- Email: [email protected]
Pituitary adenylate cyclase-activating polypeptide: a novel, long-lasting, endothelium-independent vasorelaxant
European Journal of Pharmucofogy, 197 (1991) 131- 134 @ 1991 Ekevier Science Publishers B.V. 0014-2999/91/$03.50 ADONIS 00~4299991#~~8Q
EJP 51860
John B. Warren, Louise E. Donnelly, Seamus Cullen, Blair E. Robertson *. Mohammad Stephen R. Bloom ’ and John MacDermot ’ ~e~ar~~~en~sof ~~~nico~Pho~acoIo~
A. Ghatei
t,
and medicine, Royal P~~gradua~e Medico! School, London, U.K. and ’ ~niuers~~~I Laboratory of Fh~sioIo~, Oxfold, U.K.
Received 29 November 1990, revised MS received 20 February 1991, accepted 26 February 1991
The vasoactivity of the 27- and 38-amino acid forms of the novel peptide pituitary adenylate cyclase-activating polypeptide (PACAP) was tested in vitro. Both forms of PACAP caused endothelium-jndependent vasodilation (assayed by their vasodilator action on rabbit aorta). When superfused for 1 min the relaxation EC,, of PACAP was 23 & 8 nM and of PACAP was 152 f 66 nM. PACAP was 100-fold more potent than vasoactive intestinal polypeptide (VIP) (PACAP shows 68% amino acid sequence homology with VIP), and had a prolonged duration of action, a 1 min exposure to 1 PM PACAP lasting 135 f 7 min and to I FM PACAP 108 f 3 min. Adenylate cyclase activity in homogenates of rabbit aortic smooth muscle cells was increased by PACAP and PACAP with EC&s of 4.4 and 0.73 nM, respectively. PACAP and PACAP are potent. Iong-lasting, endothelium-independent vasodilators. Pituitary adenylate cyclase-activating polypeptide; Vasodilatation; Artery: VIP (vasoactive intestinal poiypeptide): Adenylate cyclase; Endothelium; EDRF (endothelium-delved relaxing factor)
Pituitary adenylate cyclase activating polypeptide (PACAP) is a novel neuropeptide which has recently been isolated from ovine hypothala~ and found to be a potent stimulus of adenylate cyclase of anterior pituitary cells in culture (Miyata et al., 1989). Both a 27amino acid residue (PACAP27) and a 38-amino acid residue form (PACAP38, in which the first 27 residues are identical to PACAP27) have been described (Kimura et al., 1990). High levels of binding were found in rat lung and brain with low levels in the aorta (Lam et al., in press). It has furthermore been shown that PACAP mRNA is expressed in several mammalian tissues (Kimura et al., 1990). PACAP has considerable structural homology with vasoactive intestinal polypeptide (VIP), althou~ it is over 1000 times more potent than VIP at stimulating adenylate cyclase in cultured neural cells (Miyata et al., 1989). Preliminary experiments suggested that both forms of the peptide cause long-lasting hypotension when
Correspondence to: J.B. Warren, Applied Pharmacology, National Heart and Lung Institute, Dovehouse Street, London SW3 6LY. U.K.
injected into the anaesthetised rat and this has recently been reported by others (Miyata et al., 1990). We set out to determine the vasoactivity of PACAP and PACAP on isolated aortic rings with and without endothelium. Results were compared with VIP, a known endothelium-independent vasorelaxant, as the N-terminal sequence of PACAP shows 68% homology with VIP (Miyata et a1.,1989).
