Pharmacological Characterization and Cross Talk of α_1A- and α_1B-Adrenoceptors Coexpressed in Human Embryonic Kidney 293 Cells

The α₁-adrenoceptors (α₁-ARs) belong to the large super-family of G protein-coupled receptors that mediate a variety of intracellular signals. So far, three distinct α₁-AR subtypes have been found, designated as α_1A-, α_1B-, and α_1D-ARs (Cotecchia et al., 1988; Schwinn et al., 1990; Lomasney et al., 1991). Each subtype shows distinct pharmacological characteristics and couples mainly via G_q/11 to the activation of phospholipase C and mitogen-activated protein kinase (MAPK) with different efficiencies (Zhong and Minneman, 1999; Koshimizu et al., 2002; Toews et al., 2003).

The α₁-AR subtypes are widely distributed in a large number of tissues and cell types, with a large degree of overlap in their anatomical localization. For example, α_1A- and α_1B-ARs are found in different proportions to be coexpressed in brain, heart, kidney, and artery of various mammalian species (Price et al., 1994). In the rat myocardium, α_1A- and α_1B-ARs coexist in a 20 to 30%:70 to 80% ratio and mediate divergent effects on cardiac inotropy, rhythmicity, and cell growth (Michel et al., 1994a; Zhang et al., 2002). In rat tail artery, α_1A- and α_1B-ARs occur in a ratio of approximately 60%:40%, mediating adrenergic contractions (Tanaka et al., 2004). The distribution of the α_1D-AR differs from α_1A- and α_1B-ARs.

Recently, G protein-coupled receptors have been reported to exist as homomeric and/or heteromeric oligomers (Devi, 2001; Gomes et al., 2001; Jordan et al., 2001; Angers et al., 2002; Lavoie et al., 2002). Hetero-oligomers of GPCR may in principle exhibit either distinct pharmacological characteristics (Jordan and Devi, 1999) or distinct functions (George et al., 2000; Scarselli et al., 2001). Recombinant α₁-ARs have also been shown to homo- or hetero-oligomerize when coexpressed in HEK 293 cells (Vicentic et al., 2002; Stanasila et al., 2003; Uberti et al., 2003), but the functional characteristics of these multimers or importance of coexistence has not been defined. In the present study, we have expressed human α_1A- and α_1B-ARs, either alone or in combination in HEK 293 cells, to investigate their possible interaction in terms of their binding and signaling properties.

Materials and Methods

Construction of Human α_1a- and α_1b-AR Expression Vectors. Human α_1a- and α_1b-ARs were encoded in expression vector pCR3 (pCR3-α_1a and pCR3-α_1b), as described previously (Taniguchi et al., 1999). α_1a-AR (1.5-kilobase EcoRI-EcoRI fragment of pCR3-α_1a plasmid) was also inserted at the EcoRI-EcoRI sites of the mammalian expression vector pIRES-puro2, and the resulting pIRES-puro2-α_1a plasmid was used to coexpress α_1A-AR with α_1B-AR.

Expression in HEK 293 Cells. HEK 293 cells were maintained as monolayer cultures at 37°C in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 2 mM l-glutamine, 100 U/ml penicillin/streptomycin in a humidified atmosphere of 5% CO₂, 95% O₂. Cells were seeded at 2 × 10⁶ cells/100-mm dish and after 24 h transfection was performed using LipofectAMINE (Invitrogen, Carlsbad, CA). The pCR3-α_1a and pCR3-α_1b plasmids were used for generation of HEK 293 cells stably expressing α_1A- or α_1B-ARs. Cell lines stably expressing each receptor were selected in medium containing 1000 μg/ml geneticin. Equal amount of the pIRES-puro2-α_1a and pCR3-α_1b plasmids were used for generation of α_1A,1B-AR cells stably coexpressing α_1A- and α_1B-ARs (4 μg of each plasmid/100-mm dish). Cultures were maintained in the selection media containing 10 μg/ml puromycin at first and then 500 μg/ml geneticin additionally. Finally, two cell lines coexpressing α_1A- and α_1B-ARs (termed AB1 and AB2) were obtained; the total densities of α₁-ARs were 5860 ± 790 and 7800 ± 560 fmol/mg protein, respectively. The proportions of α_1A- and α_1B-ARs (and their densities) were 24%:76% (1410:4450 fmol/mg protein) in AB1 cell line and 11%:89% (860:6900 fmol/mg protein) in AB2 cell line. Cell lines expressing the same amount of each α₁-AR subtype as AB1 cell line were also chosen: 1600 ± 520 fmol/mg protein in α_1A-AR cell and 4650 ± 690 fmol/mg protein in α_1B-AR cell (see Results; Table 1).

