INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

VASCULAR SMOOTH MUSCLE CELLS (VSMC) are regulated by a variety of neurotransmitters and hormones that influence a network of signal transduction pathways. Norepinephrine and the vasoactive peptides angiotensin II and arginine vasopressin (AVP) act on cell surface receptors that couple through heterotrimeric G proteins to stimulate phospholipase C and produce diacylglycerol and inositol 1,4,5-trisphosphate [Ins(1,4,5)P₃] (2, 25). These second messengers then activate protein kinase C and elevate intracellular free Ca²⁺ to initiate contraction of VSMC (28) and, in vascular injury, may promote cellular growth (1, 9) and deposition of extracellular matrix (15). In contrast, epinephrine, acting on ₂-adrenergic receptors, and local hormones, such as prostaglandins I₂ and E₂, stimulate cAMP formation, which decreases contractility (3) and antagonizes agonist-induced hypertrophy and matrix formation (10, 12, 32). The opposing influences of these pathways are important in determining the functional state of VSMC, and integration of this signaling is a key component in vascular homeostasis.

The principal control of cAMP formation occurs with the activity of the effector enzyme adenylyl cyclase. At least nine forms of this effector have now been identified with discrete tissue distributions and unique regulatory properties (29). The various isozymes of adenylyl cyclase differentially respond to activated G_s and G_i as well as to G (30), Ca²⁺ (16), Ca²⁺/calmodulin (6, 8, 31), and phosphorylation, providing a potential focal point within the cell for the integration of diverse stimuli. The regulatory properties of adenylyl cyclases have been largely determined through studies of purified enzymes expressed in Sf9 insect cells, and it is not always clear how properties defined in vitro may function in the intact cell to integrate coincident stimuli into a defined response. As an approach to this question, we initially examined the integration of multiple signals to adenylyl cyclase after stable expression of select effector isoforms in DDT₁-MF2 smooth muscle cells (24), where integration of both stimulatory and inhibitory inputs was found to be a dynamic process depending on both enzyme type and phosphorylation state. The present study extends these findings by examining signal integration in two commonly used models of arterial smooth muscle cells, each expressing an endogenous complement of receptors and effectors. Experiments were focused on the interaction between stimuli affecting Ins(1,4,5)P₃ and cAMP signaling pathways. We report that the integration of these pathways is qualitatively different in the two cell models and determined by Ca²⁺ mobilization and the population of adenylyl cyclase isozymes expressed.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials. DMEM and amphotericin B were purchased from GIBCO-BRL (Grand Island, NY). Fetal bovine serum was obtained from HyClone Labs (Logan, UT). ¹²⁵I-labeled 2'-O-monosuccinyl cAMP tyrosyl methyl ester was synthesized using the method of Oehlenschlager et al. (26). [d(CH₂)-O-Me-Tyr²,Arg⁸]vasopressin and [d(CH₂)-D-Ile²,Ile⁴,Arg⁸,Ala⁹]- vasopressin were from Bachem (Torrance, CA). [-³²P]dCTP was purchased from DuPont-NEN (Boston, MA). The Fast Track mRNA isolation kit was obtained from Invitrogen (San Diego, CA) and Rnazol was from Tel-Test (Friendswood, TX). Isoproterenol, AVP, 3-isobutyl-1-methylxanthine (IBMX), penicillin, streptomycin, and trypsin were from Sigma (St. Louis, MO). A23187 and calphostin C were purchased from Calbiochem (San Diego, CA). All other chemicals were of the highest purity available. Adenylyl cyclase constructs were kindly provided as follows: types I and IV, Dr. Alfred Gilman (Department of Pharmacology, Southwestern University, Dallas, TX); types II and III, Dr. Randall Reed (Department of Molecular Biology and Genetics, Johns Hopkins School of Medicine, Baltimore, MD); types V and VI, Dr. Yoshihiro Ishikawa (Department of Medicine, Harvard Medical School, Boston, MA); type VII, cloned from DDT₁-MF2 cells; type VIII, Dr. John Krupinski (Geisinger Research Clinic, Danfield, PA); and type IX, Dr. Richard Premont (Duke University, Durham, NC).