2. Materials and methods 2.1. ~iaassa~ studies The bioassay tissue consisted of 3 mm rings of lower thoracic aorta taken from male New Zealand white rabbits. Animals were first overdosed with i.v. pentobarbitone and the recommendations from the declaration of Helsinki and the internationally accepted principles in the care and use of experimental animals were adhered to. Rings were mounted on hooks attached to a Grass model 79D force transducer and recorder (Grass Instrttment Co., Quincy, MA) as previously described (Warren et al., 1990). They were maintained at 37°C and continuously perfused at 2 ml/min with KrebsHenseleit buffer (mM: NaCl 118, KC1 4.7, MgSQ,
1.2, NaHCO, 25, glucose 11, with 5% CO, in air. The bioassay over a minimum period of 100 min to ~~~ai~t~~a resting tension of 2 g, then perfused with r c~~tai~~ng potassium 130 mM for 3 min and ed a feather 45 min to return to baseline before ~~~t~cting with phenyiephrine. The endothehum was removed from some rings. before nlounti~g on hooks. by gently rotating around the closed tips of a pair of fine forceps, The rings were contracted by perfusing with buffer containing I @M pheny~ephrine and the absence or presence of endothehmn confirmed by det~~i~in~ the response to 1 FM acetylcho~ine in phenytephrin~ont~~ng buffer given for 1 min. Rings ~vithout endothehum contracted with phenylephrine generated a constaut force for over 3 h. Albumin (0.3% w/v) was added to the buffer for peptide dilutions and also to the bioassay perfusate to prevent peptide sticking to plastic tubing. All peptide dilutions were perfused over the bioassay tissue for 1 min. the concentration of phenylephrine being m~nt~ned at 1 PM. The relaxation response af each bioassay ring was determined by giving 1 FM nitroprusside for 1 rnin and resufts are expressed as a percentage of this response.
Adenylate cyclase activity was measured in a homogenate of the medial layer of rabbit thoracic aorta. The whole descending thoracic aorta was removed from a New Zealand white male rabbit and kept moist with buffer. The vessel was opened Iongitudindhy and tht: e~dothe~um removed by r~eated~y wiping the intima~ surface with wet filter paper. The adventitia was scraped off with a scalpel and the remaining media put into buffer at 4OC (25 mM Tris-HCl, 0.29 mM sucrose, aprotinin 1000 U/ml, pH 7.4). This was homogenized using a Polytron (~nematica, Switzerland). Adenylate cyclase was assayed as previously described (Edwards et al.. 1987; Kelly et al., 1990; Salomon et al., 1974), Reaction mixtures of 100 ~1 contained 50 mM Tris-HCI pH 7.4. 5 mM MgCl, 20 mM creatine phosphate disodium salt, 10 IU creatine kinase, 1 mM cyclic AMP, 0.25 mM Ro20-1724 as a phosphodiesterase inhibitor, 1 mM [ti3-‘F]ATP (2 PCi), 1 pM GTP and 20-30 pg of arterial media protein. Reactions were incubated at 37OC for 15 min and then te~nated by adding 800 ,ul of 6.25% (w/v) trichloroacetic acid. To each tube was added 100 ~1 of 18-3HJcyclic AMP (appro~mate~y 10000 cpm) and the reaction mixtures centrifuged at 4°C for 20 ruin at 800 X g. The E3*P]ATP and [32P]cychc AMP were separated by a two-step c~omatography procedure using a Dowex AC 5OW-X4 (200-400 mesh, BioRad Ltd.) and neutral alumina. Losses of [ 32Pfcychc AMP were corrected for by the measnremeny of the
recovery of f zH]cyclic AMP. Each measurement an assay was performed in triplicate.
within
2 3. Drugs and chemicals PACAP and PACAP were synthesized to order by Peptide Product Ltd. (Southampton, UK) and VIP was obtained from Novabiochem, Nottingham, UK. [8- ’ Hj Adenosine 3’ : 5‘-cyclic monophosphate (23.6 Ci mmol - ’ ) and [ &“P]ATP (40-50 Ci mmol- ’ ) were obtained from Amersham international. Ro20-1724 (4-(3butoxy-~methoxybenzy~)“2-i~dazolidinone) was a kind gift from Roche Products Ltd. (Welwyn Garden City, UK). All other drugs and chemicals were from the Sigma Chemical Co. (Poole, Dorset).
