- Email: [email protected]
Cold storage of rabbit thoracic aorta in University of Wisconsin solution attenuates P2Y2 purine receptors
Cryobiology 44 (2002) 91–102 www.academicpress.com
Cold storage of rabbit thoracic aorta in University of Wisconsin solution attenuates P2Y2 purine receptorsq Sarah J. Payne, Irving S. Benjamin, and Barry Alexander* Liver Sciences Unit, Academic Department of Surgery, GKT School of Medicine, St. Thomas’ Hospital, Lambeth Palace Road, London SE1 7EH, UK Molecular Pharmacology Group, School of Health Science, University of Wolverhampton, 62–68 Lichfield Street, Wolverhampton WV11DJ, UK Received 27 November 2001; accepted 12 March 2002
Abstract Post-transplantation thrombosis may occur in donor segments of iliac arteries and livers following surgical removal and storage in University of Wisconsin (UW) solution for transplantation. We have previously suggested that purine receptors are vulnerable to denaturation after UW storage. The aims of the present study were to determine what particular subtypes of purine P2Y receptors in rabbit thoracic aorta deteriorate after 8 days of UW storage by studying vascular reactivity to acetylcholine, ATP, 2MeSATP and UTP. Ring segments of aortae from male New Zealand White rabbits were mounted upon fine-wire myographs and vasodilatation to the above agents tested on fresh tissue, and after 8 days of UW storage. Vasodilatation to ATP was attenuated by 100 lM L-NAME in fresh tissue suggesting that the relaxant response was, in part, due to nitric oxide (NO). P2Y-mediated relaxation to ATP was significantly attenuated by UW storage and cholinergic responses were not. This attenuated relaxation to ATP was not further attenuated by L-NAME, suggesting a loss of the NO-dependent mechanism. De-endothelialisation indicated that UTP-mediated vasorelaxation, via P2Y2 receptors, was endothelium-dependent. Any residual endothelium-independent relaxation to UTP was abolished by UW storage and endotheliumdependent UTP relaxation was reduced to the same level as that seen in fresh, de-endothelialised tissue. In contrast responses to 2MeSATP, via P2Y1 receptors, were predominantly endothelium-independent and were only partially attenuated by UW storage. Responses to pyridoxalphosphate-6-azophenyl-20 ; 40 -disulphonic acid (PPADS) and L-NAME suggested that vasorelaxation to 2MeSATP and UTP was mediated by P2Y1 and P2Y2 receptors, respectively. It is therefore concluded that UW storage predominantly decreases P2Y2 receptor-mediated vascular reactivity. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: Preservation; University of Wisconsin solution; P2 Y Purine receptors; ATP; Arteries
q
Dr. Sarah Payne was supported by grants provided by King’s Medical Research Trust and The British Federation of Women Graduates Charitable Foundation. * Corresponding author. Fax: 0044-0207-928-8742. E-mail address: [email protected] (B. Alexander). 0011-2240/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII: S 0 0 1 1 - 2 2 4 0 ( 0 2 ) 0 0 0 1 0 - X
92
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
Liver transplantation frequently requires segments of (iliac) arteries to be used as conduits for re-connection of the hepatic arterial supply to the donor liver. However, hepatic artery thrombosis has been reported following transplantation [29] and may not become manifest until the tenth postoperative day [22]. Therefore arterial vessel remnants are stored in University of Wisconsin (UW) solution at 4 °C for up to 10 days to cover the event of requirement for re-anastomosis to replace the thrombosed artery. The resultant thrombosis may be due to poor vessel ‘quality’ and/or deficiencies in storage and collection techniques and it has been proposed that endothelial function may become compromised by storage of these vessels in cold-preservation solutions [15]. However, it is currently believed that UW storage may be beneficial specifically for preservation of the vascular endothelium in rabbit thoracic aorta for up to 72 h
[19], although progressive endothelial degeneration after 72 h UW storage has been reported [12,27]. The presence of purine receptors has been demonstrated in the hepatic vasculature of the dog [23], pig [5], rabbit [6], and rat liver [30]. UW storage of rabbit thoracic aorta resulted in reduced endothelium-independent relaxation to sodium nitroprusside and morphologically wellpreserved vascular endothelium [18] and smooth muscle [3]. Storage in Krebs–B€ ulbring buffer, which histologically appeared to destroy the vascular endothelium, attenuated acetylcholineinduced relaxation but preserved ATP-induced relaxation. In addition, UW storage maintained endothelium-dependent relaxation to acetylcholine but attenuated endothelium-dependent relaxation to ATP [4]. ATP-induced vasorelaxation may occur via NO-dependent, endothelium-inde-
Fig. 1. Current classification of purine receptors according to ranked order of agonist potency and molecular cloning of specific receptors on each tissue based originally upon a proposal by Abbracchio and Burnstock [1]. The ranked order of agonist potencies of various P2X purine receptors, that elicit vasoconstriction on the left-hand side of the vertical dotted line, has been omitted due to spatial constraints and because the focus of the present manuscript is upon P2Y receptors. The right-hand side of the dotted line illustrates the classification of P2Y receptor subtypes that elicit vasodilatation. The classification of P2Y receptor subtypes is continually under review and, at the present time, there is a consensus that only 6 subtypes of receptor exist. Their classification varies according to the author and some believe that the P2Y3 and P2Y6 may be homologues of each other and therefore should not be separate entities [20,21]. Conversely, it has been argued that other subtypes exist such as the P2Y7 receptor which has been postulated to be a specific leukotrine B4 receptor [16] and that these should be included as an additional receptor subtype. P2Y1 and P2Y2 receptors, as the present study has shown, exist on vascular smooth muscle and endothelium. Delineation between the 2 is therefore based upon ranked order of agonist potency, molecular cloning, and action, respectively, of some putative specific inhibitors which, in the present study, was PPADS. Receptor subtypes with only 1 or 2 agonists mentioned imply that these are the major agonists above all others. Thus, the finite delineations of this area of research are in a state of flux at the present time.
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
pendent P2Y1 , and endothelium-dependent P2Y2 receptors (Fig. 1). 2-Methylthio ATP (2MeSATP) and uridine-50 -triphospate (UTP) elicit relaxation by activation of P2Y1 and P2Y2 receptor subtypes, respectively, whilst ATP is a potent agonist at both receptor subtypes [9]. Therefore previusly observed ATP-induced relaxation in (endothelium-disrupted) Krebs–B€ ulbring-stored tissue and (preserved-endothelium) UW-stored tissue may have occurred via either of these receptors. Moreover, either of these receptor subtypes may become damaged during UW storage, thus, providing an explanation for previous data, the implication being that muscarinic receptors are more resilient to UW storage. Differentiation between P2Y1 and P2Y2 purine receptor-mediated vasorelaxation may be achieved by the use of the P2Y1 antagonist PPADS [10,11,25]. The aim of the present study therefore was to determine which subtype of P2Y receptors degenerates following UW storage by the use of PPADS and the nitric oxide (NO) synthase inhibitor L -arginine methyl ester (L-NAME).
Methods Tissue preparation Male New Zealand White rabbits weighing between 2.2 and 2.9 kg were killed by a lethal injection of sodium pentobarbitone through a cannulated ear vein. Following thoracolaparotomy, the thoracic aorta was carefully excised avoiding vessel traction and cleaned of adherent fat and connective tissue in a sterile petri dish containing Krebs–B€ ulbring buffer at room temperature. Rings, 4 mm in length, were cut from a
93
20 mm section of aorta located near the heart to decrease error arising due to differential distribution of receptors [7,31]. During the course of tissue preparation, care was taken to avoid contact with the internal surface of the aorta to maintain endothelial cell integrity. Each ring was placed under 2 g tension upon a fine wire myograph, according to our previous protocols [18,4], and calibrated for isometric measurements in Krebs–B€ ulbring buffer, the composition of which is detailed in Table 1. This was heated to 37 °C and oxygenated with 95%/5% O2 =CO2 . The fine wire myographs were connected to FT03 force displacement transducers for recording with a Grass 79F Polygraph (Grass Instrument, Quincy, MA, USA). All drugs were made up in fresh distilled water. Noradrenaline was dissolved in 0.1 mM ascorbic acid (Vitamin C) to prevent oxidation. Experimental design Aortic rings of tissue were divided into intact or de-endothelialised tissue, ensuring that rings within a given group originated from different aortae [18]. Rings from each segment of aorta were either tested immediately upon fresh tissue (in Krebs–B€ ulbring buffer) upon harvesting or after storage for 8 days in sterile Krebs–B€ ulbring buffer, UW (see Table 1) or saline, n ¼ 6 per group except with fresh tissue where n ¼ 16. Eight days storage was chosen because this is the maximum time allocated at King’s College Hospital for storage of remnant arteries to cover the event of requirement for reanastomosis of thrombosed arteries following the initial transplantation [22]. De-endothelialisation was achieved by rotating the aorta around an internally inserted glass rod.
