Elevation of Endothelial Microparticles, Platelets, and Leukocyte Activation in Patients With Venous Thromboembolism

Transcription

1 Journal of the American College of Cardiology Vol. 45, No. 9, by the American College of Cardiology Foundation ISSN /05/$30.00 Published by Elsevier Inc. doi: /j.jacc Elevation of Endothelial Microparticles, Platelets, and Leukocyte Activation in Patients With Venous Thromboembolism Julio A. Chirinos, MD,* Gustavo A. Heresi, MD,* Hermes Velasquez, MD,* Wenche Jy, PHD,* Joaquin J. Jimenez, MD,* Eugene Ahn, MD,* Lawrence L. Horstman, BS,* Andres O. Soriano, MD,* Juan P. Zambrano, MD,* Yeon S. Ahn, MD* Miami, Florida OBJECTIVES BACKGROUND METHODS The purpose of this research was to determine the levels of platelet, leukocyte, and endothelial activation and markers of cellular interactions in patients with venous thromboembolism (VTE). The details of interactions between endothelium, platelets, and leukocytes in VTE are not well understood. We studied 25 patients with VTE and compared 25 healthy controls. We used flow cytometry to measure: 1) endothelial microparticles (EMP) identified by CD31 /CD42b (EMP 31 ) or E-selectin (EMP 62E ); 2) platelet microparticles (CD31 /CD42b ); 3) surface expression of P-selectin in platelets and CD11b in leukocytes; 4) EMP-monocyte conjugates (percentage of monocytes positive for E-selectin); and 5) platelet-leukocyte conjugates (PLC) expressed as percentage of leukocytes positive for CD41. RESULTS Patients with VTE had marked elevations of EMP 31 (2,193 vs. 383 counts/ l; p 0.003), EMP 62E (368 vs. 223 counts/ l; p 0.001), and EMP-monocyte conjugates (3.3% vs. 2.5%; p 0.002), as well as increased activation of platelets (35.2 vs. 5.0 fluorescence intensity units for P-selectin; p ) and leukocytes (13.9 vs. 7.7 U for CD11b; p 0.004). Also elevated in VTE were PLC (61.7% vs. 39.6%; p 0.01). Expression of CD11b in leukocytes strongly correlated with PLC (r 0.74; p ). CONCLUSIONS Marked activation of endothelium, platelets, and leukocytes occurs in VTE, and VTE, or the accompanying inflammatory process, involves the release of EMP and formation of EMP-monocyte conjugates and PLC. These findings support prior studies suggesting that release of EMP and their binding to monocytes are key events in thrombogenesis. Our findings also support the concept that the formation of PLC regulates leukocyte activation and participates in linking thrombosis with inflammation. (J Am Coll Cardiol 2005;45: ) 2005 by the American College of Cardiology Foundation The pathophysiology of venous thromboembolism (VTE) involves endothelial damage, blood stasis, and hypercoagulability (1). Significant advances have been achieved in the understanding of humoral factors of hypercoagulability. However, the details of interactions between endothelium, platelets, and leukocytes in VTE have been less well studied. See page 1472 Recently, flow-cytometric analysis of microparticles from platelets and from the endothelium has emerged as a powerful method for assessing the status of endothelial cells (EC), platelets, leukocytes, and their interactions (2 4). Endothelial microparticles (EMP) are small ( 1.5 m) vesicular fragments of the EC membrane released during EC activation or apoptosis (2,3). Similarly, platelet microparticles (PMP) are released by platelets upon activation (4). Both endothelial (5,6) and platelet (7) microparticles can From the *Department of Medicine and Wallace H. Coulter Platelet Laboratory, University of Miami, Miami, Florida. This work was supported by the Wallace H. Coulter Foundation and funds from Mary Beth Weiss, Jane and Charles Bosco, and gifts from Frank Smather. Manuscript received September 9, 2004; revised manuscript received November 11, 2004, accepted December 20, interact with leukocytes. By employing strategic combinations of fluorescent-labeled antibodies, it is possible to detect EMP-leukocyte and platelet-leukocyte conjugates (2,4). In addition, leukocyte expression of activation