- Email: [email protected]
Increased levels of platelet-derived microparticles in patients with diabetic retinopathy
Diabetes Research and Clinical Practice 68 (2005) 193–201 www.elsevier.com/locate/diabres
Increased levels of platelet-derived microparticles in patients with diabetic retinopathy Nahoko Ogataa,*, Masahito Imaizumia, Shosaku Nomurab, Akira Shozuc, Miwa Arichia, Masato Matsuokaa, Miyo Matsumuraa a
Department of Ophthalmology, Kansai Medical University, Fumizono-cho 10-15, Moriguchi, Osaka 570-8507, Japan b First Department of Internal Medicine, Kansai Medical University, Moriguchi, Osaka, Japan c Third Department of Internal Medicine, Kansai Medical University, Moriguchi, Osaka, Japan Received 10 March 2004; received in revised form 24 May 2004; accepted 13 October 2004 Available online 10 December 2004
Abstract Diabetic retinopathy is caused by capillary occlusions. Platelet-derived microparticles (PMPs) stimulate the coagulation cascade and increase leukocyte and endothelial cell adhesions, both of which are key events in the development of diabetic retinopathy. However, the correlation between the levels of PMPs and diabetic retinopathy has not been precisely determined. The PMPs levels and the expression of platelet CD62P and CD63 were measured in 92 diabetic patients. The level of PMPs was significantly correlated with the expression of CD62P (r = 0.76, P < 0.0001) and CD63 (r = 0.71, P < 0.0001). The mean level of PMPs in diabetics (507 15/104 platelets (plt), mean S.E.) was significantly higher than that in normal. The PMPs levels increased with the progression of the diabetic retinopathy; 480 28/104 plt in diabetic patients without retinopathy (n = 25), 504 40/104 plt with mild or moderate non-proliferative diabetic retinopathy (n = 13), 512 29/104 plt with severe nonproliferative diabetic retinopathy (n = 25), and 528 25/104 plt with proliferative diabetic retinopathy (n = 29). The PMPs level in patients with non-perfused retinal areas (582 27/104 plt, n = 24) was significantly higher than patients without non-perfused areas (469 23/104 plt, n = 30; P = 0.0096) and without diabetic retinopathy (P = 0.024). These high correlations indicate that increased levels of PMPs may accelerate diabetic retinopathy. # 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Platelet-derived microparticles (PMPs); Diabetic retinopathy; Platelet; Coagulation; Diabetes
1. Introduction The number of patients with diabetes is increasing dramatically, and diabetic retinopathy resulting in * Corresponding author. Tel.: +81 6 6992 1001; fax: +81 6 6993 2222. E-mail address: [email protected] (N. Ogata).
retinal ischemia [1–3] is a major cause of adult blindness. To prevent the progression of diabetic retinopathy to proliferative diabetic retinopathy (PDR), retinal photocoagulation is performed. Although photocoagulation is the only effective treatment to reduce the retinal ischemia, this treatment destroys retinal cells including photoreceptor cells and can thereby induce severe vision [4]. Therefore, a
0168-8227/$ – see front matter # 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2004.10.010
194
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
search for new and better non-invasive therapy for diabetic retinopathy is going on. Diabetic retinopathy is characterized at its early stage by areas of microvascular damage and capillary non-perfusion. Platelets play an important role in this vascular occlusive process [1,5–7], and patients with diabetes develop hypercoagulability and hyperaggreability of platelets [5,7]. Various platelet abnormalities appear to be involved in the genesis and/or the evolution of diabetic microangiopathy [5,7], as suggested by the presence of platelet aggregates and thrombi in the small vessels of the retina and the kidneys of diabetic patients [6]. Pro-coagulant activity is expressed as markers of activated platelet, e.g., platelet CD62P (P-selectin), CD63, and plateletderived microparticles (PMPs) [8,9]. PMPs were first observed as vesicles released from activated platelets following their adhesion to vessel walls [10]. PMPs contain a platelet-activating factor (PAF), and express high-affinity receptors for factor VIII and platelet/endothelium attachment receptors such as P-selectin [11,12]. In addition, PMPs increase the level of intracellular adhesion molecule (ICAM)-1, stimulate cytokine secretion, and increase tissue factor expression in endothelial cells [13]. P-selectin is believed to be involved in the initial leukocyte– endothelium interactions [14–16], and ICAM-1 plays an important role in the adhesion and transmigration of leukocytes [17,18]. Thus, PMPs activate the coagulation cascade [19], and increase leukocyte and endothelial cell adhesions [12,20], both of which are key events in the development of capillary occlusion in diabetic patients [21–24]. Therefore, the release of PMPs has been investigated and determined to be evidence of cellular activation occurring in diseases associated with diabetes and thrombotic risk [12,21–29]. However, information on the level of PMPs in patients with diabetic retinopathy is limited [21] and the data have not been well determined. The purpose of this study was to determine whether the microangiopathic changes in the retina arise from the increased level of PMPs. To accomplish this, we have measured the level of PMPs and activated platelets by determining the expression of CD62P (platelet P-selectin) and CD63. Understanding the role of PMPs in retinal ischemia may provide clues for non-invasive therapies for diabetic retinopathy.
2. Patients and methods 2.1. Patients Ninety-two type 2 diabetic patients (37 men and 55 women, ages 31–83 years, average 60.5 years) were studied. All patients who were diagnosed as having type 2 diabetes and agreed to have additional examinations were investigated. An informed consent was obtained from all patients after an explanation of the purpose and procedures of the study, and the procedures were performed to conform to the tenets of the declaration of Helsinki. Patients with severe nephropathy, under treatment by hemodialysis, or being treated with anti-thrombotic or anti-platelets agents were excluded. 2.2. Examination of diabetic retinopathy The stage of diabetic retinopathy was determined by ophthalmoscopy and fluorescein angiography prior to the examination of the blood samples, and the patients were classified according to the severity scale of diabetic retinopathy [30]. The 92 patients included 25 with no apparent diabetic retinopathy (NDR), 13 with mild or moderate non-proliferative diabetic retinopathy (mild or moderate NPDR), 25 with severe non-proliferative diabetic retinopathy (severe NPDR), and 29 with proliferative diabetic retinopathy (PDR). The 54 patients with severe NPDR and with PDR were subdivided into two groups; those with and those without non-perfused areas as determined on fluorescein angiograms. Diabetic eyes were classified as not having non-perfused areas if the retinopathy had been treated with retinal photocoagulation, and the previously documented non-perfused areas had regressed fully and were not present when the eyes were evaluated for this study. The eyes were considered to have non-perfused areas if they were not treated or even if the retinopathy had been treated with retinal photocoagulation and non-perfused areas were still present. Using these criteria, 30 patients (16 with severe NPDR and 14 with PDR) were classified as not having non-perfused areas, and 24 patients (9 with severe NPDR and 15 with PDR) were classified as having non-perfused areas. The demographics of the patients with diabetic retinopathy are shown in Table 1.
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
195
Table 1 CD62P, CD63, and PMPs in diabetic retinopathy Diabetic retinopathy
Number of patients
CD62P(%) (mean S.E.) normal level <10%
CD63 (%) (mean S.E.) normal level <10%
PMPs/104 plt (mean S.E.) normal level = 263 9
NDR MNPDR SNPDR PDR Total
25 13 25 29 92
19.4 2.0 22.6 1.1 20.7 1.3 23.8 1.0 21.0 0.7
18.7 1.2 22.3 1.3 20.2 1.3 22.8 1.2 20.4 0.6
480 28 504 40 512 29 527 25 507 15
NDR, no apparent retinopathy; MNPDR, mild or moderate non-proliferative diabetic retinopathy; SNPDR, severe non-proliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy.
