Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

-Adrenergic receptors (AR) have been among the most extensively studied members of the G protein-coupled receptor (GPCR) family. AR activation of adenylyl cyclase is one of the first hormone-activated GPCR pathways to be characterized, and these receptors are therapeutic targets for a variety of clinical conditions including hypertension, coronary artery disease, heart failure, and asthma. Three AR subtypes have been cloned (₁AR, ₂AR, and ₃AR). The ₁AR and ₂AR are pharmacologically more similar to each other than they are to the ₃AR. All three receptor subtypes are highly conserved across species, suggesting that both sequence similarities and differences between subtypes are physiologically important. The close structural and functional properties of the ₁AR and ₂AR are paradigmatic of many other GPCR families where two or more receptor subtypes respond to the same hormone and couple to the same effector systems. However, although ₁AR and ₂AR have very similar signaling properties when expressed in undifferentiated cell lines (Green et al., 1992), there is a growing body of experimental evidence suggesting that they have different signaling properties in differentiated cells in vivo (Aprigliano et al., 1997; Zhou et al., 1997). Moreover, these receptors may differ in other functional parameters such as desensitization (Michel et al., 1990). Thus, characterizing the functional differences between these highly homologous receptors in the context of differentiated cells will shed light on their physiologic function and the consequence of pharmacologic manipulation in vivo.

We have been interested understanding the specific role of ₁ and ₂AR subtypes in regulating cardiac myocyte function. Both subtypes are expressed in human and murine hearts. Our previous in vivo studies on ₁AR knockout (₁AR-KO) mice, ₂AR knockout (₂AR-KO) mice and ₁₂AR double knockout (₁₂AR-KO) mice demonstrated that the ₁AR is primarily responsible for catecholamine-induced changes in heart rate and contractility (Rohrer et al., 1996, 1999; Chruscinski et al., 1999). In an effort to understand the functional role of ₂AR in the heart, we chose to study receptor subtype-specific signaling in cultured neonatal myocytes. This experimental system has several experimental advantages. These cells contract spontaneously and cultures can be maintained for up to 1 week. The contraction rate of cultured myocytes is responsive to catecholamines added to the culture medium. Thus, by studying receptor-modulated contraction rate, we are able to examine the integrated response to all of the signaling pathways activated by the complement of  adrenergic receptors natively expressed in these cells. By examining the effect of a nonselective AR agonist in myocytes from ₁AR-KO, ₂AR-KO, and ₁₂AR-KO mice, we have been able to identify distinct signaling pathways for the ₁AR, ₂AR, and the ₃AR.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Mice. ₁AR-KO (Rohrer et al., 1996), ₂AR-KO (Chruscinski et al., 1999) and ₁₂AR-KO (Rohrer et al., 1999) mice were all constructed by gene targeting as described.

Preparation of Cultured Neonatal Mouse Ventricular Myocytes. Spontaneously beating neonatal cardiac myocytes were prepared from hearts of 1- to 2-day-old mouse pups (from wild-type mice and from ₁AR-KO, ₂AR-KO, and ₁/₂AR-KO mice). Briefly, hearts were quickly excised, the atria were cut off, then the ventricles were minced and digested at 37°C for 45 min in calcium-free HEPES-buffered Hanks' solution, pH 7.4, plus 100 µg/ml collagenase type II and 1X pancreatin (Invitrogen, Carlsbad, CA). To reduce the contribution of nonmyocardial cells, cells are preplated for 1 h. The myocyte-enriched cells remaining in suspension were plated in 35-mm tissue culture dishes for the contraction studies, or 12-well plates for cAMP accumulation assay. Culture dishes are precoated with 1.5% gelatin for 30 min. Myocytes were cultured in Dulbecco's modified Eagle's medium containing 10% horse serum, 5% fetal bovine serum, and antibiotics (1× gentamycin; Roche Molecular Biochemicals, Indianapolis, IN). Although the culture technique includes a preplating step that effectively decreases fibroblast contamination, myocytes are cultured in presence of 1× cytosine--D-arabinofuranoside (Sigma, St. Louis, MO) to block the cardiac fibroblast proliferation. Contraction rate and cAMP accumulation assays are performed in culture media containing serum and buffered with 20 mM HEPES, pH 7.4.