3. Results 3. I. Vmodilatcw acriuity When given to rings under resting tension only, neither PACAP27, PACAP nor VIP had any effect. The relaxation of preconst~cted rings was not affected by the presence of endothelium and tb.erefore only the results with de-endotheliaIized rings are shown. The upper half of fig. 1 shows a typi\:al trace of the long duration of action of PACAP on a precontracted aortic ring and the lower half shows the concentrationresponse reiationships of PACAP27, PACAP and VJP. The relaxant effect on the bioassay tissue showed an
PACAWB
Fig. 1. The upper trace shows the force generated by a de-endothelialked rabbit thoracic aortic ring which had been ~~tinuousIy contracted with 1 gM phenyleph~ne (PE) and then sup~Fusedfor 1 min with 1 PM nitroprusside (NP) followed by 300 nM PACAP given for 1 min (PACAP27). The lower graph shows the mean ( +S.E.M., n =i 6 preparations in each case) effect of PACAP27, PACAMS and VIP on a de”endotheiialized rat aortic ring which had been precontracted with phenylephrine 1 FM, Results are expressed as a W of the rel~ati~n in response to nitropru~ide (NP) 1 FM.
133 TABLE 1 Duration of action of PACAP27, PACAP and VIP superfused for 1 min over a de-endothehalized rabbit thoracic aorta which had been precontracted with 1 pM phenylephrine Peak effect (mm)
3x10-8 lo-’ 3x10-7 10-6 VIP 10-s
Duration (mm)
PACAP
PACAP
PACAP
PACAP
5.3 +0.6 8.5 f 1.6 11.3*1.0 12.8k1.4 5.OkO.2
4.6kO.6 7.0 + 0.9 lO.O+_1.8 12.3 kO.6
32+3 33+2 56+2 137&S 34&2
20+3 39*4 75&t 108&3
EC,, of 23 & 8 nM for PACAP and 152 + 66 nM for PACAP (means f S.E.M., n = 6, 100% = relaxation caused by 1 PM nitroprusside). Table 1 shows that both forms of PACAP were slow to act, the peak bioassay response occurring 12-13 min after a 1 min exposure to a concentration of 1 PM and lasting approximately 2 h. 3.2. Adenylate cyclase activity Both PACAP and PACAP caused concentration-dependent increases in adenylate cyclase activity in homogenates of smooth muscle prepared from rabbit aorta (fig. 2). The EC,, for adenylate cyclase activation was 4.4 nM for PACAP and 0.73 nM for PACAP38. This activation of adenylate cyclase was found to be linear over the first 20 min of incubation in a time course experiment. EC+ were calculated for the cyclase assay by iterative non-linear least squares regression 20aI
15IO% ii : e
5
/ O10
z
9
9
log [PACAP
10
9
8
log[PACAPB]
7
6
7
6
M
M
Fig. 2. Activation of adenylate cyclase by PACAP (a) and PACAP (b). The vertical axis shows the increase in activity (Sv) above basal activity. Results are the meansrtS.E.M. of data from a typical single experiment performed in triplicate. Each experiment was performed three times. The basal activity of adenylate cyclase was 13.7k 1.6 pmol cyclic AMP min- ’ mg protein-’ (n = 3). The EC,, was 4.4 nM for PACAP and 0.73 nM for PACAP 38.
analysis for a simple bi-molecular interaction between the ligand and a single class of receptors (GraphPAD). In a comparison of the efficacy of PACAP27, PACAP38, VIP and isoprenaline, measurements were made of adenylate cyclase activity in the presence of each compound at, or near, a saturating concentration. VIP (1 PM) induced an increase of 125 + 8% above basal enzyme activity which compared with 131 5 9% for VIP (10 PM), 134 f 14% for isoprenahne (10 PM). 237 + 25% for PACAP (1 PM) and 206 f 12% for PACAP (1 PM).