2.3. Methods 2.3.1. HbA1c The level of HbA1c was determined in all patients by standard laboratory procedures.
2.3.3. Total cholesterol The concentrations of total cholesterol were determined in all patients by standard laboratory procedures. 2.4. Statistical analyses
2.3.2. Analysis of activated platelets and plateletderived micropartities The concentration of activated platelets and PMPs in the blood was determined by flow cytometry as described [20–22]. Briefly, blood samples were collected in tubes containing 3.8% sodium citrate, and washed platelets were prepared by centrifuging platelet-rich plasma. The washed platelets were fixed with 2% paraformaldehyde, rinsed, and re-suspended in stock solution. The platelets were then incubated with a FITC-labeled monoclonal antibody against platelets GPIX, and analyzed with an ortho cytoron absolute analyzer (Ortho Diagnostic Systems, Tokyo, Japan) to measure the level of PMPs. Only cells and particles positive for GPIX were gated to separate the platelets and PMPs from electric noise. To differentiate between platelets and PMPs, the lower limit of the platelet gate was set at the limit of the forwardscatter profile of resting platelets. The number of PMPs released from 104 platelets (plt) was used as a measure of the PMP concentration. As an index of platelet activation, the expressions of CD62P and CD63 were quantified. The washed platelets were fixed with 2% paraformaldehyde, incubated with anti-CD62P antibody (CLB-thromb/ 6, Immunoteck Marseile, France) and anti-CD63 antibody (CLB-gran/12, Immunoteck Marseile, France). They were then analyzed by flow cytometry as previously described [8]. The platelets positive for CD62P and CD63 were expressed as percentages.
The results were analyzed using a one-way ANOVA and Fisher test. The correlation between PMPs and the different factors was examined by Pearson’s product moment coefficient of correlation (r). The Spearman’s rank correlation coefficient was also calculated. P-values less than 0.05 were accepted as significant.
3. Results 3.1. Correlations between PMPs, and CD62P and CD63 The level of PMPs was strongly correlated with the level of CD62P (r = 0.76, P < 0.0001, Pearson’s product moment coefficient of correlation, Fig. 1A). The Spearman’s rank correlation coefficient was also significant (P < 0.0001). The level of PMPs was significantly higher in patients with higher concentration of CD63 (r = 0.71 and P < 0.0001, Fig. 1B). These results demonstrated that the platelets were indeed activated, and in all likelihood, accounted for the higher levels of PMPs. 3.2. Correlation between PMPs and HbA1c There was a significantly higher level of PMPs in patients with higher levels of HbA1c. Calculations
196
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
Fig. 1. (A) Correlation between the level of PMPs and CD62P. (B) Correlation between the level of PMPs and CD63. The levels of CD62P and CD63 increase as the level of PMPs increases. The coefficient of correlation (r) between PMP and CD62P is r = 0.76 (P < 0.0001), and CD63 is r = 0.71 (P < 0.0001).
showed that the correlation between PMPs and HbA1c was significant (r = 0.28, P = 0.011, Fig. 2). 3.3. Correlation between PMPs and the stage of diabetic retinopathy The concentration of PMPs in all patients with diabetes was 507 15/104 plts (mean standard error of the mean, n = 92). The level of PMPs in diabetic patients was significantly higher than that in non-diabetic control subjects (263 9/104 plts, based on our previous data [21]). The level of PMPs
increased as the stage of diabetic retinopathy advanced; 480 28/104 plt in patients with NDR (n = 25), 504 40/104 plt with mild or moderate NPDR (n = 13), 512 29/104 plt with severe NPDR (n = 25), and 527 25/104 plt with PDR (n = 29). However, the differences in the PMPs levels at the different stages of diabetic retinopathy were not significant (Table 1, Fig. 3A). 3.4. PMPs levels in patients with non-perfused areas The mean level of PMPs in patients with nonperfused area was 582 27/104 plts (9 with severe NPDR and 15 with PDR), which was significantly higher than 469 23/104plt in patients without nonperfused areas (16 with severe NPDR and 14 with PDR, P = 0.0096), and also in patients with NDR (P = 0.024, Fig. 3B). 3.5. Correlation between the stage of diabetic retinopathy, and CD62P and CD63
Fig. 2. Correlation between PMPs and HbA1c. The level of HbA1c increases as the level of PMPs increases (r = 0.28, P = 0.011).