Measurement of Spontaneous Rate of Cardiac Myocyte Contraction. Measurement of spontaneous contraction rate was carried out as described previously (Johnson and Mochly-Rosen, 1995) with some modifications. Briefly, about 3 × 10⁵ cardiac cells were cultured in 35-mm Petri dishes (Corning, Palo Alto, CA, as described above) to obtain a uniformly beating syncytium. On day 4, the culture dishes were placed in a temperature-regulation apparatus positioned on the stage of an inverted microscope (Nikon, Tokyo, Japan) connected to a video camera. Cells were equilibrated at 37°C for 10 min before monitoring the contraction rate. Contraction rates of cells within the syncytium were determined at 2- to 5-min intervals for 10 min before and 30 min after the addition of isoproterenol or CL 316243. All assays were recorded on videotape.

Measurement of cAMP Accumulation. To measure intracellular cAMP, myocytes were cultured in 12-well plates (5 × 10⁵ cardiac cells per well). Cells were incubated for 30 min at 37°C with 1 mM isobutyl methylxanthine (IBMX; Sigma) immediately before the addition of the agonist isoproterenol. Cells were treated with different concentrations of isoproterenol for 5 to 15 min at room temperature. The assay was terminated by the aspiration of the incubation buffer and addition of 1 ml of 100% ethanol to each well. The cell lysates were then collected, boiled for 5 min, cooled, and stored at 80°C. Aliquots were dried in a spin-vacuum, and cAMP in the residue was determined using a radioimmunoassay (Amersham Pharmacia Biotech, Piscataway, NJ).

PTX Treatment. For contraction and cAMP measurement, cells were preincubated with PTX (0.75 mg/ml) at 37°C for at least 3 h as described previously (Xiao et al., 1995). Successful inactivation of inhibitory G proteins (Gi/Go) in PTX-treated cells was verified by the loss of the ability of adenosine (at 10⁶ M) to decrease the basal contraction rate or to reverse the positive chronotropic effect of AR stimulation by isoproterenol in WT myocytes.

PKI Treatment. Cells were preincubated with myr PKI_14-22 (20 µM) at 37°C for 10 min before isoproterenol exposure.

Drugs. Isoproterenol, adenosine, and PTX were obtained from Sigma, CL-316243 was a kind gift of Wyeth Ayerst Laboratories (Philadelphia, PA), and myr PKI_14-22 was obtained from Calbiochem (San Diego, CA).

Data Analysis. Time course experiments, two-way analysis of variance corrected for repeated measures was used to test for significance (p < 0.05) between groups. If the analysis of variance was significant, a t test using Bonferroni's method was used to compare responses at multiple time points of interest where maximal effects of treatments were observed. Analysis was done using Prism (GraphPad Software, Inc., San Diego, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Contraction Rate Response to Isoproterenol in Myocytes from Wild-Type and AR Knockout Mice. Neonatal myocytes were harvested and maintained in culture for 4 days before contraction rate studies. Figure 1A shows the response of wild-type neonatal myocytes to increasing doses of the AR selective agonist isoproterenol. We observed the maximal response with 10 µM isoproterenol. We also examined cAMP accumulation in cultures of wild-type, neonatal myocytes. In contrast to the relatively high concentrations of isoproterenol needed to stimulate contraction rate, elevations of intracellular cAMP could be observed with as little as 0.01 µM isoproterenol (Fig. 1B). This may be caused in part by the inclusion of the phosphodiesterase inhibitor IBMX in the cAMP accumulation assays.