4. Discussion Both the 27-amino acid and the 38-amino acid fotms PACAP were of similar potency as vasodilators of rabbit aortic rings and approximately 100 times more potent than VIP. This suggests that the activity of PACAP is restricted to the first 27 amino acid residues that are identical to PACAP27. Both forms of PACAP had a surprisingly long duration of action; the relaxation of the tissue after a 1 min exposure to a micromolar concentration lasted approximately 2 h. This could be explained by tight binding to the PACAP receptor. Specific antagonists would be needed to determine if endogenous PACAP contributes to blood vessel tone in vivo. Binding sites are present at low levels in rat aorta (Lam et al., in press) but it is not known if PACAP is present in blood vessels in vivo. Even small quantities of PACAP in the presence of a low density of receptor sites could, nonetheless. have a physiological role because of its high potency (IOO-fold greater than VIP). PACAP was a potent activator of adenylate cyclase in a suspension of homogenized aortic smooth muscle cells, thus confirming similar activity previously found using cultured anterior pituitary cells. It is possible that this is the mechanism whereby it causes vasorelaxation. However, the correlation between the concentration of the peptide that caused aortic relaxation or activation of adenylate cyclase was not close, possibly reflecting the shorter duration of exposure of the bioassay tissue to the peptide in the superfusion experiment. Altematively. this may reflect that there is a threshold stimulation of adenylate cyclase before vasorelaxation takes place, or that PACAP works by another intracellular mechanism. If PACAP does work via adenylate cyclase. this probably triggers the sequestration of calcium ions into the sarcoplasmic recticulum, the subsequent lowering of intracellular calcium concentration reducing the formation of the calcium-calmodulin complex hence reducing actin-myosin interaction (MacDermot. 1990). An alternative explanation, that cyclic AMP-dependent phosphorylation of myosin light chain kinase causes of
may
not incur
in viva ~~arnrn and Stull.
se from the anterior pituitary
is conspecialized circulation, which in turn ctive factors derived from the sites for PACAP have been tb~~~s and the anterior pituitary of the rat (Lam et al., in press} and this raises the ility that this potent vasodilator may play a part control of the anterior pituitary circulation. of PACAP tested were remarkably longtent vas~lators. It remains to be ascert~ned if s~f~c~ent peptide is present in the vasculature to have d ~a~opbys~olo~c~ role. but our results suggest that or& small amounts may have significant effects.
SC. was supported bv the British Heart Foundation and B.E.R. is a WeBcome Priie student. The work was supported in part by the Wellcome Trust.
eferen~s Edwards. R.J.. J. MacDermot and A.J. Wilkins, 1987, Prostacyclin analogues reduce ADP-ribosylation of the a-subuni: of the regu-
Iatory G,protein and di~nish adenosine (AZ) responsiveness of platelets. Br. J. Pharmacol. 90, 501. Kamm, W.E. and J.T. Stull. 1985, The function of myosin and myosin light kinase phosphorylation in smooth muscle. Ann. Rev. Pharmocol. Toxicol. 25, 593. Kelly, E., M. Keen, P. Nobbs and J. MacDermot, 1990, NaF and guanine nucieotides modulate adenylate cyclase activity in NGlOS-15 c&s by interacting with both G, and Gi, Br. 1. Pharmacol. 100, 223. Kimura, C., S. Ohkubo, K. Ogi, M. Hosoya, Y. Itoh, H. Onda. A. Miyata, L. Jiang. R.R. Dahl, H.H. Stibbs, A. Arimura and M. Fujino, 1990, A novel peptide which stimulates adenylate cyclase: molecular cloning and characterization of the ovine and human cDNA’s. B&hem. Biophys. Res. Co~un. 166, 81. Lam. H.-C.. K. Takahashi, M.A. Ghatei, SM. Kanse, J.M. Polak and S.R. Bfoom, Binding sites of a novel neuropeptide ‘pituitary adenylate cyclase activating polypeptide’ in the rat brain and lung, European J. B&hem. (in press), MacDermot. J.. 1990. Prostacylcin, cyclic AMP, and relaxation of vascular smooth muscle, Clin. Pharmacol. 7,49. Miyata, A., A. Arimura, R.R. D&i, N. Mina~no, A. Uehara. L. Jiang, M.D. Culler and D.H. Coy, 1989, Isolation of a novel 38 residue-hypothalamic polypeptide which stimulates adenylate cyclase in pituitary ceils, B&hem. Biophys. Res. Commun. 164, 567.