Whole Cell Binding. HEK 293 cells expressing α_1A-, α_1B-, and the both ARs were washed twice, harvested without the use of trypsin by gentle pipetting, and transferred to the glass tubes. Intact cell binding assays were performed with [³H]prazosin in Krebs-HEPES buffer (110 mM NaCl, 4.5 mM KCl, 1.3 mM CaCl_2, 1.2 mM KH₂PO_4, 1.2 mM MgSO_4, 25 mM NaHCO_3, 11.7 mM glucose, 5 mM HEPES, pH 7.4) in a final volume 1 ml for 4 h at 4°C or for 30 min at 37°C. Cell number in each incubation tube was adjusted to less than 10⁵ cells/ml to avoid radioligand depletion. Assays were performed in duplicate, and nonspecific binding was defined as the amount of radioligand bound in the presence of 10 μM phentolamine. [³H]Prazosin between 30 and 2000 pM was used in saturation binding experiments. Binding competition experiments were done in the presence of 200 pM [³H]prazosin with the addition of increasing concentrations of the unlabeled drugs. Reactions were terminated by rapid filtration with a cell harvester (Brandel Inc., Gaithersburg, MD) onto Whatman GF/C glass filter presoaked in 0.3% polyethyleneimine for 30 min. The filters were then washed three times and dried before the measurement of filter-bound radioactivity by liquid scintillation counting. Protein concentrations were quantified by the method of Bradford using bovine serum albumin as standard (Bradford, 1976).

Inositol Phosphate Determination. HEK 293 cells stably expressing α_1A-, α_1B-, and coexpressing both AR subtypes were seeded in collagen-coated 24-well plates (∼1 × 10⁵ cells/well) 24 h before each experiment. As reported previously (Israilova et al., 2002), cells were then washed twice with inositol-free Dulbecco's modified Eagle's medium without serum (fetal bovine serum) and labeled for 24 h with [³H]myoinositol at 5 μCi/ml. After labeling, cells were washed twice with Krebs-HEPES buffer containing 10 mM LiCl to remove unincorporated radioactivity. Then, cells were preincubated for 20 min in 1 ml of Krebs-HEPES buffer with 10 mM LiCl in the presence or absence of antagonists and were challenged by various concentrations of phenylephrine or methoxamine for 45 min. These procedures were done at 37°C. The reactions were terminated by addition of ice-cold 40% perchloric acid and cooling for 20 min at 4°C. The samples were neutralized with 1.6 mol/l KOH/100 mM Tris, and the resulting extracts were centrifuged. The resulting supernatant was applied to 1 ml of AG1-X8 (100-200 mesh, chloride form; Bio-Rad, Hercules, CA). Total inositol phosphates were eluted with 1 ml of 1 mol/l HCl and were quantified in a liquid scintillation counter. Single concentrations of the antagonist at concentrations 30 to 100 times higher than the binding affinity (pK_i) were used to detect any shift in the concentration-response curve for phenylephrine or methoxamine. Competition experiments in the presence of antagonist were carried out in parallel with control experiment in the absence of antagonist and the maximum IP accumulation in control experiment was taken as 100%. The pK_b values were calculated as described by Furchgott (1972).

