Human Placental Ectonucleoside Triphosphate Diphosphohydrolase Gene Transfer via Gelatin-Coated Stents Prevents In-Stent Thrombosis
Background— In-stent thrombosis is mainly triggered by adenosine diphosphate (ADP)-dependent platelet aggregation after percutanous coronary stent implantation. Ectonucleoside triphosphate diphosphohydrolase (E-NTPDase) rapidly hydrolyzes ADP to adenosine monophosphate, inhibiting platelet aggregation. We tested the hypothesis that local delivery of human placental E-NTPDase (pE-NTPDase) gene into injured arteries via gene-eluting stent could prevent subacute in-stent thrombosis.
Methods and Results— We generated gene-eluting stents by coating bare metal stents with cationic gelatin hydrogel containing pE-NTPDase cDNA (pE-NTPDase stent), and implanted the stents into rabbit femoral arteries (FA) prone to production of platelet-rich thrombi due to repeated balloon injury at 4-week intervals. After the second injury, E-NTPDase gene expression was severely decreased; however, the implantation of pE-NTPDase stent increased E-NTPDase mRNA levels and NTPDase activity to higher level than normal FA. The FAs with pE-NTPDase stents maintained patency in all rabbits (P<0.01), whereas the stent-implanted FAs without pE-NTPDase gene showed low patency rates (17% to 25%). The occlusive platelet-rich thrombi, excessive neointimal growth, and infiltration of macrophages were inhibited in stent implanted FA with pE-NTPDase gene, but not without pE-NTPDase gene.
Conclusions— Human pE-NTPDase gene transfer via cationic gelatin-coated stents inhibited subacute in-stent thrombosis and suppressed neointimal hyperplasia and inflammation without antiplatelet drugs.
In-stent thrombosis is a life-threatening adverse event that often follows coronary stent implantation. The use of polymer-coated drug-eluting stents (DES) has significantly reduced the rates of restenosis and target lesion revascularization (TLR) compared with bare metal stents (BMS). However, such stents do not reduce the incidence of subacute in-stent thrombosis and even increase late in-stent thrombosis attributable to incomplete reendothelialization and vascular healing.1–3 Therefore, long-term use of antiplatelet drugs is required after DES implantation; this may cause critical adverse events including hemorrhagic complications and thrombotic thrombocytopenic purpura in a small number of cases. Over the past couple of decades, a number of studies have focused on development of new types of coronary stents, such as gene-eluting stents, to prevent restenosis or accelerate reendothelialization4,5; however, further efforts to develop stents that prevent in-stent thrombosis without antiplatelet drugs are necessary.
In-stent thrombosis is primarily triggered by local platelet activation, culminating in platelet aggregation, the generation of coagulation factors, and the formation of a fibrin network and a stable occlusive thrombus.6,7 Plaque rupture or vascular injury resulting from stent implantation or the stent itself can induce initial platelet adhesion and activation. Subsequent intracellular signaling events trigger the release of secondary agonists, such as thromboxane A2 from membrane phospholipids and adenosine diphosphate (ADP) from dense granules, which led to the production of coagulation factors on the platelet surface and generation of thrombin.8,9 Finally, subacute in-stent thrombosis occurs. Thus, ADP is a key agonist that triggers platelet aggregation, and degrading ADP or blocking binding of ADP to purinergic receptors on the platelet surface might prevent in-stent thrombosis.10,11 This concept is clinically supported by evidence that thienopyridines, specific purinergic receptor blockers, significantly prevent subacute thrombosis after coronary artery stent implantation.12,13
Vascular ectonucleoside triphosphate diphosphohydrolase (vE-NTPDase), or CD39, is a membrane-bound enzyme that plays a crucial role in vascular endothelial function by inhibiting platelet aggregation via phosphohydrolysis of ATP and ADP to AMP.14 vE-NTPDase has 2 putative transmembrane domains and an extracellular domain containing 5 apyrase conserved regions (ACRs), of which ACR-1, -4, and -5 are important for enzymatic activity.15 Makita et al16 purified human placental E-NTPDase (pE-NTPDase), an alternative form of vE-NTPDase that differs at the N-terminal 11-aa residues. The purified pE-NTPDase has been demonstrated to have 2 putative transmembrane domains and an extracellular domain containing an enzymatically active region like vE-NTPDase. This protein also inhibits platelet aggregation induced by platelet agonists or shear stress.16,17 Recently, Furukoji et al18 reported that human pE-NTPDase gene transfer to air-injured arteries via adenovirus vector suppressed platelet aggregation, thrombus formation, and neointimal growth in mice. Based on these findings, we speculate that local delivery of the pE-NTPDase gene through a gene-eluting stent might prevent subacute in-stent thrombosis and neointimal hyperplasia without requiring antiplatelet agents.