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Contraction rate of neonatal myocytes in culture. Myocytes from wild-type (WT), 1AR-KO, 2AR-KO, and 12AR-KO mice were cultured and intrinsic contraction rate was measured as described under Materials and Methods. A, neonatal myocytes from WT mice were exposed to increasing concentrations of isoproterenol. The maximum change in contraction rate relative to basal was observed with 10 µM isoproterenol. The data represent the mean ± S.E. of four experiments. B, cAMP accumulation in cultured neonatal myocytes from wild-type mice after a 5-min treatment with different concentrations of isoproterenol. The data represent the mean ± S.E. of three experiments done in triplicate.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   A, the change in contraction rate in response to 10 µM isoproterenol in neonatal myocytes from WT, 1AR-KO, 2AR-KO, and 12AR-KO mice. The data represent the mean ± S.E. of n experiments (WT, n = 8; 1AR-KO, n = 10; 2AR-KO, n = 9; 12AR-KO, n = 8) from at least five different myocyte preparations. Individual myocyte cultures were used only once. B, maximum responses to 10 µM isoproterenol in neonatal myocytes from WT, 1AR-KO, 2AR-KO, and 12AR-KO mice. Maximal response were taken from experiments shown in Fig. 2A and compared by unpaired Student t test. *p < 0.05, significant differences from wild-type.

Dual Coupling of ₂AR in Neonatal Myocytes. The biphasic contraction rate response after stimulation of ₂AR in neonatal myocytes from ₁AR KO mice suggests sequential coupling to different signaling pathways. We therefore examined the contribution of pertussis toxin-sensitive G proteins to isoproterenol-induced changes in contraction rate. Figure 3A shows the effectiveness of the pertussis toxin treatment protocol in wild-type myocytes. We observed no response to adenosine, either before or after exposure to 10 µM isoproterenol, in pertussis toxin-treated cells. We used this protocol to study the role of pertussis toxin-sensitive G proteins in isoproterenol-induced changes in contraction rate in 1AR knockout myocytes (Fig. 3B). Compared with control myocytes, the magnitude of the rate increase is greater and stimulatory phase is prolonged in pertussis toxin-treated myocytes. No decrease below baseline is observed. Interestingly, the increase in contraction rate after isoproterenol stimulation in ₁AR-KO myocytes in the presence of pertussis toxin attenuates more rapidly than the response in ₂AR-KO myocytes, suggesting that the ₂AR desensitizes more rapidly than does the ₁AR under these experimental conditions. Stimulation of contraction rate by ₁AR in ₂AR-KO myocytes was not significantly altered by pertussis toxin treatment (Fig. 3C).

View larger version (24K): [in this window] [in a new window]   Fig. 3.   The effect of pertussis toxin treatment on the isoproterenol-stimulated contraction rate of neonatal myocytes. Neonatal myocytes were cultured and treated with pertussis toxin (PTX) 3 h before contraction rate experiments as described under Materials and Methods. The basal contraction rate of myocytes was not altered significantly by PTX treatment. A, experiments on wild-type myocytes show that pertussis toxin treatment effectively blocks the inhibitory effect of adenosine (ade) on contraction rate before and after exposure to isoproterenol (Iso). B, the effect of pertussis toxin treatment on isoproterenol-induced changes in contraction rate in myocytes from 1AR-KO mice. C, the effect of pertussis toxin treatment on isoproterenol-induced changes in contraction rate in myocytes from 2AR-KO mice. The data represent the mean ± S.E. of N experiments from at least three different myocyte preparations. *p < 0.05; a Bonferroni's t test was performed on curves that were found to be significantly different by a two-way analysis of variance corrected for repeated measures.