The mean level of CD62P in patients with diabetes was 21.0 0.7%, which was higher than that in nondiabetic control subjects (normal level < 10% [21]). The mean level of CD62P was 19.4 2.0% in patients with NDR (n = 25), 22.6 1.1% with mild or moderate NPDR (n = 13), 20.7 1.3% with severe
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
197
Fig. 3. The levels of PMPs in patients with diabetic retinopathy. (A) The level of PMPs increases as the stage of diabetic retinopathy gets worse, but the differences in the PMPs levels at the different stages of diabetic retinopathy are not significant. (B) The level of PMPs in patients with nonperfused areas is significantly higher than that in patients without non-perfused areas (P = 0.0096) and patients with NDR (P = 0.024). NDR, no apparent retinopathy; MNPDR, mild or moderate non-proliferative diabetic retinopathy; SNPDR, severe non-proliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy; non-PFA, retinas with non-perfused areas. Bars represent S.E. and the asterisks (*) indicate P < 0.05.
NPDR (n = 25), and 23.8 1.0% with PDR (n = 29, Table 1). All of these levels were significantly higher than that of controls, and the difference between in patients with NDR and PDR was significant (P = 0.029). The mean level of CD62P was 23.6 1.9% (n = 24) in patients with non-perfused areas and 20.7 1.1% (n = 30) in patients without non-perfused areas. This difference was not significant (P = 0.12) but the mean CD62P level in patients with nonperfused areas and the level in patients with NDR was significant (P = 0.039; Table 1, Fig. 4A). The mean level of CD63 in patients with diabetes was 20.4 0.6%, which was higher than that in controls (<10% [21]). The levels of CD63 were 18.7 1.2% in patients with NDR (n = 25), 22.3 1.3% with mild or moderate NPDR (n = 13), 20.2 1.3% with severe NPDR (n = 25), and 22.8 1.2% with PDR (n = 29, Table 1). The differences in the CD63 levels at the different stages of diabetic retinopathy were not significant. However, the mean CD63 level in patients with non-perfused areas (23.9 1.3%, n = 24) was sig-
nificantly higher than that in NDR patients (P = 0.006) and in patients without non-perfused areas (19.7 1.2%, n = 30, P = 0.023, Fig. 4B). 3.6. Correlation between the stage of diabetic retinopathy and total cholesterol The mean level of cholesterol in patients with diabetes was 203.7 5.7 mg/dl which is slightly high but still in the normal range (136–220 mg/dl). The mean level of cholesterol was 203.3 7.5 mg/dl in patients with NDR (n = 25), 204.1 15.4 mg/dl with mild or moderate NPDR (n = 13), 192.9 14.8 mg/dl with severe NPDR (n = 25), and 220.0 17.3 mg/dl with PDR (n = 29). The differences in the cholesterol levels at the different stages of diabetic retinopathy were not significant. The mean level of cholesterol was 195.4 10.8 mg/dl (n = 24) in patients with nonperfused areas and 218.0 15.0 mg/dl (n = 30) in patients without non-perfused areas, and this difference was not significant. In addition there was no correlations between the cholesterol levels and PMPS (r = 0.07).
