Glucose and collagen regulate human platelet activity through aldose reductase induction of thromboxane

Diabetes mellitus is associated with platelet hyperactivity, which leads to increased morbidity and mortality from cardiovascular disease. This is coupled with enhanced levels of thromboxane (TX), an eicosanoid that facilitates platelet aggregation. Although intensely studied, the mechanism underlying the relationship among hyperglycemia, TX generation, and platelet hyperactivity remains unclear. We sought to identify key signaling components that connect high levels of glucose to TX generation and to examine their clinical relevance. In human platelets, aldose reductase synergistically modulated platelet response to both hyperglycemia and collagen exposure through a pathway involving ROS/PLCγ2/PKC/p38α MAPK. In clinical patients with platelet activation (deep vein thrombosis; saphenous vein graft occlusion after coronary bypass surgery), and particularly those with diabetes, urinary levels of a major enzymatic metabolite of TX (11-dehydro-TXB₂ [TX-M]) were substantially increased. Elevated TX-M persisted in diabetic patients taking low-dose aspirin (acetylsalicylic acid, ASA), suggesting that such patients may have underlying endothelial damage, collagen exposure, and thrombovascular disease. Thus, our study has identified multiple potential signaling targets for designing combination chemotherapies that could inhibit the synergistic activation of platelets by hyperglycemia and collagen exposure.

Accelerated atherosclerosis and microvascular disease contribute to the morbidity and mortality associated with diabetes mellitus (DM) (1–3). Vascular inflammation, endothelial dysfunction associated with hyperglycemia, impaired fibrinolysis, and increased coagulation factors as well as abnormal platelet function are typical for DM, contributing to the increased thrombotic events and development of arteriosclerosis (4). Altered platelet function in DM, including altered adhesion and aggregation, may contribute to the pathogenesis of DM vascular complications by promoting microthrombus formation, contributing to enhanced risk of small vessel occlusions and accelerated atherothrombotic diseases (5, 6). Patients with type 2 DM (T2DM) exhibit platelet hyperreactivity both in vitro and in vivo, coupled with biochemical evidence of persistently increased thromboxane-dependent (TX-dependent) platelet activation (7, 8). Despite many important studies, the mechanism by which platelets transduce glucose levels into enhanced TX generation independently of endothelial and other blood cell–derived factors remains unclear. Similarly, optimal antiplatelet therapy for DM patients remains to be achieved.

Aldose reductase (AR) is the first enzyme of the polyol pathway, and it represents a minor source of glucose utilization, accounting for less than 3% of glucose consumption during euglycemia. However, during hyperglycemia, the activity of AR is substantially increased, representing up to 30% of total glucose consumption, and the abnormal activation of the polyol pathway leads to depletion of reducing equivalents and accumulation of osmotically active polyols (9, 10). Therefore, enhanced utilization of this pathway is well known to contribute to microvascular and macrovascular DM complications by increasing oxidative and osmotic stress. Also, the pharmacological inhibition of AR has been shown to reduce the frequency of polyneuropathy and retinopathy in DM patients (9–11). In fact, it has been shown that human AR expression in transgenic mice accelerates DM atherosclerosis (12), suggesting that AR may play an important role in atherothrombosis. We therefore hypothesized that AR contributes to platelet activation under hyperglycemic conditions and that the signaling mechanism may provide an explanation for the increased risk of atherothrombosis in DM patients.

In the present study, we investigated the role of AR in collagen-induced platelet aggregation and TX biosynthesis under normal-glucose (NG) and high-glucose (HG) conditions and the molecular mechanism by which AR contributes to collagen-induced platelet aggregation and TX generation. Based on these observations, we further pursued 3 case-controlled clinical studies in separate patient populations to support our mechanistic studies and to determine the clinical implications for DM patients. We demonstrate for what we believe is the first time that AR is a key transducer of hyperglycemia through a complex signaling pathway leading to the release of TX, particularly under conditions of platelet activation. Moreover, DM patients with enhanced biosynthesis of TX in vivo (as assessed by measuring a major urinary TX metabolite) despite aspirin (acetylsalicylic acid, ASA) therapy may have underlying endothelial damage and thromboembolic disease.

Growing evidence has implicated hyperactive platelets and thrombus formation as critical components in the development of DM micro- and macrovascular disease. In the following studies, we systematically investigated the relationship among glucose levels, TXA₂ generation, and platelet activation in human platelets and analyzed TX levels in DM patients with and without thrombosis. In addition to highlighting the importance of glucose regulation of platelet activity through TXA₂ generation and release, our study identifies platelet AR as a key transducer of plasma glucose in regulating platelet activity.

TX biosynthesis in vivo in DM patients versus non-DM patients (study group 1). The rationale for our studies was based upon our observation (Figure 1) and those of others (7, 8) of increased TXA₂ generation in vivo (as measured by a major urinary TX metabolite, 11-dehydro-TXB₂ [TX-M]) in patients with DM. TX-M was assessed in DM patients (n = 102) and compared with that in a small group of normal volunteers (healthy subjects [HS]; n = 10), with all subjects on low-dose ASA (selective COX-1 inhibition) reducing TXA₂ generation for prevention of cardiovascular events (13). There was a clear statistically significant increase in the DM group compared with the HS group: the median (with interquartile range [IQR; Q1–Q3]) was 77.5 (54.0 to 130.0) pg/mg creatinine for HS and (123.4 to 330.7) 208.8 pg/mg creatinine for DM patients. We proceeded to study the mechanism for glucose-induced signaling on platelet TXA₂ generation and activation.

Figure 1

Levels of urinary TX-M in HS and DM patients, with all subjects on low-dose ASA (100 mg/day). The levels of urinary TX-M were measured as an index of TX release in HS (n = 10) and DM (n = 102) patients. Data are expressed as median with IQR.

Human platelet aggregation and activation are exquisitely sensitive to glucose concentrations. The effects of glucose on activation were studied in platelets isolated from venous blood of HS. The aggregation induced by 1 μg/ml collagen was increased when platelets were preincubated with increasing concentrations of glucose (Figure 2, A and C). Interestingly, there was no effect of glucose on ADP-induced aggregation (Figure 2, B and C), as measured by percentage of light transmission. To further validate this observation, we performed flow cytometry using P selectin antibodies and observed a similar dose-dependent effect of glucose on platelet activation in response to collagen but not ADP (Figure 2, D–F). This highlights a sensitivity of human platelets to incremental doses of glucose (5.5–25 mmol/l) when stimulated by collagen. Important questions remained as to the mechanism and how this relates to our initial clinical observations of increased TX-M in DM patients despite the use of ASA (Figure 1). Interestingly, a recent study using 2D difference gel electrophoresis and mass spectrometry demonstrated that AR expression and its activity contribute to the human platelet activation after stimulation of the glycoprotein VI (GPVI) receptor by collagen (14). AR is known to metabolize glucose to sorbitol. We therefore examined the contribution of AR to collagen-induced platelet aggregation.