Cell cultures. A7r5 cells were obtained from the American Type Culture Collection. Primary cultures of aortic VSMC were prepared from 3- to 4-wk-old male Sprague-Dawley rats by the method of Ross (27) as previously described (19). Isolates from one rat were used for each preparation of VSMC, and every experiment was performed with at least three different populations of cells. Cells prepared with this procedure showed the hill-and-valley growth pattern typical of VSMC and stained positive for smooth muscle actin using a monoclonal antibody specific for the -isoform of actin. Once established, both A7r5 cells and VSMC were maintained in DMEM supplemented with 10% fetal bovine serum, 2 mM glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin in a humidified atmosphere of 10% CO₂-90% air at 37°C. The medium was routinely changed every 48 h, and cells were subcultured once weekly after detachment with 0.05% trypsin in phosphate-buffered saline.

Determination of intracellular calcium. Release of intracellular free Ca²⁺ was determined using a fluorometric imaging plate reader (FLIPR) system (Molecular Devices; Sunnyvale, CA) for high-throughput optical screening of cell-based fluorometric assays. The instrument simultaneously monitors all 96 wells of a microplate with kinetic updates in the subsecond range and includes a 6-W argon ion laser (Coherent; Santa Clara, CA), optical scanning system, temperature control (37 ± 0.1°C), and a charge-coupled device camera. Cells for intracellular Ca²⁺ measurements were subcultured into 96-well clear-bottomed black microplates (Corning Costar; Cambridge, MA). On the day of the experiment, cells were then dye loaded with 4 µM fluo 3-AM (excitation at 488 nm, emission at 540 nm, Molecular Probes; Eugene, OR) in HEPES-buffered salt solution (pH 7.4) containing 2.5 mM probenecid for 1 h at 37°C. After cells were washed four times with buffer, they were placed in the FLIPR instrument and simultaneously exposed to different test agents, and emission intensities in each well of the microplate were monitored at 1.5-s intervals over 6 min. Six wells were averaged for each treatment protocol, and experiments were performed in at least three different batches of cultured cells.

Preparation and analysis of RNA. Total cellular RNA was isolated from freshly harvested cells or brain cortex by guanidinium denaturation utilizing Rnazol (Tel-Test). Polyadenylated RNA [poly(A)+ RNA] was isolated from total RNA using an oligo(dT) cellulose affinity matrix (Fast Track mRNA isolation kit, Invitrogen). For RNA blot analysis, mRNA was subjected to electrophoresis on 1% agarose-3% formaldehyde gels followed by transfer to nylon membranes by capillary blotting with 10× saline-sodium citrate. After transfer, the membrane was baked at 80°C for 1.5 h in a vacuum oven, and RNA blot hybridizations were performed in an aqueous phosphate buffer system according to Mahmoudi and Lin (22). Briefly, nylon membranes were first prehybridized in phosphate buffer containing (in mM) 500 Na₂HPO₄ (pH 7.2), 1 EDTA (pH 8.0), 7% SDS, and 1% bovine serum albumin for 60 min at 65°C. Incubation was continued for 12-15 h in the same solution containing ³²P-labeled probes (1.6-5.7 × 10⁶ counts · min¹ · ml¹), which were generated by random priming (Multiprime DNA labeling system, Amersham). The DNA fragments used for probe generation were as follows: type I, full-length cDNA; type II, full-length cDNA; type III, nt 682-2,714; type IV, full-length cDNA; type V, nt 715-1,204; type VI, full-length cDNA; type VII, nt 2,836-3,111; type VIII, full-length cDNA; and type IX, nt 1-957. The fragments used for priming were generated with the appropriate restriction enzymes and purified using agarose gels. The hybridized blots were washed (60°C) twice in buffer containing (in mM) 40 Na₂HPO₄ (pH 7.2), 1 EDTA (pH 8.0), and 5% SDS for 60 min and once in buffer containing (in mM) 40 Na₂HPO₄ (pH 7.5), 1 EDTA (pH 8.0), and 1% SDS for 60 min. The blots were then dried and exposed to Kodak XAR film using intensifying screens at 70°C for 12-48 h.