To evaluate antithrombotic effects of the gene-eluting stent in a clinic-like setting, we used a repeated balloon injury model in the rabbit femoral artery, which is similar to acute coronary syndrome (ACS) in pathogenesis of occlusive thrombi.19,20 In this study, we generated a new coronary stent coated with biodegradable cationic gelatin hydrogel as a platform for gene elution and evaluated the effects of human pE-NTPDase gene transfer to injured arteries on prevention of subacute in-stent thrombosis and smooth muscle cell proliferation using the rabbit ACS-like thrombus model.
Please see the supplemental materials at http://atvb.ahajournals.org for detailed Methods.
We evaluated the effects of human pE-NTPDase on rabbit platelet aggregation induced by ADP as previously described.16,18
Preparation of Cationic Gelatin-Coated Stents
Cationic gelatin hydrogel was generated as previously described.21 Normal stainless steel stents (bare metal stents [BMS]) were coated with cationic gelatin hydrogels to prepare the gene-eluting stents.
Plasmid Expression Vectors and Gene-Eluting Stents
The human pE-NTPDase cDNA was FLAG-tagged at the N terminus in the pBS CAG vector.17 We also constructed a control pBS CAG vector encoding bacterial β-galactosidase (LacZ) to evaluate effective transduction of plasmid DNA from a stent platform into rabbit arterial tissue. We generated pE-NTPDase gene-eluting stent (pE-NTPDase stent) and LacZ gene-eluting stent (LacZ stent) with the use of the stent coated with cationic gelatin hydrogel (gelatin-coated stent).
Repeated-Balloon Injury of Rabbit Femoral Arteries
The Animal Experimentation Committee at Nara Medical University approved the experimental protocol used in this study. We used 76 male Japanese white rabbits (SLC Japan, Shizuoka, Japan) weighing 2.5 to 2.7 kg. Repeated-balloon injury of the right femoral artery (FA) was performed to induce platelet-rich thrombus similar to ACS as previously described, with minor modification.19,20
In Vivo Stent Implantation
Rabbits were assigned to 1 of 4 groups: 22 rabbits received BMS, 22 gelatin-coated stents, 10 LacZ stents, and 22 pE-NTPDase stents. Each stent was implanted in the injured right FA by balloon inflation.
Evaluation of Patency of Stent Implanted Arteries
Transcutaneous continuous Doppler analysis on day 3 post stent implantation (n=10 for the BMS, gelatin-coated stent, and pE-NTPDase stent groups, and n=4 for the LacZ stent group), and angiography on days 3 and 7 post stent implantation (n=10 and 12 on each day for the BMS, gelatin-coated stent, and pE-NTPDase stent groups, n=4 and 6 on each day for the LacZ stent group) were carried out to evaluate the patency of the stent implanted arteries.
X-Gal Staining, PCR, Western Blotting, and Immunohistochemistry
X-Gal staining of stent implanted FAs, evaluation of E-NTPDase and endothelial nitric oxide synthase (eNOS) mRNA expression by real-time polymerase chain reaction (PCR) and E-NTPDase protein expression by Western blotting in stent implanted FAs, and immunohistochemical examination were performed as previously described (please see the supplemental Methods).
