were seen between tissues with considerable alpha_1A-adrenoceptor fractions (cerebral cortex and kidney) and those without (liver). This is in agreement with some studies on cloned alpha₁-adrenoceptor subtypes (Hirasawa et al., 1997) but in clear contrast to the aforementioned studies on rat tissue alpha₁-adrenoceptor subtypes. Therefore, we have investigated the concentration-dependent inactivation of murine tissue alpha_1B-adrenoceptors (liver) and alpha_1A-adrenoceptors (cerebral cortex from alpha_1B-adrenoceptor knockout mice; Cavalli et al., 1997). Although these experiments demonstrated statistically significant discrimination of subtypes by chloroethylclonidine, the difference was only small, and tissue alpha_1A-adrenoceptors were readily inactivated by 10 µM chloroethylclonidine. These observations are similar to those made under identical conditions in guinea pigs (Büscher et al., 1996). Thus, our study highlights potential pitfalls in the use of chloroethylclonidine for the identification of alpha₁-adrenoceptor subtypes and gives further evidence for a very cautious interpretation of data obtained with chloroethylclonidine.

In summary, the present study has detected and characterized alpha₁-adrenoceptor subtypes at the protein level in a variety of murine tissues. The overall pharmacological profile of murine alpha₁-adrenoceptor subtypes appears to be similar to that in other species. However, the quantitative and qualitative alpha₁-adrenoceptor subtype expression in a given tissue appears to differ considerably between species. Therefore, the utility of mice for the identification of alpha₁-adrenoceptor subtypes mediating a given response may be limited.