Evidence for ₃AR in Neonatal Myocyte. We observe a small but reproducible decrease in contraction rate in myocytes from ₁₂AR-KO mice. In myocytes treated with pertussis toxin, isoproterenol induced a brief (~10 min) and small stimulation of contraction rate (Fig. 4A). This observation suggested a possible role for the ₃AR. This was confirmed by treatment of ₁₂AR-KO myocytes with the ₃AR selective agonist CL-316243, which produced a greater decrease in contraction rate in untreated myocytes than did isoproterenol. Moreover, CL-316243 stimulated an increase in contraction rate in pertussis toxin-treated myocytes similar to that stimulated by isoproterenol (Fig. 4B). Comparable inhibitory effects of CL-316243 could also be observed in myocytes from wild-type mice (Fig. 4C), indicating that the ₃AR response in ₁₂AR-KO myocytes is not caused by up-regulation of ₃AR receptors as a consequence of the loss of ₁AR and ₂AR expression.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Evidence for 3AR receptor signaling in wild-type and 12AR-KO mice. A and B, the effect of pertussis toxin treatment on the contraction rate of neonatal myocytes from 12AR-KO mice. Cultures were treated with 10 µM isoproterenol (A) or 10 µM CL316243 (B, a 3AR-selective agonist) at time = 0 min. C, The effect of CL316243 on the contraction rate of neonatal myocytes from wild-type mice. The data represent the mean ± S.E. of N experiments from at least two different myocyte preparations. *p < 0.05; a Bonferroni's t test was performed on curves that were found to be significantly different by a two-way analysis of variance corrected for repeated measures.

Differential Effect of PKA Inhibition on ₁AR and ₂AR Signaling in Neonatal Myocytes. Stimulation of both ₁AR and ₂AR increases contraction rate in neonatal myocytes and both ₁AR and ₂AR stimulate adenylyl cyclase in cultured myocytes. To determine the role of cAMP-dependent protein kinase A (PKA) in modulating contraction rate, we examined the effect of the PKA specific inhibitor PKI. As shown in Fig. 5A, PKI had a dramatic inhibitory effect on isoproterenol-stimulated contraction rate in ₂AR-KO myocytes. In contrast, the stimulation of contraction rate was slightly increased and more prolonged in ₁AR-KO myocytes after PKI treatment (Fig. 5B). Surprisingly, the magnitude of the inhibition of contraction rate was also more profound in PKI-treated ₁AR-KO myocytes at 30 min after ISO addition. PKI had no significant effect on the response to isoproterenol in ₁₂AR-KO myocytes (Fig. 5C).

View larger version (24K): [in this window] [in a new window]   Fig. 5.   The effect of the PKA inhibitor PKI on the contraction rate of neonatal myocytes from 2AR-KO mice (A), 1AR-KO mice (B), and 12AR-KO mice (C). Neonatal myocyte cultures were treated with 20 µM PKI 10 min before the addition of 10 µM isoproterenol as described under Materials and Methods. The data represent the mean ± S.E. of N experiments from at least two different myocyte preparations. *p < 0.05; a Bonferroni's t test was performed on curves that were found to be significantly different by a two-way analysis of variance corrected for repeated measures. **p < 0.01; significance by unpaired Student t test. Note that the curves in B were not found to be significantly different by two-way analysis of variance; however, the effect of PKI on the maximal isoproterenol-induced inhibition of contraction rate (**) in 1AR-KO myocytes was examined by an unpaired Student t test and found to be significant (p < 0.005).

cAMP Levels Do Not Predict the Coupling Behavior of the ₂AR in ₁AR-KO Myocytes. The biphasic response of ₁AR-KO myocyte contraction rate to stimulation by isoproterenol suggests that the ₂AR initially couples predominantly to Gs, but switches predominantly to Gi after ~15 min. To determine the role of cAMP in the biphasic modulation of contraction rate, we examined cAMP levels in ₁AR-KO myocytes at 5 and 15 min after isoproterenol stimulation. We were unable to detect cAMP accumulation in the absence of the phosphodiesterase inhibitor IBMX; therefore, the cAMP accumulation experiments do not exactly reproduce the conditions used for the contraction rate experiments. Nevertheless, we expected to see differences in accumulation following pertussis toxin treatment. As shown in Fig. 6, in control myocytes, the cAMP accumulation at 15 min was only slightly greater than at 5 min, consistent with a switch in coupling from Gs to Gi at 15 min. However, pertussis toxin treatment did not increase the cAMP accumulation at 15 min. Therefore, the cAMP studies do not predict the contraction rate behavior observed in Fig. 3B.