marker CD11b and platelet expression of activation marker CD62P can be readily measured by flow cytometry. In this study, we aimed to determine whether these markers of activation of EC, platelets, and leukocytes and their interactions are increased in patients with VTE. METHODS Study subjects. A convenience sample of 25 patients with deep vein thrombosis (DVT) and/or pulmonary embolism (PE) was studied and compared to 25 healthy controls studied in the same period of time. Patients with a clinical suspicion of VTE were prospectively enrolled at Jackson Memorial Hospital/University of Miami. The research protocol was approved by our institutional review board, and informed consent was obtained from all patients. Important demographic, clinical, and laboratory characteristics were prospectively recorded upon enrollment. Patients were followed, and results of diagnostic tests were recorded. We selected patients with DVT confirmed by

2 1468 Chirinos et al. JACC Vol. 45, No. 9, 2005 Cell Interactions in Vein Thrombosis May 3, 2005: Abbreviations and Acronyms DVT deep vein thrombosis EC endothelial cell EMP endothelial microparticles IQR interquartile range mab monoclonal antibodies PE pulmonary embolism PMP platelet microparticles TF tissue factor VTE venous thromboembolism duplex studies of the lower extremities; visualization of a clot by duplex ultrasonography and lack of compressibility of a vein by the ultrasound probe were considered indicative of DVT. Pulmonary embolism was diagnosed by ventilation/perfusion scanning, spiral computed tomographic angiography of the chest, and/or pulmonary artery angiography. Findings on ventilation/perfusion scanning were interpreted according to the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) criteria, and only scans with a high probability of PE were considered as indicative of PE. Criteria for a positive spiral computed tomography scan included a partial filling defect (defined as central or marginal intraluminal areas of low attenuation surrounded by variable amounts of contrast medium with regular or irregular borders), a complete filling defect, the railway track sign (thromboembolic masses seen floating freely in the lumen, allowing the flow of blood between the wall of the vessel and the thrombus/embolus), and mural defects (found in peripheral areas of low attenuation with arterial sections). The following findings in pulmonary angiography were established as diagnostic of PE: a constant intraluminal filling defect or a persistent acute cutoff sign of an arterial pulmonary branch seen on more than one projection. We excluded patients with conditions known or suspected to independently increase levels of EMP or PMP; specifically, patients with sepsis, acute infection, pregnancy, acute coronary syndromes, heparin-induced thrombocytopenia, antiphospholipid syndrome, thrombotic thrombocytopenic purpura, transient ischemic attacks, severe hypertension ( 160/95 mm Hg), recent cardiopulmonary bypass, or multiple sclerosis were excluded. Laboratory methods. BLOOD SAMPLING AND HAN- DLING. Blood samples consisted of two citrate Vacutainers (light-blue top; Becton-Dickinson, Franklin Lakes, New Jersey) from a peripheral vein upon enrollment. Samples were processed for assay within 4 h. One was saved for whole-blood tests, and the other was centrifuged 10 min at 160 g to prepare platelet-rich plasma; then the plateletrich plasma was centrifuged for 8 min at 1,000 g to prepare platelet-poor plasma. FREE PMP AND EMP. The method of EMP assay was previously described (8,9); PMP and EMP were measured simultaneously in platelet-poor plasma by the pair of fluorescent monoclonal antibodies (mab), PE-labeled anti- CD31, and FITC-labeled anti-cd42b, both from BD/ Pharmingen (Franklin Lakes, New Jersey). Events were counted by triggering on the red fluorescence signal of PE, above background noise, on the y-axis of the dot-plot, while the green signal of CD42b was on the x-axis (four-decade log scales, both x and