MAPK Assay. Cells were serum-starved for 6 h and treated for 5 min with the α₁-ARs agonists phenylephrine or methoxamine at 37°C. After stimulation, medium was aspirated and monolayers were lysed in 2× Laemmli sample buffer, resolved by 12.5% SDS-polyacrylamide gel electrophoresis, and transferred to polyvinylidene difluoride membrane. The activation of the p42/44 ERK was determined by immunoblotting using 1:1000 dilution of rabbit polyclonal phosphospecific antibodies followed by horseradish peroxidase-conjugated anti-rabbit IgG (Cell Signaling Technology Inc., Beverly, MA). Identical polyvinylidene difluoride membrane was immunoblotted in parallel for measuring the total amount of ERK1/2 (Santa Cruz Biotechnology, Inc., San Diego, CA) transferred to the membrane. Band intensities were determined by densitometry and were expressed as optical density units. Each sample value was normalized by dividing the phospho-ERK1/2 density by the total ERK1/2 density. For each experiment, ERK1/2 stimulation over basal was calculated by dividing the treated normalized optical density by the basal normalized optical density.

Data Analysis. Data were analyzed using commercially available software (GraphPad Prism, version 3.00; GraphPad Software Inc., San Diego, CA). Binding data were first fitted to a one- and then two-site model, and if the residual sums of squares were statistically less for a two-site fit of the data than for a one-site, as determined by an F test comparison, and then the two-site model was accepted. Nonlinear regression analyses were applied to sigmoid concentration-response curves of IP and ERK responses. pK_b value for antagonist was determined for a single concentration of antagonist by the concentration-ratio method (Furchgott, 1972). p values less than 0.05 were considered significant (Student's t test). Data are represented as mean ± S.E.M.

Drugs. The drugs used and their sources were as follows: [³H]prazosin from PerkinElmer Life Sciences, Boston, MA); (-)-1(3-hydroxypropyl)-5-((2R)-2-{[2-(2,2,2-trifluoroethyl]oxy]phenyl}oxy)ethyl]amino}propyl)-2,3-dihydro-1H-indole-7-carboxamide (KMD-3213) was from Kissei Pharmaceutical Co., Ltd. (Matsumoto, Japan); phenylephrine hydrochloride, prazosin hydrochloride, phentolamine, and methoxamine hydrochloride were from Sigma-Aldrich (St. Louis, MO). [³H]Prazosin was diluted in assay buffer. The stock solutions of prazosin and KMD-3213 were prepared with 50% ethanol and dimethyl sulfoxide, respectively, and then diluted with assay buffer in functional and binding experiments before use.

Results

Comparison of Binding Affinities in HEK 293 Cells Stably Expressing α_1A-, α_1B-AR Alone or in Combination. To establish α_1A,1B cell line coexpressing both α_1A- and α_1B-ARs, HEK 293 cells were simultaneously transfected with two different plasmids encoding the α_1a-or α_1b-ARs, and finally the two α_1A,1B cell lines (termed AB1 and AB2) stably expressing both α_1A- and α_1B-ARs were obtained. The total density of α₁-ARs in both cell lines was 5860 ± 790 and 7800 ± 560 fmol/mg protein, respectively, which was evaluated by [³H]prazosin binding on suspension of intact cells at 4°C. To subdivide α_1A/α_1B subpopulation in AB1 and AB2 cell lines, competition of specific [³H]prazosin binding from intact cells by the highly α_1A-selective antagonist KMD-3213 (Murata et al., 2000) was assessed. Competition analysis in the AB1 and AB2 cell lines demonstrated KMD-3213 to be highly discriminating (approximately 450-fold) between α_1A- and α_1B-ARs. The proportion of α_1A- and α_1B-ARs (and their densities) were 24%:76% (1410:4450 fmol/mg protein) in AB1 cell line and 11%:89% (860:6900 fmol/mg protein) in AB2 cell line. The AB1 cell line was used primarily in the present study, because a comparable expression of each AR subtype was obtained in the cell lines stably expressing the α_1A-AR and α_1B-AR, respectively (Table 1).