Salomon, Y., C. Londos and M. Rodell, 1974, A highly sensitive adenylate cyclase assay, Anal. B&hem. 58, 541. Warren, J.B., R. Loi, N.B. RendelI and G.W. Taylor, 1990, Nitric oxide is inactivated by the bacterial pigment pyocyanin, Biochem. J. 266, 921.
EJP 51860
John B. Warren, Louise E. Donnelly, Seamus Cullen, Blair E. Robertson *. Mohammad Stephen R. Bloom ’ and John MacDermot ’ ~e~ar~~~en~sof ~~~nico~Pho~acoIo~
A. Ghatei
t,
and medicine, Royal P~~gradua~e Medico! School, London, U.K. and ’ ~niuers~~~I Laboratory of Fh~sioIo~, Oxfold, U.K.
Received 29 November 1990, revised MS received 20 February 1991, accepted 26 February 1991
The vasoactivity of the 27- and 38-amino acid forms of the novel peptide pituitary adenylate cyclase-activating polypeptide (PACAP) was tested in vitro. Both forms of PACAP caused endothelium-jndependent vasodilation (assayed by their vasodilator action on rabbit aorta). When superfused for 1 min the relaxation EC,, of PACAP was 23 & 8 nM and of PACAP was 152 f 66 nM. PACAP was 100-fold more potent than vasoactive intestinal polypeptide (VIP) (PACAP shows 68% amino acid sequence homology with VIP), and had a prolonged duration of action, a 1 min exposure to 1 PM PACAP lasting 135 f 7 min and to I FM PACAP 108 f 3 min. Adenylate cyclase activity in homogenates of rabbit aortic smooth muscle cells was increased by PACAP and PACAP with EC&s of 4.4 and 0.73 nM, respectively. PACAP and PACAP are potent. Iong-lasting, endothelium-independent vasodilators. Pituitary adenylate cyclase-activating polypeptide; Vasodilatation; Artery: VIP (vasoactive intestinal poiypeptide): Adenylate cyclase; Endothelium; EDRF (endothelium-delved relaxing factor)
Pituitary adenylate cyclase activating polypeptide (PACAP) is a novel neuropeptide which has recently been isolated from ovine hypothala~ and found to be a potent stimulus of adenylate cyclase of anterior pituitary cells in culture (Miyata et al., 1989). Both a 27amino acid residue (PACAP27) and a 38-amino acid residue form (PACAP38, in which the first 27 residues are identical to PACAP27) have been described (Kimura et al., 1990). High levels of binding were found in rat lung and brain with low levels in the aorta (Lam et al., in press). It has furthermore been shown that PACAP mRNA is expressed in several mammalian tissues (Kimura et al., 1990). PACAP has considerable structural homology with vasoactive intestinal polypeptide (VIP), althou~ it is over 1000 times more potent than VIP at stimulating adenylate cyclase in cultured neural cells (Miyata et al., 1989). Preliminary experiments suggested that both forms of the peptide cause long-lasting hypotension when
Correspondence to: J.B. Warren, Applied Pharmacology, National Heart and Lung Institute, Dovehouse Street, London SW3 6LY. U.K.
injected into the anaesthetised rat and this has recently been reported by others (Miyata et al., 1990). We set out to determine the vasoactivity of PACAP and PACAP on isolated aortic rings with and without endothelium. Results were compared with VIP, a known endothelium-independent vasorelaxant, as the N-terminal sequence of PACAP shows 68% homology with VIP (Miyata et a1.,1989).