y). Because CD31 occurs on both PMP and EMP but CD42b occurs only on platelets, PMP were defined by events CD31 /CD42 and EMP by events CD31 /CD42. Values are reported as counts/ l; CD31 occurs also on some leukocyte subsets, but we observed (using PE/Cy5-labeled CD45) that leukocyte microparticles accounted for a negligible fraction of CD31 /CD42 microparticles, in both normal and thrombotic patients. To further validate this procedure, known amounts of pure PMP and EMP prepared in vitro were mixed in various proportions, and then measured by this method; results were essentially the same as measuring the pure components individually. ACTIVATION STATUS OF PLATELETS AND LEUKOCYTES. Total expression of P-selectin (CD62P) was measured by FITCanti-CD62P mab (Pharmingen) in dilute whole blood as the sum of three bit-map populations: free platelets, microaggregated platelets, and platelet-leukocyte complexes (10). The population of each was multiplied by the mean fluorescent intensity (arbitrary units) of that population, then summed, then divided by the total number of events in the three bit-maps, yielding the mean fluorescent intensity of P-selectin per event. The rationality for this is that activated platelets avidly bind to each other and to some leukocytes; therefore, measuring free platelets alone does not reflect total platelet activation status. Leukocyte activation was measured by their expression of CD11b. MEASUREMENT OF LEUKOCYTES CONJUGATED WITH PMP OR EMP. The method for measuring platelet-leukocyte conjugates has been described elsewhere (7) and is based on the co-expression of fluorescent pan-leukocyte marker CD45 (PE label) with platelet marker CD41 (FITC label); neutrophils were selected (gated) based on forward- and sidescatter; EMP-monocyte conjugate assay was based on expression of mab markers of ECs, CD62E, appearing on leukocytes (coincident with CD45); monocytes, neutrophils, and lymphocytes were selected based on forward- and side-scatter (5). Statistical analysis. Given the non-normal distribution of the markers measured, values are expressed as median and interquartile range (IQR) throughout the article and were compared between the two groups using the Mann- Whitney U test. The linear relationship between continuous variables was analyzed with linear regression, and the corresponding Pearson correlation coefficient (r) and r squared (r 2 ) values were calculated. Logarithmic transformation of continuous variables (CD11b and plateletleukocyte conjugates) was undertaken to improve normality before entering regression models. All reported probability

3 JACC Vol. 45, No. 9, 2005 May 3, 2005: Chirinos et al. Cell Interactions in Vein Thrombosis 1469 values are two-tailed. Statistical significance was defined as a p value Analyses were performed with the statistical package NCSS for Windows (Kaysville, Utah). RESULTS Table 1. Baseline Characteristics of Patients With VTE Characteristic VTE (n 25) Age Male gender 12 (48%) Prior episodes of VTE 4 (16%) PE 11 (44%) Isolated DVT 14 (56%) Malignancy 9 (36%) Recent surgery (within 30 days) 12 (48%) White cell count ( 10 3 /mm 3, SD) Diabetes mellitus 6 (24%) Hypertension 7 (28%) Platelet count ( 10 3 /mm 3, SD) Hematocrit (%, SD) Antiplatelet drug use 5 (20%) DVT deep vein thrombosis; PE pulmonary embolism; VTE venous thromboembolism. Figure 1. (A to C) Levels of EMP31 (A), EMP62E (B), and EMPmonocyte conjugates in subjects with venous thromboembolism (VTE) compared with normal controls. EMP endothelial microparticles; FL fluorescence. Baseline characteristics of our population of patients with VTE are shown in Table 1. The mean age was 54 years; PE was present in 44% of patients, whereas DVT without a suspicion or diagnosis of PE was present in the remaining 56%. About one-half of the patients had recent surgery as a precipitating factor for VTE, and one-third had an underlying malignancy. EMP and EMP-monocyte conjugate levels (Figs. 1A to 1C). Patients with VTE had markedly increased