- Email: [email protected]
P2X (Purinergic) receptor distributions in rat blood vessels
Journal of the Autonomic Nervous System 75 Ž1999. 147–155
P2X žPurinergic / receptor distributions in rat blood vessels M.A. Hansen a b
a,b
, J.L. Dutton
a,c
, V.J. Balcar
a,b
, J.A. Barden
a,b
, M.R. Bennett
a,c,)
The Institute for Biomedical Research, UniÕersity of Sydney, Sydney, NSW 2006, Australia Department of Anatomy and Histology, UniÕersity of Sydney, Sydney, NSW 2006, Australia c Department of Physiology, UniÕersity of Sydney, Sydney, NSW 2006, Australia Received 31 October 1998; accepted 23 November 1998
Abstract The distribution of purinergic ŽP2X1 and P2X2 . receptors on smooth muscle cells in relation to autonomic nerve varicosities in rat blood vessels has been determined using immunofluorescence and confocal microscopy. P2X1 and P2X2 receptors were visualised using rabbit polyclonal antibodies against the extracellular domain of the receptors and varicosities visualised using a mouse monoclonal antibody against the ubiquitous synaptic vesicle proteoglycan SV2. Two size classes of P2X1 receptor clusters were observed on the smooth muscle cells of mesenteric, renal, and pulmonary arteries as well as in the aorta and in veins: a large approximately elliptical cluster 1.32 " 0.21 mm long and 0.96 " 0.10 mm in diameter; and a smaller spherical cluster with a diameter of 0.32 " 0.05 mm. The latter occurred throughout the media of arteries of all sizes, whereas the former were restricted to the adventitial surface of the media and to endothelial cells, except for the pulmonary artery, in which large receptor clusters were found throughout the media of the vessel. At the adventitial surface, the large clusters are in general located beneath SV2 labelled varicosities. None of the small clusters was associated with varicosities. Three-dimensional reconstruction of the P2X and SV2 labelling at individual varicosities showed that the varicosities were immediately apposed to the P2X receptor clusters. P2X2 receptors were located on nerves and on endothelial cells. They were also found in low density on the smooth muscle cells in the media. These observations are discussed in relation to the mechanism of purinergic transmission to the smooth muscle cells of blood vessels. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Arteries; Immunohistochemistry; Receptors; Veins
1. Introduction Transmission from sympathetic nerves to a number of different blood vessels in the rat involves the release of ATP onto purinergic receptors. These include the aorta ŽWhite et al., 1985; Kitajima et al., 1994; Garcia-Velasco et al., 1995., the pulmonary artery ŽInoue and Kannan, 1988., renal blood vessels ŽSchwartz and Malik, 1989., mesenteric blood vessels ŽAngus et al., 1988; Yamamoto et al., 1992; Windscheif et al., 1994; Brock and Van Helden, 1995.. Autoradiographic studies of the distribution of w3 Hxa ,b-methylene ATP over the media of rat blood vessels indicates that the small and medium-size arteries have a high density of purinergic receptors compared with the elastic and large muscular arteries, with the media of ) Corresponding author. Institute for Biomedical Research, Department of Physiology, The University of Sydney, Sydney, NSW 2006, Australia. T el.: q 61-2-9351-2034; fax: q 61-2-9351-3910; e-m ail: [email protected]
the large muscular arteries showing a higher density of the receptors towards the adventitial surface compared with the intimal surface ŽBo and Burnstock, 1993; Zhao et al., 1996.. However it seems likely that most of the smooth muscle cells in the media of these vessels possess purinergic receptors, as patch-clamped cells dissociated from the media show inward currents in response to exogenous application of ATP Žvon der Weid et al., 1993; Evans and Kennedy, 1994.. Purinergic receptors are of two main subtypes, G-protein-coupled receptors Žtermed P2Y . and ATP-gated channels Žtermed P2X ., with seven of the P2X receptors now isolated and expressed ŽCollo et al., 1996.. Two of these ŽP2X1 and P2X2 . are found in rat mesenteric arterioles, with immunohistochemistry localising the P2X1 receptors to smooth muscle cells and the P2X2 receptors to nerve terminals ŽVulchanova et al., 1996.. Each of the seven P2X receptors is encoded by a different gene, with each possessing two transmembrane domains, a large extracellular domain, and intracellular N- and C-termini ŽSurprenant,
0165-1838r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 1 8 3 8 Ž 9 8 . 0 0 1 8 9 - 1
148
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
1996.. All of these proteins are about 36% to 48% identical ŽSurprenant, 1996.. The P2X1 and P2X2 receptors were originally cloned from complementary DNA that encoded these receptors in rat vas deferens ŽValera et al., 1994. and in rat pheochromocytoma ŽPC12. cells ŽBrake et al., 1994.. In the present work, we have developed antibodies against P2X1 and P2X2 receptors and localised these with respect to single varicosities using antibodies against the ubiquitous vesicle proteoglycan SV2. This has enabled the distribution of these purinergic receptors to be determined at the level of individual sympathetic varicosities at different sites within the media of rat blood vessels.
2. Methods 2.1. Materials Anti-mouse SV2 antibody ŽAbSV2. specific for the synaptic vesicle proteoglycan SV2 ŽBuckley and Kelly, 1985., was kindly provided by Dr Buckley ŽHarvard University.. Cy2 and Cy3 conjugates of donkey anti-mouse and donkey anti-rabbit IgGs, adsorbed against conspecific IgGs, were from Jackson Immunoresearch ŽWest Grove, PA.. All other reagents were purchased from Sigma ŽSt. Louis, MO.. 2.2. Antibody production The consensus sequences of the rat P2X1 ŽValera et al., 1994. and P2X2 ŽBrake et al., 1994. cloned receptors were examined for suitable epitopes and the sequences Lys199– Cys217 and Cys130–Gly153 were synthesised using standard t-BOC chemistry ŽBarden et al., 1997. on an ABI synthesiser ŽHansen et al., 1997.. Ten milligrams of HPLC-purified peptide Ždetermined from ion-spray mass spectrometry. was covalently linked to 45 mg of diphtheria toxin by Chiron Mimotopes who also raised the antibodies
in two New Zealand White rabbits. Peptide–antigen conjugates were suspended in water Ž5 mgrml. and aliquots emulsified in Complete Freund’s Adjuvant. Emulsion volumes of 1 ml containing 0.5 mg of peptide were injected intramuscularly with second, third and fourth immunisations followed at 2 week intervals using Incomplete Freund’s Adjuvant. Final bleeds were obtained at 12 weeks. Blood was incubated at 378C for 30 min, stored at 48C for 15 h and serum collected following centrifugation and stored at y208C in aliquots. Each serum was tested with an ELISA assay for specific antibodies against each peptide, either AbP2X1 or AbP2X2 ŽHansen et al., 1998.. The antibody titre, defined as the reciprocal of the serum dilution resulting in an absorbance of 1.0 above background in the ELISA assay, fell in the range 75 000 " 4000 compared with 225 " 25 for pre-immune samples. Each binds selectively and specifically as shown by Western blots ŽBerry et al., 1998. as well as immunolocalisation. The two antisera were raised against non-homologous extracellular epitopes. Each was chosen in part to ensure no cross-reaction with any of the other subtypes. The two P2X antisera used have been shown to possess similar distributions in many cases but with distinctly different distributions in other cases indicating that they do not lack specificity ŽHansen et al., 1998.. Individual specificity was checked using rat P2X1 cDNA Žkindly provided by Dr R.A. North. transfected into Xenopus oocytes. AbP2X1 was added to transfected cells in the presence and absence of either a 10 mM concentration of the P2X1 epitope or a 10 mM concentration of the P2X2 epitope. Immunolabelling and confocal imaging demonstrates that the expressed P2X1 is located, as expected, within the cell membrane ŽFig. 1A.. The presence of a 10 mM concentration of the P2X1 cognate peptide as an absorption control resulted in the blocking of P2X1 staining ŽFig. 1B. while the presence of the P2X2 cognate peptide also had no effect on AbP2X1 labelling ŽFig. 1C., as expected for a non-homologous epitope to P2X1.
Fig. 1. AbP2X 1 staining is confined largely to the surface membrane of Xenopus oocytes transfected with P2X1 cDNA ŽA.. The addition of a 10 mM concentration of the P2X 1 cognate peptide to the transfected cells resulted in the blocking of P2X1 staining ŽB.. The same concentration of the P2X2 cognate peptide added to the transfected cells in the presence of AbP2X 1 which resulted in no reduction of labelling intensity ŽC.. Calibration, 25 mm.
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
2.3. Tissue preparation Blood vessels from 12 adult rats were removed and fixed in 4% paraformaldehyde in borate–acetate buffer ŽpH 9.5. for 1 h. Tissue was cryoprotected by immersion in 30% sucrose overnight before transverse sections Ž30 mm. were cut on a freezing microtome. Tissue was placed in 0.1% dimethyl sulphoxide ŽDMSO. in phosphate buffered horse serum Ž100 ml PBS, 2 ml normal horse serum ŽNHS., 0.1 ml Triton X-100, 1 g Bovine Serum Albumin. for 30 min to permeabilise the cellular membranes. The preparation was then washed 3 times in PBS Ž10 min each. and immersed in 20% NHS in PBS for 1 h to block non-specific binding sites in the tissue. This was followed by incubation with 1:100 AbSV2 together with either 1:100 pre-immune serum ŽPIS., AbP2X1 or AbP2X2 in PBS for 48 h at 48C. Slides were rinsed in PBS followed by the addition of the relevant secondary fluorescent anti-
149
bodies for 90 min at 208C then washed 3 times in PBS Žeach for 10 min. coverslipped and sealed. Sections were viewed on a Leica TCS NT UV laser confocal microscope system, with pinhole set at 1.0. 2.4. Statistical eÕaluations Sizes of receptor clusters were measured in randomly selected areas and are expressed as mean " standard deviation.