Table 1 Composition of UW solution and Krebs–B€ ulbring buffer per litre of water, both titrated to a pH 7.4 University of Wisconsin solution Kþ lactobionate KH2 PO4 MgSO4 Raffinose Adenosine Glutathione Insulin Penicillin Dexamethasone Allopurinol Hydroxyethyl starch
Krebs’–B€ ulbring buffer 100 mmol 25 mmol 5 mmol 30 mmol 5 mmol 3 mmol 100 IU 40 IU 8 mg 1 mmol 50 g
NaCl KCl MgSO4 NaHCO3 NaH2 PO4 CaCl2 D-Glucose
1.33 mmol 4.7 mmol 0.61 mmol 16.3 mmol 1.35 mmol 2.52 mmol 7.8 mmol
94
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
Successful de-endothelialisation was measured by the lack of relaxation to acetylcholine whilst contraction to noradrenaline (2 lM) was maintained. Some aortic rings were de-endothelialised by the method described and histologically examined to confirm complete removal of the endothelial monolayer.
curves in the presence and absence of PPADS, and L-NAME were analysed by Student’s, paired t test (2-tailed). Emax (the minimum concentration of drug required to elicit the maximum response) and EC50 values were calculated from statistical analyses using the curve-fitting programme, Graph Pad Prism (Graph Pad Software, San Diego, CA, USA).
Experimental protocol All vessels were equilibrated for 1 h prior to the commencement of any experiments and flushed with fresh Krebs–B€ ulbring buffer at 37 °C every 15 min. Following equilibration, vessels were precontracted with a dose of noradrenaline (2 lM), a concentration found from a cumulative dose–response curve constructed at the start of every experiment to evoke approximately 70% of the maximum contraction. Endothelium-dependent relaxation to ATP, UTP, and 2MeSATP was studied in fresh tissue and in samples following 8-day cold storage in either UW, Krebs–B€ ulbring buffer or physiological saline (0.9% NaCl, Polyfusor UK) all at pH 7.4. Responses were tested in the presence and absence of L-NAME (100 lM) and PPADS (1 lM). Responses to acetylcholine and the nucleotides were expressed as percentage relaxation from the submaximal level of precontraction produced by noradrenaline. Drugs used The following drugs were obtained from Sigma Poole (UK): acetylcholine, uridine triphosphate (sodium salt), L-NAME, adenosine triphosphate, and noradrenaline. 2-methylthio ATP and PPADS were from Research Biochemicals Inc. UW solution was provided by Du Pont Pharmaceuticals, UK. Statistical analysis Results from concentration-effect curves were expressed as means standard error of the mean (SEM). The data were checked for a normal distribution and a one-way multivariate analysis of variance (ANOVAR) was applied to check for significant differences. Where differences were apparent, a Student’s 2-tailed, unpaired t test with Bonferroni adjustment was applied to test for differences in responses at specific concentration points, P < 0:05 taken as being significantly different. Results for paired concentration-effect
Results Cholinergic responses Acetylcholine produced dose-dependent vasorelaxation in all vessels tested following precontraction with 2 lM noradrenaline that consistently produced an average of 71:3 2:4% submaximal contraction (Fig. 2). De-endothelialisation totally abolished relaxation in fresh tissue. There were no significant differences between fresh tissue, UW-, and Krebs–B€ ulbring-stored vessels. However, responses following 8-day storage in saline were significantly attenuated compared to UW-stored, Krebs–B€ ulbring-stored, and fresh tissue vessels ðP < 0:01Þ. Responses to ATP ATP produced dose-dependent relaxation in fresh tissue and after UW-, Krebs–B€ ulbring- or saline-storage (Fig. 3a). Vasorelaxation to ATP following storage in Krebs–B€ ulbring and in fresh tissue was significantly greater than that stored in UW or saline. Emax responses to ATP were significantly greater in Krebs–B€ ulbring, stored vessels ð52:52 4:02%Þ than in UW-stored (43:73 1:48%, P < 0:001) and de-endothelialised vessels (44:20 1:57%, P < 0:05). Thus storage in Krebs– B€ ulbring-buffer preserved ATP-mediated vasorelaxation in contrast to storage in UW or saline where it was attenuated. De-endothelialisation of fresh tissue significantly attenuated vasorelaxation compared to intact tissues, although it was not abolished (Fig. 3a). This suggested that ATPmediated vasorelaxation occurred via both endothelium-dependent and -independent mechanisms. There were no significant differences in de-endothelialised tissues between fresh tissue and those stored in UW, Krebs–B€ ulbring or saline (Fig. 3b) and thus it seemed reasonable to suggest that the previously observed differences could have been attributable to endothelium-dependent mechanisms. Therefore storage in Krebs–B€ ulbring buffer
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
95
Fig. 2. Concentration-dependent relaxation to acetylcholine in ring segments of rabbit thoracic aorta fresh tissue, deendothelialised fresh tissue, and after 8 days storage in saline or UW solution. The % vasorelaxation, in this and subsequent graphs, was calculated as the % decrease from the original resting tone of 71.3% of the maximum tension (vasoconstriction) generated by noradrenaline. 2 lM noradrenaline was required to achieve this percentage of sub maximal contraction. De-endothelialisation totally abolished acetylcholine-induced vasorelaxation. There were no significant differences in responses between fresh tissue and UW-stored tissues. However, these were significantly greater than those stored in saline and, obviously, those recorded from de-endothelialised tissue ðP < 0:01Þ.
appeared to preserve endothelium-dependent vasorelaxation. Significant attenuation of ATP-induced vasorelaxation by L-NAME in freshly obtained tissue reconfirmed at least partial vasorelaxation via NO (Fig. 4a). Responses to ATP in fresh tissue were not significantly attenuated by 10 lM 8-phenyltheophylline (8-PT), the non-specific P1 A2 adenosine receptor antagonist, in either endothelium-intact or de-endothelialised tissues (data not presented) and thus confirmed that vasorelaxation to ATP was not due to catabolism to adenosine. Responses to 2MeSATP 2MeSATP Produced dose-dependent relaxation, indicative of P2Y1 purine receptor activation, in fresh tissue that was attenuated by 28% at the Emax after removal of the endothelium (Fig. 5a). This suggested that 72% of the vasorelaxation to 2MeSATP occurred via direct smooth muscle activation. The significant attenuation of UWand saline-stored vessels compared to fresh tissue,
de-endothelialised, and Krebs–B€ ulbring-stored tissue was indicative that UW- and saline-storage down-regulated a mixture of both endotheliumdependent and direct smooth muscle receptors that are activated by 2MeSATP (Fig. 5a). PPADS caused a highly significant attenuation of responses to 2MeSATP in fresh tissue suggesting P2Y1 receptor activation (Fig. 7a). L-NAME significantly attenuated the Emax of 2MeSATP in fresh tissue by 56% from 40:63 0:77 to 17:77 0:75%, P < 0:05 Students’ paired t test n ¼ 6 (Fig. 4b). This suggested that up to 56.3% of the P2Y1 receptor-mediated response was NOdependent in this tissue. De-endothelialised tissues produced no significant differences between fresh and Krebs–B€ ulbring-stored tissue (Fig. 5b) and thus it can be deduced that smooth muscle receptors remained intact. In addition, there were no significant differences between saline- and UWstored tissue compared to de-endothelialised tissue stored in Krebs–B€ ulbring or in fresh tissue (Fig. 5b) and this again suggested that the majority of smooth muscle receptors responding to 2MeSATP were not damaged.
96
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
a
b
Fig. 3. Concentration-dependent relaxation to ATP in: (a) intact tissues, fresh tissue, and after 8 days of storage in saline, Krebs–B€ ulbring or UW solution. Responses in Krebs–B€ ulbring and fresh tissue were significantly greater than those obtained from de-endothelialised, saline-, and UW-stored tissue ðP < 0:05Þ; and (b) de-endothelialised tissue, there were no significant differences between de-endothelialised stored and fresh tissue.
