255M. Nonaka et al.: Hyperacute Xenorejection and PlateletsSurg TodayJpn J Surg (1999) 29:255–259

Platelets as Participants in Hyperacute Guinea Pig-to-Rat LungXenorejection

Makoto Nonaka, Mitsutaka Kadokura, Daisuke Kataoka, Shigeru Yamamoto, Noboru Tanio, Koichi Inoue,and Toshihiro Takaba

First Department of Surgery, Showa University School of Medicine, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142, Japan

Abstract: It is well known that xenotransplantation leads toimmediate graft dysfunction. This study was designed to spe-cifically examine the role of platelets in mediating lung hyper-acute xenorejection (HXR) in a guinea pig-to-rat model. Atotal of 18 lungs were perfused with blood using an ex vivoapparatus. The animals were divided into the following fourgroups: a CE group comprising circuit only with rat blood;a SYN group comprising rat lungs and blood; a XE groupcomprising guinea pig lungs and rat blood; and an SH groupcomprising guinea pig lungs and rat blood with sarpogrelatehydrochloride, a suppresser of platelet aggregation. The plate-let and serotonin in the blood were lower and the wet/dryweight ratio of the lung (W/D) in the XE group were higherthan those in the SYN group after perfusion. The plateletcount was higher, but the serotonin and W/D were lower inthe SH group than in the XE group. These results suggestthat platelets strongly affect HXR. Thus, the administrationof drugs to suppress platelet aggregation would reducexenotransplanted lung edema.

Key Words: xenotransplantation, hyperacute xenorejection,platelet, sarpogrelate hydrochloride, serotonin

Introduction

Discordant xenotransplantation leads to immediategraft dysfunction,1 and while complement activity isknown to mediate allograft rejection,2 the mechanismsof hyperacute xenorejection (HXR) are still unknown.Recent research in the field of xenotransplantation has

Reprint requests to: M. NonakaThis paper was presented at the 13th Biennial Asian Congresson Thoracic and Cardiovascular Surgery, held in Sydney,Australia, October 12–15, 1997(Received for publication on Oct. 28, 1997; accepted onMay 15, 1998)

focused primarily on humoral factors that mediateHXR such as xenoreactive natural antibodies andcomplement.3 However, another possible factor con-tributing to HXR is the accumulation of activated plate-lets in association with microvascular thrombosis.

Sarpogrelate hydrochloride (SH) is a selective S2-serotonergic receptor antagonist, that inhibits S2-serotonergic platelet aggregation and suppresses bloodvessel constriction mediated by the S2-serotonergicreceptor.4,5 The diminution of intragraft platelet aggre-gation suggests that platelet-dependent mechanismsmay play an important role in complement recruitmentduring HXR. Therefore, we examined the effects of asuppresser on platelet aggregation. This study wasdesigned to specifically examine the role of plateletsin mediating lung HXR using an ex vivo syngeneic andxenoperfusion model.

Materials and Methods

All animals received humane care in strict compliancewith the Principles of Laboratory Animal Care of theNational Society for Medical Research, and the Guidefor the Care and Use of Laboratory Animals of theNational Institute of Health, publication no. 86-23 (re-vised 1985).

A total of 78 Wistar rats weighing 205 to 330g, with amean weight of 279g, and 12 Hartley guinea pigs weigh-ing 255 to 315 g, with a mean weight of 277 g, were usedin this study. The donor guinea pigs and rats were anes-thetized and euthanized with ether inhalation. None ofthese animals were given an intravenous heparin injec-tion. Through a median sternotomy, the heart-lung blocwas excised. The main pulmonary artery (PA) was iso-lated and a double lumen 7F catheter was inserted intothe main PA without being flushed. A 16-gauge tubewas inserted into the trachea, after which the left atrialappendage was incised and any thrombi removed. The

256 M. Nonaka et al.: Hyperacute Xenorejection and Platelets

heart-lung bloc of the guinea pig or rat was suspended,then ventilated with room air at 50breaths/min atpositive end expiratory pressure 5 0 mmHg (5 ml/kgbody weight), and perfused with 30ml blood from ratsheparinized with 40IU/ml. This perfused blood was ob-tained from anesthetized rats and withdrawn from theabdominal aorta through a laparotomy. The blood wasperfused at a flow rate of 2ml/min for 15 min using aninfusion pump, drained through the left atrium, andcollected. The pulmonary venous pressure was main-tained at 0mmHg. The pulmonary arterial pressure(PAP) and airway pressure (AWP) were continuouslymeasured and recorded each minute.