198
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
Fig. 4. The Levels of CD62P and CD63 in patients with diabetic retinopathy. (A) The mean CD62P level in patients with non-perfused areas is higher than that in patients without non-perfused areas, but the difference is not significant (P = 0.08). (B) The mean CD63 level in patients with non-perfused areas is significantly higher than that in patients without diabetic retinopathy (P = 0.006) and without non-perfused areas (P = 0.023). NDR, no apparent retinopathy; MNPDR, mild or moderate non-proliferative diabetic retinopathy; non-PFA, retinas with nonperfused areas. Bars represent S.E. and the asterisks (*) indicate P < 0.05.
4. Discussion Diabetic retinopathy is caused by capillary occlusions, and capillary occlusions result from microvascular thrombi in which erythrocytes, platelets, and leukocytes play a role [1,31]. The role of platelets in the pathogenesis of diabetic retinopathy has been pointed out [5] and diabetic patients have activated platelets [5,6,31], but the role of activated platelets in diabetic retinopathy has not been determined. Activated platelets express adhesive molecules, e.g., CD62P (platelet P-selectin) and CD63 [9,23,24]. PMPs are also released from activated platelets, and the PMPs contain inner granules, membranous microvesicles, membranous fragments produced by mechanical disruptions, and have coagulative activity [9,12,19]. Earlier, we found that the PMPs levels are correlated with presence of hypertension, nephropathy, atherosclerosis, and hyperlipidemia [12,22,23, 32–34]. There have been reported that PMPs are higher in type 2 diabetic patients [12,21–23], and we have reported on the potential role of PMPs in the
complication of diabetes such as neuropathy, nephropathy, retinopathy, and atherosclerosis [21–24]. However, the relationship between PMPs and diabetic retinopathy has not been clearly assessed. The level of PMPs in our patients with type 2 diabetes was significantly higher than that of control subjects confirming our previous reports [12,21–24]. The higher levels of PMPs most likely resulted from the activation of platelets as shown by its high correlation to CD62P and CD63, both markers of platelet activation. Interestingly, a recent study reported a significant elevation of PMPs in type l diabetes, and no significant increase in type 2 diabetic patients and the age-matched control subjects [25]. This discrepancy in the level of PMPs in type 2 diabetic patients may arise from differences in the patient and control populations (Japanese and Caucasian) and/or from the techniques used to measure the level of the PMPs. The significant correlation between the levels of PMPs and HbA1c in our type 2 diabetic patients suggests that diabetic patients with poor control of glucose levels have higher levels of PMPs.
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
It has been reported that PMPs trigger different types of biological responses after interacting with their target cells. For example, they increase the level of ICAM-1, and ICAM-1 is known to mediate leukocyte adhesion and transmigration. Thus, ICAM-1 may be associated with the circulatory stasis observed in diabetic retinopathy [35–38]. In addition, PMPs may activate endothelial cells and monocytes, or may stimulate cytokine secretion and tissue factor expression in endothelial cells [12,13,24,39]. PMPs express several platelet/endothelial attachment receptors on their surface, such as glycoprotein Ilb/IIIa (aIIb b3), factor VIII, GPIb, and GPIa/IIa, and CD62P (P-selectin) [9,11,12]. P-selectin is believed to be involved in the initial stages of leukocyte–endothelium interactions [14–16]. Thus, these properties suggest that PMPs themselves increase leukocyte–endothelial cell adhesion. As in atherogenesis, increased leukocyte–endothelial cell adhesion is a key early event in the development of capillary occlusion in diabetic retinopathy [31,36–38,40,41]. Leukocytes from