Responses to UTP UTP produced dose-dependent vasorelaxation in fresh tissue, UW-, Krebs–B€ ulbring, and salinestored tissues (Fig. 6a). Responses to UTP following storage in Krebs–B€ ulbring, saline, and fresh tissue were significantly greater than those stored in UW at all concentrations of UTP tested,
up to and including the Emax ðP < 0:05Þ. De-endothelialisation virtually abolished vasorelaxation at all concentrations of UTP tested in stored compared to fresh tissue (Fig. 6b). L-NAME significantly attenuated UTP-induced relaxation at all concentrations of UTP (Fig. 4c). The Emax of UTP was significantly attenuated from 48:55 6:38 to 12:85 2:89% P < 0:001 Stu-
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
a
97
different P2Y2 receptors. This was in complete contrast to 2MeSATP where PPADS virtually abolished all (P2Y1 -induced) vasorelaxation in UW-stored tissue.
Discussion
b
c
Fig. 4. Concentration-dependent relaxation to (a) ATP; (b) 2MeSATP and (c) UTP with (j + L-NAME) and without ( ) L-NAME) in fresh tissue. There a significant shift to left of the ATP-induced responses ðP < 0:05Þ and a highly significant attenuation of 2MeSATP- ðP < 0:05Þ and UTP-induced responses ðP < 0:001Þ.
dent’s paired t test, representing a reduction of 74.5% of the original response, thus, suggesting that up to 75% of the response to UTP was NOdependent. Therefore vasorelaxation to UTP which was predominantly endothelium- and NOdependent was significantly attenuated following storage in UW compared to storage in Krebs– B€ ulbring buffer or saline. PPADS significantly enhanced vasorelaxation to UTP in fresh tissue (Fig. 7b), thereby suggesting that UTP acted via
The data from the present study confirmed previous observations of retained acetylcholineinduced [19] and attenuated ATP-induced vasorelaxation [4] in rabbit thoracic aorta following UW storage. Preliminary observations of retained ATP-induced, endothelium-dependent, relaxation following storage in Krebs–B€ ulbring buffer and attenuation of these responses following UW storage where the vascular endothelium appeared structurally well preserved [4] were also reconfirmed in the present study. However, these earlier studies could not establish which particular subtype of, or how, purine receptors altered during UW cold storage. This was particularly relevant because the vascular endothelium appeared well preserved in UW-stored vessels yet disintegrated in Krebs–B€ ulbring buffer-stored vessels where ATP-induced vasorelaxation was conserved and similar to that in fresh tissue. The present study showed that ATP-induced relaxation was P2Y receptor-driven because; (i) this was significantly attenuated by both de-endothelialisation (Fig. 3a) and by L-NAME (Fig. 4a) in fresh tissue; (ii) residual responses to ATP were not mediated via P1 (adenosine) purine receptor activation because they were not attenuated by the P1 A2 receptor antagonist, 8-PT, in either intact or de-endothelialised tissues (data not presented); and (iii) concentration-dependent relaxation was demonstrated to 2MeSATP and UTP, agonists believed to activate P2Y1 - and P2Y2 -receptors, respectively [9]. Data from the present study demonstrated that, after UW storage, the reduced endotheliumdependent, ATP-induced, vasorelaxation was similar to that in de-endothelialised fresh tissue or storage in saline. However, de-endothelialisation demonstrated that 72% of the 2MeSATP-induced vasorelaxation was conserved in fresh tissue (Fig. 5a) and 81.6% in Krebs–B€ ulbring buffer stored tissue. L-NAME significantly attenuated fresh tissue responses to 2MeSATP by 56% thus indicating almost an equal division between P2Y1 NO-dependent and -independent receptors. However, this was in contrast to P2Y2 receptors where de-endothelialisation and/or UW storage
98
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
a
b
Fig. 5. Concentration-dependent relaxation to 2MeSATP in: (a) intact tissues; and (b) de-endothelialised tissues in fresh tissue and after storage in saline, Krebs–B€ ulbring or UW solution. There was a highly significant attenuation of responses in UW- and saline-stored tissues compared to fresh tissue-, Krebs–B€ ulbring, and de-endothelialised tissue in intact tissues ðP < 0:05Þ. Removal of the endothelium (b) abolished the presumably endothelium-dependent differences between Krebs–B€ ulbring-stored and UW- and saline-stored tissue since all 3 were attenuated compared to fresh tissue ðP < 0:05Þ.
virtually abolished responses to UTP (Fig. 6b). In addition, de-endothelialisation of fresh tissue also abolished responses to UTP, produced similar graphs to intact UW-stored tissue (Fig. 6a), and supported the hypothesis that most P2Y2 receptors were located on the endothelium. PPADS was used to further segregate between P2Y1 and P2Y2 receptor subtypes by its differential activity on 2MeSATP- and UTP-induced vasorelaxation, respectively. PPADS significantly
attenuated 2MeSATP-induced vasorelaxation in fresh tissues (Fig. 7a) but, in contrast, it actually enhanced responses to UTP (Fig. 7b) [11,25]. This enhancement of UTP-mediated responses may have been due to inhibition of ectonucleotidase activity by PPADS which effectively increases the net agonist concentration available at the receptor binding site [13,28], although this remains to be confirmed in future studies. The NO-dependent proportion of P2Y2 (UTP-mediated) receptors
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
99
a
b
Fig. 6. Concentration-dependent relaxation to UTP in (a) intact tissues and (b) de-endothelialised tissues in fresh tissue and in tissues after storage in saline, Krebs–B€ ulbring or UW solution. (a) There was a highly significant attenuation of responses in de-endothelialised and UW-stored tissue compared to fresh tissue, Krebs–B€ ulbring-, and saline-stored tissue ðP < 0:05Þ. (b) De-endothelialisation virtually abolished relaxation in all of the stored tissues compared to fresh tissue.
was much greater than that of P2Y1 receptors and accounted for up to 80% of the population according to L-NAME attenuation (Fig. 4c). These observations therefore indicated that 2MeSATP and UTP acted via different receptor subtypes and according to the present study, storage in UW predominantly inhibited the major proportion of NO- and endothelium-dependent P2Y2 receptors. Histological evidence from our previous studies suggested that the morphology of the endo-
thelium appeared well preserved following UW storage [4,18] and that cholinergic endotheliumdependent vasorelaxation is retained in UWstored tissue (Fig. 2). However, this did not necessarily prove that all vascular receptors retain their functional capacity following UW storage and that some may become denatured at a quaternary or tertiary structural level, undetectable using light or electron microscopy. Another underlying mechanism that may explain our data
100
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
a
b
original reason for inclusion of adenosine into UW solution was to replenish adenosine levels for ATP synthesis and maintain cell membrane integrity after reperfusion [26]. However, recent studies showed that long-term exposure to adenosine were detrimental because of increased generation of cytokines [14], decreased production of vascular endothelial growth factor [8] increased nitrite and nitrate, myeloperoxidase activity, and lipoperoxidation levels [24]. In addition, the present study suggests that P2Y2 receptor down regulation occurred following UW storage within the purine (ATP and UTP) signal transduction pathways that led to NO production. It is unclear how the above mechanisms may affect these pathways but the production of oxygen free radicals, possibly nitrites and nitrates via NO production following reperfusion [2], could be involved. Additional studies suggested that omission of UW colloid elements may reduce impaired endothelium-dependent hyperpolarisation factor (EDHF)-mediated activity, inhibited Ca2þ -activated and attenuated ATP-sensitive Kþ channels [17] following UW storage. However, it is impossible to ascertain precisely which particular stage(s) within the P2Y2 signal transduction pathway are altered by UW storage. Moreover, lack of vascular relaxation may promote thrombosis via creating a low flow state but it remains unknown whether this is a primary causal effect for thrombosis or if thrombosis occurs when altered reactivity is combined with other damage such as overt endothelial damage and these concepts remain the subject of future studies.