Figure 1 shows the representative [³H]prazosin binding competition curves for KMD-3213 and methoxamine in the three cell lines at 4°C. The results of these binding studies at 4°C and 37°C are summarized in Table 1. KMD-3213 and methoxamine clearly discriminated between α_1A- and α_1B-ARs with their distinct affinities, which were consistent with the values measured in α_1A,1B-AR cells. On the other hand, the binding affinities for prazosin or phenylephrine did not clearly discriminate between the three cell lines (Table 1).

Comparison of IP Responses in HEK 293 Cells Stably Expressing α_1A-, α_1B-AR alone or in Combination. Basal IP accumulation in the absence of agonist during the 45-min incubation was 600 ± 70 and 670 ± 140 dpm/10⁵ cells in α_1A-and α_1A,1B-ARs, respectively, values that were significantly higher than that (400 ± 40 dpm/10⁵ cells) measured in α_1B-AR cells (p < 0.05). Phenylephrine produced a concentration-dependent increase in IP accumulation in the three cell lines (Fig. 2A). The maximum accumulation varied greatly among the three cell lines: the maximal accumulation in α_1A,1B-AR cells was approximately 160% of that in α_1A-AR cells, whereas the accumulation in α_1B-AR cells was only 18% of that in α_1A-AR cells (Table 2). The pEC₅₀ value (7.5 ± 0.2) for α_1A,1B-AR was significantly higher than that for α_1A- and α_1B-ARs (6.8 ± 0.1 and 6.6 ± 0.2, respectively). Coculture of 10⁵ α_1A-AR cells and 10⁵ α_1B-AR cells produced essentially the same concentration-response curve for phenylephrine as that in α_1A-AR cells alone (Fig. 2A; Table 2). These results suggested that the greater response and higher potency in α_1A,1B-AR cells could not be explained by a simple summation of the responses produced separately by the α_1A- and α_1B-AR cells. The enhanced responses to phenylephrine were also observed in another cell line (AB2), which had expressed α_1A-and α_1B-AR as a ratio of 11%:89% (data not shown).

Methoxamine also produced concentration-dependent increases in IP accumulation in α_1A- and α_1A,1B-AR cells. The concentration-response curve in α_1A,1B-AR cells completely overlapped that in α_1A-AR cells. The maximum responses to methoxamine were comparable with that induced by phenylephrine in α_1A-AR cells (Fig. 2B; Table 2). In contrast, the response in α_1B-AR cells was less than 10% of those in α_1A-and α_1A,1B-AR cells, and the potency was very low (pEC₅₀ of 4.8 ± 0.2).

Prazosin at 10 nM antagonized the effects of phenylephrine and methoxamine without changing the basal IP accumulation, resulting in a rightward shift of the concentration-response curves (Figs. 3 and 4). The pK_b values for prazosin estimated in response to phenylephrine were 9.3 ± 0.3 for α_1A-AR and 11.0 ± 0.3 for α_1B-AR. An intermediate value (10.1 ± 0.5) for α_1A,1B-AR was observed. In contrast, the pK_b values in the response to methoxamine were 9.2 ± 0.1 and 9.2 ± 0.2 for α_1A- and α_1A,1B-ARs, respectively, values that were consistent with the value estimated for the response to phenylephrine in α_1A-AR cells (Table 3). The response to methoxamine in α_1B-AR cells was strongly inhibited by 10 nM prazosin (Fig. 4B).