Figure 2

Glucose-induced platelet activation and aggregation in response to 1 μg/ml collagen or 1 μM ADP. The platelet suspensions were incubated with 5.5 mmol/l, 15 mmol/l, or 25 mmol/l glucose for 90 minutes. Percentage of light transmission, an index of platelet aggregation, was measured in platelet suspensions in response to (A) 1 μg/ml collagen or (B) 1 μM ADP for 10 minutes. (C) Quantification of data presented as percentage of light transmission. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 and **P < 0.01 compared with 5.5 mM glucose; ^#P < 0.05 compared with 15 mM glucose. P selectin translocation to membrane was assessed by flow cytometry after stimulation with (D) 1 μg/ml collagen or (E) 1 μM ADP. The representative overlay plots were presented as the number of events over the log of associated fluorescence (baseline refers to the group without collagen or ADP stimulation). (F) Quantification of data presented as MFI. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 and **P < 0.01 compared with 5.5 mM glucose; ^#P < 0.05 compared with 15 mM glucose; ^§§§P < 0.001 compared with baseline.

AR contributes to collagen-induced platelet aggregation under NG and HG conditions. We next tested the effect of 5.5 or 25 mmol/l glucose (basal or hyperglycemia) on human platelets in the presence or absence of epalrestat (selective AR inhibitor [ARI]). As shown in Figure 3, glucose again potentiated collagen-stimulated aggregation in human platelets. In response to 1 μg/ml collagen, the aggregation in 25 mmol/l glucose was 25% higher than that in 5.5 mmol/l glucose and was attenuated by treatment with 1–10 μmol/l epalrestat (Figure 3, A–C), suggesting that such aggregation was abolished by inhibition of AR. The concentration of epalrestat used in the present study was based on the dose-response curve (Supplemental Figure 1; supplemental material available online with this article; doi: 10.1172/JCI59291DS1), which is consistent with previous studies (14). To corroborate the spectrophotometric aggregation assays, P selectin, the marker for platelet activation, was analyzed using flow cytometry. As in Figure 2F, collagen-induced P selectin surface expression was enhanced in HG compared with NG (Figure 3, D–F), and epalrestat effectively attenuated the increase in P selectin. In addition, genetic knockdown of AR reduced the P selectin translocation to membrane in collagen-stimulated megakaryocyte (MEG-01) platelet-like particles in NG (Supplemental Figure 2). We further reconfirmed the specificity in response to collagen activation using 2 concentrations of collagen, arachidonic acid (AA), ADP, and thrombin receptor agonist peptide (TRAP) in addition to epalrestat (Supplemental Figure 3). These results combined suggest that AR plays a central role in collagen-induced platelet aggregation under NG, but particularly under HG, conditions. Supporting these data, we found that HG alone or collagen treatment in NG increased AR activity and that the combination of collagen and HG potentiated this effect (Figure 4A). This increase in activity could not be accounted for by increased AR expression, as collagen treatment (in NG or HG, but not HG alone) induced only a modest increase in AR expression (Figure 4B), which did not correlate with the degree of activation (Figure 4A). Mannitol was used as osmotic control to demonstrate that AR expression is not affected by the increased osmolarity. Thus, collagen and HG synergize in activating AR.

Figure 3

AR contributes to collagen-induced platelet activation and aggregation. Platelet suspensions were incubated with 5.5 mmol/l glucose (NG) or 25 mmol/l glucose (HG) for 90 minutes in the absence (blue line) or presence of 1 μmol/l (red line) or 10 μmol/l (green line) epalrestat (ARI). Percentage of light transmission was measured in platelet suspensions under (A) NG or (B) HG conditions in response to 1 μg/ml collagen for 5 minutes (n = 5 HS). (C) Quantification of data from 1 μg/ml collagen expressed as aggregation in percentage of light transmission. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 and **P < 0.01 compared with the vehicle incubated in 5.5 mM glucose; ^###P < 0.001 compared with the vehicle incubated in 25 mM glucose. P selectin translocation to membrane was assessed by flow cytometry. Representative overlay plots in (D) NG and (E) HG groups in the presence or absence of 10 μmol/l epalrestat were presented as the number of events over the log of associated fluorescence (baseline refers to the group without collagen stimulation). (F) Quantification of data presented as MFI. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 compared with the control in 5.5 mM glucose; ^###P < 0.001 compared with the control incubated in 25 mM glucose.

Figure 4

AR activity is upregulated in collagen-stimulated platelets. (A) Activity of AR in the human platelets in response to 1 μg/ml collagen under NG and HG conditions. Platelet suspensions were incubated with NG or HG for 90 minutes in the presence or absence of 10 μmol/l epalrestat prior to stimulation by 1 μg/ml collagen for 10 minutes. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001, **P < 0.01, and *P < 0.05 compared with values incubated in NG alone; ^###P < 0.001 compared with values in NG and HG with the addition of 1 μg/ml collagen, respectively. (B) Expression of AR in human platelets in response to 1 μg/ml collagen. Platelet suspensions were incubated with NG or HG for 90 minutes prior to stimulation by 1 μg/ml collagen for 10 minutes, and the total cellular extract was collected for experiments. AR expression was normalized to GAPDH expression. Data are expressed as mean ± SD (n = 5 HS). **P < 0.01 compared with values incubated in NG.

Inhibition of AR attenuates the phosphorylation of p38α MAPK in human platelets in response to collagen under NG and HG conditions. In order to identify potential signaling mechanisms by which AR enhances collagen-induced platelet aggregation, we performed a screen using a phosphoproteomic kinase array (R & D Systems) (data not shown). Human platelets induced by 1 μg/ml collagen exhibited a 5-fold increase in p38α phosphorylation, which was markedly attenuated by 10 μmol/l epalrestat. A p38 phosphoproteomic array confirmed that collagen specifically induced phosphorylation of the α isoforms (Figure 5A). The p38α activation by collagen and inhibition by epalrestat in both NG and HG was confirmed by Western blot (Figure 5B). Interestingly, p38α phosphorylation was significantly increased in HG-incubated platelets even in the absence of collagen stimulation (Figure 5B), suggesting that HG incubation may potentiate platelet aggregation via p38α, perhaps through its downstream effector cytosolic phospholipase A₂ (cPLA₂) (15), leading to TX biosynthesis. Most importantly, collagen-induced platelet aggregation in both NG and HG was reduced by pretreatment with a p38 inhibitor (Figure 5C).

Figure 5

AR is required for p38α MAPK phosphorylation in collagen-stimulated platelets. Platelet suspensions were incubated with NG or HG for 90 minutes in the presence or absence of 10 μmol/l epalrestat prior to stimulation by 1 μg/ml collagen for 10 minutes, and the total extract was harvested for experiments. (A) Phosphorylation of different kinases in collagen-stimulated platelets under NG conditions was assayed by human phosphokinase array kits. Representative results are shown. Co, collagen. (B) Phosphorylation of p38α MAPK was further measured by Western blot using specific antibody for its phosphorylated form. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 compared with values incubated in NG alone; ^###P < 0.001 compared with values in NG with the addition of 1 μg/ml collagen; ^§§§P < 0.001 compared with values in HG with the addition of 1 μg/ml collagen. (C) Platelet suspensions were incubated with NG or HG for 90 minutes in the presence or absence of 50 μmol/l SB239063 prior to stimulation by 1 μg/ml collagen. Platelet aggregation was presented as the percentage of light transmission and measured in platelet suspensions under NG or HG conditions in response to 1 μg/ml collagen for 5 minutes. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 compared with vehicle incubated in NG; ^###P < 0.001 compared with vehicle incubated in HG. Gluc, glucose.