3. Results Tissue sections were exposed to either a combination of antibodies to the P2X1 receptor and the ubiquitous synaptic vesicle proteoglycan SV2 or to P2X2 receptors and SV2.
Fig. 2. The distribution of SV2 ŽA. and P2X 1 ŽB. receptors at the adventitial surface of mesenteric arteries with the media at the top of the figure. Large P2X 1 receptor clusters are colocalised with SV2 Žarrows in centre of figure.. The arrowhead at the right shows an SV2 label not associated with P2X1 . Calibration, 5 mm. The distribution of SV2 ŽC. and P2X 2 ŽD. receptors in the adventitia of intralobular arteries of the kidney with the media at the top of the figure. Distribution is very similar. Calibration, 5 mm.
150
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
3.1. Mesenteric arteries and intralobular renal arteries P2X1 immunoreactivity was prominent at the adventitial surface of a rat mesenteric artery ŽFig. 2B.. Only large clusters, greater than 1 mm diameter, could be distinguished at the adventitial surface where they were inevitably double labelled with SV2 ŽFig. 2A, arrow in centre of figure.. SV2 label was confined to the adventitial surface of the vessel with some SV2 label not associated with P2X1 ŽFig. 2A and B, arrowhead at right of figure.. Puncta labelled with both P2X1 and SV2 are observed on what appear to be the adipocytes labelled with an arrow at the bottom of Fig. 2A and B. Neither SV2 nor distinctly large P2X1 puncta, suggesting the presence of receptor clusters, were observed in the media at the top of Fig. 2A and B. The distribution of P2X2 receptors and SV2-labelled varicosities ŽFig. 2C and D. was similar in the intralobular renal arteries. Large P2X2 receptor clusters also were found at the adventitial surface of the media in high concentration, with most colocalised with SV2 stained varicosities ŽFig. 2C and D.. SV2-labelled varicosities were not observed nor were large P2X2 receptor clusters in the media of the artery. 3.2. Pulmonary artery The distribution of P2X1 receptors and of SV2 in this artery was the same as that in the mesenteric and renal arteries, with the important exception that large P2X receptor clusters could be observed throughout the media ŽFig. 3B and D.. These were often associated with colocalised SV2 immunoreactivity especially close to the adventitia ŽFig. 3A and C.. In the case of P2X1 receptors ŽFig. 3B, near adventitia and Fig. 3D, deeper in the media., these consisted of discrete large high-density clusters, elliptical in shape when reconstructed in three-dimensional view using the confocal microscope, with the ellipse possessing a major axis of 1.32 " 0.17 mm and a minor axis of 0.96 " 0.10 mm Ž n s 70.. A large number of small circular P2X1-labelled clusters were also present throughout the media ŽFig. 3D, arrowhead., possessing a diameter of 0.32 " 0.05 mm and not associated with any SV2 labelling ŽFig. 3C, see corresponding arrowhead.. P2X2 receptors also were found throughout the media in discrete clusters ŽFig. 3F and H.. High power views of these clusters labelled by the large arrowhead in Fig. 3F are shown in
151
Fig. 3H. The small P2X2-labelled clusters in Fig. 3F Žan example shown with the arrow. were not colocalised with SV2 ŽFig. 3E.. The enlargement in Fig. 3H shows more clearly that most of the large P2X2-labelled clusters are associated with SV2 Žarrowheads in Fig. 3G and H. while there are some SV2-labelled nerve structures in Fig. 3G that are not labelled by P2X2 Žarrow in Fig. 3G and H.. The P2X2-labelled clusters also fell into the same size classes as did the P2X1-labelled clusters, with the large clusters usually possessing colocalised SV2 labelling but not the small clusters ŽFig. 3F; see arrow.. 3.3. Endothelial cells The endothelial cells of all vessels were labelled with both P2X1 and P2X2 receptor clusters but not with SV2. An example taken from the endothelial cells of a mesenteric artery is shown in Fig. 4A and B. These P2X1 clusters were of the small-size category and could be discerned to cover the entire surface of the endothelial cells ŽFig. 4B.. The example shown in Fig. 4C and D is taken from the endothelial cells of the aorta in which SV2 labelling is shown in Fig. 4C and the corresponding P2X2 labelling is shown in Fig. 4D. Small-sized discrete puncta can be discerned covering the cells. 3.4. Veins Small veins in the urinary bladder were examined to see if the distribution of receptors in these vessels was different from that in arteries. Large P2X1 receptor patches could be observed on the smooth muscle cells at the adventitial surface of the media of these vessels ŽFig. 4F.. Double labelling with SV2 antibody ŽFig. 4E. showed that a large proportion of these receptor clusters was associated with varicosities ŽFig. 4E and F; see arrowheads.. P2X2 labelled clusters also were evident in the adventitia, and were always colocalised with SV2. Neither P2X1 nor P2X2 labelling was observed elsewhere in the veins.
4. Discussion 4.1. The distribution of P2 X1 and P2 X 2 receptor clusters in blood Õessels Two main transmitters have been identified for the sympathetic innervation of many arteries, namely nor-
Fig. 3. The distribution of P2X 1 and P2X2 receptors in relation to SV2 labelled nerve terminal varicosities in the pulmonary artery. ŽA. and ŽB. show SV2 and P2X 1 labelling, respectively, in the media at the adventitial surface. Calibration, 5 mm. ŽC. and ŽD. show SV2 and P2X1 labelling, respectively, deeper within the media than in ŽA.. Several small 0.32 mm clusters of P2X 1 are present around the large colocalised SV2 and P2X1 clusters. The arrow head shows a small P2X 1 cluster with no SV2. Calibration, 5 mm. ŽE. and ŽF. show SV2 and P2X2 labelling in the media towards the adventitial surface of the artery. A group of P2X 2 rSV2 clusters is indicated by the arrow head and a small P2X2 cluster not colocalised with SV2 by the arrow. Calibration, 20 mm. ŽG. and ŽH., an enlargement of the P2X 2 rSV2 clusters Žtwo arrow heads. indicated by the large arrow head in ŽE. and ŽF.. Some SV2 non-localised with P2X 2 is also seen Žarrow.. Calibration, 12 mm.
152
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
Fig. 4. The localisation of P2X 1 receptors and SV2 on endothelial cells and veins. ŽA. and ŽB. show SV2 and P2X1 labelling, respectively, of an endothelial cell from the intimal surface of a mesenteric artery. Arrow heads point to distinct P2X 1 receptor clusters on the cell. Calibration, 2 mm. ŽC. and ŽD. show SV2 and P2X 2 labelling, respectively, of an endothelial cell from the intimal surface of the aorta. Calibration bars are 2 mm. ŽE. and ŽF. show the distribution of SV2 and P2X 1 , respectively, in veins of the urinary bladder. The arrow head indicates one SV2 stained profile colocalised with a P2X1 cluster. Calibration, 10 mm.