2. Materials and methods 2.1. ~iaassa~ studies The bioassay tissue consisted of 3 mm rings of lower thoracic aorta taken from male New Zealand white rabbits. Animals were first overdosed with i.v. pentobarbitone and the recommendations from the declaration of Helsinki and the internationally accepted principles in the care and use of experimental animals were adhered to. Rings were mounted on hooks attached to a Grass model 79D force transducer and recorder (Grass Instrttment Co., Quincy, MA) as previously described (Warren et al., 1990). They were maintained at 37°C and continuously perfused at 2 ml/min with KrebsHenseleit buffer (mM: NaCl 118, KC1 4.7, MgSQ,
1.2, NaHCO, 25, glucose 11, with 5% CO, in air. The bioassay over a minimum period of 100 min to ~~~ai~t~~a resting tension of 2 g, then perfused with r c~~tai~~ng potassium 130 mM for 3 min and ed a feather 45 min to return to baseline before ~~~t~cting with phenyiephrine. The endothehum was removed from some rings. before nlounti~g on hooks. by gently rotating around the closed tips of a pair of fine forceps, The rings were contracted by perfusing with buffer containing I @M pheny~ephrine and the absence or presence of endothehmn confirmed by det~~i~in~ the response to 1 FM acetylcho~ine in phenytephrin~ont~~ng buffer given for 1 min. Rings ~vithout endothehum contracted with phenylephrine generated a constaut force for over 3 h. Albumin (0.3% w/v) was added to the buffer for peptide dilutions and also to the bioassay perfusate to prevent peptide sticking to plastic tubing. All peptide dilutions were perfused over the bioassay tissue for 1 min. the concentration of phenylephrine being m~nt~ned at 1 PM. The relaxation response af each bioassay ring was determined by giving 1 FM nitroprusside for 1 rnin and resufts are expressed as a percentage of this response.
Adenylate cyclase activity was measured in a homogenate of the medial layer of rabbit thoracic aorta. The whole descending thoracic aorta was removed from a New Zealand white male rabbit and kept moist with buffer. The vessel was opened Iongitudindhy and tht: e~dothe~um removed by r~eated~y wiping the intima~ surface with wet filter paper. The adventitia was scraped off with a scalpel and the remaining media put into buffer at 4OC (25 mM Tris-HCl, 0.29 mM sucrose, aprotinin 1000 U/ml, pH 7.4). This was homogenized using a Polytron (~nematica, Switzerland). Adenylate cyclase was assayed as previously described (Edwards et al.. 1987; Kelly et al., 1990; Salomon et al., 1974), Reaction mixtures of 100 ~1 contained 50 mM Tris-HCI pH 7.4. 5 mM MgCl, 20 mM creatine phosphate disodium salt, 10 IU creatine kinase, 1 mM cyclic AMP, 0.25 mM Ro20-1724 as a phosphodiesterase inhibitor, 1 mM [ti3-‘F]ATP (2 PCi), 1 pM GTP and 20-30 pg of arterial media protein. Reactions were incubated at 37OC for 15 min and then te~nated by adding 800 ,ul of 6.25% (w/v) trichloroacetic acid. To each tube was added 100 ~1 of 18-3HJcyclic AMP (appro~mate~y 10000 cpm) and the reaction mixtures centrifuged at 4°C for 20 ruin at 800 X g. The E3*P]ATP and [32P]cychc AMP were separated by a two-step c~omatography procedure using a Dowex AC 5OW-X4 (200-400 mesh, BioRad Ltd.) and neutral alumina. Losses of [ 32Pfcychc AMP were corrected for by the measnremeny of the
recovery of f zH]cyclic AMP. Each measurement an assay was performed in triplicate.
within
2 3. Drugs and chemicals PACAP and PACAP were synthesized to order by Peptide Product Ltd. (Southampton, UK) and VIP was obtained from Novabiochem, Nottingham, UK. [8- ’ Hj Adenosine 3’ : 5‘-cyclic monophosphate (23.6 Ci mmol - ’ ) and [ &“P]ATP (40-50 Ci mmol- ’ ) were obtained from Amersham international. Ro20-1724 (4-(3butoxy-~methoxybenzy~)“2-i~dazolidinone) was a kind gift from Roche Products Ltd. (Welwyn Garden City, UK). All other drugs and chemicals were from the Sigma Chemical Co. (Poole, Dorset).