levels of EMP 31 (2,193 counts/ l; IQR 1,000 to 6,222) compared to normal controls (383 counts/ l; IQR 243 to 2,403; p 0.003). Similarly, patients with VTE had higher levels of EMP 62E (368 counts/ l; IQR 253 to 888) than normal controls (223 counts/ l; IQR 84 to 315; p 0.001); EMP-monocyte conjugates were also significantly higher in patients with VTE (3.3 fluorescence intensity units [FLIU]; IQR 2.96 to 3.91) than in controls (2.5 FLIU; IQR 2.13 to 2.83; p 0.002). PMP, platelet P-selectin, leukocyte CD11b, and plateletleukocyte conjugates (Figs. 2A to 2C). Platelet expression of CD62P was markedly increased in patients with VTE (35.2 FLIU; IQR 23.5 to 58.2) compared to normal controls (5.0 FLIU; IQR 1.5 to 21.6; p ) Unexpectedly, levels of PMP were not significantly elevated in patients with VTE (8,212 vs. 5,395 counts/ l; p 0.52); CD11b expression in leukocytes was also higher in patients with VTE (13.9 FLIU; IQR 9.6 to 21.2) than in controls (7.7 FLIU; IQR 4.9 to 10.7; p 0.004). Increased levels of platelet-leukocyte conjugates were also seen in the VTE group (61.7%; IQR 41.4 to 73.2) compared to normal controls (39.6%; IQR 21 to 58; p 0.01). As shown in Figure 3, there was a strong and highly significant correlation between CD11b expression in leukocytes and platelet-leukocyte conjugates (r 0.74; r ; p ). Finally, to rule out the possibility that the differences observed were driven by the postoperative state present in some of our patients, we repeated all the comparisons excluding patients who had undergone surgery in the preceding 30 days (n 12). All differences were confirmed in these analyses. Probability values for these comparisons were as follows: EMP 31 (p 0.01), EMP 62E (p 0.004), EMP-monocyte conjugates (p 0.004), P-selectin expression in platelets (p ), platelet-leukocyte conjugates (p 0.01), and CD11b expression in leukocytes (p 0.005). No significant differences were found when these markers were compared between patients with VTE who had recent surgery versus those who did not. DISCUSSION This study demonstrates that marked activation of the endothelium, platelets, and leukocytes occurs in VTE, and that venous thrombosis, or the accompanying inflammatory process, involves release of circulating EMP and formation of EMP-leukocyte and platelet-leukocyte conjugates. The presence of elevated EMP in this study reflects increased endothelial activation in VTE, consistent with prior studies (11,12). Under physiological conditions, the

4 1470 Chirinos et al. JACC Vol. 45, No. 9, 2005 Cell Interactions in Vein Thrombosis May 3, 2005: Figure 3. Correlation on CD11b expression in leukocytes and levels of platelet-leukocyte conjugates in subjects with venous thromboembolism. Figure 2. (A to C) Surface expression of P-selectin (CD62P) in platelets (A), CD11b in leukocytes (B), and levels of platelet-leukocyte conjugates in subjects with venous thromboembolism (VTE) compared with healthy controls. FL fluorescence. vascular endothelium plays a potent antithrombotic role, but upon endothelial activation is shifted toward a prothrombotic state (13); EC activation is associated with EMP release (2). Additionally, activated EC express tissue factor (TF), the main trigger of thrombin generation. Endothelial microparticles express many receptors of the parent EC and can support thrombin generation by way of the TF/factor VIIa pathway (14). Coagulation is supported by the presence of phosphatidylserine (15), which is translocated from the inner to the outer leaflet of the cell membrane during EMP formation (6). Hence, EMP provides a source of TF as well as a catalytic surface for assembly of the prothrombinase complex (16). Taken collectively, these considerations suggest that EMP are not only a marker of endothelial activation in VTE, but play an active role in the thrombotic process. They may also link thrombosis to inflammation insofar as EMP, like PMP, have been shown to function as vectors for many inflammatory mediators (17). Further research