adrenaline and ATP ŽSneddon and Burnstock, 1984; Neild and Kotecha, 1986; Papanicolaou and Medgett, 1986; Abe et al., 1987.. These transmitters are stored in nerves that are usually confined to the adventitial surface of the media ŽBennett, 1972a.. However, in vessels with a large media, above about 100 mm wide, there is often an invasion of the outer 30% of the media by nerves from the adventitial surface including rat tail artery ŽCheung, 1982.; sheep
carotid artery ŽKeatinge, 1966.; dog dorsal pedal artery ŽTsunekawa et al., 1967., a result that has been obtained several times in the case of the pulmonary artery ŽCech and Dolezel, 1967; Fillenz, 1970.. These observations were confirmed in the present work with the mesenteric and renal arteries possessing varicosities only in the adventitia whereas the pulmonary artery contained varicosities throughout the media. Most of these varicosities possessed
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
colocalised P2X1 receptor clusters, whether at the adventitial surface or deeper in the media. The dimension of these clusters was clearer in the media, because of their relative isolation from each other. Clusters were about 1 mm in diameter Žellipsoidal in shape. and were precisely located beneath the varicosities, with only a few hundred nanometres of extension beyond the borders of the varicosities. It is interesting in this regard that varicosities are found that come into close contact with smooth muscle cells at the adventitial surface. There is only a single layer of basal lamina between the varicosity and muscle membranes in arteries with media ranging in size from 2 to 25 cells thick ŽLuff and McLachlan, 1989.. In the case of large muscular arteries, less than half of the varicosities within 1 mm of the muscle cells form close contacts. This situation applies even though these varicosities may be devoid of a complete Schwann cell covering, possess vesicles accumulated towards the side of the varicosity facing the muscle, and even have apparently prejunctional specialisations ŽLuff et al., 1995.. The colocalisation of P2X1 receptor clusters with varicosities might arise from the location of the receptors on the varicosities or on the Schwann cells that partially wrap around these rather than on the apposing smooth muscle cell membrane. It seems likely that the clusters are on the muscle cells. Three-dimensional reconstruction of the distribution of the receptor label with the confocal microscope showed the label confined to an ellipsoid apposing the SV2 labelled varicosity rather than wrapped around the varicosity as would be expected if the receptor was found on Schwann cells. It is also unlikely that the receptor is uniquely present on the varicosities, as the receptors are present following destruction of the sympathetic nerves with 6-hydroxydopamine treatment Žpersonal observations.. Adipose tissue in the adventitia of all the blood vessels examined in the present work also received an innervation from SV2 stained varicosities, with these possessing P2X1 receptor clusters. This confirms the recent discovery that adipocytes possess P2X receptors ŽLee and Pappone, 1997. which, when activated, increase membrane trafficking ŽPappone and Lee, 1996.. One significant observation that emerged from this study was the high density of P2X1 receptors on endothelial cells. It is known that noradrenaline can release ATP from endothelial cells in which it is stored ŽShinozuka et al., 1994; Luff et al., 1995.. ATP may also be released from the endothelial cells by stress as a consequence of increased flow rates through the vessels ŽMilner et al., 1996.. However, the P2 purinoceptor subclass that has generally been taken to mediate these effects, is of the P2Y variety ŽKennedy et al., 1985; White et al., 1985; Ralevic and Burnstock, 1988; Hansmann et al., 1997.. The role of P2X1 receptors on endothelial cells is yet to be elucidated. Smooth muscle cells in the media of all the blood vessels examined in the present work possessed small P2X1 receptor clusters. These clusters, approximately circular
153
with an apparent fluorescence diameter of about 0.3 mm, are considerably smaller than the clusters found beneath the varicosities with a diameter of about 1 mm. Such clusters may mediate the currents recorded in patchclamped smooth muscle cells dissociated from the media of vessels and exposed to ATP Žvon der Weid et al., 1993; Evans and Kennedy, 1994.. Both nerves and endothelial cells possessed P2X2 receptors, as did smooth muscle cells ŽVulchanova et al., 1996, 1997.. These authors however also claimed that P2X2 receptors were excluded from smooth muscle cells. No reports of P2X receptor labelling on endothelial cells have appeared. P2X2 receptors were found together with SV2 labelling in the media of those blood vessels for which the nerves were not confined to the adventitia, that is mostly in the outer 30% of the media of the pulmonary artery. The physiological role of P2X2 receptors on nerves and on endothelial cells is unknown. 4.2. P2 X1 receptors and junctional transmission to smooth muscle The autonomic neuromuscular junction was originally defined in terms of the effects of transmitter released from varicosities acting directly on receptors on smooth muscle, together with radiation into the cell of the electrical signs of transmission due to the action of transmitter on surrounding cells ŽBennett, 1972a,b; Bennett and Gibson, 1995.. In blood vessels, the present work shows high-density receptor clusters are found in close apposition to varicosities. However the question arises as to how the electrical signs of transmission radiate from the adventitial surface, where the receptor clusters are exclusively found in all but the large muscular arteries, to the smooth muscle cells at the intimal surface. One possibility is that excitatory junction potentials ŽEJPs. generated by the release of transmitter onto the adventitial smooth muscle cells ŽAEJPs. are conducted electrotonically through the smooth muscle syncytium to the smooth muscle cells at the intimal surface ŽBennett, 1972a.. Given the known properties of the smooth muscle syncytium ŽBennett et al., 1993, 1995; Henery et al., 1997. and that the smooth muscle cells in the media are each confined to a single plane in the media, this would seem to be possible if a large percentage of the varicosities released transmitter on arrival of the nerve impulse. In this case the adventitial surface of the media is equipotential during the AEJP, so that the syncytium effectively become one-dimensional as far as current flow is concerned. However the probability of the evoked secretion of ATP quanta is very low from varicosities ŽAstrand and Stjarne, 1989; Lavidis and Bennett, 1992., so it is ¨ uncertain if the media acts as a one-dimensional cable or as a three-dimensional syncytium during transmission. Furthermore, even given that current flow is one-dimensional, it is uncertain just how far effective transmission can be radiated from the adventitial surface to produce EJPs at the intimal surface ŽIEJPs..
154
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
The alternative to syncytial conduction of the EJP is that transmitter released at the adventitial surface can diffuse sufficiently far to affect smooth muscle cells at the intimal surface. Noradrenaline is detectable at the intimal surface of the rat tail artery after high frequency stimulation in the presence of cocaine to block neuronal uptake ŽBao et al., 1993.. Under these conditions, the concentration of noradrenaline at the intimal surface is about 20% of that at the adventitial surface, so that it seems unlikely that appreciable concentrations of noradrenaline will be reached at the intima under normal conditions of an intact uptake mechanism and low frequency stimulation. The case for ATP reaching effective concentrations at the intimal surface of a medium size vessel like the rat tail artery is even more tenuous, given that the media is 12 to 15 smooth muscle cells thick ŽSittiracha et al., 1987.. Firstly, ATP is metabolised by ecto-enzymes in the extracellular space of smooth muscles ŽWestfall et al., 1997.. Furthermore, in order for the P2X class of purinoceptors to be activated, they must bind two ATP molecules, a condition that is difficult to fulfil over long diffusion pathways ŽBennett et al., 1995.. The problem might be resolved by the observation that P2X1 receptors are found in small clusters on all smooth muscle cells in the media of all the vessels studied. There is no evidence for action potential propagation from the adventitial smooth muscle to the intimal smooth muscle. It might be then that the diffusion of ATP from varicosities some distance in from the adventitial surface onto diffuse receptors located deeper into the media towards the intima is sufficient to generate EJPs in the intimal smooth muscle that produce a physiological effect. An alternative explanation for the role of the small receptor clusters is that they are smooth muscle autoreceptors that are activated by the release of ATP from the muscle cells. Evidence for this has been recently forthcoming with the discovery that spontaneous potentials can be recorded from smooth muscle cells at the intimal surface of the media of rat blood vessels that are due to the spontaneous release of ATP from a source other than nerves ŽHenery and Bennett, 1998.. It will be of interest to carry out physiological experiments to determine the role of the P2X receptor clusters described in the present work in the control of blood vessels. Acknowledgements We wish to thank the National Health and Medical Research Council of Australia, Australian Research Council, National Heart Foundation of Australia, and the Clive and Vera Ramaciotti Foundation for financial support. References Abe, K., Matsuki, N., Kasuya, Y., 1987. Pharmacological and electrophysiological discrimination of contractile responses to selective al-
