Effects of dexamethasone on G protein levels and adenylyl cyclase activity in rat vascular smooth muscle cells A. R. McLellan, S. C. J. Kenyon
Tawil,
F.
Lyall,
G.
Milligan, J.
M. C. Connell and
MRC Blood Pressure Unit, Western Infirmary, Dumbarton Road, Glasgow Gil 6NT, U.K. *Molecular Pharmacology Group, Department of Biochemistry, University of Glasgow, Glasgow Gil 6NT, U.K. received
23 March 1992
ABSTRACT
Dexamethasone administration in vitro has been shown to increase adenylyl cyclase activity in vascular smooth muscle cells (VSMC) from renal arteries and in non-vascular cell lines. To investigate whether G proteins are involved in this response, cultured VSMC from mesenteric arteries of Sprague\p=m-\Dawleyrats were incubated in the presence and absence of 10 nm dexamethasone for 24 and 48 h. Basal and stimulated adenylyl cyclase activities were increased by approximately 50% after treatment with dexamethasone. The changes were neither specifically associated with ligands which stimulate adenylyl cyclase catalytic unit via Gs (isoproterenol and prostaglandin E1) nor with guanylylimidodiphosphate (0\m=.\1 nM), which inhibits the
INTRODUCTION
catalytic unit via Gi. This suggests that dexamethasone enhances adenylyl cyclase activity through changes at the level of the catalytic unit, rather than through the G proteins which modulate its activity. No differences were seen in immunoblotting studies of the levels of Gi\g=a\2,Gs\g=a\,Gi\g=a\3 and \g=b\ subunits. Similarly, dexamethasone had no effect on the expression of mRNA for Gi\g=a\2and Gs\g=a\. The results indicate that glucocorticoid-induced increases of adenylyl cyclase activity are due to changes at the level of the adenylyl cyclase catalytic unit rather than alteration of the levels or turnover of Gs\g=a\,Gi\g=a\2, Gi\g=a\3and \g=b\subunits in the membranes of VSMC. Journal of Molecular Endocrinology (1992) 9,
al. 1976; Handa et al. 1984; Whitworth et al. 1986; Russo Iijima Malik, 1985; et al. 1990). It is clear, however, that glucocorticoid hypertension is induced through glucocorticoid receptor-dependent mechanisms (Grunfeld et al. 1985), and glucocorticoid receptors have been iden¬ tified in arterial vessels from a number of species including rats (Meyer & Nichols, 1981) and humans (Scott et al. 1987). Post-receptor events, particularly in cultured vascular smooth muscle cells (VSMC), have not been extensively studied. Several intracellular messengers are involved in changes of vascular tone mediated by prostaglandins, vasopressin and a and ß adrenoreceptors, including adenylyl cyclase/ cyclic AMP (cAMP), the inositol phosphates and cyclic GMP. Dexamethasone administration in vitro has been shown to increase adenylyl cyclase activity in VSMC from renal arteries (Yasunari et al. 1989), and similar observations have been made in a number of non-vascular cell lines (Foster & Perkins, 1977; Johnson & Jaworski, 1983; Rizzoli et al. 1986; et
al. 1975; Schomig &
The association between glucocorticoids and hyper¬ tension is well established; for example, more than 70% of patients with Cushing's syndrome are hyper¬ tensive (Plotz et al. 1952; Orth & Liddle, 1971). Furthermore, hypertension can be induced experi¬ mentally in rats (Knowlton & Loeb, 1957; Krakoff et al. 1975; Haack et al. 1977; Suzuki et al. 1982) and in many other species, using a range of glucocor¬ ticoids including dexamethasone (Suzuki et al. 1982). Although the mechanisms underlying glucocorticoid hypertension remain to be fully elucidated, a number of glucocorticoid-mediated effects have been implicated; these include increased plasma vol¬ ume (Haack et al. 1977), stimulation of the reninangiotensin system by increased plasma renin substrate concentration (Krakoff et al. 1975), decreased prostaglandin synthesis (Handa et al. 1983) and enhancement of vascular reactivity to vari¬ ous pressor agents (Mendlowitz et al. 1958; Krakoff
237-244
et
Rodan & Rodan, 1986; Chang & Bourne, 1987). As G proteins have key roles in transducing membrane signals from receptors to intracellular messengers, the possibility that glucocorticoids exert their effects on adenylyl cyclase activity by alteration of G pro¬ teins has been explored. Two G proteins, Gsa and Gia2 (Bushfield et al. 1990c), modulate the activity of adenylyl cyclase; these respectively mediate stimula¬ tion and inhibition of the enzyme. Glucocorticoids have been shown to have contrasting effects on these G proteins when studied in vitro (Rodan & Rodan, 1986; Chang & Bourne, 1987) and in vivo (Haigh et al. 1990). Increases in Gso (Rodan & Rodan, 1986; Chang & Bourne, 1987), with either no change (Chang & Bourne, 1987) or an increase (Rodan & Rodan, 1986) in G¡, have been reported to be associ¬ ated with increased adenylyl cyclase activity (Chang & Bourne, 1987) following glucocorticoid exposure in vitro in two non-vascular cell lines. However, glucocorticoid administration to adrenalectomized animals in vivo has been reported to decrease the Gs : G¡ ratio in rat aortic plasma membranes. This would tend to inhibit adenylyl cyclase activity, favouring vasoconstriction with increased pressor sensitivity (Haigh et al. 1990). The purpose of this study was to reassess the effects of glucocorticoids on cultured VSMC in vitro, and to ascertain whether changes in adenylyl cyclase activity are associated with an alteration in G protein levels.