3. Results 3. I. Vmodilatcw acriuity When given to rings under resting tension only, neither PACAP27, PACAP nor VIP had any effect. The relaxation of preconst~cted rings was not affected by the presence of endothelium and tb.erefore only the results with de-endotheliaIized rings are shown. The upper half of fig. 1 shows a typi\:al trace of the long duration of action of PACAP on a precontracted aortic ring and the lower half shows the concentrationresponse reiationships of PACAP27, PACAP and VJP. The relaxant effect on the bioassay tissue showed an
PACAWB
Fig. 1. The upper trace shows the force generated by a de-endothelialked rabbit thoracic aortic ring which had been ~~tinuousIy contracted with 1 gM phenyleph~ne (PE) and then sup~Fusedfor 1 min with 1 PM nitroprusside (NP) followed by 300 nM PACAP given for 1 min (PACAP27). The lower graph shows the mean ( +S.E.M., n =i 6 preparations in each case) effect of PACAP27, PACAMS and VIP on a de”endotheiialized rat aortic ring which had been precontracted with phenylephrine 1 FM, Results are expressed as a W of the rel~ati~n in response to nitropru~ide (NP) 1 FM.
133 TABLE 1 Duration of action of PACAP27, PACAP and VIP superfused for 1 min over a de-endothehalized rabbit thoracic aorta which had been precontracted with 1 pM phenylephrine Peak effect (mm)
3x10-8 lo-’ 3x10-7 10-6 VIP 10-s
Duration (mm)
PACAP
PACAP
PACAP
PACAP
5.3 +0.6 8.5 f 1.6 11.3*1.0 12.8k1.4 5.OkO.2
4.6kO.6 7.0 + 0.9 lO.O+_1.8 12.3 kO.6
32+3 33+2 56+2 137&S 34&2
20+3 39*4 75&t 108&3
EC,, of 23 & 8 nM for PACAP and 152 + 66 nM for PACAP (means f S.E.M., n = 6, 100% = relaxation caused by 1 PM nitroprusside). Table 1 shows that both forms of PACAP were slow to act, the peak bioassay response occurring 12-13 min after a 1 min exposure to a concentration of 1 PM and lasting approximately 2 h. 3.2. Adenylate cyclase activity Both PACAP and PACAP caused concentration-dependent increases in adenylate cyclase activity in homogenates of smooth muscle prepared from rabbit aorta (fig. 2). The EC,, for adenylate cyclase activation was 4.4 nM for PACAP and 0.73 nM for PACAP38. This activation of adenylate cyclase was found to be linear over the first 20 min of incubation in a time course experiment. EC+ were calculated for the cyclase assay by iterative non-linear least squares regression 20aI
15IO% ii : e
5
/ O10
z
9
9
log [PACAP
10
9
8
log[PACAPB]
7
6
7
6
M
M
Fig. 2. Activation of adenylate cyclase by PACAP (a) and PACAP (b). The vertical axis shows the increase in activity (Sv) above basal activity. Results are the meansrtS.E.M. of data from a typical single experiment performed in triplicate. Each experiment was performed three times. The basal activity of adenylate cyclase was 13.7k 1.6 pmol cyclic AMP min- ’ mg protein-’ (n = 3). The EC,, was 4.4 nM for PACAP and 0.73 nM for PACAP 38.
analysis for a simple bi-molecular interaction between the ligand and a single class of receptors (GraphPAD). In a comparison of the efficacy of PACAP27, PACAP38, VIP and isoprenaline, measurements were made of adenylate cyclase activity in the presence of each compound at, or near, a saturating concentration. VIP (1 PM) induced an increase of 125 + 8% above basal enzyme activity which compared with 131 5 9% for VIP (10 PM), 134 f 14% for isoprenahne (10 PM). 237 + 25% for PACAP (1 PM) and 206 f 12% for PACAP (1 PM).