pha 1- and alpha 2-adrenoceptor agonists in rat tail artery. Jpn. J. Pharmacol. 45, 249–261. Angus, J.A., Broughton, A., Mulvaney, M.J., 1988. Role of alpha-adrenoceptors in constrictor responses of rat, guinea-pig and rabbit small arteries to neural activation. J. Physiol. 403, 495–510. Astrand, P., Stjarne, L., 1989. On the secretory activity of single varicosi¨ ties in the sympathetic nerves innervating the rat tail artery. J. Physiol. 409, 207–220. Bao, J.X., Gonon, F., Stjarne, L., 1993. Frequency- and train length-de¨ pendent variation in the roles of postjunctional alpha 1- and alpha 2-adrenoceptors for the field stimulation-induced neurogenic contraction of rat tail artery. Naunyn Schmiedebergs’ Arch. Pharmacol. 347, 601–616. Barden, J.A., Cuthbertson, R.M., Jia-Zhen, W., Moseley, J.M., Kemp, B.E., 1997. Solution structure of parathyroid hormone related protein Žresidues 1–34. containing an Ala substituted for an Ile in position 15 ŽPTHrPwAla15x-Ž1–34.. J. Biol. Chem. 272, 29572–29578. Bennett, M.R., 1972a. Structure and electrical properties of the autonomic neuromuscular junction. Phil. Trans. R. Soc. Lond. Ser. B 265, 25–34. Bennett, M.R., 1972. Autonomic Neuromuscular Transmission. Monographs. Physiological, Society 30. Cambridge Univ. Press. Bennett, M.R., Gibson, W.G., 1995. On the contribution of quantal secretion from close-contact and loose-contact varicosities to the synaptic potentials in the vas deferens. Phil. Trans. R. Soc. Lond. Ser. B 347, 187–204. Bennett, M.R., Gibson, W.G., Poznanski, R.R., 1993. Extracellular current flow and potential during quantal transmission from varicosities in a smooth muscle syncytium. Phil. Trans. R. Soc. Lond. Ser. B 342, 89–99. Bennett, M.R., Farnell, L., Gibson, W.G., Karunanithi, S., 1995. Quantal transmission at purinergic junctions: stochastic interaction between ATP and its receptors. Biophys. J. 68, 925–935. Berry, D., Yao, M., Barden, J.A., Balcar, V.J., Hansen, M.A., Bennett, M.R., Keogh, A., dos Remedios, C.G., 1998. Alterations in the expression of P2X1 receptors in failing and non-diseased human atria. Electrophoresis 19, 856–859. Bo, X., Burnstock, G., 1993. Heterogeneous distribution of w3 Hx a ,bmethylene ATP binding sites in blood vessels. J. Vasc. Res. 30, 87–101. Brake, A.J., Wagenbach, M.J., Julius, D., 1994. New structural motif for ligand-gated ion channels defined by an ionotropic ATP receptor. Nature 371, 519–523. Brock, J.A., Van Helden, D.F., 1995. Enhanced excitatory junction potentials in mesenteric arteries from spontaneous hypertensive rats. Pflug. Arch. Eur. J. Physiol. 430, 901–908. Buckley, K., Kelly, R.B., 1985. Identification of a transmembrane glycoprotein specific for secretory vesicles of neural and endocrine cells. J. Cell. Biol. 100, 1284–1294. Cech, S., Dolezel, S., 1967. Monoaminergic innervation of the pulmonary vessels in various laboratory animals Žrat, rabbit, cat.. Experientia 23, 114–115. Cheung, D.W., 1982. Two components in the cellular response of rat tail arteries to nerve stimulation. J. Physiol. 328, 401–408. Collo, G., North, R.A., Kawashima, E., Merlo-Pich, E., Neidhart, S., Surprenant, A., Buell, G., 1996. Cloning of P2X5 and P2X6 receptors and the distribution and properties of an extended family of ATP-gated ion channels. J. Neurosci. 16, 2495–2507. Evans, R.J., Kennedy, C., 1994. Characterisation of P2-purinoceptors in the smooth muscle of the rat tail artery: a comparison between contractile and electrophysiological responses. Br. J. Pharmacol. 113, 853–860. Fillenz, M., 1970. Innervation of pulmonary and bronchial blood vessels of the dog. J. Anat. 106, 449–461. Garcia-Velasco, G., Sanchez, M., Hidalgo, A., Garcia de Boto, M.J., 1995. Pharmacological dissociation of UTP- and ATP-elicited con-
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155 tractions and relaxations in isolated rat aorta. Eur. J. Pharmacol. 294, 521–529. Hansen, M.A., Barden, J.A., Balcar, V.J., Keay, K.A., Bennett, M.R., 1997. Structural Motif and Characteristics of the Extracellular Domain of P2X Receptors. Biochem. Biophys. Res. Commun. 236, 670–675. Hansen, M.A., Barden, J.A., Balcar, V.J., Bennett, M.R., 1998. The distribution of single P2X 1 -receptor clusters on smooth muscle cells in relation to nerve varicosities in the rat urinary bladder. J. Neurocytol. 27, 529–539. Hansmann, G., Bultmann, R., Tuluc, F., Starke, K., 1997. Characterisation by antagonists of P2-receptors mediating endothelium-dependent relaxation in the rat aorta. Naunyn Schmiedebergs’ Arch. Pharmacol. 356, 641–652. Henery, R.J., Bennett, M.R., 1998. Synaptic and spontaneous potentials in smooth muscle cells of the endothelial surface of the rat tail artery. Proc. Aust. Neurosci. Soc. 9, 138. Henery, R.J., Gibson, W.G., Bennett, M.R., 1997. Quantal currents and potential in the three-dimensional anisotropic bi-domain model of smooth muscle. Bull. Math. Biol. 59, 1047–1075. Inoue, T., Kannan, M.S., 1988. Nonadrenergic and noncholinergic excitatory neurotransmission in rat intrapulmonary artery. Am. J. Physiol. 254, H1142–H1148. Keatinge, W.R., 1966. Electrical and mechanical response of arteries to stimulation of sympathetic nerves. J. Physiol. 185, 701–715. Kennedy, C., Delbro, D., Burnstock, G., 1985. P2-purinoceptors mediate both vasodilation Žvia the endothelium. and vasoconstriction of the isolated rat femoral artery. Eur. J. Pharmacol. 107, 161–168. Kitajima, S., Ozaki, H., Karaki, H., 1994. Role of different subtypes of P2 purinoceptor on cytosolic Ca2q levels in rat aortic smooth muscle. Eur. J. Pharmacol. 266, 263–267. Lavidis, N., Bennett, M.R., 1992. Probabilistic secretion of quanta from visualized sympathetic nerve varicosities in mouse vas deferens. J. Physiol. 454, 9–26. Lee, S.C., Pappone, P.A., 1997. Effects of P2 purinergic receptor stimulation in brown adipocytes. Am. J. Physiol. 273, C679–C686. Luff, S.E., McLachlan, E.M., 1989. Frequency of neuromuscular junctions on arteries of different dimensions in the rabbit, guinea pig and rat. Blood Vessels 26, 95–106. Luff, S.E., Young, S.B., McLachlan, E.M., 1995. Proportions and structure of contacting and non-contacting varicosities in the perivascular plexus of the rat tail artery. J. Comp. Neurol. 361, 699–709. Milner, P., Boden, P., Burnstock, G., 1996. Long-term guanethidine sympathectomy suppresses flow-induced release of ATP and endothelin from endothelial cells isolated from adult rat aorta. J. Vasc. Res. 33, 139–145. Neild, T.O., Kotecha, N., 1986. Effects of alpha beta methylene ATP on membrane potential, neuromuscular transmission and smooth muscle contraction in the rat tail artery. Gen. Pharmacol. 17, 461–464. Papanicolaou, M., Medgett, I.C., 1986. Effects of blockade of alpha 1and alpha 2-adrenoceptors on vasoconstrictor responses to single and two pulse stimulation in rat tail artery. Eur. J. Pharmacol. 131, 211–218. Pappone, P.A., Lee, S.C., 1996. Purinergic receptor stimulation increases membrane trafficking in brown adipocytes. J. Gen. Physiol. 108, 393–404.
155
Ralevic, V., Burnstock, G., 1988. Actions mediated by P2 purinoceptor subtypes in the isolated perfused mesenteric bed of the rat. Br. J. Pharmacol. 95, 637–645. Schwartz, D.D., Malik, K.U., 1989. Renal periarterial nerve stimulationinduced vasoconstriction at low frequencies is primarily due to release of a purinergic transmitter in the rat. J. Pharmacol. Exp. Ther. 250, 764–771. Shinozuka, K., Hashimoto, M., Masumura, S., Bjur, R.A., Westfall, D.P., Hattori, K., 1994. In vitro studies of release of adenine nucleotides and adenosine from rat vascular endothelium in response to alpha 1-adrenoceptor stimulation. Br. J. Pharmacol. 113, 1203–1208. Sittiracha, T., McLachlan, E.M., Bell, C., 1987. The innervation of the caudal artery of the rat. Neuroscience 21, 647–659. Sneddon, P., Burnstock, G., 1984. ATP as a co-transmitter in rat tail artery. Eur. J. Pharmacol. 106, 149–152. Surprenant, A., 1996. Functional properties of native and cloned P2X receptors. Ciba Found. Symp. 198, 208–219. Tsunekawa, K., Mohri, K., Ikeda, M., Ohgushi, N., Fujiwara, M., 1967. Histochemical demonstration of adrenergic fibers in the smooth muscle layer of media of dorsal–pedal artery in dog. Experientia 23, 842–843. Valera, S., Hussy, N., Evans, R.J., Adami, N., North, R.A., Surprenant, A., Buell, G., 1994. A new class of ligand-gated ion channel defined by P2x receptor for extracellular ATP. Nature 371, 516–591. von der Weid, P.Y., Serebryakov, V.N., Orallo, F., Bergmann, C., Snetkov, V.A., Takeda, K., 1993. Effects of ATP on cultured smooth muscle cells from rat aorta. Br. J. Pharmacol. 108, 638–645. Vulchanova, L., Arvidsson, U., Riedl, M., Wang, J., Buell, G., Surprenant, A., North, R.A., Elde, R., 1996. Differential distribution of two ATP-gated channels ŽP2X receptors. determined by immunocytochemistry. Proc. Natl Acad. Sci. USA 93, 8063–8067. Vulchanova, L., Riedl, M.S., Shuster, S.J., Buell, G., Surprenant, A., North, R.A., Elde, R., 1997. Immunohistochemical study of the P2X2 and P2X3 receptor subunits in rat and monkey sensory neurones and their central terminals. Neuropharmacology 36, 1229–1242. Westfall, T.D., Kennedy, C., Sneddon, P., 1997. The ecto-ATPase inhibitor ARL 67156 enhances parasympathetic neurotransmission in the guinea-pig urinary bladder. Eur. J. Pharmacol. 329, 169–173. White, T.D., Chaudhry, A., Vohra, M.M., Webb, D., Leslie, R.A., 1985. Characteristics of P2 Žnucleotide. receptors mediating contraction and relaxation of rat aortic strips: possible physiological relevance. Eur. J. Pharmacol. 118, 37–44. Windscheif, U., Ralevic, V., Baumert, H.G., Mutschler, E., Lambrecht, G., Burnstock, G., 1994. Vasoconstrictor and vasodilator responses to various agonists in the rat perfused mesenteric arterial bed: selective inhibition by PPADS of contractions mediated via P2X-purinoceptors. Br. J. Pharmacol. 113, 1015–1021. Yamamoto, R., Takasaki, K., Nickols, G.A., 1992. Purinergic vasoconstrictor component revealed by moderate cooling in the isolated mesenteric vasculature of Sprague–Dawley rats. J. Pharmacol. Exp. Ther. 262, 1133–1138. Zhao, M., Bo, X., Burnstock, G., 1996. Distribution of w3 Hxa ,b-methylene ATP binding sites in pulmonary blood vessels of different species. Pulmonary Pharmacol. 9, 167–174.