Acknowledgments Fig. 7. Concentration-dependent relaxations to (a) 2MeSTP and (b) UTP with (j + PPADS) and without ( ) PPADS) in fresh tissue. PPADS caused a highly significant attenuation to responses above 7:0 log (M) 2MeSATP (P < 0:01, Student’s paired t test).
may be down-regulation of P2Y2 -receptors, specifically during storage, that may occur due to the unique composition of UW solution. UW solution has a particularly high adenosine and Kþ ion concentration and although the tissues were incubated in fresh Krebs–B€ ulbring solution for 1 h and flushed regularly before experiments commenced, the long-lasting effects of these components, particularly following ischaemia reperfusion injury, are currently unknown. The
We thank Du Pont Pharmaceuticals (UK, Ltd.) for their gift of UW solution. Dr. C.A. Brown is thanked for his invaluable comments and assistance in the preparation of this manuscript.
References [1] M.P. Abbracchio, G. Burnstock, Purinoceptors: Are there families of P2X and P2Y purinoceptors? Pharmacol. Ther. 64 (1994) 445–475. [2] B. Alexander, The role of nitric oxide in hepatic metabolism, Nutrition 14 (1998) 376–390. [3] B. Alexander, J.V. Gryf-Lowczowski, J.S. Marway, J. Salisbury, D. Sherlock, V. Preedy, I.S. Benjamin,
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
Functional and compositional studies of arteries stored in University of Wisconsin solution compared with Krebs–B€ ulbring buffer, Exp. Mol. Pathol. 71 (2001) 265–273. B. Alexander, J.V. Gryf-Lowczowski, D. Sherlock, J. Salisbury, I.S. Benjamin, Paradoxical cholinergic and purinergic vascular reactivity of rabbit thoracic aorta cold-stored in University of Wisconsin solution, J. Pharm. Pharmacol. 51 (1999) 623–630. B. Alexander, R.T. Mathie, Diminished hyperaemic response of the hepatic artery to portal venous occlusion (the buffer response) in Asian hybrid minipigs: a comparison of the response to that observed in dogs, J. Comp. Physiol. [B] 163 (1993) 5–10. B. Alexander, R.T. Mathie, V. Ralevic, G. Burnstock, An isolated dual-perfused rabbit liver preparation for the study of hepatic blood flow regulation, J. Pharmacol. Toxicol. Methods 27 (1992) 17–22. B.M. Altura, B.T. Altura, Heterogeneity of drug receptors in different segments of rabbit thoracic aorta, Eur. J. Pharmacol. 12 (1970) 44–52. A.J. Archambault, M.G. Sirois, P.N. Bernatchez, C. Fiset, P.S. Haddad, Vascular endothelial growth factor production by isolated rat hepatocytes after cold ischemia-warm reoxygenation, Liver Transpl. 7 (2001) 988–997. M.R. Boarder, S.M. Hourani, The regulation of vascular function by P2 receptors: multiple sites and multiple receptors, Trends Pharmacol. Sci. 19 (1998) 99–107. C.A. Brown, S.J. Charlton, M.R. Boarder, Enhancement of the response to purinergic agonists in P2Y1 transfected 1321N1 cells by antagonists suramin and PPADS, Br. J. Pharmacol. 120 (1997) 1049–1052. S.J. Charlton, C.A. Brown, G.A. Weisman, J.T. Turner, L. Erb, M.R. Boarder, Cloned and transfected P2Y4 receptors: characterization of a suramin and PPADS-insensitive response to UTP, Br. J. Pharmacol. 119 (1996) 1301–1303. E.G. Chedrawy, J.A. Sullivan, G. Li, M.M. Vohra, Effects of University of Wisconsin preservation solution on rabbit aortic endothelium and smooth muscle, Br. J. Pharmacol. 116 (1995) P167. B.C. Chen, C.M. Lee, W.W. Lin, Inhibition of ectoATPase by PPADS, uramin and reactive blue in endothelial cells, C6 glioma cells and RAW 264.7 macrophages, Br. J. Pharmacol. 119 (1996) 1628– 1634. C.H. Chiang, C.P. Yu, C.P. Wu, H.C. Yan, W.C. Perng, Cytokine up-regulation in ischaemic/reperfused lungs perfused with University of Wisconsin solution and normal saline, Clin. Sci. (Lond) 101 (2001) 285–294. M.G. Clemens, K. Chun, E. Miescher, D. Jones, J. Zhang, Leukocyte-dependent and -independent hepatic microvascular injury during reperfusion after
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
101
warm ischaemia, Prog. Appl. Microcirc. 19 (1993) 139–151. V.R. Dasari, J. Jin, S.P. Kunapuli, Distribution of leukotriene B4 receptors in human hematopoietic cells, Immunopharmacology 48 (2000) 157–163. Z.D. Ge, G.W. He, Comparison of University of Wisconsin and St. Thomas’ Hospital solutions on endothelium-derived hyperpolarizing factor-mediated function in coronary micro-arteries, Transplantation 70 (2000) 22–31. J.V. Gryf-Lowczowski, D. Sherlock, J. Salisbury, I.S. Benjamin, B. Alexander, Cold-storage of rabbit thoracic aorta in University of Wisconsin solution reduces endothelium-independent vasodilation, J. Pharm. Pharmacol. 49 (1997) 1096–1101. S. Karatapanis, P.A. McCormick, M. Jacobs, G. Stansby, K. Rolles, N. McIntyre, A.K. Burroughs, J.Y. Jeremy, Effect of liver and kidney preservation solutions on vascular viability: studies on contraction and relaxation of rat aorta in vitro, Transplant. Proc. 25 (1993) 2148–2152. Q. Li, M. Olesky, R.K. Palmer, T.K. Harden, R.A. Nicholas, Evidence that the p2y3 receptor is the avian homologue of the mammalian P2Y6 receptor, Mol. Pharmacol 54 (1998) 541–546. Q. Li, J.B. Schachter, T.K. Harden, R.A. Nicholas, The 6H1 orphan receptor, claimed to be the p2y5 receptor, does not mediate nucleotide-promoted second messenger responses, Biochem. Biophys. Res. Commun. 236 (1997) 455–460. W.C. Marujo, A.N. Langnas, R.P. Wood, R.J. Stratta, S. Li, B.W. Shaw Jr., Vascular complications following orthotopic liver transplantation: outcome and the role of urgent revascularization, Transplant. Proc. 23 (1991) 1484–1486. R.T. Mathie, B. Alexander, The role of adenosine in the hyperaemic response of the hepatic artery to portal vein occlusion (the ‘buffer response’), Br. J. Pharmacol. 100 (1990) 626–630. F. Pi, F. Badosa, A. Sola, C.J. Rosello, C. Xaus, N. Prats, E. Gelpi, G. Hotter, Effects of adenosine on ischaemia-reperfusion injury associated with rat pancreas transplantation, Br. J. Surg. 88 (2001) 1366–1375. V. Ralevic, G. Burnstock, Discrimination by PPADS between endothelial P2Y- and P2U-purinoceptors in the rat isolated mesenteric arterial bed, Br. J. Pharmacol. 118 (1996) 428–434. J.H. Southard, T.M. van Gulik, M.S. Ametani, P.K. Vreugdenhil, S.L. Lindell, B.L. Pienaar, F.O. Belzer, Important components of the UW solution, Transplantation 49 (1990) 251–257. M.M. Vohra, E.G. Chedrawy, G. Li, J.A. Sullivan, Functional assessment of rabbit thoracic aorta stored in University of Wisconsin, St. Thomas’ Hospital and Krebs solutions, Br. J. Pharmacol. 120 (1997) P153. U. Windscheif, O. Pfaff, A.U. Ziganshin, C.H. Hoyle, H.G. Baumert, E. Mutschler, G. Burnstock,
102
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
G. Lambrecht, Inhibitory action of PPADS on relaxant responses to adenine nucleotides or electrical field stimulation in guinea-pig taenia coli and rat duodenum, Br. J. Pharmacol. 115 (1995) 1509–1517. [29] K. Yanaga, G. Lebeau, J.W. Marsh, R.D. Gordon, L. Makowka, A.G. Tzakis, S. Todo, A.C. Stieber, S. Iwatsuki, T.E. Starzl, Hepatic artery reconstruction for hepatic artery thrombosis after orthotopic liver transplantation, Arch. Surg. 125 (1990) 628–631.
[30] W. Yang, I.S. Benjamin, B. Alexander, Localisation of hepatic vascular resistance sites in the isolated dual-perfused rat liver, Eur. J. Pharmacol. 364 (1999) 13–21. [31] W. Yang, I.S. Benjamin, B. Alexander, Nitric oxide modulates acetylcholine-induced vasodilatation in the hepatic arterial vasculature of the dual-perfused rat liver, Acta Physiol. Scand. 171 (2001) 413– 418.