MATERIALS AND METHODS
All chemicals
supplied by Sigma Chemical Co., Poole, Dorset, U.K., with the exception of GTP, guanylylimidodiphosphate (GppNHp), creatine kinase and creatine phosphate which were obtained from Boehringer Mannheim, Lewes, East Sussex, U.K., and the radioisotopes [32P]ATP, [5',83H]cAMP and ,25I-labelled donkey anti-rabbit IgG, which were supplied by Amersham International pic, Amersham, Bucks, U.K. Horseradish-labelled donkey anti-rabbit IgG was supplied by the Scottish Antibody Production Unit, Law Hospital, Carluke, Strathclyde, U.K. were
Cell culture
VSMC were isolated from the mesenteric beds of male Sprague-Dawley rats (weighing 300-350 g) and cultured as previously described (Balmforth et al. 1988). Briefly, fat was removed by blunt dissec¬ tion and the vascular tree transferred to an enzyme dissociation mixture (10 ml) of collagenase (1 -25 mg/ ml), elastase (005 mg/ml) and soya bean trypsin
inhibitor (0-1%) in Hank's balanced salt solution. After incubation at 37 °C for 10—15 min, the arteries were transferred to fresh Dulbecco's modified Eagle's medium (DMEM) and any remaining fat, adventitia and endothelial cells were removed by careful dissection. The cleaned arteries were cut into small pieces, transferred to fresh enzyme dissocia¬ tion mixture and incubated at 37 °C with periodic trituration until a single cell suspension was obtained. The cell suspension was centrifuged (5 °C, 200 £ for 5 min) and the pellet resuspended in DMEM containing 10% fetal calf serum (FCS) and 10% horse serum, 2mM L-glutamine, 100 U penicillin/ml and 100 pg streptomycin/ml. The dis¬ persed cells were plated in 25 cm culture flasks and maintained in a humidified atmosphere of 5% C02, 95% air at 37 °C. Cells were subcultured as required up to the 7th passage. The effects of dexamethasone on cultured cells were tested when cells were 75% confluent. The medium was changed to one contain¬ ing 10% FCS±10nM dexamethasone and the cells were incubated for 24—48 h.
Plasma membrane
preparation
Cells were scraped from flasks and homogenized using a Polytron (Kinematica, Lucern, Switzerland) at setting 10 for 15 s. The homogenate was first cen¬ trifuged at 1000 j' for 10 min and the supernatant was centrifuged again at 37 000 £ for 30 min. The final plasma membrane pellet was suspended in 10 mM Tris-HCl buffer (pH 7-4) and stored in aliquots at —80 °C. Protein concentrations in the membrane preparations were measured against bovine serum albumin (BSA) standards by the method of Peterson (1977). The purity of the plasma membrane preparations was assessed by measuring 5' nucleotidase (EC 3.1.3.5) activity (Dixon & Purdom, 1954).