Measurement of NTPDase Activity
NTPDase activity of whole FAs of stent implanted site was measured by luciferin-luciferase bioluminescence assay using an ATP assay system (TOYO B-Net CO, LTD) on days 3 and 7 post stent implantation (n=3 and 4 on each day, respectively, for the BMS and pE-NTPDase stent groups). The data were expressed as the ratio of activity in stent implanted FAs to activity in contralateral normal FA.
Data are expressed as means±SD. Differences between individual groups were evaluated using the unpaired Student t test or ANOVA with Bonferroni multiple comparisons. Values of P<0.05 were considered significant.
Effects of Human Placental E-NTPDase on Rabbit Platelet Aggregation
First, we investigated whether platelet aggregation was induced by ADP in rabbits, as well as the effects of human pE-NTPDase on rabbit platelet aggregation. Rabbit platelet aggregation was induced by ADP, and this was significantly suppressed by human pE-NTPDase (supplemental Figure I), indicating that human pE-NTPDase might inhibit ADP-dependent platelet aggregation also in rabbits.
Successful Gene Transfer via Gelatin-Coated Stent in Vivo
To evaluate gene transfer and protein expression in the injured arterial tissue, we implanted LacZ stents into rabbit FAs after repeated-balloon injury, and performed X-Gal staining on days 3 and 7 post implantation in the BMS, gelatin-coated stent, and LacZ stent groups (n=3 on each day). In the LacZ stent group, strong X-Gal staining was observed in the implanted FA on days 3 and 7 (supplemental Figure II), whereas no staining was observed in the arteries of the BMS or gelatin-coated stent groups on day 3 (supplemental Figure II) or 7 (data not shown). These data suggest successful and continuous protein expression after gene transfer from cationic gelatin-coated stents.
E-NTPDase Gene Expression in Stent Implanted Arteries
We examined the expression of E-NTPDase mRNA in the FAs on days 3 and 7 post implantation by real-time PCR and expressed the data as the ratio of E-NTPDase mRNA to β-actin mRNA. As shown in Figure 1A, human pE-NTPDase mRNA expression in the FA was observed 3 days after pE-NTPDase stent implantation and persisted until 7 days after implantation; human pE-NTPDase mRNA was not detected immediately after stent implantation (day 0). In contrast, human pE-NTPDase expression was not observed at any time point in either the BMS or gelatin-coated stent group. We also examined total E-NTPDase mRNA expression of both exogenous human pE-NTPDase and endogenous rabbit vE-NTPDase by real-time PCR (Figure 1B). In the BMS and gelatin-coated stent groups, after the second balloon injury the total E-NTPDase mRNA level was reduced to about 10% of that in normal vessels on day 3 and then recovered to 50% to 60% of normal levels on day 7. However, in the pE-NTPDase stent group, total mRNA levels of E-NTPDase increased by 40% on day 3 and decreased slightly to levels comparable to normal at day 7. These results indicate that human pE-NTPDase gene transfer allowed maintenance of total E-NTPDase mRNA levels in injured vessel walls despite denudation of endothelial cells.
E-NTPDase Protein Expression and Enzymatic Activity
Immunostaining for human pE-NTPDase (YH34) revealed that human pE-NTPDase was expressed mainly in smooth muscle cells widely from the media to the surface of neointima but not in the adventitial cells only in the pE-NTPDase stent implanted FAs, though was not detected in the other stent implanted FAs either on day 3 (data not shown) or 7 (Figure 1C). We evaluated pE-NTPDase protein expression in FAs by Western blotting using anti-FLAG antibody, because YH34 did not work in Western blotting. As expected, FLAG was detected only in the pE-NTPDase stent group, and not in the BMS or gelatin-coated stent groups or in normal arteries on day 3 (data not shown) or 7 (Figure 1D) post implantation. Consistent with the E-NTPDase gene and protein expression, NTPDase activity in FAs implanted with the pE-NTPDase stents was similar to that in contralateral normal FAs, and was significantly higher than in FAs implanted with BMS on both days 3 and 7 (P<0.05, Figure 1E).