Cold storage of rabbit thoracic aorta in University of Wisconsin solution attenuates P2Y2 purine receptorsq Sarah J. Payne, Irving S. Benjamin, and Barry Alexander* Liver Sciences Unit, Academic Department of Surgery, GKT School of Medicine, St. Thomas’ Hospital, Lambeth Palace Road, London SE1 7EH, UK Molecular Pharmacology Group, School of Health Science, University of Wolverhampton, 62–68 Lichfield Street, Wolverhampton WV11DJ, UK Received 27 November 2001; accepted 12 March 2002
Abstract Post-transplantation thrombosis may occur in donor segments of iliac arteries and livers following surgical removal and storage in University of Wisconsin (UW) solution for transplantation. We have previously suggested that purine receptors are vulnerable to denaturation after UW storage. The aims of the present study were to determine what particular subtypes of purine P2Y receptors in rabbit thoracic aorta deteriorate after 8 days of UW storage by studying vascular reactivity to acetylcholine, ATP, 2MeSATP and UTP. Ring segments of aortae from male New Zealand White rabbits were mounted upon fine-wire myographs and vasodilatation to the above agents tested on fresh tissue, and after 8 days of UW storage. Vasodilatation to ATP was attenuated by 100 lM L-NAME in fresh tissue suggesting that the relaxant response was, in part, due to nitric oxide (NO). P2Y-mediated relaxation to ATP was significantly attenuated by UW storage and cholinergic responses were not. This attenuated relaxation to ATP was not further attenuated by L-NAME, suggesting a loss of the NO-dependent mechanism. De-endothelialisation indicated that UTP-mediated vasorelaxation, via P2Y2 receptors, was endothelium-dependent. Any residual endothelium-independent relaxation to UTP was abolished by UW storage and endotheliumdependent UTP relaxation was reduced to the same level as that seen in fresh, de-endothelialised tissue. In contrast responses to 2MeSATP, via P2Y1 receptors, were predominantly endothelium-independent and were only partially attenuated by UW storage. Responses to pyridoxalphosphate-6-azophenyl-20 ; 40 -disulphonic acid (PPADS) and L-NAME suggested that vasorelaxation to 2MeSATP and UTP was mediated by P2Y1 and P2Y2 receptors, respectively. It is therefore concluded that UW storage predominantly decreases P2Y2 receptor-mediated vascular reactivity. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: Preservation; University of Wisconsin solution; P2 Y Purine receptors; ATP; Arteries
q
Dr. Sarah Payne was supported by grants provided by King’s Medical Research Trust and The British Federation of Women Graduates Charitable Foundation. * Corresponding author. Fax: 0044-0207-928-8742. E-mail address: [email protected] (B. Alexander). 0011-2240/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII: S 0 0 1 1 - 2 2 4 0 ( 0 2 ) 0 0 0 1 0 - X
92
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
Liver transplantation frequently requires segments of (iliac) arteries to be used as conduits for re-connection of the hepatic arterial supply to the donor liver. However, hepatic artery thrombosis has been reported following transplantation [29] and may not become manifest until the tenth postoperative day [22]. Therefore arterial vessel remnants are stored in University of Wisconsin (UW) solution at 4 °C for up to 10 days to cover the event of requirement for re-anastomosis to replace the thrombosed artery. The resultant thrombosis may be due to poor vessel ‘quality’ and/or deficiencies in storage and collection techniques and it has been proposed that endothelial function may become compromised by storage of these vessels in cold-preservation solutions [15]. However, it is currently believed that UW storage may be beneficial specifically for preservation of the vascular endothelium in rabbit thoracic aorta for up to 72 h
[19], although progressive endothelial degeneration after 72 h UW storage has been reported [12,27]. The presence of purine receptors has been demonstrated in the hepatic vasculature of the dog [23], pig [5], rabbit [6], and rat liver [30]. UW storage of rabbit thoracic aorta resulted in reduced endothelium-independent relaxation to sodium nitroprusside and morphologically wellpreserved vascular endothelium [18] and smooth muscle [3]. Storage in Krebs–B€ ulbring buffer, which histologically appeared to destroy the vascular endothelium, attenuated acetylcholineinduced relaxation but preserved ATP-induced relaxation. In addition, UW storage maintained endothelium-dependent relaxation to acetylcholine but attenuated endothelium-dependent relaxation to ATP [4]. ATP-induced vasorelaxation may occur via NO-dependent, endothelium-inde-
Fig. 1. Current classification of purine receptors according to ranked order of agonist potency and molecular cloning of specific receptors on each tissue based originally upon a proposal by Abbracchio and Burnstock [1]. The ranked order of agonist potencies of various P2X purine receptors, that elicit vasoconstriction on the left-hand side of the vertical dotted line, has been omitted due to spatial constraints and because the focus of the present manuscript is upon P2Y receptors. The right-hand side of the dotted line illustrates the classification of P2Y receptor subtypes that elicit vasodilatation. The classification of P2Y receptor subtypes is continually under review and, at the present time, there is a consensus that only 6 subtypes of receptor exist. Their classification varies according to the author and some believe that the P2Y3 and P2Y6 may be homologues of each other and therefore should not be separate entities [20,21]. Conversely, it has been argued that other subtypes exist such as the P2Y7 receptor which has been postulated to be a specific leukotrine B4 receptor [16] and that these should be included as an additional receptor subtype. P2Y1 and P2Y2 receptors, as the present study has shown, exist on vascular smooth muscle and endothelium. Delineation between the 2 is therefore based upon ranked order of agonist potency, molecular cloning, and action, respectively, of some putative specific inhibitors which, in the present study, was PPADS. Receptor subtypes with only 1 or 2 agonists mentioned imply that these are the major agonists above all others. Thus, the finite delineations of this area of research are in a state of flux at the present time.
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
pendent P2Y1 , and endothelium-dependent P2Y2 receptors (Fig. 1). 2-Methylthio ATP (2MeSATP) and uridine-50 -triphospate (UTP) elicit relaxation by activation of P2Y1 and P2Y2 receptor subtypes, respectively, whilst ATP is a potent agonist at both receptor subtypes [9]. Therefore previusly observed ATP-induced relaxation in (endothelium-disrupted) Krebs–B€ ulbring-stored tissue and (preserved-endothelium) UW-stored tissue may have occurred via either of these receptors. Moreover, either of these receptor subtypes may become damaged during UW storage, thus, providing an explanation for previous data, the implication being that muscarinic receptors are more resilient to UW storage. Differentiation between P2Y1 and P2Y2 purine receptor-mediated vasorelaxation may be achieved by the use of the P2Y1 antagonist PPADS [10,11,25]. The aim of the present study therefore was to determine which subtype of P2Y receptors degenerates following UW storage by the use of PPADS and the nitric oxide (NO) synthase inhibitor L -arginine methyl ester (L-NAME).
Methods Tissue preparation Male New Zealand White rabbits weighing between 2.2 and 2.9 kg were killed by a lethal injection of sodium pentobarbitone through a cannulated ear vein. Following thoracolaparotomy, the thoracic aorta was carefully excised avoiding vessel traction and cleaned of adherent fat and connective tissue in a sterile petri dish containing Krebs–B€ ulbring buffer at room temperature. Rings, 4 mm in length, were cut from a
93
20 mm section of aorta located near the heart to decrease error arising due to differential distribution of receptors [7,31]. During the course of tissue preparation, care was taken to avoid contact with the internal surface of the aorta to maintain endothelial cell integrity. Each ring was placed under 2 g tension upon a fine wire myograph, according to our previous protocols [18,4], and calibrated for isometric measurements in Krebs–B€ ulbring buffer, the composition of which is detailed in Table 1. This was heated to 37 °C and oxygenated with 95%/5% O2 =CO2 . The fine wire myographs were connected to FT03 force displacement transducers for recording with a Grass 79F Polygraph (Grass Instrument, Quincy, MA, USA). All drugs were made up in fresh distilled water. Noradrenaline was dissolved in 0.1 mM ascorbic acid (Vitamin C) to prevent oxidation. Experimental design Aortic rings of tissue were divided into intact or de-endothelialised tissue, ensuring that rings within a given group originated from different aortae [18]. Rings from each segment of aorta were either tested immediately upon fresh tissue (in Krebs–B€ ulbring buffer) upon harvesting or after storage for 8 days in sterile Krebs–B€ ulbring buffer, UW (see Table 1) or saline, n ¼ 6 per group except with fresh tissue where n ¼ 16. Eight days storage was chosen because this is the maximum time allocated at King’s College Hospital for storage of remnant arteries to cover the event of requirement for reanastomosis of thrombosed arteries following the initial transplantation [22]. De-endothelialisation was achieved by rotating the aorta around an internally inserted glass rod.