Determination of cAMP. At the end of the experiments, cells were incubated with 0.1 N HCl at 4°C for 15 min to extract intracellular cAMP as previously described (19). The acid extract was then assayed for cAMP by radioimmunoassay as described by Brooker et al. (5). No cAMP was detected in the extracellular incubation buffer.

Protein determination. The concentration of protein in each culture well was determined by the method of Lowry et al. (21) with bovine serum albumin as the standard.

Data analysis. Data are presented as means ± SE and were analyzed using analysis of variance followed by Duncan's multiple-range test for detecting differences, with P < 0.05 considered as significant. The number of observations reported represents the number of individual experiments conducted, each performed in triplicate.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of AVP on isoproterenol-stimulated cAMP accumulation in VSMC and A7r5 cells. The influence of AVP on agonist-stimulated cAMP formation in primary cultures of VSMC is shown in Fig. 1. Incubation of VSMC with AVP (0.3 µM) had no significant effect on basal cAMP levels but synergistically enhanced isoproterenol-stimulated cAMP accumulation. Stimulation of VSMC with isoproterenol (1 µM) alone for 60 s increased cAMP levels by 331 ± 42 pmol/mg protein. In comparison, simultaneous exposure of cells to isoproterenol in combination with AVP (0.3 µM) produced an increase in cAMP of 815 ± 110 pmol/mg protein, a value over twofold greater than that observed in cells treated only with -agonist. Surprisingly, AVP had a totally opposite effect on agonist-induced cAMP accumulation in A7r5 cells. Treatment of A7r5 cells with AVP (0.3 µM) again had no effect on basal cAMP but significantly decreased isoproterenol-stimulated cAMP levels (Fig. 1). Stimulation of A7r5 cells with isoproterenol (1 µM) increased intracellular cAMP by 454 ± 41 pmol/mg protein in control cells. However, when cells were coincubated with isoproterenol in combination with AVP, isoproterenol-stimulated cAMP was only 243 ± 28 pmol/mg protein, a 47% decrease from the control stimulation. In parallel experiments, pretreatment of cells with pertussis toxin (100 ng/ml) for 24 h, which causes ADP ribosylation of G_i, effectively disrupting G protein coupling, did not alter the action of AVP to affect isoproterenol-stimulated cAMP in either cell type (data not shown).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Effect of arginine vasopressin (AVP) on isoproterenol (Iso)-stimulated cAMP formation in cultured vascular smooth muscle cells (VSMC) and A7r5 cells. Cells were incubated for 60 s in HEPES-buffered salt solution with 0.3 µM AVP, 1 µM Iso, or Iso + AVP in the presence of 0.5 mM 3-isobutyl-1-methylxanthine (IBMX). At the end of the incubation, the test buffer was removed, and cells were incubated at 4°C for 15 min in 0.1 N HCl to extract cAMP, which was then quantitated as described in MATERIALS AND METHODS. Values are means ± SE; n = 8-16 experiments, each performed in triplicate. *Significantly different from Iso, P < 0.05.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Concentration-response curves for AVP effects on Iso-stimulated cAMP formation in VSMC () and A7r5 cells (). Cells were incubated for 60 s in HEPES-buffered salt solution containing 0.5 mM IBMX with 1 µM Iso and the indicated concentrations of AVP. To determine the effect of 1 µM [d(CH2)-O-Me-Tyr2,Arg8]vasopressin, VSMC () or A7r5 cells () were preincubated with the V1 receptor antagonist for 5 min before incubation for 60 s with Iso + AVP. At the end of the incubation, the test buffer was removed, and cells were incubated at 4°C for 15 min in 0.1 N HCl to extract cAMP, which was then quantitated as described in MATERIALS AND METHODS. Values are means ± SE; n = 4-7 experiments, each performed in triplicate. *Significantly different from 10 nM AVP and AVP + V1 receptor antagonist, P < 0.05.