is needed to define whether the circulating concentrations of EMP in VTE are sufficient to facilitate thrombus formation and mediate inflammation. The increased levels of platelet activation marker P-selectin (CD62P) (Fig. 2) are consistent with previous reports (18 20). Previous studies measured soluble P-selectin, as opposed to P-selectin expression on platelets as we did. However, the majority of so-called soluble P-selectin is in reality largely bound to microparticles (2). Although venous thrombosis has been traditionally associated with red blood cells and fibrin-rich red clot, platelets are key in the pathophysiology of VTE (21). Our study demonstrates increased formation of platelet-leukocyte conjugates in VTE. Furthermore, we found a strong correlation between CD11b expression in leukocytes and plateletleukocyte conjugates in these patients (Fig. 3). This finding is consistent with in vitro studies showing that platelet adhesion to leukocytes upregulates Mac-1 (CD11b/Cd18), which itself supports interactions with platelets (7,22), and support the concept that the formation of platelet-leukocyte conjugates regulate leukocyte activation and participate in linking thrombosis with inflammation in vivo. Activation of human platelets by diverse stimuli results in shedding of microparticles (4,23). In vitro studies have suggested that microparticles shed by aggregating platelets may, in turn, cause platelet activation and EC activation through the transcellular delivery of arachidonic acid (23)or other mediators (17), and that PMP provide a catalytic phospholipid surface for assembly of the prothrombinase complex (24,25) as discussed above. In addition, we have shown that PMP bind and activate neutrophils in vitro (7). The finding of normal circulating levels of PMP in patients with VTE in our study is intriguing. This finding is not necessarily inconsistent with the above cited studies, because platelet activation in VTE may result in PMP formation followed by their rapid binding to leukocytes. Finally, our study demonstrates markedly increased leukocyte activation (Fig. 2B) and binding of EMP to monocytes in patients with VTE (Fig. 1C). Platelets, leukocytes, and ECs co-localize and interact in the milieu of a forming thrombus (26). The release of EMP by EC has the potential to amplify coagulation events through the induction of TF (6). The activation of leukocytes and their adhesive interaction with EC are key events in vein wall inflammation and thrombus organization.

5 JACC Vol. 45, No. 9, 2005 May 3, 2005: Chirinos et al. Cell Interactions in Vein Thrombosis 1471 Conclusions. Our study has demonstrated that activation of the endothelium, platelets, and leukocytes occurs in patients with VTE, and that venous thrombosis (or the accompanying inflammatory process) involves release of EMP and formation of EMP-leukocyte and plateletleukocyte conjugates. To the best of our knowledge, this is the first study to demonstrate elevated circulating EMP, platelet-leukocyte conjugates, and EMP-leukocyte conjugates in VTE. Microparticles released from EC and their interactions with leukocytes are likely to play a central role on clot formation, but further research is needed to clarify this role for the different species studied here (e.g., EMP 31, EMP 62E ) of EMP and their interactions with leukocytes in VTE and whether the formation of EMP represents a therapeutic target to prevent and limit thrombus formation. Reprint requests and correspondence: Dr. Julio A. Chirinos, 7601 East Treasure Drive, 709, Miami, Florida jchirinos@med.miami.edu. REFERENCES 1. Turpie AG, Chin BS, Lip GY. Venous thromboembolism: pathophysiology, clinical features, and prevention. BMJ 2002;325: Horstman LL, Jy W, Jimenez JJ, Ahn YS. Endothelial microparticles as markers of endothelial dysfunction. Front Biosci 2004;9: Jimenez JJ, Jy W, Mauro L, Soderland C, Horstman LL, Ahn YS. Endothelial cells release phenotypically and quantitatively distinct microparticles in activation and