Adenylyl cyclase assays Adenylyl cyclase activity was assessed by measuring
[a-32P]cAMP
released from
[a-32P]ATP
substrate
(Salomon, 1979). The method consisted essentially of two phases: incubation of plasma membranes with various ligands in the presence of labelled ATP, followed by Chromatographie separation of the reaction products. [3H]cAMP was added prior to chromatography, to allow correction for recovery (usually 70-80%). The reaction mix (50 pi) contained: 1 ) a phosphateregenerating system consisting of 5 mM creatine phosphate (sodium salt) and 50 U creatine kinase/ ml; 2) 5 mM magnesium acetate; 3) 0-5 mM ATP; 4)
mM cAMP; 5) 1 mM dithiothreitol; 6) 01 mg BSA/ml and 7) [a-32P]ATP (to give 2x 106 c.p.m./ assay) in 25 mM Tris—acetate buffer (pH 7-6). When added, the guanine nucleotides GTP and GppNHp
005
used at final concentrations of 10 pM and 01 pM respectively. Adenylyl cyclase activity was measured under basal conditions and in the presence of MnCl2 (20 mM), NaF (10 mM), prostaglandin E] (10 pM), isoproterenol (01 mM), forskolin (10 pM) or forskolin (10 pM)/GppNHp (0-1 nM). Experi¬ ments were conducted both in the presence and absence of 10 mM 3-isobutyl-l-methyl xanthine (IBMX). All assays were performed in quadrupli¬ cate. The interassay coefficient of variation of this assay was 10%.
were
dodecyl sulphate (SDS) electrophoresis and immunoblotting Plasma membrane protein was acid-precipitated, solubilized in Laemmli buffer (Laemmli, 1970) and boiled for 3 min before resolution by SDSpolyacrylamide gel electrophoresis (PAGE) on gels containing 10% acrylamide and 0-8% N,N'-bismethylene acrylamide. Immunoblotting with anti-peptide antisera was then performed as detailed previously (Gawler et al. 1987; Bushfield et al. 1990a). Briefly, proteins were transferred to nitrocellulose paper (Towbin et al. 1979) which was Sodium
then blocked for 2 h in 5% dried skimmed milk in buffer comprising 50 mM NaCl in 20 mM Tris-HCl (pH 7-5). After washing, the blot was incubated at 37 °C overnight with antiserum (at a 1: 200 dilution in 1% gelatin in phosphate-buffered saline). Follow¬ ing a further wash, the blot was then incubated for 2 h with the second antibody (horseradish peroxidI-labelled ase-labelled donkey anti-rabbit IgG or anti-rabbit IgG). The peroxidase-labelled bands were visualized with o-dianisidine (Milligan et al. 1987). Autoradiography using Kodak X-Omat AR Film (Eastman Kodak Co. Ltd, Rochester, NY, U.S.A.) was used to locate I-labelled bands, which were then cut out and counted on a NE 1612 gamma counter. The counts obtained were expressed rela¬ tive to the counts obtained by cutting out and counting the bind obtained by subjecting a standard human platelet membrane preparation to
electrophoresis.
Identical amounts of sample and control mem¬ branes were loaded. The quantities were determined by subjecting various amounts of plasma membrane protein to SDS-PAGE to obtain linear relationships between applied protein and radioactivity detected upon counting excised labelled bands, identified using I-labelled second antibody. Membrane pro¬ tein was loaded in 100 pg aliquots to achieve optimal
High molecular weight protein standards (Gibco BRL, Life Technologies Ltd, Paisley, Strathclyde, U.K.) were run in parallel with the samples under study. In addition, a standard plasma membrane preparation derived from human plate¬ lets was included on all blots, to facilitate the com¬ parison of data from different blots. The binding characteristics of the primary anti¬ bodies CS1, SGI, 13B and BN1 have all been described in full previously (Gawler et al. 1987; Goldsmith et al. 1987; Milligan, 1988; Bushfield et al. 1990a,b). Antiserum CS1 was raised against the C-terminal decapeptide of Gsct, and recognizes the 44 kDa and 42 kDa forms of Gsa. Antiserum SGI, raised against the C-terminal decapeptide of rod transducin, was used to detect Gia2. The decapeptide corresponding to amino acids 345-354 of Gia3 was identified specifically with the antibody 13B. BN1 antibody to the N-terminal decapeptide recognized blots.
and ß2 subunits. After blotting, the residual protein on the gels, which had not been transferred electrophoretically, was visualized by staining with Coomassie blue (0-25% in 45% methanol/10% glacial acetic acid), with subsequent destaining in 45% methanol/10% glacial acetic acid.