Role of calcium in signal integration in VSMC and A7r5 cells. V₁ receptors couple to phospholipase C and, when activated, lead to the generation of Ins(1,4,5)P₃ and diacylglycerol (25). The effect of AVP to enhance cAMP regulation in cultured VSMC has been previously shown to be independent of protein kinase C activation but, instead, was related to the release of Ca²⁺ from intracellular stores (36). A similar mechanism has been found for the effect of AVP in A7r5 cells. Treatment of A7r5 cells with the phorbol ester phorbol 12-myristate 13-acetate (PMA; 0.1 µM) for 10 min did not alter either basal or stimulated cAMP levels, and inhibition of protein kinase C activation with calphostin C (0.1 µM) also had no effect on the action of AVP to inhibit isoproterenol stimulation (data not shown). However, the action of AVP in A7r5 cells was found to be totally dependent on intracellular Ca²⁺ mobilization. To address this, A7r5 cells were incubated in either normal Ca²⁺ buffer (control) or Ca²⁺-free buffer or treated with the intracellular chelator 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA) to attenuate Ca²⁺-dependent agonist effects. The influence of each intervention on AVP-induced free Ca²⁺ transients is shown in Fig. 3A. In control buffer, exposure of A7r5 cells to AVP evoked a sharp release of intracellular Ca²⁺, which peaked within seconds, followed by a decline and then a secondary phase generally associated with influx from the extracellular media. Consistent with this interpretation, removal of extracellular Ca²⁺ failed to alter the initial peak of Ca²⁺ elevation but essentially eliminated the delayed phase of influx, and free intracellular Ca²⁺ returned to basal levels by 2 min. In turn, loading of cells with BAPTA predictably attenuated both phases of AVP-induced free Ca²⁺ elevation. The comparative effects of these interventions on cAMP stimulation are illustrated in Fig. 3B. In this series, incubation of control cells with AVP (0.3 µM) reduced isoproterenol-stimulated cAMP levels to 60 ± 5% of those produced in cells incubated with only isoproterenol. In Ca²⁺-free buffer, AVP remained effective and decreased isoproterenol-stimulation to 46 ± 6% of that observed with -agonist alone, a value that was not significantly different from the effect of AVP in control cells. In contrast, pretreatment of cells with BAPTA-AM (30 µM) for 30 min eliminated the effect of AVP to attenuate agonist-induced cAMP production. In cells pretreated with BAPTA, isoproterenol-stimulated cAMP formation in the presence of AVP was 97 ± 4% of that observed in cells exposed only to isoproterenol. The composite of these data suggests for both A7r5 cells and cultured VSMC that AVP elevates intracellular free Ca²⁺, which then differentially modulates the response of adenylyl cyclase to -adrenergic receptor stimulation.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Calcium dependence of AVP inhibition of Iso-stimulated cAMP formation in A7r5 cells. A: cells were preincubated for 30 min in either calcium-free HEPES-buffered salt solution containing 2 mM EGTA (calcium-free) or HEPES-buffered salt solution containing 30 µM 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA)-AM. At time 0, 0.3 µM AVP was added, and the intracellular Ca2+ concentration ([Ca2+]i) was monitored by a fluorometric imaging plate reader (FLIPR) over a 6-min time interval as described in MATERIALS AND METHODS. Each response represents the average of 6 wells simultaneously treated with AVP and is representative of 3 different experiments. B: cells were pretreated as in A and then incubated with 0.3 µM AVP, 1 µM Iso, or Iso + AVP in the presence of 0.5 mM IBMX. For cells incubated in calcium-free solution, test agents were maintained in calcium-free buffer with 2 mM EGTA. At the end of the incubation, the test buffer was removed, and cells were incubated at 4°C for 15 min in 0.1 N HCl to extract cAMP, which was then quantitated as described in MATERIALS AND METHODS. Iso-stimulated cAMP was 433 ± 52, 397 ± 19, and 386 ± 32 pmol/mg protein for control, calcium-free, and BAPTA cells, respectively. Values are means ± SE; n = 3-6 experiments, each performed in triplicate. *Significantly different from control, P < 0.05.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Effects of A23187 and AVP on [Ca2+]i in A7r5 cells and cultured VSMC. A: A7r5 cells were simultaneously exposed to 1 µm A23187 or 0.3 µm AVP, and [Ca2+]i was monitored by a FLIPR over a 6-min time interval as described in MATERIALS AND METHODS. Each response represents the average of 6 wells simultaneously treated with test agent and is representative of 3 experiments. B: VSMC were simultaneously exposed to 1 µm A23187 or 0.3 µm AVP, and [Ca2+]i was monitored as in A. Each response represents the average of 6 wells simultaneously treated with test agent and is representative of 3 experiments.