We next aimed to understand the mechanism by which collagen activates p38α. We have demonstrated that collagen activates AR (Figure 4A). As AR can lead to ROS generation, which can regulate p38α (16), we measured oxidative stress production by using specific fluorescent probes. The levels of oxidative stress and superoxide were significantly increased upon collagen-induced platelet aggregation and were even higher in HG compared with NG (Figure 6A). The increases were significantly attenuated by epalrestat, suggesting that AR contributes to oxidative stress in collagen-stimulated platelets. Treatment with 10 mmol/l N-acetyl-cysteine (NAC, ROS scavenger) or 100 μmol/l apocynin (NADPH oxidase inhibitor) also significantly reduced levels of oxidative stress and superoxide. Moreover, HG increased oxidative stress and superoxide production even in the absence of collagen. Notably, NAC and apocynin attenuated collagen-induced platelet aggregation and TX release to an extent similar to that of ARI, indicating that the oxidative stress likely transduced the AR signal (Figure 6, B and C). Pyocyanin (Pyo), an ROS inducer, served as a positive control in the experiment (Figure 6A).

Figure 6

AR contributes to oxidative stress in collagen-stimulated platelets. Platelet suspensions were incubated with NG or HG for 90 minutes in the presence or absence of 10 μmol/l epalrestat, 1 mmol/l NAC, and 100 μmol/l apocynin (Apo). (A) Washed platelets were incubated with 1 μmol/l ROS/Superoxide Detection Mix for 60 minutes at 37°C in the presence or absence of 1 μg/ml collagen for 10 minutes. The quantification of data for ROS is shown in the left panel, and that for superoxide is shown in the right panel. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001, **P < 0.01, and *P < 0.05 compared with values incubated in NG alone; ^###P < 0.001, ^##P < 0.01, and ^#P < 0.05 compared with values in NG with the addition of 1 μg/ml collagen; ^§§§P < 0.001 compared with values in HG with the addition of 1 μg/ml collagen. (B) Aggregation expressed as the percentage of light transmission and measured in platelet suspensions under NG or HG conditions. (C) Levels of TXB₂ assayed in the supernatant. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 compared with values incubated in NG with the addition of 1 μg/ml collagen; ^###P < 0.001 compared with values in HG with the addition of 1 μg/ml collagen.

PLCγ2 is phosphorylated and activated in response to oxidative stress, which is thought to play an important role in platelet activation (17). As shown in Figure 7A, under both NG and HG conditions, the phosphorylation of PLCγ2 was significantly enhanced upon collagen-induced platelet aggregation. Moreover, PLCγ2 phosphorylation was increased under HG conditions even without collagen stimulation. As shown in Figure 7B, treatment with the PLC inhibitor U73122 or the ROS scavenger NAC significantly reduced collagen-induced phosphorylation of p38α MAPK, providing evidence that p38α MAPK activation was regulated by PLCγ2 during collagen-induced platelet aggregation and that this process is associated with oxidative stress. In addition, treatment with another AR inhibitor, quercetin (structurally distinct from epalrestat and with broader specificity), also attenuated collagen-induced phosphorylation of p38α MAPK, further confirming the effect of AR inhibition (Figure 7B). Taken together, these results suggest that AR contributes to collagen-induced platelet activation via oxidative stress–induced tyrosine phosphorylation of PLCγ2.

Figure 7

AR is required for PLC phosphorylation in collagen-stimulated platelets. (A) Platelet suspensions were incubated with NG or HG for 90 minutes in the presence or absence of 10 μmol/l epalrestat prior to stimulation by 1 μg/ml collagen for 10 minutes, and the total extract was harvested for experiments. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 and **P < 0.01 compared with values incubated in NG alone; ^###P < 0.001 compared with values in NG with the addition of 1 μg/ml collagen; ^§§P < 0.01 compared with values in HG with the addition of 1 μg/ml collagen. p-PLCγ2, phospho-PLCγ2. (B) Inhibition of the upstream pathway of p38α MAPK attenuated its phosphorylation in collagen-stimulated platelets. Platelet suspensions were incubated with NG or HG for 90 minutes in the presence of 5 μmol/l U73122, 1 mmol/l NAC, and 10 μmol/l quercetin prior to stimulation by 1 μg/ml collagen for 10 minutes, and the total extract was harvested for experiments. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 and *P < 0.05 compared with values incubated in NG or HG alone, respectively; ^###P < 0.001 compared with values in NG or HG with the addition of 1 μg/ml collagen, respectively. p-p38α, phospho-p38α.

Inhibition of AR prevents the translocation and phosphorylation of PKC isoforms (α, βII, and δ) in human platelets in response to collagen under NG and HG conditions. PLCγ leads to the generation of diacylglycerol (DAG), which activates the PKC isoforms (18, 19). p38α MAPK is activated downstream of PKC in platelets (20). We therefore studied the effect of AR on PKC isoforms (α, βII, and δ) known to be present in platelets. As shown in Figure 8, A and B, PKCα was highly expressed in the cytosolic fraction in the human platelets under both NG and HG conditions. Upon collagen stimulation, the level of PKCα was increased in the membrane compared with the cytosolic fraction, indicating PKCα translocation to a membrane during platelet aggregation. Inhibition of AR with 10 μmol/l epalrestat prevented collagen-induced translocation of PKCα under both NG and HG conditions. Similar to PKCα, PKCβII and PKCδ were highly expressed in the cytosolic fraction and translocated to the membrane upon collagen-induced platelet aggregation under NG conditions (Figure 8, A, C, and D). However, under HG conditions, the levels of PKCβII and PKCδ were higher in the membrane fraction than in the cytosolic fraction with or without 1 μg/ml collagen, which was attenuated by the treatment with 10 μmol/l epalrestat. The blots were probed with antibodies against Na⁺/K⁺ ATPase and GAPDH, which served as the membrane and cytosolic markers, respectively (Figure 8A). Since the subcellular translocation of PKCα, PKCβII, and PKCδ was prevented by AR inhibitor, we next determined whether inhibition of AR prevents the phosphorylation of PKCα, PKC βII, and PKCδ by using phosphospecific antibodies that correlate with kinase activity in total extracts. Consistent with the previous genetic knockout and inhibitor studies on the role of PKC isoforms in platelet aggregation (21, 22), the phosphorylation of different PKC isoforms was in line with the findings in their translocation (Supplemental Figure 4), indicating that PKCα, PKCβII, and PKCδ translocation from cytoplasm to membrane and phosphorylation are required for collagen-induced aggregation. In addition, these results suggest that HG potentiates the activation of PKCβII and PKCδ similarly to the effect of collagen in an AR-dependent manner.

Figure 8

AR is required for the intracellular translocation of PKC isoforms (α, βII, and δ) in collagen-stimulated platelets. (A) Platelet suspensions were incubated with NG or HG for 90 minutes in the presence or absence of 10 μmol/l epalrestat prior to stimulation by 1 μg/ml collagen for 10 minutes, and the membrane and cytosolic fractions were harvested for Western blot. Representative blots are shown. Expressions of PKC isoforms in cytosolic and membrane fractions were normalized to GAPDH and Na⁺/K⁺ ATPase expression, respectively. Quantitative results of (B) PKCα, (C) PKCβII, and (D) PKCδ expression were plotted in histograms. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 compared with values incubated in NG alone; ^###P < 0.001 and ^##P < 0.01 compared with values in NG with the addition of 1 μg/ml collagen; ^§§§P < 0.001 and ^§§P < 0.01 compared with values in HG with the addition of 1 μg/ml collagen.