ßi
probes Synthetic 3 3-mer oligodeoxynucleotides comple¬ mentary to bases of the mRNAs encoding the a subGeneration of oligonucleotide
units of Gia2 (amino acids 125-135) and Gs (amino acids 147-157) were generated as previously described (Bushfield et al. 1990a; Griffiths et al. 1990). These probes were labelled at the 5' end with P (6000 Ci/mol; Amersham International). The incorporated radioactivity averaged 10 000Bq/pg
probe.
VSMC were cultured and exposed to 10 nM dexamethasone or saline vehicle for 24 and 48 h, exactly as described above. Total RNA was extracted from cells using the RNAzol B method (Biogenesis Ltd, Bournemouth, Dorset, U.K.) according to the manufacturer's instructions. RNA extraction and Northern
blotting Total cellular RNA was denatured by incubation at 65 °C with 2 M formaldehyde and deionized 50% (v/v) formamide. It was resolved in 1 -2% agarose gels containing 2-2 M formaldehyde before transfer to Hybond N nylon filters (Amersham International). Filters were prehybridized at 37 °C for 4 h in a mix¬ ture containing: 1) 30% formamide; 2) 5 x strength buffer containing 900 mM NaCl, 50 mM NaH2PO+
and 5 mM EDTA, pH 7-4; 3) 5 x Denhardt's solu¬ tion containing 1 mg/ml each of Ficoll 44, BSA and polyvinylpyrrolidone; 4) 01% SDS and 5) 200 pg denatured salmon testes DNA/ml. The RNA on the filters was then hybridized (42 °C, 16 h) in the same buffer with the relevant oligodeoxynucleotide probes
(5x 106 c.p.m./ml).
The blots were washed for 30 min at 55 °C to a final stringency of 0-5 x strength buffer containing 75 mM NaCl, 7-5 mM trisodium citrate, pH 70, and 0-1% SDS. The bound probes were localized by autoradiography using Kodak X-Omat AR film at -80 °C.
presence of
MnCl2 or NaF tended to be higher in the dexamethasone-treated cells, although the differences did not achieve statistical significance. Similar results were obtained after exposure to dexamethasone for 48 h. For example, the results of one experiment performed under basal conditions, in the presence of 10 pM prostaglandin E] and in the presence of 01 mM isoproterenol were (control vs dexamethasone) 931 vs 2408, 1398 vs 2955 and 1242 vs 2951 pmol cAMP/15 min per mg respectively. Inclusion of the cAMP phosphodiesterase inhibitor IBMX increased adenylyl cyclase activities by about 10%, but the increment was similar in both dexamethasone-treated and control membranes.
Statistics
Non-parametric data Whitney U tests.
were
compared using Mann-
RESULTS
Adenylyl cyclase studies Adenylyl cyclase activities are shown in Fig. 1. Exposure of VSMC to 10 nM dexamethasone for 24 h was associated with significantly greater adenylyl cyclase activity under basal conditions and in the presence of prostaglandin E), isoproterenol or forskolin. Adenylyl cyclase activity measured in the
Western blotting and Northern blotting for protein subunits
Data obtained from the Western blot studies are shown in Figs 2, 3 and 4. No differences were seen in Gia2 subunit levels between dexamethasone-treated and control cell membranes (dexamethasone vs control; meansi s.E.M., w 8; 2193-7±62-3 vs 2255T ± 38 c.p.m.). Similarly, no differences were seen in either the 44 kDa (dexamethasone vs con¬ trol; 1564-9±65-8 vs 1497-4±46 c.p.m.) or the 42 kDa bands of Gsa (1331-24=181-2 vs 1355-4± 165-7 c.p.m.), or in Gia3 (dexamethasone vs control; 1832-6±91-2 vs 1648-6±57-4 c.p.m.). The overall =
figure 1. Adenylyl cyclase activities (means ± s.n.M.) in plasma membranes from dexamethasone-treated (10 nM for 24 h; shaded bars) and control (solid bars) vascular smooth muscle cells. Assay conditions and ligands studied are indicated along the abscissa; PGE] prostaglandin Ej and GppNHp guanylylimidodiphosphate. The number of experiments (n) 5, except for basal conditions (w 6) and in the presence of forskolin 4-0-1 nM GppNHp (« 3). Statistical comparisons were by MannWhitney U test; *P
Tawil,
F.