View larger version (42K): [in this window] [in a new window]   Fig. 6.   cAMP accumulation in neonatal myocytes from 1AR KO mice. Myocyte cultures were treated with pertussis toxin (PTX) or vehicle control for 3 h before the addition of 10 µM isoproterenol. cAMP accumulation was determined at 5 and 15 min as described under Materials and Methods. The data represent the average of three experiments done in triplicate.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our studies indicate that all three AR subtypes are expressed in neonatal myocytes and demonstrate that each subtype couples to distinct signaling pathways that influence the rate of spontaneous beating of myocytes in culture. Stimulation of the ₁AR produces the greatest chronotropic effect through a PKA-dependent mechanism. The inhibition of contraction rate by the PKA inhibitor PKI suggests that PKA phosphorylation of the L-type calcium channel (Yue et al., 1990) is a likely mechanism for the increase in contraction rate mediated by the ₁AR. The difference in maximal contraction rate between wild-type and ₂AR-KO myocytes (Fig. 2) was not found to be significant, but the observed trend could be explained by the loss of the stimulatory component of ₂AR activation. The mechanism by which the ₂AR stimulates contraction rate during the first 10 min after agonist activation seems to differ from that used by the ₁AR. Although ₂AR stimulation causes a rise in intracellular cAMP when assayed using a cAMP accumulation assay, ₂AR stimulation of rate is insensitive to PKI (Fig. 5). The stimulation of contraction rate by the ₂AR may involve direct interactions between Gs and a channel (Imoto et al., 1988; Yatani and Brown, 1989; Yatani et al., 1990). It is also possible that the ₂AR stimulates contraction rate via the cAMP-sensitive, nonselective cation channel (pacemaker channel, I_f channel) (Ludwig et al., 1998). The lack of effect of PKI on ₂AR stimulated contraction rate suggests that the increase in cAMP induced by ₂AR stimulation is physically compartmentalized and either does not activate PKA or the PKA activated after ₂AR stimulation does not have access to the L-type calcium channel. Evidence for compartmentalization of cAMP signaling has been observed in adult rat myocytes (Zhou et al., 1997; Kuschel et al., 1999a) and adult canine myocytes (Kuschel et al., 1999b) where both ₁AR and ₂AR stimulate cAMP accumulation, but only ₁ activation leads to phosphorylation of phospholamban by PKA.

Several studies have implicated caveolae as potential signaling domains for -adrenergic receptors. Both ₁AR and ₂AR co-purify with caveolin in sucrose gradient fractions from COS-7 cells overexpressing the receptors (Schwencke et al., 1999). These results suggest that caveolar localization would not be a mechanism for differential signaling by the ₁AR and ₂AR. Yet, this may be attributed to overexpression of recombinant receptors in a highly undifferentiated cell line. Studies in cultured cardiac myocytes provide more support for compartmentalization of signaling components. Adenylyl cyclase type 6 and ₁AR were both preferentially localized to caveolae when overexpressed by recombinant adenovirus in neonatal rat cardiac myocytes (Ostrom et al., 2000). However, studies of native receptors in neonatal rat cardiac myocytes provide evidence for preferential localization of ₂AR in caveolae, whereas the majority of ₁ARs were found in noncoveolar membrane fractions (Rybin et al., 2000).