Inhibition of AR blocks TX release in human platelets in response to collagen under NG and HG conditions. cPLA₂ generates AA, the substrate for prostanoid synthesis. As cPLA₂ is a downstream effector of p38α MAPK in platelets (15, 16), we postulated that AR contributes to collagen-induced aggregation via enhanced TXA₂ signaling. We assessed the generation of TXB₂ (nonenzymatic metabolite of TXA₂) and the levels of TX receptor (TP) in the cytosolic and membrane fractions. During collagen-induced platelet aggregation, TXB₂ generation was increased by 10-fold (from ~500 to ~5000 pg/ml) under NG conditions and by nearly 12-fold (from ~650 to ~8000 pg/ml) under HG conditions (Figure 9A). TXB₂ required AR activity, as treatment with 10 μmol/l epalrestat significantly reduced the levels of TXB₂ released from platelets. The surface expression of TP was also significantly increased, and its expression in the cytosolic fraction was decreased upon collagen-induced platelet aggregation in the HG group but not in the NG group (Figure 9B). Treatment with 10 μmol/l epalrestat significantly abolished the changes in surface expression. ROS also contributed to the TXB₂ induction (Figure 6, A and C) and increased platelet activity (Figure 6B). In addition, the HG-induced surface expression of TP was also significantly attenuated by treatment with NAC and apocynin (Figure 9C), supporting a role for ROS. Collectively, these results suggest that AR and oxidative stress are required for hyperglycemia-induced TX generation and TP surface expression.

Figure 9

AR is involved in TX signaling in platelets. Role of AR in (A) TX generation and (B) surface expression of TP in collagen-stimulated platelets. Platelet suspensions were incubated with NG or HG for 90 minutes in the presence or absence of 10 μmol/l epalrestat prior to stimulation with 1 μg/ml collagen for 10 minutes. Levels of TXB₂ were assayed in the supernatant. Expression of TP was assessed in the membrane and cytosolic fractions. Expression of TP in cytosolic and membrane fraction was normalized to GAPDH and Na⁺/K⁺ ATPase expression, respectively. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 compared with values incubated in NG alone; ^###P < 0.001 and ^##P < 0.01 compared with values in NG with the addition of 1 μg/ml collagen; ^§§§P < 0.001 and ^§§P < 0.01 compared with values in HG with the addition of 1 μg/ml collagen. (C) Expression of TP was assessed in the membrane and cytosolic fractions. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 compared with values incubated in HG alone; ^###P < 0.001 compared with values in HG with the addition of 1 μg/ml collagen.

Correlation of TXB₂ generation with increased platelet aggregation. We have delineated a pathway in which AR contributes to platelet activation via PLC-dependent activation of p38α, with subsequent TXA₂ generation. To validate this pathway, platelet aggregation and TXB₂ levels were measured in the presence of U73122 (PLC inhibitor), SB239063 (p38 inhibitor), quercetin, and SQ29548 (TP antagonist). Treatment with PLC and p38 inhibitors attenuated collagen-induced TXB₂ release, suggesting that PLCγ2 and its downstream effector, p38α MAPK, contributes to TXB₂ release. Moreover, all of these inhibitors reduced platelet aggregation in response to 1 μg/ml collagen. Importantly, treatment with SQ29548 also attenuated collagen-induced platelet aggregation, suggesting an important role for TP in increasing platelet aggregation in NG or HG (Figure 10A). SQ29548 also attenuated collagen-induced TXB₂ release, suggesting a potential positive feedback mechanism (Figure 10B). Collectively, our results support a model in which AR activity increases oxidative stress and mediates PLCγ2 phosphorylation and PKC activation, which leads to p38α MAPK activation, resulting in collagen-induced TXA₂ generation, TP activation, and aggregation in platelets.

Figure 10

Activation of p38α MAPK signaling pathway is required for TX generation and aggregation in collagen-stimulated platelets. Platelet suspensions were incubated with NG or HG for 90 minutes in the presence or absence of 50 μmol/l SB239063, 5 μmol/l U73122, 10 μmol/l quercetin, and 10 μmol/l SQ29548 prior to stimulation by 1 μg/ml collagen. (A) The platelet aggregation was presented as the percentage of light transmission and measured in platelet suspensions under NG or HG condition in response to 1 μg/ml collagen for 5 minutes. (B) After stimulation, the supernatant was harvested for the measurement of TXB₂ generation. Data are expressed as mean ± SD (n = 5 HS). ***P < 0.001 compared with vehicle incubated in NG; ^###P < 0.001 compared with vehicle incubated in HG. (C) Correlation of TX release with increased platelet aggregation in collagen-stimulated human platelets. Platelet suspensions were incubated with NG or HG for 90 minutes prior to stimulation by 1 μg/ml collagen for 10 minutes; the percentage of light transmission was also measured in platelet suspensions, and the TX release was assayed in the supernatant.

In order to assess the association between TXA₂ generation and increased platelet aggregation, we plotted the percentage of aggregation against the level of released TXB₂. As shown in Figure 10C, a linear relationship with positive slope (R² = 0.8439; slope = 6.192) occurred between TXB₂ levels and the percentage of aggregation, indicating that increased platelet aggregation is closely correlated with TXA₂ generation in our experimental human models. In addition, this result supports the concept that platelet aggregation increases TXA₂ biosynthesis and increased TXA₂ biosynthesis increases platelet aggregation, leading to an amplifying positive feedback mechanism. This important observation provides an explanation for increased platelet activity during hyperglycemia by AR-dependent induction of TXA₂ generation.

Increased urinary TX-M levels in patients with deep vein thrombosis (study group 2). We demonstrated that TXA₂ generation is increased with collagen-induced activation of normal human platelets ex vivo and that acute hyperglycemia further enhanced TXA₂ biosynthesis in vitro. To determine whether glucose levels and collagen-dependent thrombosis induce TXA₂ production in patients, we measured urinary levels of TX-M in a case-controlled study of patients with or without deep vein thrombosis (DVT). DVT is known to be caused in part by endothelial damage and collagen exposure (23). The DVT patients (n = 34) and controls (n = 23) were characterized by comparable age, cardiovascular risk factors (cigarette smoking, DM, hypertensive status, lipid levels), BMI, and systemic inflammatory status index (i.e., C-reactive protein [CRP] and fibrinogen). As shown in Figure 11A, there was a statistically significant increase of TX-M in the DVT patients versus controls (for non-DM patients without DVT, the median [with IQR (Q1–Q3)] was 551.5 [436.2 to 600] pg/mg creatinine; for non-DM patients with DVT, it was 942 [657.5 to 1319] pg/mg creatinine). There were 8 DM patients out of 57 total patients, but unfortunately the data on urinary levels of TX-M were only available in 5 DM patients. Interestingly, these few DM patients that were presented had very high TX-M only in the presence of DVT. A further study was necessary in a patient population with a higher incidence of DM to assess the link among DM (hyperglycemia), thrombosis (platelet hyperactivity), and TXA₂ generation in vivo (urinary levels of TX-M).

Figure 11

Increased urinary TX metabolites in patients with DVT or CABG thrombosis. (A) For patients with DVT, the level of urinary TX metabolite was measured as an index of TX release in DM patients with or without DVT (n = 2 and n = 3, respectively) compared with non-DM patients with or without DVT (n = 32 and n = 20, respectively). None of the patients were on ASA therapy. (B) For patients with CABG thrombosis (RIGOR patients), the level of urinary TX metabolite was measured in DM patients with or without CABG thrombosis (n = 28 and n = 75, respectively) compared with non-DM patients with or without CABG thrombosis (n = 68 and n = 120, respectively). All the patients were treated with ASA (325 mg once daily) upon discharge from surgery as described in Methods. Data are expressed as median with IQR.