These animals were divided into four groups. Inthe xenotransplantation model (XE group, n 5 6), theheart-lung bloc of a guinea pig was perfused with theblood of rats. In an antiplatelet xenotransplantationmodel (SH group, n 5 6), the heart-lung bloc of aguinea pig was perfused with the blood of rats to whichhad been added a suppresser of platelet aggregation,0.1mM sarpogrelate hydrochloride (SH) (MitsubishiChemical, Tokyo, Japan), the chemical compositionof which was (6)-1-[0-[2-(m-methoxyphenyl)ethyl]-phenoxy]-3-(dimethylamino)-2-propyl hydrogen succi-nate hydrochloride. In a syngeneic model (SYN group,n 5 6), the heart-lung bloc of a rat was perfused with theblood of rats, the purpose of which was to demonstratean ischemia-reperfusion injury (I/R). In an ex vivo cir-

culating model (CE group, n 5 6), the circuit withoutthe heart-lung bloc was perfused with the blood of rats,representing a lung circuit to evaluate the influence ofsuch a system on circulatory rat blood.

The white blood cell (WBC) count, polymorpho-nuclear leukocyte (PMN) count, platelet count, comple-ment activity, serotonin level, and activated partialthromboplastin time (APTT) in the rat blood were mea-sured immediately pre- and postperfusion. Comple-ment activity was measured by the total hemolyticcomplement assay (CH50), which was performed usingstandard methods. The serotonin concentration wasmeasured in whole blood.

After the perfusion, the left lung was removed, wetweighed, and placed in a drying oven (90°C). After 48 h,it was weighed again dry. The wet/dry weight ratio (W/D) was used as an estimate of the extent of lung edema.

All data are expressed as mean 6 standard error ofthe mean (SEM). The t-test and analysis of variancewere used for statistical analysis. A P-value of less than0.05 was considered significant.

Results

In the CE group, the WBC, PMN, and platelet counts,complement activity, and serotonin level were not sig-nificantly different pre- and postperfusion (Table 1).

Table 1. White blood cell (WBC) counts (/µl), polymorphonuclear leukocyte (PMN)counts (/µl), platelet counts (3100/mm3), complement activity (CH50) (U/ml), andserotonin levels (ng/ml) in the CE, SYN, XE, and SH groups

Group Preperfusion Postperfusion Post/pre (%)

WBC CE 8 800 6 529 8083 6 396 93 6 6SYN 8 167 6 558 4417 6 352 54 6 2XE 6 883 6 718 4117 6 710 61 6 9SH 8 883 6 189 6100 6 875 70 6 11

PMN CE 2 799 6 224 2255 6 272 81 6 6SYN 2 292 6 175 793 6 155 35 6 6XE 1 359 6 187 760 6 311 55 6 20SH 2 999 6 595 404 6 149 15 6 7

Platelet CE 8 438 6 371 7987 6 130 95 6 4SYN 7 952 6 241 8023 6 265 101 6 1XE 8 047 6 409 6182 6 314 77 6 1SH 8 622 6 469 7545 6 439 88 6 1

Complement CE 42.3 6 0.5 39.3 6 0.7 93 6 1SYN 44.1 6 1.4 42.3 6 3.9 96 6 1XE 40.5 6 1.8 38.1 6 2.5 94 6 5SH 40.1 6 2.6 29.6 6 4.2 73 6 9

Serotonin CE 1 457 6 48 1493 6 13 98 6 3SYN 1 548 6 69 1506 6 72 97 6 1XE 1 335 6 136 997 6 108 76 6 6SH 1 589 6 81 1119 6 60 71 6 6

Values are expressed as the mean 6 SEMCE, ex vivo circulating model; SYN, syngeneic model; XE, xenotransplantation model; SH,antiplatelet xenotransplantation model

257M. Nonaka et al.: Hyperacute Xenorejection and Platelets

The WBC count (P , 0.01) and the PMN count (P ,0.01) were decreased in the SYN group compared withthose in the CE group. The WBC count (P , 0.05),platelet count (P , 0.01), and serotonin level (P,0.01)were decreased in the XE group compared with those inthe CE group. The PMN count (P , 0.01) and serotoninlevel (P , 0.01) were decreased in the SH group com-pared with those in the CE group. The third column ofTable 1 shows the changes in the WBC, PMN, andplatelet counts and in the complement activity andserotonin level from the preperfusion values in all thegroups. These values are expressed as percentages ofthe preperfusion values.