View larger version (11K): [in this window] [in a new window]   Fig. 5.   Effect of A23187 on Iso-stimulated cAMP formation in A7r5 cells and cultured VSMC. A: A7r5 cells were preincubated for 5 min with 1 µM A23187 in HEPES-buffered salt solution containing 0.5 mM IBMX and then incubated for 60 s with 1 µM Iso. Values are means ± SE; n = 3 experiments, each performed in triplicate. B: VSMC were preincubated for 5 min with 1 µM A23187 in HEPES-buffered salt solution containing 0.5 mM IBMX and then incubated for 60 s with 1 µM Iso. Values are means ± SE; n = 7 experiments, each performed in triplicate. *Significantly different from basal values, P < 0.05; **significantly different from Iso, P < 0.05.

Differential expression of adenylyl cyclases in A7r5 cells and VSMC. The qualitatively different signal integration observed in A7r5 cells and VSMC may reflect the adenylyl cyclase isoforms expressed by the two different models. To examine this possibility, each cell type was characterized in terms of the population of adenylyl cyclase transcripts expressed. Poly(A)+-enriched RNA from A7r5 cells and VSMC was evaluated by high-stringency RNA blot hybridization using probes for effector enzyme types I-IX. RNA isolated from the brain was used as a positive control for blot analysis because each of the adenylyl cyclase isoforms is expressed in the brain. Types I, II, VII, and IX were evident in the brain but were not detected in either A7r5 cells or VSMC (data not shown). Types III, V, and VI were expressed in A7r5 cells and VSMC (Fig. 6). In both cell groups, the level of type V transcript was relatively low and required increased exposure of the film for detection. Interestingly, adenylyl cyclases types IV and VIII were expressed in VSMC but were not detected in A7r5 cells (Fig. 6). These results clearly indicate a differential expression of adenylyl cyclase isozymes in the two different models. Most notably, A7r5 cells express transcripts encoding types III, V, and VI, whereas VSMC express types III, V, and VI but also the type IV and VIII forms of the effector, each of which possesses unique regulatory properties that could distinguish VSMC in regard to the integration of AVP signaling and cAMP regulation.

View larger version (60K): [in this window] [in a new window]   Fig. 6.   RNA blot analysis of mRNA isolated from the rat brain, VSMC, and A7r5 cells. Polyadenylated RNA was isolated, electrophoresed, and transferred to nylon membrane. Each lane contained 8 µg of polyadenylated RNA. Random-primed probes labeled with 32P were generated from cDNAs encoding adenylyl cyclase (AC) isozymes [types III (AC III), VI (AC VI), V (AC V), IV (AC IV), and VIII (AC VIII)] and were hybridized to the blot as described in MATERIALS AND METHODS. The position of coelectrophoresed RNA size markers (in kb) is indicated.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

VSMC process multiple external stimuli that influence the contractile state of blood vessels and the response of the vascular wall to injury. This myriad of external signals must be integrated by the cell to produce the desired physiological response. Variations in signal processing may contribute to the manifestations of vascular disease, and an understanding of the components of signal integration in the cell could present new avenues of intervention. In the present study, we examined the influence of the vasoactive peptide AVP on cAMP regulation in primary cultures of VSMC and in A7r5 cells, a cell line derived from the rat thoracic aorta (18). AVP had no effect on basal cAMP but differentially modified the response to isoproterenol in the two cell models.