KMD-3213 (10 nM) also shifted the concentration-response curves for phenylephrine and methoxamine to the right without affecting the basal IP level. However, the shift was clearly evident in α_1A-AR cells and minimal in α_1B-AR cells (Figs. 3 and 4). Thus, the estimated pK_b values for KMD-3213 in response to phenylephrine and methoxamine were high (10.5 or 10.4) for α_1A-AR and low (8.1) for α_1B-AR, values that were consistent with the binding affinities (Tables 1 and 3). The pK_b value for KMD-3213 in α_1A,1B-AR cells was high (10.8 ± 0.2) in response to methoxamine, whereas the value in response to phenylephrine was low (9.4 ± 0.3). The pK_b value in response to phenylephrine was intermediate between the values (10.5 ± 0.2 and 8.1 ± 0.2) estimated in α_1A-and α_1B-AR cells, respectively.

Comparison of ERK Activities in HEK 293 Cells Stably Expressing α_1A-, α_1B-AR Alone or in Combination. Because the IP response to phenylephrine was greatly enhanced by coexpression of α_1A- and α_1B-ARs, the MAPK pathway was also examined. Phenylephrine stimulation of α_1A-, α_1B-, and α_1A,1B-AR cells resulted in rapid and robust increases in ERK1/2 activities that peaked at 5 min and had returned to close to basal levels within 30 min (data not shown). The stimulatory effect of phenylephrine was concentration-dependent (assessed at 5 min) (Fig. 5A), generating similar pEC₅₀ values and maximum activations for α_1A- and α_1B-AR cells (Table 4). On the other hand, two important differences were observed in the α_1A,1B-AR cell line. First, in keeping with the stimulation of IP production, the magnitude of the ERK1/2 response (relative to basal values) was greater by about 2-fold, compared with either the α_1A-or α_1B-AR cells alone. Second, the concentration-effect curve for ERK1/2 activation by phenylephrine was approximately 5-fold left-shifted (pEC₅₀ of 7.6 ± 0.2; p < 0.05) compared with those of α_1A- and α_1B-AR cells (pEC₅₀ of 6.8 ± 0.1 and 6.7 ± 0.2; Table 4). Methoxamine also stimulated ERK1/2 activation in all cell lines tested. Compared with the response to phenylephrine, the magnitude of maximum activation and pEC₅₀ value were comparable with those in α_1A-AR cells but were significantly lower in α_1B-AR cell line. Both parameters in α_1A,^1B-AR cells were consistent with those in α_1A-AR cells, suggesting that neither agonist sensitivity nor responsiveness was increased by coexpression in the case of ERK1/2 activation in response to methoxamine (Table 4; Fig. 5B).

Discussion

Recombinant α_1A- and α_1B-ARs subtypes were recently shown to homo- and hetero-oligomerize when coexpressed in HEK 293 cells (Vicentic et al., 2002; Stanasila et al., 2003; Uberti et al., 2003), implying that there may be a functional consequence of this interaction. However, despite the strong evidence for a physical interaction between α_1A- and α_1B-ARs in individual cells, functional consequences of the coexpression were not evaluated in detail. In the present study, we have coexpressed α_1A- and α_1B-ARs in HEK 293 cells and investigated whether coexpression produces pharmacological profiles that may differ from the parameters for the same receptors when expressed alone. The α_1A,1B-AR cell lines used in this study, which demonstrated a coexpression of these receptor populations, both exhibited α_1A-AR < α_1B-AR densities. This kind of subpopulation ratio has been seen in many tissues (Minneman et al., 1988; Lazou et al., 1994; Michel et al., 1994a; Zhang et al., 2002), but a distinct opposite population ratio has been observed in other tissues (Zhang et al., 2002; Tanaka et al., 2004). At first, we examined the α_1A,1B-AR cell line by a ligand binding approach and compared its properties with those of cells expressing α_1A- or α_1B-AR alone. Whole cell binding experiments were done at 4°C and 37°C because the binding affinities were compared with the functional affinities obtained at 37°C. As shown in Table 1, prazosin showed a high affinity for α₁-ARs in all cell lines tested and did not discriminate α_1A- and α_1B-ARs in α_1A,^1B-AR cells. The same results were obtained for the affinities of phenylephrine. In contrast, KMD-3213 and methoxamine showed clearly different affinities for α_1A- and α_1B-ARs, in agreement with previous reports (Murata et al., 2000; Zhang et al., 2002) and could discriminate individually the α_1A- and α_1B-subtypes in α_1A,1B-AR cells with the same affinities as those in the cells individually expressing each receptor subtype. These results show that the binding profiles of each subtype expressed in α_1A,1B-AR cells are not modified by coexpression. The present binding study furthermore suggests that methoxamine may be used as a selective agonist of α_1A-AR in cells coexpressing both receptors, whereas, as expected, phenylephrine can costimulate both α_1A- and α_1B-ARs in α_1A,1B-AR cells without any discrimination.