Increased urinary levels of TX-M in DM patients with saphenous vein coronary artery bypass graft thrombosis treated with ASA (study group 3). Because of our interesting DVT findings, we performed another case-controlled study assessing another patient population with a much higher prevalence of DM: a severe coronary artery disease population with coronary artery bypass graft (CABG) thrombosis (Table 1). Not surprisingly, with such severe disease, all patients were on 325 mg daily ASA (to inhibit platelet COX-1–dependent production of TXA₂) for secondary prevention of further cardiovascular events. The groups were comparable for age and cardiovascular risk factors. As shown in Figure 11B, there were no TX-M differences between DM and non-DM patients without graft occlusion while on ASA (for non-DM patients without CABG thrombosis, the median [with IQR (Q1–Q3)] was 328 pg/mg [233 to 424.5] creatinine; for DM patients without CABG thrombosis, it was 312 pg/mg [226 and 495] creatinine). However, among those with thrombosis, there was a significant difference between DM and non-DM patients (for non-DM patients with CABG thrombosis, the median [with IQR (Q1–Q3)] was 304 [206 and 450] pg/mg creatinine; for DM patients with CABG thrombosis, it was 466.5 [275 and 718] pg/mg creatinine). This supports our experimental observation that TXA₂ biosynthesis in vivo is enhanced in patients with both thrombosis and DM despite the use of ASA. The mechanism for this apparent insensitivity to ASA in the presence of thrombosis and DM required further exploration.

Table 1

Cardiovascular risk factors for the populations studied

ASA ex vivo does not cause a complete saturation of platelet COX-1 activity. In order to verify whether enhanced generation of TXA₂ in vivo was a consequence of inadequate inhibition of platelet COX-1 (ex vivo) by ASA administration, we compared residual generation of TXB₂ 12 hours after the last dose of ASA in whole blood allowed to clot for 1 hour at 37°C (serum TXB₂), which is an index of maximal capacity of platelets to generate TXB₂.

We detected enhanced median values of residual generation of platelet TXB₂ in DM patients versus HS; the differences were not statistically significant. However, 12% of DM patients had higher values of residual TXB₂ (i.e., 4 ng/ml) than the upper extreme values of serum TXB₂ detected in control subjects after ASA administration. In order to verify whether in these patients higher generation of platelet TXB₂ was due to the phenomenon of ASA resistance, which implies that the ASA target, i.e., COX-1, is less sensitive to inactivation by ASA, we assessed residual generation of TXB₂ after adding ASA (50 μg/ml) in vitro. As shown in Figure 12, in vitro addition of ASA caused a further significant reduction of TXB₂ generation in both DM patients and HS. Importantly, only 2% of patients had residual TXB₂ levels higher than 4 ng/ml in DM groups with ASA in vitro. Our results (Figure 12) show that in almost all DM patients, ASA is able to reduce serum TXB₂ to a similar extent when in vivo followed by in vitro ASA treatment of patients’ blood is used. Thus, the apparent ASA insensitivity and platelet hyperactivity is, in most cases, not due to ASA resistance, but rather a combination of inadequate ASA bioavailability and increased TX production (and receptor expression) via the AR/ROS pathway. It is important to point out that the observed residual COX-1 activity is present even after ASA treatment. This COX-1 activity likely converts the increased AA substrate generated from the p38α MAPK/PLA₂ activation pathway, leading to enhanced platelet TXA₂ generation and increased platelet activation under conditions of acute hyperglycemia and collagen activation.

Figure 12

Effect of in vitro administration of ASA on the residual TXB₂ generation in HS and DM patients. Serum TXB₂ levels, markers of platelet COX-1 activity, were assessed in HS (n = 14) and DM (n = 102) patients with low-dose ASA (100 mg/day). Residual generation of TXB₂ was assessed in whole blood allowed to clot for 1 hour at 37°C 12 hours after the last dose of ASA. Data are expressed as median with IQR. **P < 0.01 and *P < 0.05 compared with values in the corresponding group without ASA in vitro.

Analysis of human TP genetic variants in cardiovascular patients. In order to ensure and confirm that the increase in TX levels was not secondary to a dysfunctional or defective TP, genomic DNA from patients and volunteers from the Yale (human platelet signaling studies), Chieti (study groups 1 and 2), and Johns Hopkins (study group 3) studies was sequenced for variants in the TP gene (TBXA2R). No dysfunctional nonsynonymous mutations could account for the observed results (data not shown). This result demonstrates that the difference in TX-M levels between the DM and non-DM patients could not be attributed to structural defects in the human TP (hTP).

We set out to discover the mechanism leading to TX release and platelet hyperactivity in DM. We demonstrate for what we believe is the first time that AR plays a central role in synergistically trans­ducing hyperglycemia and collagen exposure to TX release, increased cell surface TP expression, and enhanced platelet activity. This effect is associated with ROS generation and the activation of PLCγ2, PKCβII, PKCδ, and p38α MAPK (Figure 13). Our clinical studies support these results, as increased TXA₂ generation in vivo, particularly in DM patients, reflected underlying thrombovascular disease (DVT and CABG thrombosis). These mechanistic insights may explain in part the poor prognosis observed with patients presenting with hyperglycemia and acute coronary syndrome (24–30). Enhanced platelet activity due to a combination of hyperglycemia and collagen exposure may predispose to further thromboembolic disease.

Figure 13

Proposed mechanism for hyperglycemia-induced platelet hyperactivity through TX signaling mediated by the AR pathway. Hyperglycemia synergizes with collagen stimulation in platelets to activate the AR-dependent polyol pathway, leading to increased ROS production (via GSH depletion and NADH oxidase [Nox]). Excessive ROS can increase TXA₂ production by potentiating PLCγ2/PKC/p38α MAPK signaling, thus providing more arachidonic acid substrate for COX-1 and TXA₂ synthase (TS) to produce TXA₂, leading to platelet hyperactivity. The oxidative stress can also lead to enhanced surface expression of TP, which increases the potency of circulating TXA2.

Complex regulatory pathway required for the synergistic effects of glucose on collagen-stimulated aggregation. The enhanced utilization of the AR pathway during hyperglycemia is well known to contribute to disease in many tissues, including microvascular and macrovascular complications, polyneuropathy, and retinopathy in DM patients (10, 11). Surprisingly, its role and relationship to TX release and platelet hyperactivity has been unexplored. We now demonstrate that AR plays a critical role in transducing collagen signaling in platelet aggregation and that this is further enhanced in conditions of hyperglycemia. Our mechanistic studies are supported by a recent study using 2D difference gel electrophoresis and mass spectrometry, which demonstrated that the expression of AR was significantly increased in the activated human platelets after stimulation of the GPVI receptor (14). The binding of collagen to GPVI results in receptor clustering and phosphorylation of the tyrosine kinase Syk and the activation of PLC (31). In platelets, many PLC isoforms have been identified, and evidence suggests that the signaling through collagen-induced aggregation is mediated by the receptor-bound tyrosine kinases and activation of PLCγ (32, 33). We have now demonstrated that PLCγ2 phosphorylation is greatly increased in the glucose-treated platelets and collagen-induced platelet aggregation and that PLC inhibition using U73122 effectively prevented collagen-induced TXB₂ generation and aggregation, suggesting that PLCγ2 phosphorylation is required for collagen-induced platelet aggregation. Looking further downstream (Figure 13), in HG- and collagen-stimulated platelets, PKCα, PKCβII, and PKCδ were translocated from cytosol to membrane, with concomitant increase in their phosphorylated forms, thus supporting a role of PKCα, PKCβII, and PKCδ in this process. These findings are also supported by previous studies that demonstrate PKCα, PKCβII, and PKCδ as playing a crucial role in collagen-induced platelet aggregation (21, 22). Thus, a complex web of signaling components is involved in platelet activation upon exposure to collagen and hyperglycemia (Figure 13). Our study unifies many of the platelet-dependent signaling components in the AR transduction of glucose signaling. AR appears to serve as a master switch that causes collagen-induced platelet aggregation to be enhanced by the presence of HG.