Exposure of A7r5 cells to AVP decreased isoproterenol-stimulated cAMP accumulation in a concentration-dependent manner. A 50% reduction in agonist-stimulated cAMP was observed in cells treated with 300 nM peptide. Because all experiments were performed in the presence of the phosphodiesterase inhibitor IBMX, AVP appears to have an action to reduce production of cAMP by the effector enzyme adenylyl cyclase. The effect of AVP to diminish isoproterenol-stimulated cAMP formation was not altered in A7r5 cells that were pretreated with pertussis toxin. Consequently, this effect of the peptide cannot be attributed to an action on the inhibitory guanine nucleotide-binding protein G_i. The results suggest that AVP activates a signal pathway in A7r5 cells to yield second messenger molecules, which then attenuate the effectiveness of receptor-mediated adenylyl cyclase activation.

The V₁ subtype of the AVP receptor is coupled to phospholipase C and, upon activation, promotes the formation of Ins(1,4,5)P₃ and diacylglycerol (25). In aortic smooth muscle cells, including the A7r5 cell, the V₁ receptor is the predominant subtype expressed (4, 33). Consistent with this observation, pretreatment of A7r5 cells with the selective V₁ receptor antagonist [d(CH₂)-O-Me-Tyr²,Arg⁸] vasopressin completely blocked the action of AVP to decrease isoproterenol-stimulated cAMP formation. This finding indicates that AVP acts through V₁ receptors and raises the possibility that activation of protein kinase C and/or Ca²⁺ mobilization is the signal pathway through which AVP attenuates receptor-mediated adenylyl cyclase stimulation in A7r5 cells.

Activation of protein kinase C with phorbol esters has varying effects on different isoforms of adenylyl cyclase. Basal or stimulated activities of adenylyl cyclase types I, II, and V are enhanced by protein kinase C stimulation (14, 17, 35) and, in one report (20), agonist activation of type VI was decreased. The type VI effector was found to be expressed in A7r5 cells, and, conceivably, stimulation of protein kinase C by diacylglycerol produced in response to AVP could contribute to the effect of the peptide on agonist-induced cAMP formation. However, direct activation of protein kinase C with PMA failed to alter either basal or isoproterenol-stimulated cAMP levels in A7r5 cells. Furthermore, inhibitors of protein kinase C activation did not diminish the action of AVP to reduce cAMP production. Such observations argue against a role for protein kinase C in this action of the peptide. In contrast, the ability of AVP to mobilize Ca²⁺ appears to be key to its effect on adenylyl cyclase activation. Removal of Ca²⁺ from the extracellular buffer did not alter AVP inhibition of isoproterenol-stimulated cAMP in A7r5 cells. However, pretreatment of cells with BAPTA to chelate and buffer AVP-induced elevation of intracellular free Ca²⁺ eliminated the action of the peptide to attenuate agonist-stimulated cAMP production. In addition, elevation of intracellular Ca²⁺ with the calcium ionophore A23187, like AVP, effectively decreased isoproterenol-stimulated cAMP formation.

The basic mechanism of AVP signaling in cultured VSMC is the same as that described for A7r5 cells. However, when VSMC were incubated with AVP in combination with isoproterenol, AVP synergistically increased, rather than decreased, cAMP production. This effect was again blocked by pretreatment of cells with a V₁ receptor antagonist and is consistent with previous reports (19, 36) where both angiotensin II and AVP have been shown to enhance agonist-stimulated cAMP formation in cultured VSMC and in freshly isolated strips of the rat aorta. For each peptide, the effect was found to be independent of protein kinase C activation but directly related to the mobilization of intracellular Ca²⁺. In the present study, the importance of Ca²⁺ was emphasized further when VSMC were treated with the calcium ionophore A23187. Incubation of VSMC with A23187 alone appeared to activate adenylyl cyclase activity and increased cAMP above the basal levels found in untreated control cells. Moreover, when cells were incubated with A23187 in combination with isoproterenol, cAMP production was greater than that observed with either agonist alone, although the interaction was only slightly greater than additive. The absence of the marked synergism observed when isoproterenol is combined with AVP may reflect the pool of Ca²⁺ mobilized when cells are activated with AVP compared with the more generalized effects of an ionophore in that some studies have suggested that specific Ca²⁺ sources may be linked to different effector isoforms (7, 11). The differential time course of Ca²⁺ elevation by A23187 compared with AVP may also contribute to the difference observed. Alternatively, peptide hormones may have an additional action that may be key to the synergistic enhancement of adenylyl cyclase stimulation.

