Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Dec 5;115(12):997-1006.
doi: 10.1161/CIRCRESAHA.115.304591. Epub 2014 Oct 6.

Thymidine Phosphorylase Participates in Platelet Signaling and Promotes Thrombosis

Free PMC article

Thymidine Phosphorylase Participates in Platelet Signaling and Promotes Thrombosis

Wei Li et al. Circ Res. .
Free PMC article


Rationale: Platelets contain abundant thymidine phosphorylase (TYMP), which is highly expressed in diseases with high risk of thrombosis, such as atherosclerosis and type II diabetes mellitus.

Objective: To test the hypothesis that TYMP participates in platelet signaling and promotes thrombosis.

Methods and results: By using a ferric chloride (FeCl3)-induced carotid artery injury thrombosis model, we found time to blood flow cessation was significantly prolonged in Tymp(-/-) and Tymp(+/-) mice compared with wild-type mice. Bone marrow transplantation and platelet transfusion studies demonstrated that platelet TYMP was responsible for the antithrombotic phenomenon in the TYMP-deficient mice. Collagen-, collagen-related peptide-, adenosine diphosphate-, or thrombin-induced platelet aggregation were significantly attenuated in Tymp(+/-) and Tymp(-/-) platelets, and in wild type or human platelets pretreated with TYMP inhibitor KIN59. Tymp deficiency also significantly decreased agonist-induced P-selectin expression. TYMP contains an N-terminal SH3 domain-binding proline-rich motif and forms a complex with the tyrosine kinases Lyn, Fyn, and Yes in platelets. TYMP-associated Lyn was inactive in resting platelets, and TYMP trapped and diminished active Lyn after collagen stimulation. Tymp/Lyn double haploinsufficiency diminished the antithrombotic phenotype of Tymp(+/-) mice. TYMP deletion or inhibition of TYMP with KIN59 dramatically increased platelet-endothelial cell adhesion molecule 1 tyrosine phosphorylation and diminished collagen-related peptide- or collagen-induced AKT phosphorylation. In vivo administration of KIN59 significantly inhibited FeCl3-induced carotid artery thrombosis without affecting hemostasis.

Conclusions: TYMP participates in multiple platelet signaling pathways and regulates platelet activation and thrombosis. Targeting TYMP might be a novel antiplatelet and antithrombosis therapy.

Keywords: intracellular signaling; platelet; thrombosis; thymidine phosphorylase.