Lyall,
G.
Milligan, J.
M. C. Connell and
MRC Blood Pressure Unit, Western Infirmary, Dumbarton Road, Glasgow Gil 6NT, U.K. *Molecular Pharmacology Group, Department of Biochemistry, University of Glasgow, Glasgow Gil 6NT, U.K. received
23 March 1992
ABSTRACT
Dexamethasone administration in vitro has been shown to increase adenylyl cyclase activity in vascular smooth muscle cells (VSMC) from renal arteries and in non-vascular cell lines. To investigate whether G proteins are involved in this response, cultured VSMC from mesenteric arteries of Sprague\p=m-\Dawleyrats were incubated in the presence and absence of 10 nm dexamethasone for 24 and 48 h. Basal and stimulated adenylyl cyclase activities were increased by approximately 50% after treatment with dexamethasone. The changes were neither specifically associated with ligands which stimulate adenylyl cyclase catalytic unit via Gs (isoproterenol and prostaglandin E1) nor with guanylylimidodiphosphate (0\m=.\1 nM), which inhibits the
INTRODUCTION
catalytic unit via Gi. This suggests that dexamethasone enhances adenylyl cyclase activity through changes at the level of the catalytic unit, rather than through the G proteins which modulate its activity. No differences were seen in immunoblotting studies of the levels of Gi\g=a\2,Gs\g=a\,Gi\g=a\3 and \g=b\ subunits. Similarly, dexamethasone had no effect on the expression of mRNA for Gi\g=a\2and Gs\g=a\. The results indicate that glucocorticoid-induced increases of adenylyl cyclase activity are due to changes at the level of the adenylyl cyclase catalytic unit rather than alteration of the levels or turnover of Gs\g=a\,Gi\g=a\2, Gi\g=a\3and \g=b\subunits in the membranes of VSMC. Journal of Molecular Endocrinology (1992) 9,
al. 1976; Handa et al. 1984; Whitworth et al. 1986; Russo Iijima Malik, 1985; et al. 1990). It is clear, however, that glucocorticoid hypertension is induced through glucocorticoid receptor-dependent mechanisms (Grunfeld et al. 1985), and glucocorticoid receptors have been iden¬ tified in arterial vessels from a number of species including rats (Meyer & Nichols, 1981) and humans (Scott et al. 1987). Post-receptor events, particularly in cultured vascular smooth muscle cells (VSMC), have not been extensively studied. Several intracellular messengers are involved in changes of vascular tone mediated by prostaglandins, vasopressin and a and ß adrenoreceptors, including adenylyl cyclase/ cyclic AMP (cAMP), the inositol phosphates and cyclic GMP. Dexamethasone administration in vitro has been shown to increase adenylyl cyclase activity in VSMC from renal arteries (Yasunari et al. 1989), and similar observations have been made in a number of non-vascular cell lines (Foster & Perkins, 1977; Johnson & Jaworski, 1983; Rizzoli et al. 1986; et
al. 1975; Schomig &
The association between glucocorticoids and hyper¬ tension is well established; for example, more than 70% of patients with Cushing's syndrome are hyper¬ tensive (Plotz et al. 1952; Orth & Liddle, 1971). Furthermore, hypertension can be induced experi¬ mentally in rats (Knowlton & Loeb, 1957; Krakoff et al. 1975; Haack et al. 1977; Suzuki et al. 1982) and in many other species, using a range of glucocor¬ ticoids including dexamethasone (Suzuki et al. 1982). Although the mechanisms underlying glucocorticoid hypertension remain to be fully elucidated, a number of glucocorticoid-mediated effects have been implicated; these include increased plasma vol¬ ume (Haack et al. 1977), stimulation of the reninangiotensin system by increased plasma renin substrate concentration (Krakoff et al. 1975), decreased prostaglandin synthesis (Handa et al. 1983) and enhancement of vascular reactivity to vari¬ ous pressor agents (Mendlowitz et al. 1958; Krakoff
237-244
et
Rodan & Rodan, 1986; Chang & Bourne, 1987). As G proteins have key roles in transducing membrane signals from receptors to intracellular messengers, the possibility that glucocorticoids exert their effects on adenylyl cyclase activity by alteration of G pro¬ teins has been explored. Two G proteins, Gsa and Gia2 (Bushfield et al. 1990c), modulate the activity of adenylyl cyclase; these respectively mediate stimula¬ tion and inhibition of the enzyme. Glucocorticoids have been shown to have contrasting effects on these G proteins when studied in vitro (Rodan & Rodan, 1986; Chang & Bourne, 1987) and in vivo (Haigh et al. 1990). Increases in Gso (Rodan & Rodan, 1986; Chang & Bourne, 1987), with either no change (Chang & Bourne, 1987) or an increase (Rodan & Rodan, 1986) in G¡, have been reported to be associ¬ ated with increased adenylyl cyclase activity (Chang & Bourne, 1987) following glucocorticoid exposure in vitro in two non-vascular cell lines. However, glucocorticoid administration to adrenalectomized animals in vivo has been reported to decrease the Gs : G¡ ratio in rat aortic plasma membranes. This would tend to inhibit adenylyl cyclase activity, favouring vasoconstriction with increased pressor sensitivity (Haigh et al. 1990). The purpose of this study was to reassess the effects of glucocorticoids on cultured VSMC in vitro, and to ascertain whether changes in adenylyl cyclase activity are associated with an alteration in G protein levels.