The results of the XE group were compared withthose of the SYN group. The mean PAP in the XEgroup was higher than that in the SYN group during theperfusion (P , 0.01; Fig. 1). The WBC count (P , 0.01)

and PMN count (P , 0.01) were significantly decreasedin the SYN group, whereas the WBC count (P , 0.05),platelet count (P , 0.01), and serotonin level (P , 0.05)were decreased in the XE group. The platelet count (P, 0.01) and serotonin level (P , 0.01) were decreasedin the XE group compared with those in the SYN group(Table 1). After the perfusion, the W/D in the XE groupwas higher than that in the SYN group at 6.15 6 0.08and 5.64 6 0.07, respectively (P , 0.01; Fig. 2).

In the SH group, the WBC (P , 0.05), PMN (P ,0.01), and platelet (P , 0.01) counts, as well as thecomplement activity (P , 0.05) and serotonin level(P , 0.01), were decreased. Comparing the SH groupwith the XE group, the mean PAP in the SH groupwas lower than that in the XE group during the perfu-sion (P , 0.05; Fig. 1). The platelet count in the XEgroup was decreased compared with that in the SHgroup (P , 0.01; Table 1), whereas the serotonin levelin the SH group was decreased compared with that inthe XE group (P , 0.05). The W/D in the SH group,being 5.58 6 0.09, was lower than that in the XE groupafter the perfusion (P , 0.01; Fig. 2).

The APTT was over 300s in every blood sample fromeach group.

Discussion

A striking reduction in platelet counts was observed inthe pulmonary venous effluent in the XE group of thepresent study. Since additional SH in the XE modelsignificantly decreased the W/D and reduced plateletdeposition, we can conclude that platelets played aprominent role during HXR. HXR may be character-

Fig. 1. Changes in airway pressure (AWP, mmHg) and pul-monary arterial pressure (PAP, mmHg). Circles, syngeneicmodel (SYN) group; squares, xenotransplantation model(XE) group; triangles, antiplatelet xenotransplantation model(SH) group

Fig. 2. Wet/dry weight ratios (W/D) in SYN, XE, and SHgroups

258 M. Nonaka et al.: Hyperacute Xenorejection and Platelets

ized by the formation of platelet adhesion and thrombi.Thus, the adhesion, aggregation, and activation ofplatelets could participate in the pathogenesis of theendothelial changes associated with HXR. To clarifythe effects of platelet aggregation on many componentsof the vascular system and lung edema, we studied thePMN counts, complement activity, and serotoninlevels in syngeneic grafts, xenografts, and xenograftswith SH.

Platelet aggregation to the graft causes vascular in-jury which activates and aggregates platelets.6 BecauseSH inhibits platelet aggregation and suppresses bloodvessel constriction,4 theoretically, it should inhibit graftfailure; however, few studies have demonstrated thattransplanted grafts treated with platelet inhibitors7,8

have prolonged survival. In fact, Adachi et al. reportedno prolongation of xenograft survival.9 Furthermore,platelets may protect the graft by adhering to dis-continuities in endothelium and by contributing to therepair of damaged blood vessels,10 although platelet ag-gregation causes the formation of platelet and fibrinthrombi, leading to graft ischemia.11 Recently, Bustosand Platt demonstrated platelet-endothelial cell interac-tions in a xenograft model and described the impor-tance of these interactions;12 however, the effect ofplatelets against xenografts is still controversial.

Activation of the entire complement cascade to yieldthe membrane attack complex could lead to the lysis ofendothelial cells. If cell lysis is not an essential event inthe pathogenesis of HXR, the membrane attack com-plex may still play a critical role. The best characterizedchange induced by the membrane attack complex is theincrease in intracellular calcium ions, reflecting, at leastin part, ion flux across the cell membranes. The increasein intercellular calcium ions causes endothelial cells toelaborate on the von Willebrand factor which in turncontributes to platelet aggregation.13 It is well knownthat the consumption of complement in the graft causessevere endothelial injury. In allorejection, there is animmediate fixation of the recipient’s complement to theendothelium of the donor graft, the local release ofplatelet activating factor, and the massive accumulation,aggregation, and degranulation of platelets in the vascu-lature.10 However, in our model, the role of complementactivity in HXR appeared to be different when plateletinhibitor was added to the blood. More complementwas deposited in the SH lungs than in the XE lungs,although the difference was not significant. Further-more, the edema in the SH lungs was less than that inthe XE lungs. Under platelet inhibition, more comple-ment may be deposited in the SH group than in the XEgroup, to facilitate the accumulation of platelets forprotecting the graft by adhering to discontinuities in theendothelium. In the XE group, platelets filled in gaps inthe endothelial cell junctions, but the degranulation of

platelets would make the lung edema worse than that inthe SH group. Although platelets may influence HXRmore than complement, it is still not clear whether theeffects of SH are only related to the inhibition of plate-let aggregation.