The inhibitory effect of ₂AR activation on contraction rate can be abolished by pertussis toxin (Fig. 3), indicating that ₂ARs are capable of coupling to Gi/o proteins in cardiac myocytes. Other investigators have failed to detect evidence for ₂AR coupling to Gi proteins in neonatal murine cardiac myocytes as assayed by cAMP accumulation, calcium transients, and contractile function (Sabri et al., 2000). These conflicting results can be explained by the fact that different functional assays are used. More likely, Gi coupling may be observed only when ₂ARs are maximally stimulated by a full agonist. This is technically difficult to do in wild-type myocytes without also activating ₁ARs. Selective activation of only ₂ARs in wild-type mice requires submaximal doses of subtype-selective agonists such as zinterol, which is a partial agonist relative to isoproterenol. By using neonatal myocytes from ₁AR-KO mice, we are able to use maximal doses of the full agonist isoproterenol and observe the relatively subtle effects of ₂AR coupling to Gi on myocyte contraction rate.

Of particular interest is the biphasic coupling of the ₂AR, with predominant Gs coupling during the first 10 min of stimulation and predominant Gi coupling from 15 min onward. Dual coupling of the ₂AR to both Gs and Gi has been observed in 293 cells (Daaka et al., 1997) and in adult cardiac myocytes (Xiao et al., 1995, 1999a; Communal et al., 1999), but this is the first demonstration of sequential Gs, Gi coupling in cardiac myocytes. Moreover, the mechanism of the sequential coupling in 293 cells appears to differ from that in murine neonatal myocytes. In 293 cells evidence suggests that the ₂AR coupling to Gi requires PKA phosphorylation of the ₂AR (Daaka et al., 1997). Based on this model one would expect that treatment of neonatal myocytes from 1AR-KO mice with the PKA selective inhibitor PKI would reduce or abolish the inhibition of contraction rate by the ₂AR. However, PKI treatment resulted in an exaggeration of both stimulatory and inhibitory effects of isoproterenol on myocyte contraction rate in ₁AR-KO myocytes (Fig. 5B). Thus, it is unlikely that PKA phosphorylation plays a significant role in the switch of ₂AR coupling from Gs to Gi/o proteins in neonatal myocytes. The fact that PKI augments coupling to both Gs and Gi/o proteins suggests that PKA mediated receptor phosphorylation may impair coupling to both G proteins. The mechanism of this apparent switch in coupling from Gs to Gi/o has yet to be determined, but may be caused by differences in the relative abundance of Gi/o and Gs in myocytes. Although the ₂AR has a higher affinity for Gs, Gi may be more abundant. Precoupling of the ₂AR to Gs may account for the initial stimulation in contraction rate. However, after Gs dissociates from the activated receptor, the receptor is more likely to encounter Gi than another Gs. The mechanism by which pertussis toxin-sensitive G proteins decrease contraction rate after ₂AR activation is not known. It could be due to inhibition of adenylyl cyclase or through a direct effect of the activated Gi subunit on an inwardly rectifying potassium channel (IKACh) (Wickman et al., 1994).

Although cAMP plays a critical role in regulating myocyte function, we did not observe a strong correlation between cAMP accumulation and myocyte contraction rate in response to isoproterenol (Figs. 1 and 6). An increase in cAMP accumulation was detected after exposure to 0.01 µM isoproterenol, whereas 1 µM isoproterenol was needed to induce a detectable increase in contraction rate (Fig. 1). Moreover, the cAMP accumulation assay did not predict the biphasic effect of ₂AR stimulation on contraction rate in ₁AR-KO myocytes (Figs. 3 and 6). These discrepancies are most likely caused by technical differences in the assays. To detect cAMP accumulation in myocytes, cultures must be preincubated with a phosphodiesterase inhibitor (IBMX) to prevent hydrolysis of cAMP. This increases the sensitivity of the cAMP accumulation assay, and may make it difficult to observe a change in coupling from Gs to Gi. In addition, the myocyte cultures contain a small population of fibroblasts, which may contribute disproportionately to cAMP accumulation. This may be particularly difficult for myocytes from ₁AR-KO mice, in which we observe the lowest levels of isoproterenol-induced cAMP accumulation.