MATERIALS AND METHODS
All chemicals
supplied by Sigma Chemical Co., Poole, Dorset, U.K., with the exception of GTP, guanylylimidodiphosphate (GppNHp), creatine kinase and creatine phosphate which were obtained from Boehringer Mannheim, Lewes, East Sussex, U.K., and the radioisotopes [32P]ATP, [5',83H]cAMP and ,25I-labelled donkey anti-rabbit IgG, which were supplied by Amersham International pic, Amersham, Bucks, U.K. Horseradish-labelled donkey anti-rabbit IgG was supplied by the Scottish Antibody Production Unit, Law Hospital, Carluke, Strathclyde, U.K. were
Cell culture
VSMC were isolated from the mesenteric beds of male Sprague-Dawley rats (weighing 300-350 g) and cultured as previously described (Balmforth et al. 1988). Briefly, fat was removed by blunt dissec¬ tion and the vascular tree transferred to an enzyme dissociation mixture (10 ml) of collagenase (1 -25 mg/ ml), elastase (005 mg/ml) and soya bean trypsin
inhibitor (0-1%) in Hank's balanced salt solution. After incubation at 37 °C for 10—15 min, the arteries were transferred to fresh Dulbecco's modified Eagle's medium (DMEM) and any remaining fat, adventitia and endothelial cells were removed by careful dissection. The cleaned arteries were cut into small pieces, transferred to fresh enzyme dissocia¬ tion mixture and incubated at 37 °C with periodic trituration until a single cell suspension was obtained. The cell suspension was centrifuged (5 °C, 200 £ for 5 min) and the pellet resuspended in DMEM containing 10% fetal calf serum (FCS) and 10% horse serum, 2mM L-glutamine, 100 U penicillin/ml and 100 pg streptomycin/ml. The dis¬ persed cells were plated in 25 cm culture flasks and maintained in a humidified atmosphere of 5% C02, 95% air at 37 °C. Cells were subcultured as required up to the 7th passage. The effects of dexamethasone on cultured cells were tested when cells were 75% confluent. The medium was changed to one contain¬ ing 10% FCS±10nM dexamethasone and the cells were incubated for 24—48 h.
Plasma membrane
preparation
Cells were scraped from flasks and homogenized using a Polytron (Kinematica, Lucern, Switzerland) at setting 10 for 15 s. The homogenate was first cen¬ trifuged at 1000 j' for 10 min and the supernatant was centrifuged again at 37 000 £ for 30 min. The final plasma membrane pellet was suspended in 10 mM Tris-HCl buffer (pH 7-4) and stored in aliquots at —80 °C. Protein concentrations in the membrane preparations were measured against bovine serum albumin (BSA) standards by the method of Peterson (1977). The purity of the plasma membrane preparations was assessed by measuring 5' nucleotidase (EC 3.1.3.5) activity (Dixon & Purdom, 1954).