diabetic patients have been shown to be more adherent to endothelial cells [42], and in experimental diabetes, they tend to be easily entrapped in retinal capillaries which then leads to the areas of capillary nonperfusion and endothelial cell damage [43]. These properties may also be modulated by PMPs. Significantly higher levels of CD62P and CD63 were found in our diabetic patients, and especially in those with non-perfused areas. In addition, the levels of PMPs increased with the advancement in the stage of diabetic retinopathy and especially in diabetic retinopathy with non-perfused areas. Some risk factors for the development of retinopathy in diabetic patients have been reported, i.e., hypertension, hypercholesterolemia, and HbA1c levels [44,45]. These factors are also potent stimulators for the release of PMPs [12,33,34]. However in our patients, we did not find a significant correlation between the levels of total cholesterol and the stage of diabetic retinopathy and the level of PMPs. Thus, these findings together with our results suggest that the level of PMPs is a sensitive indicator of microvascular occlusion in diabetic retinopathy. PMPs may be involved in the normal hemostatic response to vascular injury because they have prothrombinase activity [46]. At present, the procoagulant activity of the PMPs in the circulation of
199
diabetic patients has not been studied. PMPs are reported to have both pro- and anti-coagulant activity [12,19], and many studies have reported that PMPs are rich in membrane receptors for coagulation factor Va and provide a catalytic surface for the assembly of the prothrombinase complex [9,11,12]. Thus, PMPs produced by circulating blood cells could play a role in the dissemination of potential pro-coagulants. We found relatively higher levels of PMPs, CD62P, and CD63 in patients with mild or moderate NPDR than in patients with NDR or with diabetic retinopathy without non-perfused areas. There were some patients with mild or moderate NPDR who showed high levels of these factors. Because PMPs bind to fibrin and act as pro-coagulants [19], PMPs activate hypercoagulation and induce platelet hyperaggreability which leads to premature atherosclerosis in diabetic patients [1,20,22]. Then these patients, who showed high levels of these factors, may have poorer prognosis for retinopathy, although further study will be necessary. It has been reported that the platelet aggregation inhibitors, and treatment of hypertension and hyperlipidemia are useful for preventing PMP-dependent vascular damages in type 2 diabetic patients [33,34,47–49]. Such non-invasive therapy may be also effective for diabetic retinopathy. In conclusion, our findings demonstrated that there was a tendency for an increase in the PMPs levels according to the stage of diabetic retinopathy, and the significantly higher levels of PMPs were detected in diabetic retinopathy with non-perfused areas. This would suggest that the high levels of PMPs may accelerate the progression of diabetic retinopathy and may be a prognostic factor for the progression of diabetic retinopathy.
Acknowledgement This study was supported in part by a Grant-in Aid for Scientific Research from the Ministry of Education in Japan.
References [1] S. Schroder, W. Palinski, G.W. Schmid-Schobein, Activated monocytes and granulocytes, capillary non-perfusion and neo-
200
[2] [3]
[4]
[5] [6]
[7]
[8]
[9]
[10] [11]
[12] [13]
[14]
[15]
[16]
[17] [18]
[19]
[20]