Patency of Stent Implanted Arteries
We evaluated blood flow in the stent implanted FAs by transcutaneous continuous Doppler on day 3 and angiography on days 3 and 7 post implantation. Peak flow velocity measured by continuous Doppler on day 3 showed normal blood flow and patency in all of the arteries implanted with pE-NTPDase stents (10 of 10), whereas normal flow was observed in only 1 of 10 arteries in each of the BMS and gelatin-coated stent groups. Normal flow was not observed in any artery in the LacZ stent group (0 of 4; Table). The rate of angiographic patency in stent implanted FAs in the pE-NTPDase group was significantly higher than in the other 3 groups on both day 3 (100% in the pE-NTPDase stent group versus 50% in the other groups) and day 7 (100% in the pE-NTPDase stent group versus 25% in the BMS and gelatin-coated stent groups, 17% in the LacZ stent group; Table). Representative Doppler flow patterns are shown in supplemental Figure III, and representative angiography are in Figure 2A and supplemental movies. The difference in patency rate between continuous Doppler and angiographic data in the BMS, gelatin-coated stent, and LacZ stent groups might be attributable to collateral flow to the distal site from the proximal site of stent implantation affecting the continuous Doppler measurement.
Inhibition of In-Stent Thrombosis by pE-NTPDase Gene Eluting Stents
As shown in Figure 2B, histological examination demonstrated occlusive in-stent thrombosis in the BMS or gelatin-coated stent implanted FAs, whereas no thrombi were observed in FAs implanted with E-NTPDase stents on day 7 post implantation. Additionally, immunostaining for GP IIb/IIIa showed that the occlusive thrombi in the BMS and gelatin-coated stent groups contained a large amount of platelets. These results suggest that human pE-NTPDase gene eluting stent suppressed platelet aggregation in the injured arteries, inhibiting subacute in-stent thrombosis without antiplatelet drugs.
Suppression of Neointimal Growth and Inflammatory Cell Infiltration by pE-NTPDase Gene Eluting Stents
To investigate the effect of E-NTPDase on neointimal formation after vascular injury, we performed immunostaining for α-smooth muscle actin to evaluate smooth muscle cell proliferation and neointimal growth (Figure 3A), and for macrophages to evaluate infiltration of inflammatory cells (Figure 3C) in FAs on day 7 post stent implantation. The neointimal growth was assessed by the average of the ratio of neointima area to the area of neointima plus media in 3 transverse parts of each stent implanted FAs. The ratio of neointima area was significantly smaller in the pE-NTPDase stent group than in the BMS group (Figure 3B; P<0.05). On the other hand, stent struts were all covered with neointima partially including smooth muscle cells in the pE-NTPDase stent group, but not in the BMS and gelatin-coated stent groups (Figures 2B and 3⇓A). Additionally, in the BMS and gelatin-coated stent groups, infiltration of macrophages into the arterial tissues was observed especially around the implanted stents, but almost none in the pE-NTPDase stent group (Figure 3C).
Increase of eNOS Gene Expression in the Stent Implanted Arteries
We evaluated the eNOS mRNA expression in the neointima of stent-implanted FAs to speculate the reendothelialization after stent implantation, because several antibodies that recognize markers of endothelial cells, such as CD31, von willebrand factor, or eNOS, did not work for immunostaining in plastic resin-embedded samples. The levels of eNOS mRNA expression in the neointima of FAs in the pE-NTPDase stent group were higher than in the BMS and gelatin-coated stent groups on both day 3 and 7, and recovered to 80% of normal level on day 7 (Figure 4). In the BMS and gelatin-coated stent group, the levels of eNOS mRNA expression were extremely low on day 3 and recovered to 40% to 50% of normal level on day 7. These results indicate that reendothelialization is accelerated in the pE-NTPDase stent implanted arteries, and that the recovery of the E-NTPDase mRNA expression on day 7 in the BMS and gelatin-coated stent groups (Figure 1B) is caused partly by the newly regenerated endothelial cells.