Table 1 Composition of UW solution and Krebs–B€ ulbring buffer per litre of water, both titrated to a pH 7.4 University of Wisconsin solution Kþ lactobionate KH2 PO4 MgSO4 Raffinose Adenosine Glutathione Insulin Penicillin Dexamethasone Allopurinol Hydroxyethyl starch
Krebs’–B€ ulbring buffer 100 mmol 25 mmol 5 mmol 30 mmol 5 mmol 3 mmol 100 IU 40 IU 8 mg 1 mmol 50 g
NaCl KCl MgSO4 NaHCO3 NaH2 PO4 CaCl2 D-Glucose
1.33 mmol 4.7 mmol 0.61 mmol 16.3 mmol 1.35 mmol 2.52 mmol 7.8 mmol
94
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
Successful de-endothelialisation was measured by the lack of relaxation to acetylcholine whilst contraction to noradrenaline (2 lM) was maintained. Some aortic rings were de-endothelialised by the method described and histologically examined to confirm complete removal of the endothelial monolayer.
curves in the presence and absence of PPADS, and L-NAME were analysed by Student’s, paired t test (2-tailed). Emax (the minimum concentration of drug required to elicit the maximum response) and EC50 values were calculated from statistical analyses using the curve-fitting programme, Graph Pad Prism (Graph Pad Software, San Diego, CA, USA).
Experimental protocol All vessels were equilibrated for 1 h prior to the commencement of any experiments and flushed with fresh Krebs–B€ ulbring buffer at 37 °C every 15 min. Following equilibration, vessels were precontracted with a dose of noradrenaline (2 lM), a concentration found from a cumulative dose–response curve constructed at the start of every experiment to evoke approximately 70% of the maximum contraction. Endothelium-dependent relaxation to ATP, UTP, and 2MeSATP was studied in fresh tissue and in samples following 8-day cold storage in either UW, Krebs–B€ ulbring buffer or physiological saline (0.9% NaCl, Polyfusor UK) all at pH 7.4. Responses were tested in the presence and absence of L-NAME (100 lM) and PPADS (1 lM). Responses to acetylcholine and the nucleotides were expressed as percentage relaxation from the submaximal level of precontraction produced by noradrenaline. Drugs used The following drugs were obtained from Sigma Poole (UK): acetylcholine, uridine triphosphate (sodium salt), L-NAME, adenosine triphosphate, and noradrenaline. 2-methylthio ATP and PPADS were from Research Biochemicals Inc. UW solution was provided by Du Pont Pharmaceuticals, UK. Statistical analysis Results from concentration-effect curves were expressed as means standard error of the mean (SEM). The data were checked for a normal distribution and a one-way multivariate analysis of variance (ANOVAR) was applied to check for significant differences. Where differences were apparent, a Student’s 2-tailed, unpaired t test with Bonferroni adjustment was applied to test for differences in responses at specific concentration points, P < 0:05 taken as being significantly different. Results for paired concentration-effect
Results Cholinergic responses Acetylcholine produced dose-dependent vasorelaxation in all vessels tested following precontraction with 2 lM noradrenaline that consistently produced an average of 71:3 2:4% submaximal contraction (Fig. 2). De-endothelialisation totally abolished relaxation in fresh tissue. There were no significant differences between fresh tissue, UW-, and Krebs–B€ ulbring-stored vessels. However, responses following 8-day storage in saline were significantly attenuated compared to UW-stored, Krebs–B€ ulbring-stored, and fresh tissue vessels ðP < 0:01Þ. Responses to ATP ATP produced dose-dependent relaxation in fresh tissue and after UW-, Krebs–B€ ulbring- or saline-storage (Fig. 3a). Vasorelaxation to ATP following storage in Krebs–B€ ulbring and in fresh tissue was significantly greater than that stored in UW or saline. Emax responses to ATP were significantly greater in Krebs–B€ ulbring, stored vessels ð52:52 4:02%Þ than in UW-stored (43:73 1:48%, P < 0:001) and de-endothelialised vessels (44:20 1:57%, P < 0:05). Thus storage in Krebs– B€ ulbring-buffer preserved ATP-mediated vasorelaxation in contrast to storage in UW or saline where it was attenuated. De-endothelialisation of fresh tissue significantly attenuated vasorelaxation compared to intact tissues, although it was not abolished (Fig. 3a). This suggested that ATPmediated vasorelaxation occurred via both endothelium-dependent and -independent mechanisms. There were no significant differences in de-endothelialised tissues between fresh tissue and those stored in UW, Krebs–B€ ulbring or saline (Fig. 3b) and thus it seemed reasonable to suggest that the previously observed differences could have been attributable to endothelium-dependent mechanisms. Therefore storage in Krebs–B€ ulbring buffer
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
95
Fig. 2. Concentration-dependent relaxation to acetylcholine in ring segments of rabbit thoracic aorta fresh tissue, deendothelialised fresh tissue, and after 8 days storage in saline or UW solution. The % vasorelaxation, in this and subsequent graphs, was calculated as the % decrease from the original resting tone of 71.3% of the maximum tension (vasoconstriction) generated by noradrenaline. 2 lM noradrenaline was required to achieve this percentage of sub maximal contraction. De-endothelialisation totally abolished acetylcholine-induced vasorelaxation. There were no significant differences in responses between fresh tissue and UW-stored tissues. However, these were significantly greater than those stored in saline and, obviously, those recorded from de-endothelialised tissue ðP < 0:01Þ.
appeared to preserve endothelium-dependent vasorelaxation. Significant attenuation of ATP-induced vasorelaxation by L-NAME in freshly obtained tissue reconfirmed at least partial vasorelaxation via NO (Fig. 4a). Responses to ATP in fresh tissue were not significantly attenuated by 10 lM 8-phenyltheophylline (8-PT), the non-specific P1 A2 adenosine receptor antagonist, in either endothelium-intact or de-endothelialised tissues (data not presented) and thus confirmed that vasorelaxation to ATP was not due to catabolism to adenosine. Responses to 2MeSATP 2MeSATP Produced dose-dependent relaxation, indicative of P2Y1 purine receptor activation, in fresh tissue that was attenuated by 28% at the Emax after removal of the endothelium (Fig. 5a). This suggested that 72% of the vasorelaxation to 2MeSATP occurred via direct smooth muscle activation. The significant attenuation of UWand saline-stored vessels compared to fresh tissue,
de-endothelialised, and Krebs–B€ ulbring-stored tissue was indicative that UW- and saline-storage down-regulated a mixture of both endotheliumdependent and direct smooth muscle receptors that are activated by 2MeSATP (Fig. 5a). PPADS caused a highly significant attenuation of responses to 2MeSATP in fresh tissue suggesting P2Y1 receptor activation (Fig. 7a). L-NAME significantly attenuated the Emax of 2MeSATP in fresh tissue by 56% from 40:63 0:77 to 17:77 0:75%, P < 0:05 Students’ paired t test n ¼ 6 (Fig. 4b). This suggested that up to 56.3% of the P2Y1 receptor-mediated response was NOdependent in this tissue. De-endothelialised tissues produced no significant differences between fresh and Krebs–B€ ulbring-stored tissue (Fig. 5b) and thus it can be deduced that smooth muscle receptors remained intact. In addition, there were no significant differences between saline- and UWstored tissue compared to de-endothelialised tissue stored in Krebs–B€ ulbring or in fresh tissue (Fig. 5b) and this again suggested that the majority of smooth muscle receptors responding to 2MeSATP were not damaged.
96
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
a
b
Fig. 3. Concentration-dependent relaxation to ATP in: (a) intact tissues, fresh tissue, and after 8 days of storage in saline, Krebs–B€ ulbring or UW solution. Responses in Krebs–B€ ulbring and fresh tissue were significantly greater than those obtained from de-endothelialised, saline-, and UW-stored tissue ðP < 0:05Þ; and (b) de-endothelialised tissue, there were no significant differences between de-endothelialised stored and fresh tissue.