Adenylyl cyclase assays Adenylyl cyclase activity was assessed by measuring
[a-32P]cAMP
released from
[a-32P]ATP
substrate
(Salomon, 1979). The method consisted essentially of two phases: incubation of plasma membranes with various ligands in the presence of labelled ATP, followed by Chromatographie separation of the reaction products. [3H]cAMP was added prior to chromatography, to allow correction for recovery (usually 70-80%). The reaction mix (50 pi) contained: 1 ) a phosphateregenerating system consisting of 5 mM creatine phosphate (sodium salt) and 50 U creatine kinase/ ml; 2) 5 mM magnesium acetate; 3) 0-5 mM ATP; 4)
mM cAMP; 5) 1 mM dithiothreitol; 6) 01 mg BSA/ml and 7) [a-32P]ATP (to give 2x 106 c.p.m./ assay) in 25 mM Tris—acetate buffer (pH 7-6). When added, the guanine nucleotides GTP and GppNHp
005
used at final concentrations of 10 pM and 01 pM respectively. Adenylyl cyclase activity was measured under basal conditions and in the presence of MnCl2 (20 mM), NaF (10 mM), prostaglandin E] (10 pM), isoproterenol (01 mM), forskolin (10 pM) or forskolin (10 pM)/GppNHp (0-1 nM). Experi¬ ments were conducted both in the presence and absence of 10 mM 3-isobutyl-l-methyl xanthine (IBMX). All assays were performed in quadrupli¬ cate. The interassay coefficient of variation of this assay was 10%.
were
dodecyl sulphate (SDS) electrophoresis and immunoblotting Plasma membrane protein was acid-precipitated, solubilized in Laemmli buffer (Laemmli, 1970) and boiled for 3 min before resolution by SDSpolyacrylamide gel electrophoresis (PAGE) on gels containing 10% acrylamide and 0-8% N,N'-bismethylene acrylamide. Immunoblotting with anti-peptide antisera was then performed as detailed previously (Gawler et al. 1987; Bushfield et al. 1990a). Briefly, proteins were transferred to nitrocellulose paper (Towbin et al. 1979) which was Sodium
then blocked for 2 h in 5% dried skimmed milk in buffer comprising 50 mM NaCl in 20 mM Tris-HCl (pH 7-5). After washing, the blot was incubated at 37 °C overnight with antiserum (at a 1: 200 dilution in 1% gelatin in phosphate-buffered saline). Follow¬ ing a further wash, the blot was then incubated for 2 h with the second antibody (horseradish peroxidI-labelled ase-labelled donkey anti-rabbit IgG or anti-rabbit IgG). The peroxidase-labelled bands were visualized with o-dianisidine (Milligan et al. 1987). Autoradiography using Kodak X-Omat AR Film (Eastman Kodak Co. Ltd, Rochester, NY, U.S.A.) was used to locate I-labelled bands, which were then cut out and counted on a NE 1612 gamma counter. The counts obtained were expressed rela¬ tive to the counts obtained by cutting out and counting the bind obtained by subjecting a standard human platelet membrane preparation to
electrophoresis.
Identical amounts of sample and control mem¬ branes were loaded. The quantities were determined by subjecting various amounts of plasma membrane protein to SDS-PAGE to obtain linear relationships between applied protein and radioactivity detected upon counting excised labelled bands, identified using I-labelled second antibody. Membrane pro¬ tein was loaded in 100 pg aliquots to achieve optimal
High molecular weight protein standards (Gibco BRL, Life Technologies Ltd, Paisley, Strathclyde, U.K.) were run in parallel with the samples under study. In addition, a standard plasma membrane preparation derived from human plate¬ lets was included on all blots, to facilitate the com¬ parison of data from different blots. The binding characteristics of the primary anti¬ bodies CS1, SGI, 13B and BN1 have all been described in full previously (Gawler et al. 1987; Goldsmith et al. 1987; Milligan, 1988; Bushfield et al. 1990a,b). Antiserum CS1 was raised against the C-terminal decapeptide of Gsct, and recognizes the 44 kDa and 42 kDa forms of Gsa. Antiserum SGI, raised against the C-terminal decapeptide of rod transducin, was used to detect Gia2. The decapeptide corresponding to amino acids 345-354 of Gia3 was identified specifically with the antibody 13B. BN1 antibody to the N-terminal decapeptide recognized blots.
and ß2 subunits. After blotting, the residual protein on the gels, which had not been transferred electrophoretically, was visualized by staining with Coomassie blue (0-25% in 45% methanol/10% glacial acetic acid), with subsequent destaining in 45% methanol/10% glacial acetic acid.