P2X žPurinergic / receptor distributions in rat blood vessels M.A. Hansen a b
a,b
, J.L. Dutton
a,c
, V.J. Balcar
a,b
, J.A. Barden
a,b
, M.R. Bennett
a,c,)
The Institute for Biomedical Research, UniÕersity of Sydney, Sydney, NSW 2006, Australia Department of Anatomy and Histology, UniÕersity of Sydney, Sydney, NSW 2006, Australia c Department of Physiology, UniÕersity of Sydney, Sydney, NSW 2006, Australia Received 31 October 1998; accepted 23 November 1998
Abstract The distribution of purinergic ŽP2X1 and P2X2 . receptors on smooth muscle cells in relation to autonomic nerve varicosities in rat blood vessels has been determined using immunofluorescence and confocal microscopy. P2X1 and P2X2 receptors were visualised using rabbit polyclonal antibodies against the extracellular domain of the receptors and varicosities visualised using a mouse monoclonal antibody against the ubiquitous synaptic vesicle proteoglycan SV2. Two size classes of P2X1 receptor clusters were observed on the smooth muscle cells of mesenteric, renal, and pulmonary arteries as well as in the aorta and in veins: a large approximately elliptical cluster 1.32 " 0.21 mm long and 0.96 " 0.10 mm in diameter; and a smaller spherical cluster with a diameter of 0.32 " 0.05 mm. The latter occurred throughout the media of arteries of all sizes, whereas the former were restricted to the adventitial surface of the media and to endothelial cells, except for the pulmonary artery, in which large receptor clusters were found throughout the media of the vessel. At the adventitial surface, the large clusters are in general located beneath SV2 labelled varicosities. None of the small clusters was associated with varicosities. Three-dimensional reconstruction of the P2X and SV2 labelling at individual varicosities showed that the varicosities were immediately apposed to the P2X receptor clusters. P2X2 receptors were located on nerves and on endothelial cells. They were also found in low density on the smooth muscle cells in the media. These observations are discussed in relation to the mechanism of purinergic transmission to the smooth muscle cells of blood vessels. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Arteries; Immunohistochemistry; Receptors; Veins
1. Introduction Transmission from sympathetic nerves to a number of different blood vessels in the rat involves the release of ATP onto purinergic receptors. These include the aorta ŽWhite et al., 1985; Kitajima et al., 1994; Garcia-Velasco et al., 1995., the pulmonary artery ŽInoue and Kannan, 1988., renal blood vessels ŽSchwartz and Malik, 1989., mesenteric blood vessels ŽAngus et al., 1988; Yamamoto et al., 1992; Windscheif et al., 1994; Brock and Van Helden, 1995.. Autoradiographic studies of the distribution of w3 Hxa ,b-methylene ATP over the media of rat blood vessels indicates that the small and medium-size arteries have a high density of purinergic receptors compared with the elastic and large muscular arteries, with the media of ) Corresponding author. Institute for Biomedical Research, Department of Physiology, The University of Sydney, Sydney, NSW 2006, Australia. T el.: q 61-2-9351-2034; fax: q 61-2-9351-3910; e-m ail: [email protected]
the large muscular arteries showing a higher density of the receptors towards the adventitial surface compared with the intimal surface ŽBo and Burnstock, 1993; Zhao et al., 1996.. However it seems likely that most of the smooth muscle cells in the media of these vessels possess purinergic receptors, as patch-clamped cells dissociated from the media show inward currents in response to exogenous application of ATP Žvon der Weid et al., 1993; Evans and Kennedy, 1994.. Purinergic receptors are of two main subtypes, G-protein-coupled receptors Žtermed P2Y . and ATP-gated channels Žtermed P2X ., with seven of the P2X receptors now isolated and expressed ŽCollo et al., 1996.. Two of these ŽP2X1 and P2X2 . are found in rat mesenteric arterioles, with immunohistochemistry localising the P2X1 receptors to smooth muscle cells and the P2X2 receptors to nerve terminals ŽVulchanova et al., 1996.. Each of the seven P2X receptors is encoded by a different gene, with each possessing two transmembrane domains, a large extracellular domain, and intracellular N- and C-termini ŽSurprenant,
0165-1838r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 1 8 3 8 Ž 9 8 . 0 0 1 8 9 - 1
148
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
1996.. All of these proteins are about 36% to 48% identical ŽSurprenant, 1996.. The P2X1 and P2X2 receptors were originally cloned from complementary DNA that encoded these receptors in rat vas deferens ŽValera et al., 1994. and in rat pheochromocytoma ŽPC12. cells ŽBrake et al., 1994.. In the present work, we have developed antibodies against P2X1 and P2X2 receptors and localised these with respect to single varicosities using antibodies against the ubiquitous vesicle proteoglycan SV2. This has enabled the distribution of these purinergic receptors to be determined at the level of individual sympathetic varicosities at different sites within the media of rat blood vessels.
2. Methods 2.1. Materials Anti-mouse SV2 antibody ŽAbSV2. specific for the synaptic vesicle proteoglycan SV2 ŽBuckley and Kelly, 1985., was kindly provided by Dr Buckley ŽHarvard University.. Cy2 and Cy3 conjugates of donkey anti-mouse and donkey anti-rabbit IgGs, adsorbed against conspecific IgGs, were from Jackson Immunoresearch ŽWest Grove, PA.. All other reagents were purchased from Sigma ŽSt. Louis, MO.. 2.2. Antibody production The consensus sequences of the rat P2X1 ŽValera et al., 1994. and P2X2 ŽBrake et al., 1994. cloned receptors were examined for suitable epitopes and the sequences Lys199– Cys217 and Cys130–Gly153 were synthesised using standard t-BOC chemistry ŽBarden et al., 1997. on an ABI synthesiser ŽHansen et al., 1997.. Ten milligrams of HPLC-purified peptide Ždetermined from ion-spray mass spectrometry. was covalently linked to 45 mg of diphtheria toxin by Chiron Mimotopes who also raised the antibodies
in two New Zealand White rabbits. Peptide–antigen conjugates were suspended in water Ž5 mgrml. and aliquots emulsified in Complete Freund’s Adjuvant. Emulsion volumes of 1 ml containing 0.5 mg of peptide were injected intramuscularly with second, third and fourth immunisations followed at 2 week intervals using Incomplete Freund’s Adjuvant. Final bleeds were obtained at 12 weeks. Blood was incubated at 378C for 30 min, stored at 48C for 15 h and serum collected following centrifugation and stored at y208C in aliquots. Each serum was tested with an ELISA assay for specific antibodies against each peptide, either AbP2X1 or AbP2X2 ŽHansen et al., 1998.. The antibody titre, defined as the reciprocal of the serum dilution resulting in an absorbance of 1.0 above background in the ELISA assay, fell in the range 75 000 " 4000 compared with 225 " 25 for pre-immune samples. Each binds selectively and specifically as shown by Western blots ŽBerry et al., 1998. as well as immunolocalisation. The two antisera were raised against non-homologous extracellular epitopes. Each was chosen in part to ensure no cross-reaction with any of the other subtypes. The two P2X antisera used have been shown to possess similar distributions in many cases but with distinctly different distributions in other cases indicating that they do not lack specificity ŽHansen et al., 1998.. Individual specificity was checked using rat P2X1 cDNA Žkindly provided by Dr R.A. North. transfected into Xenopus oocytes. AbP2X1 was added to transfected cells in the presence and absence of either a 10 mM concentration of the P2X1 epitope or a 10 mM concentration of the P2X2 epitope. Immunolabelling and confocal imaging demonstrates that the expressed P2X1 is located, as expected, within the cell membrane ŽFig. 1A.. The presence of a 10 mM concentration of the P2X1 cognate peptide as an absorption control resulted in the blocking of P2X1 staining ŽFig. 1B. while the presence of the P2X2 cognate peptide also had no effect on AbP2X1 labelling ŽFig. 1C., as expected for a non-homologous epitope to P2X1.
Fig. 1. AbP2X 1 staining is confined largely to the surface membrane of Xenopus oocytes transfected with P2X1 cDNA ŽA.. The addition of a 10 mM concentration of the P2X 1 cognate peptide to the transfected cells resulted in the blocking of P2X1 staining ŽB.. The same concentration of the P2X2 cognate peptide added to the transfected cells in the presence of AbP2X 1 which resulted in no reduction of labelling intensity ŽC.. Calibration, 25 mm.
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
2.3. Tissue preparation Blood vessels from 12 adult rats were removed and fixed in 4% paraformaldehyde in borate–acetate buffer ŽpH 9.5. for 1 h. Tissue was cryoprotected by immersion in 30% sucrose overnight before transverse sections Ž30 mm. were cut on a freezing microtome. Tissue was placed in 0.1% dimethyl sulphoxide ŽDMSO. in phosphate buffered horse serum Ž100 ml PBS, 2 ml normal horse serum ŽNHS., 0.1 ml Triton X-100, 1 g Bovine Serum Albumin. for 30 min to permeabilise the cellular membranes. The preparation was then washed 3 times in PBS Ž10 min each. and immersed in 20% NHS in PBS for 1 h to block non-specific binding sites in the tissue. This was followed by incubation with 1:100 AbSV2 together with either 1:100 pre-immune serum ŽPIS., AbP2X1 or AbP2X2 in PBS for 48 h at 48C. Slides were rinsed in PBS followed by the addition of the relevant secondary fluorescent anti-
149
bodies for 90 min at 208C then washed 3 times in PBS Žeach for 10 min. coverslipped and sealed. Sections were viewed on a Leica TCS NT UV laser confocal microscope system, with pinhole set at 1.0. 2.4. Statistical eÕaluations Sizes of receptor clusters were measured in randomly selected areas and are expressed as mean " standard deviation.