Oxidative stress plays an important role in TX-induced platelet hyperactivity. Glucose flux through AR is well recognized to generate oxidative stress by several mechanisms, including NADPH depletion, decreasing glutathione (GSH) levels (NADPH is the cofactor for GSH reductase that regenerates GSH from GSH disulfide [GSSG]), and increasing advanced glycation end products, generating ROS (10). Previous studies have demonstrated that TXA₂ biosynthesis is upregulated by oxidant formation (e.g., hydrogen peroxide) during platelet aggregation (34–36). We and others (16, 20) have demonstrated that hyperglycemia increases oxidative stress in platelets. We show that this potentiates p38α MAPK/cPLA₂ signaling, which catalyzes AA release, producing TXA₂. Interestingly, ROS derived from platelet activation also plays an important role in signaling upon collagen-induced platelet aggregation. Free radical species act as secondary messengers that increase cytosolic Ca²⁺ during the initial phase of platelet activation processes, and PKC is involved in receptor-mediated ROS production in platelets (36). Previous studies also showed that ROS formation is involved in collagen-induced platelet aggregation, which is dependent on AA release and its metabolism (34); thus, ROS may be needed for platelet activation (37). Indeed, our studies demonstrate that oxidative stress is significantly increased in collagen-stimulated platelets, and inhibition of AR partly attenuated collagen-induced oxidative stress, supporting an important role for AR-exacerbated oxidative stress in collagen-dependent signaling. Furthermore, our current study and those of others have now demonstrated oxidative stress–induced stabilization and relocation of TP to the cell surface (38, 39).

Positive feedback cycle of TX release and platelet hyperactivity. Urinary TX-M, reflecting the whole biosynthesis of TXA₂ in the body, mainly by platelets with a minor contribution from extraplatelet sources, is significantly higher in DM, with the absolute post-ASA values in DM patients being comparable to those of controls who did not receive ASA (7, 8). Many believe enhanced TXA₂ biosynthesis in vivo in T2DM is predominantly derived from platelets (7), and such studies support our findings that platelet degranulation and synthesis of TXA₂ mediate further platelet activation in DM. This then raises the issue of identification of the initiating events that lead to this cascade of hyperactivation, thrombosis, and hyperactivation. As described, it may be that the trigger is endothelial damage (collagen exposure) and that, when combined with hyperglycemia-induced oxidative stress, ROS initiates an amplifying cycle of increased PLC/PKC/p38α MAPK signaling, TXA₂ generation, and platelet hyperactivation. Interestingly, previous animal studies demonstrated that enhanced platelet aggregation and TX synthesis may be detected within days of triggering DM with streptozotocin, suggesting that alterations in platelets and endothelial damage (with collagen exposure) may occur early in the DM state (8). Extraplatelet factors such as NO and prostacyclin are keys to breaking such a cycle; however, there is growing evidence that these naturally occurring antiaggregatory agents are reduced in DM (40).

As our DM patients were on ASA, it was necessary to determine why TX production remained elevated, particularly in the patients from study group 1 (Figure 1). We offer 2 major reasons based upon our results. First, as described, hyperglycemia synergizes with collagen stimulation in platelets to activate the AR-dependent polyol pathway, which increases oxidative stress. The oxidative stress can increase TXA₂ production by potentiating p38α MAPK/cPLA₂ signaling, thus providing more substrates (AA) for uninhibited COX-1 to produce TXA₂. Additionally, hyperglycemia- and collagen-induced oxidative stress from the AR-dependent polyol pathway can also lead to enhanced expression of TP, increasing the potency of any circulating TXA₂. When combined with reduced bioavailability, as observed with the 100 mg (enteric coated) daily usage patients, and thus incomplete inhibition of COX-1 (Figure 12), the combination results in a marked increase in TXA₂ over that in control patients (Figure 1). Such differences are less marked when patients are on 325 mg ASA (study group 3, Reduction in Graft Occlusion Rates [RIGOR] group; Figure 11B). Thus AR-dependent signaling pathways upon glucose/collagen platelet activation may provide, in part, an explanation for apparent ASA insensitivity in DM patients and may shed light on the concept of ASA insensitivity in general.

AR inhibitors as antiplatelet therapy? Our study highlights the importance of concurrent clinical studies to substantiate in vitro findings. DM patients with high urinary levels of TX-M despite ASA treatment may have underlying occult endothelial damage and thrombovascular disease. We propose that in the context of vascular damage with collagen exposure and HG, unacetylated platelet COX-1 despite ASA may participate in TXA₂ biosynthesis on a background of increased p38α MAPK–induced PLA₂ activity and thus enhanced AA substrate. Higher doses of ASA or low-dose ASA given twice daily may translate into an appropriate suppression of TXA₂ in vivo (assessing the levels of urinary TX-M). Alternatively, the addition of AR inhibitors should be considered to mitigate enhanced TXA₂ generation despite ASA treatment in DM patients. Epalrestat used in the present study is an AR inhibitor that is approved in Japan to prevent or slow the progression of DM neuropathy (41, 42). Our study using human platelets demonstrates that hyperglycemia-induced TX signaling and hyperaggregability were effectively attenuated by treatment with epal­restat. The use of AR inhibitors may be particularly effective against postprandial hyperglycemia, which is associated with enhanced lipid peroxidation and platelet activation in early T2DM (43) and increased platelet aggregation in DM patients with microangiopathy (44). The addition of AR inhibitors may be a new therapeutic option to prevent thrombotic events in DM patients.

Limitations of the present studies. There are a number of limitations to our current studies. Foremost, the sizes of our studies are relatively small. Further larger studies assessing both arterial and venous thrombosis in the setting of hyperglycemia and DM are required to corroborate our results. Although we have focused on the platelet-dependent mechanism, this is a reductionist approach, as many other factors from blood cells, the endothelium, and vascular smooth muscle cells clearly play an important role in the development of the DM platelet response. Our studies also focus on acute increases in glucose without addressing the effects of chronic increases. Moreover, there are many other important components to the DM response (e.g., protein glycation) besides AR that likely contribute to increased platelet activity. The final picture is likely to be far more complex, as not all studies in DM patients have demonstrated consistent increases in TX production (45). Despite these limitations, our study provides strong evidence for a multitargeted approach and the foundation and justification for further clinical therapeutic trials to prevent atherothrombotic disease arising from platelet hyperactivity in patients with DM.