4. Discussion Both the 27-amino acid and the 38-amino acid fotms PACAP were of similar potency as vasodilators of rabbit aortic rings and approximately 100 times more potent than VIP. This suggests that the activity of PACAP is restricted to the first 27 amino acid residues that are identical to PACAP27. Both forms of PACAP had a surprisingly long duration of action; the relaxation of the tissue after a 1 min exposure to a micromolar concentration lasted approximately 2 h. This could be explained by tight binding to the PACAP receptor. Specific antagonists would be needed to determine if endogenous PACAP contributes to blood vessel tone in vivo. Binding sites are present at low levels in rat aorta (Lam et al., in press) but it is not known if PACAP is present in blood vessels in vivo. Even small quantities of PACAP in the presence of a low density of receptor sites could, nonetheless. have a physiological role because of its high potency (IOO-fold greater than VIP). PACAP was a potent activator of adenylate cyclase in a suspension of homogenized aortic smooth muscle cells, thus confirming similar activity previously found using cultured anterior pituitary cells. It is possible that this is the mechanism whereby it causes vasorelaxation. However, the correlation between the concentration of the peptide that caused aortic relaxation or activation of adenylate cyclase was not close, possibly reflecting the shorter duration of exposure of the bioassay tissue to the peptide in the superfusion experiment. Altematively. this may reflect that there is a threshold stimulation of adenylate cyclase before vasorelaxation takes place, or that PACAP works by another intracellular mechanism. If PACAP does work via adenylate cyclase. this probably triggers the sequestration of calcium ions into the sarcoplasmic recticulum, the subsequent lowering of intracellular calcium concentration reducing the formation of the calcium-calmodulin complex hence reducing actin-myosin interaction (MacDermot. 1990). An alternative explanation, that cyclic AMP-dependent phosphorylation of myosin light chain kinase causes of
may
not incur
in viva ~~arnrn and Stull.
se from the anterior pituitary
is conspecialized circulation, which in turn ctive factors derived from the sites for PACAP have been tb~~~s and the anterior pituitary of the rat (Lam et al., in press} and this raises the ility that this potent vasodilator may play a part control of the anterior pituitary circulation. of PACAP tested were remarkably longtent vas~lators. It remains to be ascert~ned if s~f~c~ent peptide is present in the vasculature to have d ~a~opbys~olo~c~ role. but our results suggest that or& small amounts may have significant effects.
SC. was supported bv the British Heart Foundation and B.E.R. is a WeBcome Priie student. The work was supported in part by the Wellcome Trust.
eferen~s Edwards. R.J.. J. MacDermot and A.J. Wilkins, 1987, Prostacyclin analogues reduce ADP-ribosylation of the a-subuni: of the regu-
Iatory G,protein and di~nish adenosine (AZ) responsiveness of platelets. Br. J. Pharmacol. 90, 501. Kamm, W.E. and J.T. Stull. 1985, The function of myosin and myosin light kinase phosphorylation in smooth muscle. Ann. Rev. Pharmocol. Toxicol. 25, 593. Kelly, E., M. Keen, P. Nobbs and J. MacDermot, 1990, NaF and guanine nucieotides modulate adenylate cyclase activity in NGlOS-15 c&s by interacting with both G, and Gi, Br. 1. Pharmacol. 100, 223. Kimura, C., S. Ohkubo, K. Ogi, M. Hosoya, Y. Itoh, H. Onda. A. Miyata, L. Jiang. R.R. Dahl, H.H. Stibbs, A. Arimura and M. Fujino, 1990, A novel peptide which stimulates adenylate cyclase: molecular cloning and characterization of the ovine and human cDNA’s. B&hem. Biophys. Res. Co~un. 166, 81. Lam. H.-C.. K. Takahashi, M.A. Ghatei, SM. Kanse, J.M. Polak and S.R. Bfoom, Binding sites of a novel neuropeptide ‘pituitary adenylate cyclase activating polypeptide’ in the rat brain and lung, European J. B&hem. (in press), MacDermot. J.. 1990. Prostacylcin, cyclic AMP, and relaxation of vascular smooth muscle, Clin. Pharmacol. 7,49. Miyata, A., A. Arimura, R.R. D&i, N. Mina~no, A. Uehara. L. Jiang, M.D. Culler and D.H. Coy, 1989, Isolation of a novel 38 residue-hypothalamic polypeptide which stimulates adenylate cyclase in pituitary ceils, B&hem. Biophys. Res. Commun. 164, 567.
Salomon, Y., C. Londos and M. Rodell, 1974, A highly sensitive adenylate cyclase assay, Anal. B&hem. 58, 541. Warren, J.B., R. Loi, N.B. RendelI and G.W. Taylor, 1990, Nitric oxide is inactivated by the bacterial pigment pyocyanin, Biochem. J. 266, 921.