3. Results Tissue sections were exposed to either a combination of antibodies to the P2X1 receptor and the ubiquitous synaptic vesicle proteoglycan SV2 or to P2X2 receptors and SV2.
Fig. 2. The distribution of SV2 ŽA. and P2X 1 ŽB. receptors at the adventitial surface of mesenteric arteries with the media at the top of the figure. Large P2X 1 receptor clusters are colocalised with SV2 Žarrows in centre of figure.. The arrowhead at the right shows an SV2 label not associated with P2X1 . Calibration, 5 mm. The distribution of SV2 ŽC. and P2X 2 ŽD. receptors in the adventitia of intralobular arteries of the kidney with the media at the top of the figure. Distribution is very similar. Calibration, 5 mm.
150
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
3.1. Mesenteric arteries and intralobular renal arteries P2X1 immunoreactivity was prominent at the adventitial surface of a rat mesenteric artery ŽFig. 2B.. Only large clusters, greater than 1 mm diameter, could be distinguished at the adventitial surface where they were inevitably double labelled with SV2 ŽFig. 2A, arrow in centre of figure.. SV2 label was confined to the adventitial surface of the vessel with some SV2 label not associated with P2X1 ŽFig. 2A and B, arrowhead at right of figure.. Puncta labelled with both P2X1 and SV2 are observed on what appear to be the adipocytes labelled with an arrow at the bottom of Fig. 2A and B. Neither SV2 nor distinctly large P2X1 puncta, suggesting the presence of receptor clusters, were observed in the media at the top of Fig. 2A and B. The distribution of P2X2 receptors and SV2-labelled varicosities ŽFig. 2C and D. was similar in the intralobular renal arteries. Large P2X2 receptor clusters also were found at the adventitial surface of the media in high concentration, with most colocalised with SV2 stained varicosities ŽFig. 2C and D.. SV2-labelled varicosities were not observed nor were large P2X2 receptor clusters in the media of the artery. 3.2. Pulmonary artery The distribution of P2X1 receptors and of SV2 in this artery was the same as that in the mesenteric and renal arteries, with the important exception that large P2X receptor clusters could be observed throughout the media ŽFig. 3B and D.. These were often associated with colocalised SV2 immunoreactivity especially close to the adventitia ŽFig. 3A and C.. In the case of P2X1 receptors ŽFig. 3B, near adventitia and Fig. 3D, deeper in the media., these consisted of discrete large high-density clusters, elliptical in shape when reconstructed in three-dimensional view using the confocal microscope, with the ellipse possessing a major axis of 1.32 " 0.17 mm and a minor axis of 0.96 " 0.10 mm Ž n s 70.. A large number of small circular P2X1-labelled clusters were also present throughout the media ŽFig. 3D, arrowhead., possessing a diameter of 0.32 " 0.05 mm and not associated with any SV2 labelling ŽFig. 3C, see corresponding arrowhead.. P2X2 receptors also were found throughout the media in discrete clusters ŽFig. 3F and H.. High power views of these clusters labelled by the large arrowhead in Fig. 3F are shown in
151
Fig. 3H. The small P2X2-labelled clusters in Fig. 3F Žan example shown with the arrow. were not colocalised with SV2 ŽFig. 3E.. The enlargement in Fig. 3H shows more clearly that most of the large P2X2-labelled clusters are associated with SV2 Žarrowheads in Fig. 3G and H. while there are some SV2-labelled nerve structures in Fig. 3G that are not labelled by P2X2 Žarrow in Fig. 3G and H.. The P2X2-labelled clusters also fell into the same size classes as did the P2X1-labelled clusters, with the large clusters usually possessing colocalised SV2 labelling but not the small clusters ŽFig. 3F; see arrow.. 3.3. Endothelial cells The endothelial cells of all vessels were labelled with both P2X1 and P2X2 receptor clusters but not with SV2. An example taken from the endothelial cells of a mesenteric artery is shown in Fig. 4A and B. These P2X1 clusters were of the small-size category and could be discerned to cover the entire surface of the endothelial cells ŽFig. 4B.. The example shown in Fig. 4C and D is taken from the endothelial cells of the aorta in which SV2 labelling is shown in Fig. 4C and the corresponding P2X2 labelling is shown in Fig. 4D. Small-sized discrete puncta can be discerned covering the cells. 3.4. Veins Small veins in the urinary bladder were examined to see if the distribution of receptors in these vessels was different from that in arteries. Large P2X1 receptor patches could be observed on the smooth muscle cells at the adventitial surface of the media of these vessels ŽFig. 4F.. Double labelling with SV2 antibody ŽFig. 4E. showed that a large proportion of these receptor clusters was associated with varicosities ŽFig. 4E and F; see arrowheads.. P2X2 labelled clusters also were evident in the adventitia, and were always colocalised with SV2. Neither P2X1 nor P2X2 labelling was observed elsewhere in the veins.
4. Discussion 4.1. The distribution of P2 X1 and P2 X 2 receptor clusters in blood Õessels Two main transmitters have been identified for the sympathetic innervation of many arteries, namely nor-
Fig. 3. The distribution of P2X 1 and P2X2 receptors in relation to SV2 labelled nerve terminal varicosities in the pulmonary artery. ŽA. and ŽB. show SV2 and P2X 1 labelling, respectively, in the media at the adventitial surface. Calibration, 5 mm. ŽC. and ŽD. show SV2 and P2X1 labelling, respectively, deeper within the media than in ŽA.. Several small 0.32 mm clusters of P2X 1 are present around the large colocalised SV2 and P2X1 clusters. The arrow head shows a small P2X 1 cluster with no SV2. Calibration, 5 mm. ŽE. and ŽF. show SV2 and P2X2 labelling in the media towards the adventitial surface of the artery. A group of P2X 2 rSV2 clusters is indicated by the arrow head and a small P2X2 cluster not colocalised with SV2 by the arrow. Calibration, 20 mm. ŽG. and ŽH., an enlargement of the P2X 2 rSV2 clusters Žtwo arrow heads. indicated by the large arrow head in ŽE. and ŽF.. Some SV2 non-localised with P2X 2 is also seen Žarrow.. Calibration, 12 mm.
152
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
Fig. 4. The localisation of P2X 1 receptors and SV2 on endothelial cells and veins. ŽA. and ŽB. show SV2 and P2X1 labelling, respectively, of an endothelial cell from the intimal surface of a mesenteric artery. Arrow heads point to distinct P2X 1 receptor clusters on the cell. Calibration, 2 mm. ŽC. and ŽD. show SV2 and P2X 2 labelling, respectively, of an endothelial cell from the intimal surface of the aorta. Calibration bars are 2 mm. ŽE. and ŽF. show the distribution of SV2 and P2X 1 , respectively, in veins of the urinary bladder. The arrow head indicates one SV2 stained profile colocalised with a P2X1 cluster. Calibration, 10 mm.