This study is the first to report that: (1) a plasmid vector encoding the pE-NTPDase gene is reliably incorporated into and released from newly designed cationic gelatin-coated stents, (2) local delivery of the pE-NTPDase gene into injured arteries enhances the mRNA and protein expression of pE-NTPDase, thereby (3) preventing subacute in-stent thrombosis in a rabbit model of ACS-like thrombus formation without antiplatelet therapy, and (4) the locally enhanced expression of pE-NTPDase may suppress neointimal hyperplasia and inflammation, and accelerate reendothelialization in arteries after stent implantation.
E-NTPDase, which is primarily expressed in the endothelial cells of normal vessels, rapidly hydrolyzes extracellular ATP and ADP to AMP, which is further converted to the antithrombotic and antiinflammatory mediator adenosine by 5′-nucleotidase. Thus, E-NTPDase plays a critical role in the inhibition of platelet aggregation in the vasculature. However, in injured vessels, E-NTPDase activity decreases, causing accumulation of ADP and thrombus formation. In fact, in the present study, vE-NTPDase mRNA levels were decreased to about one tenth of that in normal vessels (Figure 1B) after repetitive injury. The E-NTPDase mRNA levels in the FAs implanted with BMS or gelatin-coated stent recovered to 50% to 60% of normal levels on day 7 though without human pE-NTPDase gene transfer. On the other hand, the mRNA expression of eNOS, which is specifically expressed in endothelial cells, was recovered to about 45% of normal level on day 7 in the BMS or gelatin-coated stent implanted FAs (Figure 4). Therefore, in the stent implanted arteries without human pE-NTPDase gene, the recovery of E-NTPDase mRNA levels was comparatively parallel with that of eNOS mRNA levels. Based on these data, we speculate that the recovery of E-NTPDase gene expression on day 7 in the BMS or gelatin-coated stent implanted FAs was partly induced by the newly regenerated endothelial cells.
On the other hand, as shown in Figure 1B, human pE-NTPDase gene transfer augmented the total levels of E-NTPDase mRNA to levels similar to or higher than those observed in normal vessels throughout the 7-day experimental period, resulting in inhibition of in-stent thrombosis. Several previous observations22,23 showed that overexpression of the E-NTPDase gene by adenoviral transfection or genetic engineering methods could result in resistance to thrombosis, supposing that high expression level of E-NTPDase is necessary to block thrombosis. However, the present study suggests that keeping the physiological levels of E-NTPDase is enough to prevent in-stent thrombosis after stenting. Immunohistochmical examination showed pE-NTPDase protein was mainly expressed in smooth muscle cells from media to the surface of neointima, suggesting pE-NTPDase expressed in the surface of the injured vessel prevents platelet aggregation.
The data of early recovery of eNOS mRNA expression in neointima in the pE-NTPDase stent implanted FAs indicates the acceleration of reendothelialization by the human pE-NTPDase gene transfer, which may play a part in preserving total E-NTPDase gene expression in the pE-NTPDase stent implanted arteries in subacute phase. Additionally, immunohistochemical examination (Figure 3) demonstrated that neointimal hyperplasia attributable to smooth muscle cell proliferation and macrophage infiltration after stent implantation was inhibited by the pE-NTPDase gene-eluting stent. Therefore, the pE-NTPDase gene-eluting stent prevents in-stent thrombosis in subacute phase, and may be able to prevent restenosis and late thrombosis by both acceleration of reendothelialization and suppression of neointimal hyperplasia and inflammation. Several recent studies18,24,25 have indicated effects of E-NTPDase beyond the regulation of platelet aggregation, which support the data of our present study. A mouse model showed that gene transfer of human E-NTPDase into injured arteries inhibited smooth muscle cell proliferation.18 In a mouse model of angiogenesis induced by VEGF and fibroblast growth factor, E-NTPDase knockout resulted in the failure of migration and recruitment of monocytes and endothelial cells to the angiogenic site and consequently impaired angiogenesis.24 Mizumoto et al also reported that inflammatory response was exacerbated in E-NTPDase knockout mice.25
To date, no stent has been developed that focuses on prevention of in-stent thrombosis, although first-generation drug-eluting stents developed to reduce restenosis are in widespread clinical use. In addition, a number of new types of stent that accelerate reendothelialilzation are under development. Considering that in-stent thrombosis remains a major cause of death and morbidity after percutaneous coronary intervention, future stents should be developed to have antithrombotic as well as antirestenosis functions. In this context, E-NTPDase represents a promising target for drug development.