The PMN is thought to play a central role in I/R. Thepresent study demonstrated that the PMN count in theSYN group was significantly lower than that in the CEgroup, but it was not low in the XE group. This findingindicates that PMN may not be an essential element inthe pathogenesis of HXR. Zehr et al. showed that PMNdepletion had no impact on the prognosis of HXR.14 Infact, the PMN count in the SH group was significantlydecreased during xenoperfusion and was lower than inCE group. Moreover, the PMN count in the SH groupwas lower than that in the XE and SYN groups, al-though the difference was not significant. It is suggestedthat SH administration during xenoperfusion decreaseslung edema and HXR but increases I/R.

Serotonin is accumulated in platelets, stored in densegranules, and later released when the platelets arestimulated. Our experiment showed that the plateletcount and serotonin level in whole blood from the XEgroup were decreased in comparison to those from theSYN group after the perfusion. Since almost all seroto-nin in blood is stored in the platelets,15 the decreasedserotonin concentration in the XE group may be associ-ated with the decreased platelet count. However, whenwe compared the serotonin concentration and plateletcounts between the XE and SH groups, the results werevery complicated. Because SH inhibits the release ofserotonin in association with platelet aggregation,4 andbecause the platelet count in the XE group was lowerthan that in the SH group, it was suggested that theconcentration of serotonin in whole blood from the SHgroup was higher than that in whole blood from the XEgroup. However, our study showed that the serotoninconcentration in blood from the SH group was lowerthan that in blood from the XE group. It is well knownthat more than 90% of the serotonin in plasma can becleaned by passage through a normal lung.16 Our pre-liminary study showed that serotonin was not stained inthe lung tissue (n 5 2 in each group), indicating that theserotonin in plasma is not deposited in the lung or in thePA. The fact that the serotonin in plasma was not ableto be metabolized completely in the XE group could bebecause the endothelium of the PA in the XE group wasinjured more severely than that in the SH group. Al-though free serotonin in plasma is rapidly metabolizedby the endothelial monoamine oxidase enzymatic sys-tem in the normal lung,17 injured pulmonary tissue maynot maintain this capacity. Another explanation is thataggregated and damaged platelets cannot accumulatefree serotonin. Although serotonin in plasma is rapidlyaccumulated into the platelets under normal condi-

259M. Nonaka et al.: Hyperacute Xenorejection and Platelets

tions,15 because the platelets in the XE group may havebeen more damaged than those in the SH group, theywere not able to accumulate the serotonin in theplasma. To verify this hypothesis, the serotonin concen-tration in plasma and platelets should be measured.Thus, an additional study is warranted.

The PAP in the XE group was statistically higherthan that in the SYN and SH groups during the perfu-sion. Not only platelet deposition, but also serotoninraises the PAP by constricting the PA after releasingCa21 from the intracellular stores and promoting Ca21

influx through Ca21 channels in the PA smooth musclecells.18 However, it is well known that the guinea piglung has about zero serotonin activity of the strongvasoconstrictor.19 Therefore, the elevated PAP in theXE group may be due to the platelet deposition whencompared with the SYN and SH groups.

To observe the initial changes in HXR in this study,the perfusion flow rate was very low and the perfusiontime was short. This low flow rate may increase theplatelet deposition and emphasize the effect ofantiplatelet deposition. Furthermore, heparin wasadded to the blood, which would have affected the re-sults because the coagulation system is strongly relatedto HXR.

The present study demonstrated that interfering withHXR by specifically targeting platelet deposition dimin-ished lung edema. The W/D in the XE group was higherthan that in the SYN group, and the additional SH inthe XE group reduced the W/D. These results suggestthat platelets affect lung edema and that the diminutionof intragraft platelet aggregation reduces the lungedema in HXR.

In conclusion, the results of the present studydemonstrated that in discordant xenotransplantation,platelets affect HXR more strongly than in syngeneictransplantation, in this ex vivo guinea pig-to-rat andrat-to-rat perfusion model. Furthermore, drugs to sup-press platelet aggregation improved lung edema inHXR.