adrenaline and ATP ŽSneddon and Burnstock, 1984; Neild and Kotecha, 1986; Papanicolaou and Medgett, 1986; Abe et al., 1987.. These transmitters are stored in nerves that are usually confined to the adventitial surface of the media ŽBennett, 1972a.. However, in vessels with a large media, above about 100 mm wide, there is often an invasion of the outer 30% of the media by nerves from the adventitial surface including rat tail artery ŽCheung, 1982.; sheep
carotid artery ŽKeatinge, 1966.; dog dorsal pedal artery ŽTsunekawa et al., 1967., a result that has been obtained several times in the case of the pulmonary artery ŽCech and Dolezel, 1967; Fillenz, 1970.. These observations were confirmed in the present work with the mesenteric and renal arteries possessing varicosities only in the adventitia whereas the pulmonary artery contained varicosities throughout the media. Most of these varicosities possessed
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
colocalised P2X1 receptor clusters, whether at the adventitial surface or deeper in the media. The dimension of these clusters was clearer in the media, because of their relative isolation from each other. Clusters were about 1 mm in diameter Žellipsoidal in shape. and were precisely located beneath the varicosities, with only a few hundred nanometres of extension beyond the borders of the varicosities. It is interesting in this regard that varicosities are found that come into close contact with smooth muscle cells at the adventitial surface. There is only a single layer of basal lamina between the varicosity and muscle membranes in arteries with media ranging in size from 2 to 25 cells thick ŽLuff and McLachlan, 1989.. In the case of large muscular arteries, less than half of the varicosities within 1 mm of the muscle cells form close contacts. This situation applies even though these varicosities may be devoid of a complete Schwann cell covering, possess vesicles accumulated towards the side of the varicosity facing the muscle, and even have apparently prejunctional specialisations ŽLuff et al., 1995.. The colocalisation of P2X1 receptor clusters with varicosities might arise from the location of the receptors on the varicosities or on the Schwann cells that partially wrap around these rather than on the apposing smooth muscle cell membrane. It seems likely that the clusters are on the muscle cells. Three-dimensional reconstruction of the distribution of the receptor label with the confocal microscope showed the label confined to an ellipsoid apposing the SV2 labelled varicosity rather than wrapped around the varicosity as would be expected if the receptor was found on Schwann cells. It is also unlikely that the receptor is uniquely present on the varicosities, as the receptors are present following destruction of the sympathetic nerves with 6-hydroxydopamine treatment Žpersonal observations.. Adipose tissue in the adventitia of all the blood vessels examined in the present work also received an innervation from SV2 stained varicosities, with these possessing P2X1 receptor clusters. This confirms the recent discovery that adipocytes possess P2X receptors ŽLee and Pappone, 1997. which, when activated, increase membrane trafficking ŽPappone and Lee, 1996.. One significant observation that emerged from this study was the high density of P2X1 receptors on endothelial cells. It is known that noradrenaline can release ATP from endothelial cells in which it is stored ŽShinozuka et al., 1994; Luff et al., 1995.. ATP may also be released from the endothelial cells by stress as a consequence of increased flow rates through the vessels ŽMilner et al., 1996.. However, the P2 purinoceptor subclass that has generally been taken to mediate these effects, is of the P2Y variety ŽKennedy et al., 1985; White et al., 1985; Ralevic and Burnstock, 1988; Hansmann et al., 1997.. The role of P2X1 receptors on endothelial cells is yet to be elucidated. Smooth muscle cells in the media of all the blood vessels examined in the present work possessed small P2X1 receptor clusters. These clusters, approximately circular
153
with an apparent fluorescence diameter of about 0.3 mm, are considerably smaller than the clusters found beneath the varicosities with a diameter of about 1 mm. Such clusters may mediate the currents recorded in patchclamped smooth muscle cells dissociated from the media of vessels and exposed to ATP Žvon der Weid et al., 1993; Evans and Kennedy, 1994.. Both nerves and endothelial cells possessed P2X2 receptors, as did smooth muscle cells ŽVulchanova et al., 1996, 1997.. These authors however also claimed that P2X2 receptors were excluded from smooth muscle cells. No reports of P2X receptor labelling on endothelial cells have appeared. P2X2 receptors were found together with SV2 labelling in the media of those blood vessels for which the nerves were not confined to the adventitia, that is mostly in the outer 30% of the media of the pulmonary artery. The physiological role of P2X2 receptors on nerves and on endothelial cells is unknown. 4.2. P2 X1 receptors and junctional transmission to smooth muscle The autonomic neuromuscular junction was originally defined in terms of the effects of transmitter released from varicosities acting directly on receptors on smooth muscle, together with radiation into the cell of the electrical signs of transmission due to the action of transmitter on surrounding cells ŽBennett, 1972a,b; Bennett and Gibson, 1995.. In blood vessels, the present work shows high-density receptor clusters are found in close apposition to varicosities. However the question arises as to how the electrical signs of transmission radiate from the adventitial surface, where the receptor clusters are exclusively found in all but the large muscular arteries, to the smooth muscle cells at the intimal surface. One possibility is that excitatory junction potentials ŽEJPs. generated by the release of transmitter onto the adventitial smooth muscle cells ŽAEJPs. are conducted electrotonically through the smooth muscle syncytium to the smooth muscle cells at the intimal surface ŽBennett, 1972a.. Given the known properties of the smooth muscle syncytium ŽBennett et al., 1993, 1995; Henery et al., 1997. and that the smooth muscle cells in the media are each confined to a single plane in the media, this would seem to be possible if a large percentage of the varicosities released transmitter on arrival of the nerve impulse. In this case the adventitial surface of the media is equipotential during the AEJP, so that the syncytium effectively become one-dimensional as far as current flow is concerned. However the probability of the evoked secretion of ATP quanta is very low from varicosities ŽAstrand and Stjarne, 1989; Lavidis and Bennett, 1992., so it is ¨ uncertain if the media acts as a one-dimensional cable or as a three-dimensional syncytium during transmission. Furthermore, even given that current flow is one-dimensional, it is uncertain just how far effective transmission can be radiated from the adventitial surface to produce EJPs at the intimal surface ŽIEJPs..
154
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155
The alternative to syncytial conduction of the EJP is that transmitter released at the adventitial surface can diffuse sufficiently far to affect smooth muscle cells at the intimal surface. Noradrenaline is detectable at the intimal surface of the rat tail artery after high frequency stimulation in the presence of cocaine to block neuronal uptake ŽBao et al., 1993.. Under these conditions, the concentration of noradrenaline at the intimal surface is about 20% of that at the adventitial surface, so that it seems unlikely that appreciable concentrations of noradrenaline will be reached at the intima under normal conditions of an intact uptake mechanism and low frequency stimulation. The case for ATP reaching effective concentrations at the intimal surface of a medium size vessel like the rat tail artery is even more tenuous, given that the media is 12 to 15 smooth muscle cells thick ŽSittiracha et al., 1987.. Firstly, ATP is metabolised by ecto-enzymes in the extracellular space of smooth muscles ŽWestfall et al., 1997.. Furthermore, in order for the P2X class of purinoceptors to be activated, they must bind two ATP molecules, a condition that is difficult to fulfil over long diffusion pathways ŽBennett et al., 1995.. The problem might be resolved by the observation that P2X1 receptors are found in small clusters on all smooth muscle cells in the media of all the vessels studied. There is no evidence for action potential propagation from the adventitial smooth muscle to the intimal smooth muscle. It might be then that the diffusion of ATP from varicosities some distance in from the adventitial surface onto diffuse receptors located deeper into the media towards the intima is sufficient to generate EJPs in the intimal smooth muscle that produce a physiological effect. An alternative explanation for the role of the small receptor clusters is that they are smooth muscle autoreceptors that are activated by the release of ATP from the muscle cells. Evidence for this has been recently forthcoming with the discovery that spontaneous potentials can be recorded from smooth muscle cells at the intimal surface of the media of rat blood vessels that are due to the spontaneous release of ATP from a source other than nerves ŽHenery and Bennett, 1998.. It will be of interest to carry out physiological experiments to determine the role of the P2X receptor clusters described in the present work in the control of blood vessels. Acknowledgements We wish to thank the National Health and Medical Research Council of Australia, Australian Research Council, National Heart Foundation of Australia, and the Clive and Vera Ramaciotti Foundation for financial support. References Abe, K., Matsuki, N., Kasuya, Y., 1987. Pharmacological and electrophysiological discrimination of contractile responses to selective al-