Summary. It is currently a subject of debate whether enhanced platelet activity associated with DM is a cause (thrombosis, e.g., plaque rupture) or consequence (metabolic disturbance) of disease. Our studies indicate that both of these are true. The combination of hyperglycemia during collagen activation leads to a positive feedback cycle of release of platelet TX and enhanced platelet aggregation (cause and consequence). Moreover, we demonstrate that AR leading to enhanced ROS production and TX generation plays a major role in this synergistic response. Our findings support the use of higher doses of ASA; however, inhibition of vascular prostacyclin biosynthesis makes this choice inadvisable. The foundation for antiplatelet therapy in DM patients may very well be a combination of ASA with AR inhibitors (and/or TX inhibitors) for both primary and secondary prevention of cardiovascular events.

View Supplemental data

This work was supported in part by the NIH (RO1 HL074190 to J. Hwa), an American Heart Association Established Investigator Award (to J. Hwa), the Ministero dell’Istruzione, dell’Università e della Ricerca (PRIN 2006 to P. Patrignani), a Fondazione Carichieti-Fondazione Negri Sud ONLUS (to V. Evangelista), and by the European Community Sixth Framework Programme (Eicosanox grant LSMH-CT-2004-005033 to P. Patrignani). We also thank Todd Mackenzie (Statistician, Department of Medicine, Dartmouth Medical School) for his statistical advice and analysis, and Stefano Manarini for her technical assistance.

Address correspondence to: John Hwa, Department of Internal Medicine, Section of Cardiovascular Medicine, Yale School of Medicine, Cardiovascular Research Center, 300 George Street, Rm. 759H, New Haven, Connecticut 06511, USA. Phone: 203.737.5583; Fax: 203.737.6118; E-mail: John.Hwa@yale.edu.

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J Clin Invest doi:10.1172/JCI59291.