ßi
probes Synthetic 3 3-mer oligodeoxynucleotides comple¬ mentary to bases of the mRNAs encoding the a subGeneration of oligonucleotide
units of Gia2 (amino acids 125-135) and Gs (amino acids 147-157) were generated as previously described (Bushfield et al. 1990a; Griffiths et al. 1990). These probes were labelled at the 5' end with P (6000 Ci/mol; Amersham International). The incorporated radioactivity averaged 10 000Bq/pg
probe.
VSMC were cultured and exposed to 10 nM dexamethasone or saline vehicle for 24 and 48 h, exactly as described above. Total RNA was extracted from cells using the RNAzol B method (Biogenesis Ltd, Bournemouth, Dorset, U.K.) according to the manufacturer's instructions. RNA extraction and Northern
blotting Total cellular RNA was denatured by incubation at 65 °C with 2 M formaldehyde and deionized 50% (v/v) formamide. It was resolved in 1 -2% agarose gels containing 2-2 M formaldehyde before transfer to Hybond N nylon filters (Amersham International). Filters were prehybridized at 37 °C for 4 h in a mix¬ ture containing: 1) 30% formamide; 2) 5 x strength buffer containing 900 mM NaCl, 50 mM NaH2PO+
and 5 mM EDTA, pH 7-4; 3) 5 x Denhardt's solu¬ tion containing 1 mg/ml each of Ficoll 44, BSA and polyvinylpyrrolidone; 4) 01% SDS and 5) 200 pg denatured salmon testes DNA/ml. The RNA on the filters was then hybridized (42 °C, 16 h) in the same buffer with the relevant oligodeoxynucleotide probes
(5x 106 c.p.m./ml).
The blots were washed for 30 min at 55 °C to a final stringency of 0-5 x strength buffer containing 75 mM NaCl, 7-5 mM trisodium citrate, pH 70, and 0-1% SDS. The bound probes were localized by autoradiography using Kodak X-Omat AR film at -80 °C.
presence of
MnCl2 or NaF tended to be higher in the dexamethasone-treated cells, although the differences did not achieve statistical significance. Similar results were obtained after exposure to dexamethasone for 48 h. For example, the results of one experiment performed under basal conditions, in the presence of 10 pM prostaglandin E] and in the presence of 01 mM isoproterenol were (control vs dexamethasone) 931 vs 2408, 1398 vs 2955 and 1242 vs 2951 pmol cAMP/15 min per mg respectively. Inclusion of the cAMP phosphodiesterase inhibitor IBMX increased adenylyl cyclase activities by about 10%, but the increment was similar in both dexamethasone-treated and control membranes.
Statistics
Non-parametric data Whitney U tests.
were
compared using Mann-
RESULTS
Adenylyl cyclase studies Adenylyl cyclase activities are shown in Fig. 1. Exposure of VSMC to 10 nM dexamethasone for 24 h was associated with significantly greater adenylyl cyclase activity under basal conditions and in the presence of prostaglandin E), isoproterenol or forskolin. Adenylyl cyclase activity measured in the
Western blotting and Northern blotting for protein subunits
Data obtained from the Western blot studies are shown in Figs 2, 3 and 4. No differences were seen in Gia2 subunit levels between dexamethasone-treated and control cell membranes (dexamethasone vs control; meansi s.E.M., w 8; 2193-7±62-3 vs 2255T ± 38 c.p.m.). Similarly, no differences were seen in either the 44 kDa (dexamethasone vs con¬ trol; 1564-9±65-8 vs 1497-4±46 c.p.m.) or the 42 kDa bands of Gsa (1331-24=181-2 vs 1355-4± 165-7 c.p.m.), or in Gia3 (dexamethasone vs control; 1832-6±91-2 vs 1648-6±57-4 c.p.m.). The overall =
figure 1. Adenylyl cyclase activities (means ± s.n.M.) in plasma membranes from dexamethasone-treated (10 nM for 24 h; shaded bars) and control (solid bars) vascular smooth muscle cells. Assay conditions and ligands studied are indicated along the abscissa; PGE] prostaglandin Ej and GppNHp guanylylimidodiphosphate. The number of experiments (n) 5, except for basal conditions (w 6) and in the presence of forskolin 4-0-1 nM GppNHp (« 3). Statistical comparisons were by MannWhitney U test; *P