pha 1- and alpha 2-adrenoceptor agonists in rat tail artery. Jpn. J. Pharmacol. 45, 249–261. Angus, J.A., Broughton, A., Mulvaney, M.J., 1988. Role of alpha-adrenoceptors in constrictor responses of rat, guinea-pig and rabbit small arteries to neural activation. J. Physiol. 403, 495–510. Astrand, P., Stjarne, L., 1989. On the secretory activity of single varicosi¨ ties in the sympathetic nerves innervating the rat tail artery. J. Physiol. 409, 207–220. Bao, J.X., Gonon, F., Stjarne, L., 1993. Frequency- and train length-de¨ pendent variation in the roles of postjunctional alpha 1- and alpha 2-adrenoceptors for the field stimulation-induced neurogenic contraction of rat tail artery. Naunyn Schmiedebergs’ Arch. Pharmacol. 347, 601–616. Barden, J.A., Cuthbertson, R.M., Jia-Zhen, W., Moseley, J.M., Kemp, B.E., 1997. Solution structure of parathyroid hormone related protein Žresidues 1–34. containing an Ala substituted for an Ile in position 15 ŽPTHrPwAla15x-Ž1–34.. J. Biol. Chem. 272, 29572–29578. Bennett, M.R., 1972a. Structure and electrical properties of the autonomic neuromuscular junction. Phil. Trans. R. Soc. Lond. Ser. B 265, 25–34. Bennett, M.R., 1972. Autonomic Neuromuscular Transmission. Monographs. Physiological, Society 30. Cambridge Univ. Press. Bennett, M.R., Gibson, W.G., 1995. On the contribution of quantal secretion from close-contact and loose-contact varicosities to the synaptic potentials in the vas deferens. Phil. Trans. R. Soc. Lond. Ser. B 347, 187–204. Bennett, M.R., Gibson, W.G., Poznanski, R.R., 1993. Extracellular current flow and potential during quantal transmission from varicosities in a smooth muscle syncytium. Phil. Trans. R. Soc. Lond. Ser. B 342, 89–99. Bennett, M.R., Farnell, L., Gibson, W.G., Karunanithi, S., 1995. Quantal transmission at purinergic junctions: stochastic interaction between ATP and its receptors. Biophys. J. 68, 925–935. Berry, D., Yao, M., Barden, J.A., Balcar, V.J., Hansen, M.A., Bennett, M.R., Keogh, A., dos Remedios, C.G., 1998. Alterations in the expression of P2X1 receptors in failing and non-diseased human atria. Electrophoresis 19, 856–859. Bo, X., Burnstock, G., 1993. Heterogeneous distribution of w3 Hx a ,bmethylene ATP binding sites in blood vessels. J. Vasc. Res. 30, 87–101. Brake, A.J., Wagenbach, M.J., Julius, D., 1994. New structural motif for ligand-gated ion channels defined by an ionotropic ATP receptor. Nature 371, 519–523. Brock, J.A., Van Helden, D.F., 1995. Enhanced excitatory junction potentials in mesenteric arteries from spontaneous hypertensive rats. Pflug. Arch. Eur. J. Physiol. 430, 901–908. Buckley, K., Kelly, R.B., 1985. Identification of a transmembrane glycoprotein specific for secretory vesicles of neural and endocrine cells. J. Cell. Biol. 100, 1284–1294. Cech, S., Dolezel, S., 1967. Monoaminergic innervation of the pulmonary vessels in various laboratory animals Žrat, rabbit, cat.. Experientia 23, 114–115. Cheung, D.W., 1982. Two components in the cellular response of rat tail arteries to nerve stimulation. J. Physiol. 328, 401–408. Collo, G., North, R.A., Kawashima, E., Merlo-Pich, E., Neidhart, S., Surprenant, A., Buell, G., 1996. Cloning of P2X5 and P2X6 receptors and the distribution and properties of an extended family of ATP-gated ion channels. J. Neurosci. 16, 2495–2507. Evans, R.J., Kennedy, C., 1994. Characterisation of P2-purinoceptors in the smooth muscle of the rat tail artery: a comparison between contractile and electrophysiological responses. Br. J. Pharmacol. 113, 853–860. Fillenz, M., 1970. Innervation of pulmonary and bronchial blood vessels of the dog. J. Anat. 106, 449–461. Garcia-Velasco, G., Sanchez, M., Hidalgo, A., Garcia de Boto, M.J., 1995. Pharmacological dissociation of UTP- and ATP-elicited con-
M.A. Hansen et al.r Journal of the Autonomic NerÕous System 75 (1999) 147–155 tractions and relaxations in isolated rat aorta. Eur. J. Pharmacol. 294, 521–529. Hansen, M.A., Barden, J.A., Balcar, V.J., Keay, K.A., Bennett, M.R., 1997. Structural Motif and Characteristics of the Extracellular Domain of P2X Receptors. Biochem. Biophys. Res. Commun. 236, 670–675. Hansen, M.A., Barden, J.A., Balcar, V.J., Bennett, M.R., 1998. The distribution of single P2X 1 -receptor clusters on smooth muscle cells in relation to nerve varicosities in the rat urinary bladder. J. Neurocytol. 27, 529–539. Hansmann, G., Bultmann, R., Tuluc, F., Starke, K., 1997. Characterisation by antagonists of P2-receptors mediating endothelium-dependent relaxation in the rat aorta. Naunyn Schmiedebergs’ Arch. Pharmacol. 356, 641–652. Henery, R.J., Bennett, M.R., 1998. Synaptic and spontaneous potentials in smooth muscle cells of the endothelial surface of the rat tail artery. Proc. Aust. Neurosci. Soc. 9, 138. Henery, R.J., Gibson, W.G., Bennett, M.R., 1997. Quantal currents and potential in the three-dimensional anisotropic bi-domain model of smooth muscle. Bull. Math. Biol. 59, 1047–1075. Inoue, T., Kannan, M.S., 1988. Nonadrenergic and noncholinergic excitatory neurotransmission in rat intrapulmonary artery. Am. J. Physiol. 254, H1142–H1148. Keatinge, W.R., 1966. Electrical and mechanical response of arteries to stimulation of sympathetic nerves. J. Physiol. 185, 701–715. Kennedy, C., Delbro, D., Burnstock, G., 1985. P2-purinoceptors mediate both vasodilation Žvia the endothelium. and vasoconstriction of the isolated rat femoral artery. Eur. J. Pharmacol. 107, 161–168. Kitajima, S., Ozaki, H., Karaki, H., 1994. Role of different subtypes of P2 purinoceptor on cytosolic Ca2q levels in rat aortic smooth muscle. Eur. J. Pharmacol. 266, 263–267. Lavidis, N., Bennett, M.R., 1992. Probabilistic secretion of quanta from visualized sympathetic nerve varicosities in mouse vas deferens. J. Physiol. 454, 9–26. Lee, S.C., Pappone, P.A., 1997. Effects of P2 purinergic receptor stimulation in brown adipocytes. Am. J. Physiol. 273, C679–C686. Luff, S.E., McLachlan, E.M., 1989. Frequency of neuromuscular junctions on arteries of different dimensions in the rabbit, guinea pig and rat. Blood Vessels 26, 95–106. Luff, S.E., Young, S.B., McLachlan, E.M., 1995. Proportions and structure of contacting and non-contacting varicosities in the perivascular plexus of the rat tail artery. J. Comp. Neurol. 361, 699–709. Milner, P., Boden, P., Burnstock, G., 1996. Long-term guanethidine sympathectomy suppresses flow-induced release of ATP and endothelin from endothelial cells isolated from adult rat aorta. J. Vasc. Res. 33, 139–145. Neild, T.O., Kotecha, N., 1986. Effects of alpha beta methylene ATP on membrane potential, neuromuscular transmission and smooth muscle contraction in the rat tail artery. Gen. Pharmacol. 17, 461–464. Papanicolaou, M., Medgett, I.C., 1986. Effects of blockade of alpha 1and alpha 2-adrenoceptors on vasoconstrictor responses to single and two pulse stimulation in rat tail artery. Eur. J. Pharmacol. 131, 211–218. Pappone, P.A., Lee, S.C., 1996. Purinergic receptor stimulation increases membrane trafficking in brown adipocytes. J. Gen. Physiol. 108, 393–404.
155
Ralevic, V., Burnstock, G., 1988. Actions mediated by P2 purinoceptor subtypes in the isolated perfused mesenteric bed of the rat. Br. J. Pharmacol. 95, 637–645. Schwartz, D.D., Malik, K.U., 1989. Renal periarterial nerve stimulationinduced vasoconstriction at low frequencies is primarily due to release of a purinergic transmitter in the rat. J. Pharmacol. Exp. Ther. 250, 764–771. Shinozuka, K., Hashimoto, M., Masumura, S., Bjur, R.A., Westfall, D.P., Hattori, K., 1994. In vitro studies of release of adenine nucleotides and adenosine from rat vascular endothelium in response to alpha 1-adrenoceptor stimulation. Br. J. Pharmacol. 113, 1203–1208. Sittiracha, T., McLachlan, E.M., Bell, C., 1987. The innervation of the caudal artery of the rat. Neuroscience 21, 647–659. Sneddon, P., Burnstock, G., 1984. ATP as a co-transmitter in rat tail artery. Eur. J. Pharmacol. 106, 149–152. Surprenant, A., 1996. Functional properties of native and cloned P2X receptors. Ciba Found. Symp. 198, 208–219. Tsunekawa, K., Mohri, K., Ikeda, M., Ohgushi, N., Fujiwara, M., 1967. Histochemical demonstration of adrenergic fibers in the smooth muscle layer of media of dorsal–pedal artery in dog. Experientia 23, 842–843. Valera, S., Hussy, N., Evans, R.J., Adami, N., North, R.A., Surprenant, A., Buell, G., 1994. A new class of ligand-gated ion channel defined by P2x receptor for extracellular ATP. Nature 371, 516–591. von der Weid, P.Y., Serebryakov, V.N., Orallo, F., Bergmann, C., Snetkov, V.A., Takeda, K., 1993. Effects of ATP on cultured smooth muscle cells from rat aorta. Br. J. Pharmacol. 108, 638–645. Vulchanova, L., Arvidsson, U., Riedl, M., Wang, J., Buell, G., Surprenant, A., North, R.A., Elde, R., 1996. Differential distribution of two ATP-gated channels ŽP2X receptors. determined by immunocytochemistry. Proc. Natl Acad. Sci. USA 93, 8063–8067. Vulchanova, L., Riedl, M.S., Shuster, S.J., Buell, G., Surprenant, A., North, R.A., Elde, R., 1997. Immunohistochemical study of the P2X2 and P2X3 receptor subunits in rat and monkey sensory neurones and their central terminals. Neuropharmacology 36, 1229–1242. Westfall, T.D., Kennedy, C., Sneddon, P., 1997. The ecto-ATPase inhibitor ARL 67156 enhances parasympathetic neurotransmission in the guinea-pig urinary bladder. Eur. J. Pharmacol. 329, 169–173. White, T.D., Chaudhry, A., Vohra, M.M., Webb, D., Leslie, R.A., 1985. Characteristics of P2 Žnucleotide. receptors mediating contraction and relaxation of rat aortic strips: possible physiological relevance. Eur. J. Pharmacol. 118, 37–44. Windscheif, U., Ralevic, V., Baumert, H.G., Mutschler, E., Lambrecht, G., Burnstock, G., 1994. Vasoconstrictor and vasodilator responses to various agonists in the rat perfused mesenteric arterial bed: selective inhibition by PPADS of contractions mediated via P2X-purinoceptors. Br. J. Pharmacol. 113, 1015–1021. Yamamoto, R., Takasaki, K., Nickols, G.A., 1992. Purinergic vasoconstrictor component revealed by moderate cooling in the isolated mesenteric vasculature of Sprague–Dawley rats. J. Pharmacol. Exp. Ther. 262, 1133–1138. Zhao, M., Bo, X., Burnstock, G., 1996. Distribution of w3 Hxa ,b-methylene ATP binding sites in pulmonary blood vessels of different species. Pulmonary Pharmacol. 9, 167–174.