1. Beckman JA, Creager MA, Libby P. Diabetes and atherosclerosis: epidemiology, pathophysiology, and management. JAMA. 2002;287(19):2570–2581.
   View this article via: PubMed CrossRef Google Scholar
2. Morel O, Kessler L, Ohlmann P, Bareiss P. Diabetes and the platelet: toward new therapeutic paradigms for diabetic atherothrombosis. Atherosclerosis. 2010;212(2):367–376.
   View this article via: PubMed CrossRef Google Scholar
3. Ferreiro JL, Angiolillo DJ. Diabetes and antiplatelet therapy in acute coronary syndrome. Circulation. 2011;123(7):798–813.
   View this article via: PubMed CrossRef Google Scholar
4. Carr ME. Diabetes mellitus: a hypercoagulable state. J Diabetes Complications. 2001;15(1):44–54.
   View this article via: PubMed CrossRef Google Scholar
5. Bern MM. Platelet functions in diabetes mellitus. Diabetes. 1978;27(3):342–350.
   View this article via: PubMed Google Scholar
6. Brownlee M. Biochemistry and molecular cell biology of diabetic complications. Nature. 2001;414(6865):813–820.
   View this article via: PubMed Google Scholar
7. Davi G, et al. Thromboxane biosynthesis and platelet function in type II diabetes mellitus. N Engl J Med. 1990;322(25):1769–1774.
   View this article via: PubMed CrossRef Google Scholar
8. Ferroni P, Basili S, Falco A, Davi G. Platelet activation in type 2 diabetes mellitus. J Thromb Haemost. 2004;2(8):1282–1291.
   View this article via: PubMed CrossRef Google Scholar
9. Bhatnagar A, Srivastava SK. Aldose reductase: congenial and injurious profiles of an enigmatic enzyme. Biochem Med Metab Biol. 1992;48(2):91–121.
   View this article via: PubMed CrossRef Google Scholar
10. Chung SS, Ho EC, Lam KS, Chung SK. Contribution of polyol pathway to diabetes-induced oxidative stress. J Am Soc Nephrol. 2003;14(8 suppl 3):S233–S236.
    View this article via: PubMed CrossRef Google Scholar
11. Baynes JW, Thorpe SR. Role of oxidative stress in diabetic complications: a new perspective on an old paradigm. Diabetes. 1999;48(1):1–9.
    View this article via: PubMed CrossRef Google Scholar
12. Vikramadithyan RK, et al. Human aldose reductase expression accelerates diabetic atherosclerosis in transgenic mice. J Clin Invest. 2005;115(9):2434–2443.
    View this article via: JCI PubMed CrossRef Google Scholar
13. Marcus AJ, Broekman MJ, Pinsky DJ. COX inhibitors and thromboregulation. N Engl J Med. 2002;347(13):1025–1026.
    View this article via: PubMed Google Scholar
14. Schulz C, et al. Identification of novel downstream targets of platelet glycoprotein VI activation by differential proteome analysis: implications for thrombus formation. Blood. 2010;115(20):4102–4110.
    View this article via: PubMed Google Scholar
15. Kramer RM, et al. p38 mitogen-activated protein kinase phosphorylates cytosolic phospholipase A2 (cPLA2) in thrombin-stimulated platelets. Evidence that proline-directed phosphorylation is not required for mobilization of arachidonic acid by cPLA2. J Biol Chem. 1996;271(44):27723–27729.
    View this article via: PubMed CrossRef Google Scholar
16. Coulon L, Calzada C, Moulin P, Vericel E, Lagarde M. Activation of p38 mitogen-activated protein kinase/cytosolic phospholipase A2 cascade in hydroperoxide-stressed platelets. Free Radic Biol Med. 2003;35(6):616–625.
    View this article via: PubMed Google Scholar
17. Leoncini G, Bruzzese D, Signorello MG. A role for PLCgamma2 in platelet activation by homocysteine. J Cell Biochem. 2007;100(5):1255–1265.
    View this article via: PubMed CrossRef Google Scholar
18. Lee MW, Severson DL. Signal transduction in vascular smooth muscle: diacylglycerol second messengers and PKC action. Am J Physiol. 1994;267(3 Pt 1):C659–C678.
    View this article via: PubMed Google Scholar
19. Nishizuka Y. Intracellular signaling by hydrolysis of phospholipids and activation of protein kinase C. Science. 1992;258(5082):607–614.
    View this article via: PubMed CrossRef Google Scholar
20. Saklatvala J, et al. Role for p38 mitogen-activated protein kinase in platelet aggregation caused by collagen or a thromboxane analogue. J Biol Chem. 1996;271(12):6586–6589.
    View this article via: PubMed CrossRef Google Scholar
21. Yacoub D, et al. Essential role of protein kinase C delta in platelet signaling, alpha IIb beta 3 activation, and thromboxane A2 release. J Biol Chem. 2006;281(40):30024–30035.
    View this article via: PubMed Google Scholar
22. Gilio K, et al. Functional divergence of platelet protein kinase C (PKC) isoforms in thrombus formation on collagen. J Biol Chem. 2010;285(30):23410–23419.
    View this article via: PubMed CrossRef Google Scholar
23. Saha P, et al. Leukocytes and the natural history of deep vein thrombosis: current concepts and future directions. Arterioscler Thromb Vasc Biol. 2011;31(3):506–512.
    View this article via: PubMed Google Scholar
24. Deedwania P, et al. Hyperglycemia and acute coronary syndrome: a scientific statement from the American Heart Association Diabetes Committee of the Council on Nutrition, Physical Activity, and Metabolism. Circulation. 2008;117(12):1610–1619.
    View this article via: PubMed CrossRef Google Scholar
25. Kosiborod M, et al. Admission glucose and mortality in elderly patients hospitalized with acute myocardial infarction: implications for patients with and without recognized diabetes. Circulation. 2005;111(23):3078–3086.
    View this article via: PubMed CrossRef Google Scholar
26. Foo K, et al. A single serum glucose measurement predicts adverse outcomes across the whole range of acute coronary syndromes. Heart. 2003;89(5):512–516.
    View this article via: PubMed CrossRef Google Scholar
27. Iwakura K, et al. Association between hyperglycemia and the no-reflow phenomenon in patients with acute myocardial infarction. J Am Coll Cardiol. 2003;41(1):1–7.
    View this article via: PubMed Google Scholar
28. Wahab NN, et al. Is blood glucose an independent predictor of mortality in acute myocardial infarction in the thrombolytic era? J Am Coll Cardiol. 2002;40(10):1748–1754.
    View this article via: PubMed CrossRef Google Scholar
29. Svensson A-M, McGuire DK, Abrahamsson P, Dellborg M. Association between hyper- and hypoglycaemia and 2 year all-cause mortality risk in diabetic patients with acute coronary events. Eur Heart J. 2005;26(13):1255–1261.
    View this article via: PubMed CrossRef Google Scholar
30. Suleiman M, et al. Fasting glucose is an important independent risk factor for 30-day mortality in patients with acute myocardial infarction: a prospective study. Circulation. 2005;111(6):754–760.
    View this article via: PubMed CrossRef Google Scholar
31. Yanaga F, et al. Syk interacts with tyrosine-phosphorylated proteins in human platelets activated by collagen and cross-linking of the Fc gamma-IIA receptor. Biochem J. 1995;311(pt 2):471–478.
    View this article via: PubMed Google Scholar
32. Rhee SG, Suh PG, Ryu SH, Lee SY. Studies of inositol phospholipid-specific phospholipase C. Science. 1989;244(4904):546–550.
    View this article via: PubMed CrossRef Google Scholar
33. Banno Y, Yada Y, Nozawa Y. Purification and characterization of membrane-bound phospholipase C specific for phosphoinositides from human platelets. J Biol Chem. 1988;263(23):11459–11465.
    View this article via: PubMed Google Scholar
34. Caccese D, et al. Superoxide anion and hydroxyl radical release by collagen-induced platelet aggregation--role of arachidonic acid metabolism. Thromb Haemost. 2000;83(3):485–490.
    View this article via: PubMed Google Scholar
35. Leo R, et al. Platelet activation by superoxide anion and hydroxyl radicals intrinsically generated by platelets that had undergone anoxia and then reoxygenated. Circulation. 1997;95(4):885–891.
    View this article via: PubMed Google Scholar
36. Wachowicz B, Olas B, Zbikowska HM, Buczynski A. Generation of reactive oxygen species in blood platelets. Platelets. 2002;13(3):175–182.
    View this article via: PubMed CrossRef Google Scholar
37. Pignatelli P, Pulcinelli FM, Lenti L, Gazzaniga PP, Violi F. Vitamin E inhibits collagen-induced platelet activation by blunting hydrogen peroxide. Arterioscler Thromb Vasc Biol. 1999;19(10):2542–2547.
    View this article via: PubMed CrossRef Google Scholar
38. Wilson SJ, Cavanagh CC, Lesher AM, Frey AJ, Russell SE, Smyth EM. Activation-dependent stabilization of the human thromboxane receptor: role of reactive oxygen species. J Lipid Res. 2009;50(6):1047–1056.
    View this article via: PubMed CrossRef Google Scholar
39. Ball SK, Field MC, Tippins JR. Regulation of thromboxane receptor signaling at multiple levels by oxidative stress-induced stabilization, relocation and enhanced responsiveness. PLoS One. 2010;5(9):e12798.
    View this article via: PubMed Google Scholar
40. Modesti PA, Fortini A, Gensini GF, Vanni D, Prisco D, Abbate R. Human prostacyclin platelet receptors in diabetes mellitus. Thromb Res. 1991;63(5):541–548.
    View this article via: PubMed CrossRef Google Scholar
41. Ramirez MA, Borja NL. Epalrestat: an aldose reductase inhibitor for the treatment of diabetic neuropathy. Pharmacotherapy. 2008;28(5):646–655.
    View this article via: PubMed CrossRef Google Scholar
42. Hotta N, et al. Long-term clinical effects of epalrestat, an aldose reductase inhibitor, on diabetic peripheral neuropathy: the 3-year, multicenter, comparative Aldose Reductase Inhibitor-Diabetes Complications Trial. Diabetes Care. 2006;29(7):1538–1544.
    View this article via: PubMed CrossRef Google Scholar
43. Santilli F, et al. Postprandial hyperglycemia is a determinant of platelet activation in early type 2 diabetes mellitus. J Thromb Haemost. 2010;8(4):828–837.
    View this article via: PubMed CrossRef Google Scholar
44. Kajita K, et al. Increased platelet aggregation in diabetic patients with microangiopathy despite good glycemic control. Platelets. 2001;12(6):343–351.
    View this article via: PubMed CrossRef Google Scholar
45. Alessandrini P, McRae J, Feman S, FitzGerald GA. Thromboxane biosynthesis and platelet function in type I diabetes mellitus. N Engl J Med. 1988;319(4):208–212.
    View this article via: PubMed Google Scholar
46. Saxena SP, Brandes LJ, Becker AB, Simons KJ, LaBella FS, Gerrard JM. Histamine is an intracellular messenger mediating platelet aggregation. Science. 1989;243(4898):1596–1599.
    View this article via: PubMed Google Scholar
47. Efendic S, Luft R, Wajngot A. Aspects of the pathogenesis of type 2 diabetes. Endocr Rev. 1984;5(3):395–410.
    View this article via: PubMed CrossRef Google Scholar
48. Theoret JF, Bienvenu JG, Kumar A, Merhi Y. P-selectin antagonism with recombinant p-selectin glycoprotein ligand-1 (rPSGL-Ig) inhibits circulating activated platelet binding to neutrophils induced by damaged arterial surfaces. J Pharmacol Exp Ther. 2001;298(2):658–664.
    View this article via: PubMed Google Scholar
49. Shinmura K, et al. Aldose reductase is an obligatory mediator of the late phase of ischemic preconditioning. Circ Res. 2002;91(3):240–246.
    View this article via: PubMed CrossRef Google Scholar
50. Tsakiris DA, Scudder L, Hodivala-Dilke K, Hynes RO, Coller BS. Hemostasis in the mouse (Mus musculus): a review. Thromb Haemost. 1999;81(2):177–188.
    View this article via: PubMed Google Scholar
51. Nunn B. Some characteristics of mouse platelet aggregation and a comparison of the activity of a range of compounds in mouse and human platelet-rich plasma in vitro. Thromb Haemost. 1981;45(1):1–5.
    View this article via: PubMed Google Scholar
52. Ciabattoni G, Pugliese F, Davi G, Pierucci A, Simonetti BM, Patrono C. Fractional conversion of thromboxane B2 to urinary 11-dehydrothromboxane B2 in man. Biochim Biophys Acta. 1989;992(1):66–70.
    View this article via: PubMed Google Scholar
53. Gluckman TJ, et al. Effects of aspirin responsiveness and platelet reactivity on early vein graft thrombosis after coronary artery bypass graft surgery. J Am Coll Cardiol. 2011;57(9):1069–1077.
    View this article via: PubMed CrossRef Google Scholar
54. Gluckman TJ, et al. Effect of anti-platelet factor-4/heparin antibody induction on early saphenous vein graft occlusion after coronary artery bypass surgery. J Thromb Haemost. 2009;7(9):1457–1464.
    View this article via: PubMed CrossRef Google Scholar