Acknowledgment. The authors wish to thank Ms. Miki Takiand Mr. Nobukazu Nishino for their technical assistance andMs. Misa Izumikawa for her secretarial support. Sarpogrelatehydrochloride (MCI-9042) was provided by MitsubishiChemical Corporation.

References

1. Lin SS, Platt JL (1996) Immunologic barriers to xenotrans-plantation. J Heart Lung Transplant 15:547–555

2. Egan TM, Cooper JD (1991) The lung following transplantation.In: Crystal RG, West JB, et al. (eds) The lung. Raven, New York,pp 2205–2215

3. Kroshus TJ, Bolman RM III, Dalmasso AP (1996) Selective IgMdepletion prolongs organ survival in an ex vivo model of pig-to-human xenotransplantation. Transplantation 62:5–12

4. Hara H, Osakabe M, Kitajima A, Tamao Y, Kikumoto R (1991)MCT-9042, a new antiplatelet agent is a selective S2-serotonergicreceptor antagonist. Thromb Haemost 65:415–420

5. Tsuchihashi H, Yagi N, Kimura M, Shirota K, Kitami J,Nagatomo T (1991) Binding characteristics of [3H]ketanserin forserotonin-2 receptor in the rabbit platelet. J Pharmacobiol Dyn14:461–466

6. Iuliano L, Colavita AR, Leo R, Pratico D, Violi F (1997) Oxygenfree radicals and platelet activation. Free Radic Biol Med 22:999–1006

7. Colman RB, Habal M, Hollenberg NK, Birtch AG, Busch GJ(1976) Hyperacute renal allograft rejection in the primate: thera-peutic limitations of antiplatelet agents alone and combined withheparin. Am J Pathol 82:25–42

8. Shons AR, Najarian JS (1974) Modification of xenograft rejectionby aspirin, dextran, and cinanserin: the importance of platelets inhyperacute rejection. Transplant Proc 6:435–440

9. Adachi H, Rosengard BR, Hutchins GM, Hall TS, BaumgartnerWA, Borkon AM, Reitz BA (1987) Effects of cyclosporine,aspirin, and cobra venom factor on discordant cardiac xenograftsurvival in rats. Transplant Proc 19:1145–1148

10. Ito S, Camussi G, Tetta C, Milgrom K, Andres G (1984) Hyper-acute renal allograft rejection in the rabbit: the role of platelet-activating factor and of cationic proteins derived frompolymorphonuclear leukocytes and from platelets. Lab Invest51:148–161

11. Cywes R, Mullen JBM, Stratis MA, Greig PD, Levy GA, HarveyPRC, Strasberg SM (1993) Prediction of the outcome of trans-plantation in man by platelet adherence in donor liver allografts:evidence of the importance of preservation injury. Transplanta-tion 56:316–323

12. Bustos M, Platt JL (1997) Platelet-endothelial cell interactions ina xenograft model. Transplant Proc 29:886

13. Hattori R, Hamilton KK, Fugate RD, McEver RP, Sims PJ (1989)Stimulated secretion of endothelial von Willebrand factor isaccompanied by rapid redistribution to the cell surface of theintracellular granule membrane protein GMP-140. J Biol Chem264:7768–7771

14. Zehr KJ, Herskowitz A, Lee PC, Kumar P, Gillinov AM,Baumgartner WA (1994) Neutrophil adhesion and complementinhibition prolongs survival of cardiac xenografts in discordantspecies. Transplantation 57:900–906

15. Frishman WH, Huberfeld S, Okin S, Wang YH, Kumar A,Shareef B (1995) Serotonin and serotonin antagonism in cardio-vascular and noncardiovascular disease. J Clin Pharmacol 35:541–572

16. Thomas DP, Vane JR (1967) 5-Hydroxytryptamine in the circula-tion of the dog. Nature 216:335–338

17. Herve P, Launay JM, Scrobohaci ML, Brenot F, Simonneau G,Petitpretz P, Poubeau P, Cerrina J, Duroux P, Drouet L (1995)Increased plasma serotonin in pulmonary hypertension. Am JMed 99:249–254

18. Yuan XJ, Bright RT, Aldinger AM, Rubin LJ (1997) Nitric oxideinhibits serotonin-induced calcium release in pulmonary arterysmooth muscle cell. Am J Physiol 272 (Lung Cell Mol Physiol16):L44–L50

19. Sheldon WS, Crawford MP, Brewer NR (1979) Morphophy-siology. In: Baker HJ, Lindsey JR, Weisbroth SH (eds) Thelaboratory rat. Academic, New York, pp 83–85