Stents are an ideal platform for localized delivery of drugs or genes to the vascular wall. For this purpose, stainless balloon-expandable stents coated with nonerodable polymers or phosphorylcholine, containing drugs or plasmids were developed.4,5 First generation drug-eluting stents, which use nonerodable polymers reservoirs, have recently been reported to provoke arterial hypersensitivity reactions, such as eosinophilic infiltration and in-stent thrombosis, in a small number of cases.1,2,26 Instead of nonerodable polymers or phosphorylcholine, we adopted cationic gelatin hydrogel to generate the gene-eluting stents used in the present study. Gelatin is biodegradable and has been proven to be biologically innocuous, it has been widely used for medical and pharmaceutical applications, and its biosafety has been proven through long clinical use as a surgical biomaterial. In addition, we have recently succeeded in controlling the gradual release of plasmid DNA from cationic gelatin hydrogel in vivo.21 In this study, successful gene transfer was demonstrated by the expression of human pE-NTPDase and control β-galactosidase in stent-implanted arteries for 7 days after implantation. Histological examination revealed that infiltration of macrophages was not observed in the pE-NTPDase stent implanted FAs (Figure 3C), and was not accelerated by gelatin-coated stents compared to BMS, although the follow-up period is probably not sufficiently long to draw conclusions. Furthermore, based on angiographic data showing no difference in occlusion rate between BMS and gelatin-coated groups, gelatin itself does not seem to accelerate platelet activation or thrombus formation. Therefore, cationic gelatin-coated stents may eventually prove to be safe and useful for gene transfer after coronary intervention.
Indeed the pE-NTPDase protein-eluting stents may be expected to be more effective than gene-eluting stents in point of early exertion of antithrombotic effect in early phase after stent implantation, but we had not generated the adequate gelatin for protein release in vivo. So, we tested the effects of the pE-NTPDase gene-eluting stents in this study. Further investigations are needed to demonstrate the effects of the pE-NTPDase protein-eluting stents.
In summary, human pE-NTPDase gene transfer via cationic gelatin-coated stents prevented subacute in-stent thrombosis by preserving local NTPDase activity, suppressed neointimal hyperplasia and inflammation, and might accelerate reendothelialization. These findings provide a starting point to develop next generation stents that are not susceptible to in-stent thrombosis.
Sources of Funding
This study was supported by a grant-in-aid from the Ministry of Education, Science, and Culture of Japan (No. 19790538) to Dr Kawata.
Received October 9, 2008; revision accepted March 9, 2009.
Virmani R, Guagliumi G, Farb A, Musumeci G, Grieco N, Motta T, Mihalcsik L, Tespili M, Valsecchi O, Kolodgie FD. Localized hypersensitivity and late coronary thrombosis secondary to a sirolimus-eluting stent: should we be cautious? Circulation. 2004; 109: 701–705.
Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, O'Shaughnessy C, Caputo RP, Kereiakes DJ, Williams DO, Teirstein PS, Jaeger JL, Kuntz RE, for the SIRIUS Investigators. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med. 2003; 349: 1315–1323.
Johnson TW, Wu YX, Herdeg C, Baumbach A, Newby AC, Karsch KR, Oberhoff M. Stent-based delivery of tissue inhibitor of metalloproteinase-3 adenovirus inhibits neointimal formation in porcine coronary arteries. Arterioscler Thromb Vasc Biol. 2005; 25: 754–759.