Next, we examined the functional profiles of the coexpressed receptors in α_1A,1B-AR cells. As suggested above, methoxamine predominantly stimulated the IP response and ERK1/2 activation in α_1A-AR and α_1A,1B-AR cells. This stimulation was antagonized by KMD-3213 with a high affinity for the α_1A-AR. The magnitude of the maximum responses and pEC₅₀ for methoxamine in α_1A,1B-AR cells were in accord with those in α_1A-AR cells. These results suggest that irrespective of receptor coexpression, methoxamine selectively stimulates α_1A-ARs in α_1A,1B-AR cells, similarly resulting in α_1A-AR-mediated responses. However, the phenylephrine responses in α_1A,1B-AR cells seem to be produced via a coactivation of α_1A- and α_1B-ARs, because phenylephrine did not discriminate between α_1A- and α_1B-ARs, and the responses to phenylephrine in α_1A,1B-AR cells were antagonized by prazosin and KMD-3213 with intermediate affinities between those for α_1A- and α_1B-ARs (Table 2). The latter result must be also recognized as pharmacologically important evidence that the antagonist affinity estimated in cells that coexpress the receptors can deviate from the antagonist potency observed in cells that express either receptor alone.

Another interesting finding in the present study is that the concentration dependence for the IP response and ERK1/2 activation by phenylephrine, a nonselective α₁-AR agent, were left-shifted and the maximal responses were increased in α_1A,1B-AR cells. Such increases in agonist sensitivity (supersensitivity) and responsiveness were seen only in the response to phenylephrine, but not to methoxamine. This result also supports the hypothesis that phenylephrine (but not methoxamine) coactivates the α_1A- and α_1B-ARs in the α_1A,1B-AR cell line. The enhanced responses to phenylephrine are unlikely to result from simple additivity of the individual response of each receptor subtype or an increase in receptor number, because when simply cocultured, 10⁵ α_1A-AR cells and 10⁵ α_1B-AR cells yielded neither an increase in agonist sensitivity nor responsiveness equivalent to the responses of 10⁵ α_1A,1B-AR cells.

Recently, Stanasila et al. (2003) have suggested that when the α₁-AR subtypes are expressed at comparable levels in HEK 293 cells, the trend to oligomerize is greater for homo-oligomers than for the hetero-oligomers. Thus, it is likely that homo- and hetero-oligomers both coexist in the present α_1A,1B-AR cells. One possibility to cause supersensitivity and enhancement of responsiveness may be an enhanced coactivation of hetero-oligomerized α_1A- and α_1B-ARs by a subtype nonselective agonist phenylephrine, in addition to the activation of residual homo-oligomers. In contrast to this, the results with methoxamine suggest that the α_1A-AR-selective agonist would affect the α_1A-AR only, irrespective to hetero- and homo-oligomers, failing to cause an enhanced activation.