Figure 1
Figure 1. Deletion of thymidine phosphorylase (Tymp) protects against arterial thrombosis in mice
A. Representative video images of carotid artery thrombi formation after 7.5% FeCl3 treatment. Platelets were labeled by direct jugular vein injection of rhodamine 6G and arteries were imaged in real time by fluorescent intravital microscopy. B. Blood flow cessation time. Data are presented as Mean ± SEM, n=9 for each group. C, D and E are contingency tables showing mice number based on whether occluded thrombus formed or not within the observed 30 min, and Fisher’s exact test was used for statistical analysis. Two side p=0.009 in C, p=0.029 in D and p=1 in E. F. WT and Tymp−/− recipient mice were exposed to 10.5 Gy of external beam irradiation, and then received bone marrow (BM) transplantation as indicated. The successful BM engrafted mice were subjected to 7.5% FeCl3 induced injury on carotid arteries 4 weeks later (n=4~6). G. Tymp+/−, Tymp−/− or WT recipient mice were exposed to 11 Gy of irradiation and allowed to live for 5 days to induce serious thrombocytopenia. Donor platelets isolated from WT, Tymp+/− or Tymp−/− mice were labeled with Rhodamine 6G, and total 109 platelets in 200 μl saline were transfused into the recipient mouse via jugular vein injection. The mice were then subjected to 7.5% FeCl3 induced carotid artery injury. N=5 in each group.
Figure 2
Figure 2. TYMP deficiency attenuates platelet activation in vitro
A & B. Platelet-rich plasma isolated from Tymp−/−, Tymp+/− or WT mice was used for platelet aggregation assay induced by (A) 2.5 μM ADP, and (B) 1 μg/ml collagen using a standard turbidimetric assay. N=6. C & D. Washed murine platelets were used for aggregation assay induced by (C) 0.05 U/ml thrombin or (D) 0.5 μg/ml collage-related peptide (CRP) (n=4 or 5). E. Platelets in platelet-rich plasma (20 μl) were mixed with Tyrode’s buffer (80 μl), and stimulated with indicated agonists for 5 min at room temperature. Reactions were stopped by adding 2% formaldehyde, 1 mM EDTA in PBS. P-select in expression was stained by FITC-conjugated antibody and examined by flow cytometry. *p<0.05 WT vs. Tymp+/− and Tymp−/−; # P<0.05, WT vs. Tymp−/−, N=3.
Figure 3
Figure 3. TYMP forms a complex with Src family kinases in platelets
A. A consensus SH3-binding sequence APPAP was found in different primate TYMPs; and a core binding motif PxxP was also found in murine and other orders. B, C and D. Immunoprecipitation-immunoblotting assays were performed using indicated antibodies to determine the interaction of TYMP and Src family kinases including Lyn, Fyn, Src and Yes in human platelets.
Figure 4
Figure 4. TYMP interacts with Lyn and discriminately regulates phosphorylation of Lyn, PECAM-1 (CD31) and AKT in platelets
A. Human platelet lysates were used for immunoprecipitation of Lyn and then immunoblotting was performed for TYMP and Lyn. Blots represent at least 3 independent experiments. *p<0.05 vs. corresponding times of KIN59 treatments. B. Human platelets were pretreated with vehicle or KIN59 and then stimulated with collagen for indicated times. Platelets lysates were analyzed by immunoprecipitation and immunoblotting assays using indicated antibodies. Bar graph shows Lyn activity from 3 independent experiments. *p<0.05 vs. corresponding times of KIN59 treatments. C. Human platelets treated as in panel A were analyzed by Western blot using indicated antibodies. Bar graph represents 3 independent experiments. D. Mouse platelets pooled from 10–12 mice were divided into 3 groups and stimulated with CRP as indicated. Platelet lysates were used for immunoprecipitation and immunoblotting assays using indicated antibodies. Bar graph shows the ratio of phosphorylated AKT and PECAM1 to their total proteins.
Figure 5
Figure 5. Lyn deficiency reverses the anti-thrombotic phenotype of Tymp deficient mice
A. Platelet GPVI signaling plays an important role in the FeCl3 induced thrombosis model. B. The time to blood flow cessation was assessed in Tymp+/−;Lyn+/− and Lyn−/− mice (N=4 in each group) using the FeCl3 injury induced carotid artery thrombosis model and data were compared with Tymp+/−, Tymp−/− and WT mice.
Figure 6
Figure 6. Pharmacologic inhibition of TYMP activity inhibits platelet aggregation and thrombosis without affecting hemostasis
A. The effect of different concentrations of KIN59 on different doses of collagen induced human platelet aggregation. N=6~9. B. The effect of KIN59 (250 μM) on murine platelet aggregation in response to 1 μg/ml collagen stimulation, N=6. C. The effect of KIN59 on thrombosis was assessed on eight weeks old WT mice that were treated with KIN59 (30 mg/kg/day) for 3 days. The mice were then subjected to 7.5% FeCl3 induced carotid artery thrombosis model, n=6. D. aPTT assay was performed using platelet-poor-plasma from 5 different mice. E. Tail bleeding time was assessed in anesthetized WT, Tymp+/− and Tymp−/− mice by cutting 1 cm of tail from the tip, or in WT mice received KIN59 (30 mg/kg/day) or vehicle treatment for 3 days. N=6.
Figure 7
Figure 7. TYMP functions on platelet GPVI signaling and promote platelet activation
Red arrows indicate stimulatory and blue arrows indicate inhibitory effects.

Similar articles

See all similar articles

Cited by 7 articles

See all "Cited by" articles

Publication types

MeSH terms