Egashira K, Nakano K, Ohtani K, Funakoshi K, Zhao G, Ihara Y, Koga J, Kimura S, Tominaga R, Sunagawa K. Local delivery of anti-monocyte chemoattractant protein-1 by gene-eluting stents attenuates in-stent stenosis in rabbits and monkeys. Arterioscler Thromb Vasc Biol. 2007; 27: 2563–2568.
Farb A, Sangiorgi G, Carter AJ, Walley VM, Edwards WD, Schwartz RS, Virmani R. Pathology of acute and chronic coronary stenting in humans. Circulation. 1999; 99: 44–52.
Gawaz M. Role of platelets in coronary thrombosis and reperfusion of ischemic myocardium. Cardiovasc Res. 2004; 61: 498–511.
Gordon EL, Pearson JD, Slakey LL. The hydrolysis of extracellular adenosine nucleotides by cultured endothelial cells from pig aorta. J Biol Chem. 1986; 261: 15496–15504.
Zimmermann H. 5′-Nucleotidase: molecular structure and functional aspects. Biochem J. 1992; 285: 345–365.
Gachet C. ADP receptors of platelets and their inhibition. Thromb Haemost. 2001; 6: 222–232.
Leopold JA, Antman EM. Dual antiplatelet therapy for coronary stenting: a clear path for a research agenda. Circulation. 2005; 111: 1097–1099.
Marcus AJ, Johan Broekman M, Drosopoulos JH, Islam N, Alyonycheva TN, Safier LB, Hajjar KA, Posnett DN, Schoenborn MA, Schooley KA, Gayle RB, Maliszewski CR. The endothelial cell ecto-ADPase responsible for inhibition of platelet function is CD39. J Clin Invest. 1997; 99: 1351–1360.
Makita K, Shimoyama T, Sakurai Y, Yagi H, Matsumoto M, Narita N, Sakamoto Y, Saito S, Ikeda Y, Suzuki M, Titani K, Fujimura Y. Placental ecto-ATP diphosphohydrolase: its structural feature distinct from CD39, localization and inhibition on shear-induced platelet aggregation. Int J Hematol. 1998; 68: 297–310.
Furukoji E, Matsumoto M, Yamashita A, Yagi H, Sakurai Y, Marutsuka K, Hatakeyama K, Morishita K, Fujimura Y, Tamura S, Asada Y. Adenovirus-mediated transfer of human placental ectonucleoside triphosphate diphosphohydrolase to vascular smooth muscle cells suppresses platelet aggregation in vitro and arterial thrombus formation in vivo. Circulation. 2005; 111: 808–815.
Yamashita A, Furukoji E, Marutsuka K, Hatakeyama K, Yamamoto H, Tamura S, Ikeda Y, Sumiyoshi A, Asada Y. Increased vascular wall thrombogenicity combined with reduced blood flow promotes occlusive thrombus formation in rabbit femoral artery. Arterioscler Thromb Vasc Biol. 2004; 24: 2420–2424.
Dwyer KM, Robson SC, Nandurkar HH, Campbell DJ, Gock H, Murray-Segal LJ, Fisicaro N, Mysore TB, Kaczmarek E, Cowan PJ, d'Apice AJ. Thromboregulatory manifestations in human CD39 transgenic mice and the implications for thrombotic disease and transplantation. J Clin Invest. 2004; 113: 1440–1446.
Goepfert C, Sundberg C, Sévigny J, Enjyoji K, Hoshi T, Csizmadia E, Robson S. Disordered cellular migration and angiogenesis in cd39-null mice. Circulation. 2001; 104: 3109–3115.
Lüscher TF, Steffel J, Eberli FR, Joner M, Nakazawa G, Tanner FC, Virmani R. Drug-eluting stent and coronary thrombosis: biological mechanisms and clinical implications. Circulation. 2007; 115: 1051–1058.