Responses to UTP UTP produced dose-dependent vasorelaxation in fresh tissue, UW-, Krebs–B€ ulbring, and salinestored tissues (Fig. 6a). Responses to UTP following storage in Krebs–B€ ulbring, saline, and fresh tissue were significantly greater than those stored in UW at all concentrations of UTP tested,
up to and including the Emax ðP < 0:05Þ. De-endothelialisation virtually abolished vasorelaxation at all concentrations of UTP tested in stored compared to fresh tissue (Fig. 6b). L-NAME significantly attenuated UTP-induced relaxation at all concentrations of UTP (Fig. 4c). The Emax of UTP was significantly attenuated from 48:55 6:38 to 12:85 2:89% P < 0:001 Stu-
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
a
97
different P2Y2 receptors. This was in complete contrast to 2MeSATP where PPADS virtually abolished all (P2Y1 -induced) vasorelaxation in UW-stored tissue.
Discussion
b
c
Fig. 4. Concentration-dependent relaxation to (a) ATP; (b) 2MeSATP and (c) UTP with (j + L-NAME) and without ( ) L-NAME) in fresh tissue. There a significant shift to left of the ATP-induced responses ðP < 0:05Þ and a highly significant attenuation of 2MeSATP- ðP < 0:05Þ and UTP-induced responses ðP < 0:001Þ.
dent’s paired t test, representing a reduction of 74.5% of the original response, thus, suggesting that up to 75% of the response to UTP was NOdependent. Therefore vasorelaxation to UTP which was predominantly endothelium- and NOdependent was significantly attenuated following storage in UW compared to storage in Krebs– B€ ulbring buffer or saline. PPADS significantly enhanced vasorelaxation to UTP in fresh tissue (Fig. 7b), thereby suggesting that UTP acted via
The data from the present study confirmed previous observations of retained acetylcholineinduced [19] and attenuated ATP-induced vasorelaxation [4] in rabbit thoracic aorta following UW storage. Preliminary observations of retained ATP-induced, endothelium-dependent, relaxation following storage in Krebs–B€ ulbring buffer and attenuation of these responses following UW storage where the vascular endothelium appeared structurally well preserved [4] were also reconfirmed in the present study. However, these earlier studies could not establish which particular subtype of, or how, purine receptors altered during UW cold storage. This was particularly relevant because the vascular endothelium appeared well preserved in UW-stored vessels yet disintegrated in Krebs–B€ ulbring buffer-stored vessels where ATP-induced vasorelaxation was conserved and similar to that in fresh tissue. The present study showed that ATP-induced relaxation was P2Y receptor-driven because; (i) this was significantly attenuated by both de-endothelialisation (Fig. 3a) and by L-NAME (Fig. 4a) in fresh tissue; (ii) residual responses to ATP were not mediated via P1 (adenosine) purine receptor activation because they were not attenuated by the P1 A2 receptor antagonist, 8-PT, in either intact or de-endothelialised tissues (data not presented); and (iii) concentration-dependent relaxation was demonstrated to 2MeSATP and UTP, agonists believed to activate P2Y1 - and P2Y2 -receptors, respectively [9]. Data from the present study demonstrated that, after UW storage, the reduced endotheliumdependent, ATP-induced, vasorelaxation was similar to that in de-endothelialised fresh tissue or storage in saline. However, de-endothelialisation demonstrated that 72% of the 2MeSATP-induced vasorelaxation was conserved in fresh tissue (Fig. 5a) and 81.6% in Krebs–B€ ulbring buffer stored tissue. L-NAME significantly attenuated fresh tissue responses to 2MeSATP by 56% thus indicating almost an equal division between P2Y1 NO-dependent and -independent receptors. However, this was in contrast to P2Y2 receptors where de-endothelialisation and/or UW storage
98
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
a
b
Fig. 5. Concentration-dependent relaxation to 2MeSATP in: (a) intact tissues; and (b) de-endothelialised tissues in fresh tissue and after storage in saline, Krebs–B€ ulbring or UW solution. There was a highly significant attenuation of responses in UW- and saline-stored tissues compared to fresh tissue-, Krebs–B€ ulbring, and de-endothelialised tissue in intact tissues ðP < 0:05Þ. Removal of the endothelium (b) abolished the presumably endothelium-dependent differences between Krebs–B€ ulbring-stored and UW- and saline-stored tissue since all 3 were attenuated compared to fresh tissue ðP < 0:05Þ.
virtually abolished responses to UTP (Fig. 6b). In addition, de-endothelialisation of fresh tissue also abolished responses to UTP, produced similar graphs to intact UW-stored tissue (Fig. 6a), and supported the hypothesis that most P2Y2 receptors were located on the endothelium. PPADS was used to further segregate between P2Y1 and P2Y2 receptor subtypes by its differential activity on 2MeSATP- and UTP-induced vasorelaxation, respectively. PPADS significantly
attenuated 2MeSATP-induced vasorelaxation in fresh tissues (Fig. 7a) but, in contrast, it actually enhanced responses to UTP (Fig. 7b) [11,25]. This enhancement of UTP-mediated responses may have been due to inhibition of ectonucleotidase activity by PPADS which effectively increases the net agonist concentration available at the receptor binding site [13,28], although this remains to be confirmed in future studies. The NO-dependent proportion of P2Y2 (UTP-mediated) receptors
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
99
a
b
Fig. 6. Concentration-dependent relaxation to UTP in (a) intact tissues and (b) de-endothelialised tissues in fresh tissue and in tissues after storage in saline, Krebs–B€ ulbring or UW solution. (a) There was a highly significant attenuation of responses in de-endothelialised and UW-stored tissue compared to fresh tissue, Krebs–B€ ulbring-, and saline-stored tissue ðP < 0:05Þ. (b) De-endothelialisation virtually abolished relaxation in all of the stored tissues compared to fresh tissue.
was much greater than that of P2Y1 receptors and accounted for up to 80% of the population according to L-NAME attenuation (Fig. 4c). These observations therefore indicated that 2MeSATP and UTP acted via different receptor subtypes and according to the present study, storage in UW predominantly inhibited the major proportion of NO- and endothelium-dependent P2Y2 receptors. Histological evidence from our previous studies suggested that the morphology of the endo-
thelium appeared well preserved following UW storage [4,18] and that cholinergic endotheliumdependent vasorelaxation is retained in UWstored tissue (Fig. 2). However, this did not necessarily prove that all vascular receptors retain their functional capacity following UW storage and that some may become denatured at a quaternary or tertiary structural level, undetectable using light or electron microscopy. Another underlying mechanism that may explain our data
100
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
a
b
original reason for inclusion of adenosine into UW solution was to replenish adenosine levels for ATP synthesis and maintain cell membrane integrity after reperfusion [26]. However, recent studies showed that long-term exposure to adenosine were detrimental because of increased generation of cytokines [14], decreased production of vascular endothelial growth factor [8] increased nitrite and nitrate, myeloperoxidase activity, and lipoperoxidation levels [24]. In addition, the present study suggests that P2Y2 receptor down regulation occurred following UW storage within the purine (ATP and UTP) signal transduction pathways that led to NO production. It is unclear how the above mechanisms may affect these pathways but the production of oxygen free radicals, possibly nitrites and nitrates via NO production following reperfusion [2], could be involved. Additional studies suggested that omission of UW colloid elements may reduce impaired endothelium-dependent hyperpolarisation factor (EDHF)-mediated activity, inhibited Ca2þ -activated and attenuated ATP-sensitive Kþ channels [17] following UW storage. However, it is impossible to ascertain precisely which particular stage(s) within the P2Y2 signal transduction pathway are altered by UW storage. Moreover, lack of vascular relaxation may promote thrombosis via creating a low flow state but it remains unknown whether this is a primary causal effect for thrombosis or if thrombosis occurs when altered reactivity is combined with other damage such as overt endothelial damage and these concepts remain the subject of future studies.
Acknowledgments Fig. 7. Concentration-dependent relaxations to (a) 2MeSTP and (b) UTP with (j + PPADS) and without ( ) PPADS) in fresh tissue. PPADS caused a highly significant attenuation to responses above 7:0 log (M) 2MeSATP (P < 0:01, Student’s paired t test).
may be down-regulation of P2Y2 -receptors, specifically during storage, that may occur due to the unique composition of UW solution. UW solution has a particularly high adenosine and Kþ ion concentration and although the tissues were incubated in fresh Krebs–B€ ulbring solution for 1 h and flushed regularly before experiments commenced, the long-lasting effects of these components, particularly following ischaemia reperfusion injury, are currently unknown. The
We thank Du Pont Pharmaceuticals (UK, Ltd.) for their gift of UW solution. Dr. C.A. Brown is thanked for his invaluable comments and assistance in the preparation of this manuscript.