The physiologic role of ₂AR in the adult heart is not well understood and may depend on the species being studied (Steinberg 1999; Xiao et al., 1999b). The Steinberg lab has observed differences in ₂AR signaling between mice and rats (Sabri et al., 2000) and between neonatal and adult rat cardiac myocytes (Kuznetsov et al., 1995). In ₁AR-KO mice there is no detectable effect of ₂AR stimulation on heart rate or contractility in vivo (Rohrer et al., 1996). Although studies in humans suggest that the ₂AR plays a significant role in regulating contractile function (Brodde, 1991; Kaumann et al., 1996, 1999; Molenaar et al., 2000). There is a growing body of evidence linking ₂AR stimulation to mitogen-activated kinase signaling pathways suggesting a possible role in myocyte growth and apoptosis (Communal et al., 1999; Communal et al., 2000; Singh et al., 2000).

Analysis of ₂AR coupling in ₁AR-KO mice is complicated by the presence of ₃ARs; however, examination of ₁/₂AR-KO mice allows us to estimate the contribution of ₃AR signaling in ₁AR-KO mice. Stimulation of ₃ARs has a very small and relatively brief inhibitory effect on contraction rate. Contraction rate returns to baseline after 10 min of isoproterenol exposure and there is no effect of a second addition of isoproterenol at 30 min (data not shown). Thus, ₃AR mediated inhibition of contraction rate may reduce the initial stimulatory effect of ₂AR activation on contraction rate in myocytes from ₁AR-KO mice, but probably has no significant effect on the subsequent inhibitory phase, which is greatest after 15 min of isoproterenol exposure. The activation of contraction rate by the ₃AR in pertussis toxin-treated myocytes from ₁/₂AR-KO mice is also very brief, returning to baseline 10 min after exposure to either isoproterenol or CL-316243 (Fig. 4). In contrast, in ₂AR-KO myocytes and pertussis toxin-treated ₁AR-KO myocytes, the contraction rate is near maximally stimulated at 10 min after isoproterenol stimulation. Thus, ₃AR signaling through both Gs and Gi seems to desensitize more rapidly than does signaling by either the ₁AR or the ₂AR. This is somewhat unexpected in light of earlier studies that failed to detect desensitization in ₃AR expressed in CHW or LTK cells (Nantel et al., 1993), although others have observed ₃AR desensitization to be depend on the cell line used (Chaudhry and Granneman, 1994).

Another interesting observation regarding ₃AR signaling in ₁/₂AR-KO myocytes is that the relative efficacy of the agonists isoproterenol and CL-316243 depends on the signaling pathway. When coupling to Gi/o proteins, CL-316243 is more efficacious at inhibiting contraction rate than is isoproterenol (compare solid lines in Fig. 4, A and B), whereas CL-316243 and isoproterenol are equally efficacious at stimulating contraction rate in cells treated with pertussis toxin (compare dashed lines in Fig. 4, A and B). This suggests that agonist efficacy may be G protein-specific, as has been demonstrated for other GPCRs (Spengler et al., 1993; Kenakin 1995). Thus, isoproterenol may be a full agonist for the ₃AR when coupled to Gs but not when coupled to Gi.

In conclusion, we examined isoproterenol-stimulated changes in contraction rate in neonatal myocytes from mice having disruptions in the genes for the ₁AR, the ₂AR, and both ₁AR and ₂AR. This experimental system allows us to examine subtype-specific signaling in differentiated cells in which receptors are expressed at physiologic levels. Our results indicate that all three AR subtypes are expressed in neonatal myocytes and activate distinct signaling pathways. Of particular interest is the biphasic effect of ₂AR stimulation on myocyte contraction rate and the fact that the ₂AR and the ₁AR stimulate contraction rate through different mechanisms. The different signaling pathways used by the ₁AR and the ₂AR suggest that these proteins are physically separated in the myocyte membrane providing further evidence for functional signaling microdomains.