apoptosis. Thromb Res 2003;109: Horstman LL, Ahn YS. Platelet microparticles: a wide-angle perspective. Crit Rev Oncol Hematol 1999;30: Jy W, Jimenez JJ, Mauro L, et al. Interaction of endothelial microparticles (EMP) with leukocytes: potential roles of EMP in thrombosis and inflammation (abstr). Blood 2001;98:226a. 6. Sabatier F, Roux V, Anfosso F, Camoin L, Sampol J, Dignat-George F. Interaction of endothelial microparticles with monocytic cells in vitro induces tissue factor-dependent procoagulant activity. Blood 2002;99: Jy W, Mao WW, Horstman LL, Tao J, Ahn YS. Platelet microparticles bind, activate and aggregate neutrophils in vitro. Blood Cells Mol Dis 1995;21: Jy W, Horstman LL, Arce M, Ahn YS. Clinical significance of platelet microparticles in autoimmune thrombocytopenias. J Lab Clin Med 1992;119: Jimenez J, Jy W, Mauro L, Horstman L, Ahn Y. Elevated endothelial microparticles in thrombotic thrombocytopenic purpura (TTP): findings from brain and renal microvascular cell culture and patients with active disease. Br J Haematol 2001;112: Jy W, Mao WW, Horstman LL, Valant PA, Ahn YS. A flow cytometric assay of platelet activation marker P-selectin (CD62p) distinguishes heparin-induced thrombocytopenia (HIT) from HITwith-thrombosis (HITT). Thromb Haemost 1999;82: Smith A, Quarmby JW, Collins M, Lockhart SM, Burnand KG. Changes in the levels of soluble adhesion molecules and coagulation factors in patients with deep vein thrombosis. Thromb Haemost 1999;82: Musolino C, Alonci A, Allegra A, et al. Increased levels of the soluble adhesion molecule E-selectin in patients with chronic myeloproliferative disorders and thromboembolic complications. Am J Hematol 1998;57: Bombeli T, Mueller M, Haeberli A. Anticoagulant properties of the vascular endothelium. Thromb Haemost 1997;77: Combes V, Simon AC, Grau GE, et al. In vitro generation of endothelial microparticles and possible prothrombotic activity in patients with lupus anticoagulant. J Clin Invest 1999;104: Pei G, Powers DD, Lentz BR. Specific contribution of different phospholipid surfaces to the activation of prothrombin by the fully assembled prothrombinase. J Biol Chem 1993;268: Hamilton KK, Hattori R, Esmon CT, Sims PJ. Complement proteins C5b-9 induce vesiculation of the endothelial plasma membrane and expose catalytic surface for assembly of the prothrombinase enzyme complex. J Biol Chem 1990;5: Freyssinet JM. Cellular microparticles: what are they bad or good for? J Thromb Haemost 2003;1: Bozic M, Blinc A, Stegnar M. D-dimer, other markers of haemostasis activation and soluble adhesion molecules in patients with different clinical probabilities of deep vein thrombosis. Thromb Res 2002;108: Myers DD, Hawley AE, Farris DM, et al. P-selectin and leukocyte microparticles are associated with venous thrombogenesis. J Vasc Surg 2003;38: Blann AD, Noteboom WM, Rosendaal FR. Increased soluble P-selectin levels following deep venous thrombosis: cause or effect? Br J Haematol 2000;108: Sobieszczyk P, Fishbein MC, Goldhaber SZ. Acute pulmonary embolism: don t ignore the platelet. Circulation 2002;106: Neumann FJ, Zohlnhofer D, Fakhoury L, et al. Effect of glycoprotein IIb/IIIa receptor blockade on platelet-leukocyte interaction and surface expression of the leukocyte integrin Mac-1 in acute myocardial infarction. J Am Coll Cardiol 1999;34: Barry OP, Pratico D, Lawson JA, Fitzgerald GA. Transcellular activation of platelets and endothelial cells by bioactive lipids in platelet microparticles. J Clin Invest 1997;99: Sims PJ, Wiedmer T, Esmon CT, Weiss HJ, Shattil SJ. Assembly of the platelet prothrombinase complex is linked to vesiculation of the platelet plasma membrane. J Biol Chem 1989;264: Jy W, Horstman LL, Wang F, Duncan R, Ahn YS. Platelet factor 3 in plasma fractions: its relation to microparticle size and thromboses. Thromb Res 1995;80: Lorenzet R, Napoleone E, Celi A, Pellegrini G, Di Santo A. Cell-cell interaction and tissue factor expression. Blood Coagul Fibrinolysis 1998;9 Suppl 1:S49 59.