References [1] M.P. Abbracchio, G. Burnstock, Purinoceptors: Are there families of P2X and P2Y purinoceptors? Pharmacol. Ther. 64 (1994) 445–475. [2] B. Alexander, The role of nitric oxide in hepatic metabolism, Nutrition 14 (1998) 376–390. [3] B. Alexander, J.V. Gryf-Lowczowski, J.S. Marway, J. Salisbury, D. Sherlock, V. Preedy, I.S. Benjamin,
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
Functional and compositional studies of arteries stored in University of Wisconsin solution compared with Krebs–B€ ulbring buffer, Exp. Mol. Pathol. 71 (2001) 265–273. B. Alexander, J.V. Gryf-Lowczowski, D. Sherlock, J. Salisbury, I.S. Benjamin, Paradoxical cholinergic and purinergic vascular reactivity of rabbit thoracic aorta cold-stored in University of Wisconsin solution, J. Pharm. Pharmacol. 51 (1999) 623–630. B. Alexander, R.T. Mathie, Diminished hyperaemic response of the hepatic artery to portal venous occlusion (the buffer response) in Asian hybrid minipigs: a comparison of the response to that observed in dogs, J. Comp. Physiol. [B] 163 (1993) 5–10. B. Alexander, R.T. Mathie, V. Ralevic, G. Burnstock, An isolated dual-perfused rabbit liver preparation for the study of hepatic blood flow regulation, J. Pharmacol. Toxicol. Methods 27 (1992) 17–22. B.M. Altura, B.T. Altura, Heterogeneity of drug receptors in different segments of rabbit thoracic aorta, Eur. J. Pharmacol. 12 (1970) 44–52. A.J. Archambault, M.G. Sirois, P.N. Bernatchez, C. Fiset, P.S. Haddad, Vascular endothelial growth factor production by isolated rat hepatocytes after cold ischemia-warm reoxygenation, Liver Transpl. 7 (2001) 988–997. M.R. Boarder, S.M. Hourani, The regulation of vascular function by P2 receptors: multiple sites and multiple receptors, Trends Pharmacol. Sci. 19 (1998) 99–107. C.A. Brown, S.J. Charlton, M.R. Boarder, Enhancement of the response to purinergic agonists in P2Y1 transfected 1321N1 cells by antagonists suramin and PPADS, Br. J. Pharmacol. 120 (1997) 1049–1052. S.J. Charlton, C.A. Brown, G.A. Weisman, J.T. Turner, L. Erb, M.R. Boarder, Cloned and transfected P2Y4 receptors: characterization of a suramin and PPADS-insensitive response to UTP, Br. J. Pharmacol. 119 (1996) 1301–1303. E.G. Chedrawy, J.A. Sullivan, G. Li, M.M. Vohra, Effects of University of Wisconsin preservation solution on rabbit aortic endothelium and smooth muscle, Br. J. Pharmacol. 116 (1995) P167. B.C. Chen, C.M. Lee, W.W. Lin, Inhibition of ectoATPase by PPADS, uramin and reactive blue in endothelial cells, C6 glioma cells and RAW 264.7 macrophages, Br. J. Pharmacol. 119 (1996) 1628– 1634. C.H. Chiang, C.P. Yu, C.P. Wu, H.C. Yan, W.C. Perng, Cytokine up-regulation in ischaemic/reperfused lungs perfused with University of Wisconsin solution and normal saline, Clin. Sci. (Lond) 101 (2001) 285–294. M.G. Clemens, K. Chun, E. Miescher, D. Jones, J. Zhang, Leukocyte-dependent and -independent hepatic microvascular injury during reperfusion after
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
101
warm ischaemia, Prog. Appl. Microcirc. 19 (1993) 139–151. V.R. Dasari, J. Jin, S.P. Kunapuli, Distribution of leukotriene B4 receptors in human hematopoietic cells, Immunopharmacology 48 (2000) 157–163. Z.D. Ge, G.W. He, Comparison of University of Wisconsin and St. Thomas’ Hospital solutions on endothelium-derived hyperpolarizing factor-mediated function in coronary micro-arteries, Transplantation 70 (2000) 22–31. J.V. Gryf-Lowczowski, D. Sherlock, J. Salisbury, I.S. Benjamin, B. Alexander, Cold-storage of rabbit thoracic aorta in University of Wisconsin solution reduces endothelium-independent vasodilation, J. Pharm. Pharmacol. 49 (1997) 1096–1101. S. Karatapanis, P.A. McCormick, M. Jacobs, G. Stansby, K. Rolles, N. McIntyre, A.K. Burroughs, J.Y. Jeremy, Effect of liver and kidney preservation solutions on vascular viability: studies on contraction and relaxation of rat aorta in vitro, Transplant. Proc. 25 (1993) 2148–2152. Q. Li, M. Olesky, R.K. Palmer, T.K. Harden, R.A. Nicholas, Evidence that the p2y3 receptor is the avian homologue of the mammalian P2Y6 receptor, Mol. Pharmacol 54 (1998) 541–546. Q. Li, J.B. Schachter, T.K. Harden, R.A. Nicholas, The 6H1 orphan receptor, claimed to be the p2y5 receptor, does not mediate nucleotide-promoted second messenger responses, Biochem. Biophys. Res. Commun. 236 (1997) 455–460. W.C. Marujo, A.N. Langnas, R.P. Wood, R.J. Stratta, S. Li, B.W. Shaw Jr., Vascular complications following orthotopic liver transplantation: outcome and the role of urgent revascularization, Transplant. Proc. 23 (1991) 1484–1486. R.T. Mathie, B. Alexander, The role of adenosine in the hyperaemic response of the hepatic artery to portal vein occlusion (the ‘buffer response’), Br. J. Pharmacol. 100 (1990) 626–630. F. Pi, F. Badosa, A. Sola, C.J. Rosello, C. Xaus, N. Prats, E. Gelpi, G. Hotter, Effects of adenosine on ischaemia-reperfusion injury associated with rat pancreas transplantation, Br. J. Surg. 88 (2001) 1366–1375. V. Ralevic, G. Burnstock, Discrimination by PPADS between endothelial P2Y- and P2U-purinoceptors in the rat isolated mesenteric arterial bed, Br. J. Pharmacol. 118 (1996) 428–434. J.H. Southard, T.M. van Gulik, M.S. Ametani, P.K. Vreugdenhil, S.L. Lindell, B.L. Pienaar, F.O. Belzer, Important components of the UW solution, Transplantation 49 (1990) 251–257. M.M. Vohra, E.G. Chedrawy, G. Li, J.A. Sullivan, Functional assessment of rabbit thoracic aorta stored in University of Wisconsin, St. Thomas’ Hospital and Krebs solutions, Br. J. Pharmacol. 120 (1997) P153. U. Windscheif, O. Pfaff, A.U. Ziganshin, C.H. Hoyle, H.G. Baumert, E. Mutschler, G. Burnstock,
102
S.J. Payne et al. / Cryobiology 44 (2002) 91–102
G. Lambrecht, Inhibitory action of PPADS on relaxant responses to adenine nucleotides or electrical field stimulation in guinea-pig taenia coli and rat duodenum, Br. J. Pharmacol. 115 (1995) 1509–1517. [29] K. Yanaga, G. Lebeau, J.W. Marsh, R.D. Gordon, L. Makowka, A.G. Tzakis, S. Todo, A.C. Stieber, S. Iwatsuki, T.E. Starzl, Hepatic artery reconstruction for hepatic artery thrombosis after orthotopic liver transplantation, Arch. Surg. 125 (1990) 628–631.
[30] W. Yang, I.S. Benjamin, B. Alexander, Localisation of hepatic vascular resistance sites in the isolated dual-perfused rat liver, Eur. J. Pharmacol. 364 (1999) 13–21. [31] W. Yang, I.S. Benjamin, B. Alexander, Nitric oxide modulates acetylcholine-induced vasodilatation in the hepatic arterial vasculature of the dual-perfused rat liver, Acta Physiol. Scand. 171 (2001) 413– 418.