Three distinct α₁-AR subtypes couple to G_q/11 family of G proteins and agonist binding leads to the activation of phospholipase C and stimulation of IP hydrolysis, with an increase in intracellular calcium and activations of protein kinase C and MAPK as major downstream signaling events (Zhong and Minneman, 1999; Chalothorn et al., 2002; Waldrop et al., 2002). However, the three different subtypes have been found to have different coupling efficiencies or distinct patterns of activation for different responses (Theroux et al., 1996; Zhong and Minneman, 1999; Toews et al., 2003). In general, the recombinant α_1A-AR expressed in cultured cells seemed to be more efficient than the α_1B- and α_1D-ARs in activating IP formation and causing a release of intracellular calcium. MAPK is more strongly or equally activated via α_1A-AR compared with α_1B-AR, whereas α_1D-AR is less effective. Also, n the present study stimulation of α_1A- and α_1B-ARs caused comparable activations of ERK1/2, whereas the IP response was more marked for α_1A-AR than α_1B-AR. The mechanisms for the different efficiencies between subtypes in terms of either activating MAPK or increasing IP remain to be determined. Notwithstanding, the significant increases in agonist sensitivity and responsiveness observed in the α_1A,^1B-AR cells suggest a synergism between α_1A- and α_1B-ARs and/or cross talk of signaling pathways downstream of these receptors in the same cells, even though both subtypes can couple to the same G protein.

Cross talk of α_1A- and α_1B-ARs has been reported previously in rat myocardium (Michel et al., 1994b; Deng et al., 1998). Results with double α_1a- and α_1b-AR knockout mice have indicated that both α_1a- and α_1b-ARs may be required for cardiac hypertrophy (O'Connell et al., 2003). ERK1/2 pathways are implicated in physiological hypertrophic signaling. The present study suggests that coexpression of α_1A-and α_1B-ARs may be involved in the amplification of the hypertrophic response via enhanced ERK1/2 signaling. Although receptor dimerization represents one possible mechanism for the synergy, the detailed mechanisms underlying the synergistic effects and the amplification of cellular responses remain to be determined. Recently, the same type of cross talk as observed in the present study was reported to occur at the levels of both IP and ERK1/2 regulation in the Chinese hamster ovary cells coexpressing G_i-coupled M2 and G_q/11-coupled M3 acetylcholine receptors (Hornigold et al., 2003). Cross talk between β-adrenergic and bradykinin B₂ receptors results in cooperative regulation of cyclic AMP accumulation and MAPK activation (Hanke et al., 2001).

The basal accumulation of IPs in the three cell lines used was not affected by treatment with prazosin and KMD-3213. Prazosin has been shown to be an inverse agonist of α_1A- and α_1B-ARs (Rossier et al., 1999; Zhu et al., 2000), whereas KMD-3213 is an α_1A-AR-selective neutral antagonist (Zhu et al., 2000; Zhang et al., 2002). Therefore, it is likely that α_1A-and α_1B-ARs and their combined expression in HEK 293 cells have little constitutive activity.

Except for the responses to methoxamine in α_1B-AR cells, the pEC₅₀ values for phenylephrine and methoxamine were significantly larger than their binding affinities. This result implies the presence of a significantly greater proportion of “spare receptors” for phenylephrine and methoxamine (Nickerson, 1956; Colucci et al., 1985; Taniguchi et al., 1999). Supersensitivity induced by the coexpression of α_1A- and α_1B-ARs would further increase the population of spare receptors. Thus, it is likely that adrenergic responses may be triggered with higher efficiency and with a larger receptor reserve in tissues coexpressing α_1A- and α_1B-ARs. However, it must be kept in mind that the responses observed in the present study were produced in cells overexpressing α₁-ARs. Whether such synergy occurs in cells coexpressing a lower abundance of receptors as may be found in vivo remains to be established.

In summary, the present study with HEK 293 cells coexpressing α_1A- and α_1B-ARs demonstrates that coactivation of both receptors is caused by a subtype-nonselective agonist, resulting in an enhanced activation of either the IP or ERK1/2 response, where both the sensitivity and responsiveness to stimulation by a nonselective receptor agonist are increased. The results further suggest that the response induced by coactivation of two coexpressed receptors may be inhibited by antagonists with affinities that seem to differ from the functional affinities observed for each receptor expressed on its own.