In both cultured VSMC and A7r5 cells, AVP activates V₁ receptors to elevate intracellular Ca²⁺. However, elevation of Ca²⁺ in the two cell populations has opposite effects on the increase in cellular cAMP elicited by cell surface receptors coupled to G_s. This difference may be attributable to the adenylyl cyclase isoforms expressed by the two different preparations. In the present study, expression of effector isoforms in the two cell models was evaluated by RNA blot hybridization using specific probes for adenlyl cyclase types I-IX. Because the mRNA level is not necessarily a direct predictor of protein expression, no attempt was made to make quantitative comparisons of adenylyl cyclase transcripts in the two cell models. Nonetheless, qualitative differences in the transcripts expressed were quite revealing. A7r5 cells were found to express transcripts encoding adenylyl cyclase types III, V, and VI , each of which may be influenced by Ca²⁺. Type III adenylyl cyclase is stimulated by high concentrations of Ca²⁺/calmodulin when examined in cell-free membrane preparations (8) but, when expressed in intact cells, appears to be inhibited by Ca²⁺ elevation (34). Consequently, the physiological regulation of this form of the effector remains unclear. However, types V and VI are readily inhibited by physiological concentrations of Ca²⁺ (16) and likely are the predominant effector isoforms inhibited by AVP signaling in the A7r5 cell. Cultured VSMC were observed to also express transcripts for adenylyl cyclase types III, V, and VI. Whereas types V and VI are inhibitable by Ca²⁺, this effect clearly is not dominant in the cultured cell preparation, perhaps reflecting the amount of these enzymes actually produced by the cell or, alternatively, selective coupling of receptor and effector isoforms. On the other hand, cultured VSMC also express adenylyl cyclase types IV and VIII, which were not detected in A7r5 cells. Type IV adenylyl cyclase can be activated by the -subunit of dissociated heterotrimeric guanine nucleotide-binding proteins, and this effect is synergistic with activated G_s (13, 30). The effect usually emanates from pertussis toxin-sensitive G proteins, however, and does not appear to play a role in the cell-specific influence of AVP on cAMP regulation in the present study. Of particular interest is the type VIII enzyme, which can be activated by Ca²⁺/calmodulin in the absence of other stimuli and is stimulated synergistically by Ca²⁺/calmodulin in the presence of activated G_s (6, 29). It may be the presence of the type VIII effector that allows Ca²⁺-mobilizing peptides to enhance agonist-induced adenylyl cyclase stimulation in VSMC. A more direct test of these ideas must await the development of isoform-specific inhibitors or appropriate "knockout" models.

The integration of Ca²⁺ signaling with cAMP regulation has significant implications related to vascular homeostasis. Elevation of intracellular Ca²⁺ by peptide hormones initiates contraction of VSMC and, in instances of vascular injury, may promote cell growth and deposition of extracellular matrix (1, 9, 15). These responses to Ca²⁺ mobilization are modulated, in turn, by the counterregulatory action of local hormones (such as prostaglandin I₂) that stimulate cAMP production, which then serves to relax vascular smooth muscle, attenuate vascular hypertrophy, and decrease synthesis of matrix proteins (3, 10, 12). Consequently, a process wherein acute elevation of intracellular Ca²⁺ amplifies the response of VSMC to agonists that activate adenylyl cyclase should facilitate feedback modulation of the effects of vasoactive peptides. Conversely, an effect of Ca²⁺ to attenuate adenylyl cyclase stimulation should compromise the effectiveness of this feedback. The final readout of such cross-talk for a specific cell population and set of conditions will likely depend on the profile of adenylyl cyclases expressed.