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201 vascularization in diabetic retinopathy, Am. J. Pathol. 139 (1991) 81–100. M.D. Davis, Diabetic retinopathy: a clinical overview, Diabetes Care 15 (1993) 1844–1873. E.M. Kohner, R. Chibber, Diabetic retinopathy, in: J.E. Tooke (Ed.), Diabetic Angiography, Oxford University Press, Oxford UK, 1997, pp. 233–247. Early Treatment Diabetic Retinopathy Study Research Group, Early photocoagulation for diabetic retinopathy. ETDRS report number 9, Ophthalmology 98 (1999) 766–785. J.F. Mustrard, M.A. Packham, Platelets and diabetes mellitus, N. Engl. J. Med. 311 (1984) 665–666. J.G. Dobbie, H.C. Kwann, J. Colwell, N. Suwanwela, Role of platelets in pathogenesis of diabetic retinopathy, Arch. Ophthalmol. 91 (1974) 107–109. I. Aoki, K. Shimoyama, N. Aoki, et al. Platelet-dependent thrombin generation in patients with diabetes mellitus: effects of glycemic control on coagulability in diabetes, J. Cell Cardiol. 27 (1996) 560–566. T. Murakami, Y. Komiyama, M. Masuda, et al. Flow cytometric analysis of platelet activation markers CD62P and CD63 in patients with coronary artery disease, Eur. J. Clin. Invest. 26 (1996) 996–1003. Y. Miyazaki, S. Nomura, T. Miyake, et al. High shear stress can initiate both platelet aggregation and shedding of procoagulant containing microparticles, Blood 88 (1996) 3456–3464. P. Wolf, The nature and significance of platelet products in human plasma, Br. J. Hematol. 53 (1972) 206–215. E. Gilbert, P.J. Sims, T. Wiedmer, et al. Platelet-derived microparticles express high affinity receptors for factor VIII, J. Biol. Chem. 266 (1991) 17261–17268. S. Nomura, Function and clinical significance of plateletderived microparticles, Int. J. Hematol. 74 (2001) 397–404. O.P. Barry, D. Pratico, R.C. Savani, et al. Modulation of monocyte–endothelial cell interactions by platelet microparticles, J. Clin. Invest. 102 (1998) 136–144. M. Dore’, R.J. Korthuis, D.N. Granger, et al. P-selectin mediates spontaneous leucocyte rolling in vivo, Blood 82 (1993) 1308–1316. M.B. Lawrence, T.A. Springer, Leucocytes roll on selectin at physiologic flow rates: distribution from and prerequisite for adhesion through integrins, Cell 65 (1991) 859–873. T.N. Mayadas, R.C. Johnson, H. Raybum, R.O. Hyner, D.D. Wagner, Leucocyte rolling and extravasation are severely compromised in P-selectin-deficient mice, Cell 74 (1993) 541–554. L. Osborn, Leucocyte adhesion to endothelium in inflammation, Cell 62 (1990) 3–6. J.E. Sligh, C.M. Ballantyne, S.S. Rich, et al. Inflammatory and immune responses are impaired in mice deficient in intercellular adhesion molecule 1, Proc. Natl. Acad. Sci. U.S.A. 90 (1993) 8529–8533. P. Sijander, O. Carpen, R. Lassila, Platelet-derived microparticles associate with fibrin during thrombosis, Blood 87 (1996) 4651–4663. S. Nomura, N.N. Tandon, T. Nakamura, et al. High-shearstress-induced activation of platelets and microparticles
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
enhances expression of cell adhesion molecules in THP-1 and endothelial cells, Atherosclerosis 158 (2001) 277–287. S. Omoto, S. Nomura, A. Shouzu, et al. Detection of monocyte-derived microparticles in patients with type II diabetes mellitus, Diabetologia 45 (2002) 550–555. S. Nomura, A. Suzuki, K. Katsura, et al. Platelet-derived microparticles may influence the development of atherosclerosis in diabetes mellitus, Atherosclerosis 116 (1995) 235–240. S. Omoto, S. Nomura, A. Shozu, et al. Significance of plateletderived microparticles and activated platelet in diabetic nephropathy, Nephron 81 (1999) 271–277. S. Nomura, A. Shouzu, S. Omoto, M. Nishikaw, S. Fukuhara, Significance of chemokines and activated platelets in patients with diabetes, Clin. Exp. Immunol. 121 (2000) 437–443. F. Sabatier, P. Darmon, B. Hugel, et al. Type 1 and type 2 diabetic patients display different patterns of cellular microparticles, Diabetes 51 (2002) 2820–2845. M. Galli, A. Grassi, T. Barbui, Platelet-derived microvesicles in thrombotic thrombocytopenic purpura and hemolytic uremic syndrome, Thromb. Haemost. 75 (1996) 427–431. B. Hugel, G. Socie, T. Vu, et al. Elevated levels of circulating procoagulant microparticles in patients with paroxysmal nocturnal hemoglobinuria and aplastic anemia, Blood 93 (1999) 3451–3456. Z. Mallat, H. Benamer, B. Hugel, et al. Elevated levels of shed membrane microparticles with procoagulant potential in the peripheral circulating blood of patients with acute coronary syndromes, Circulation 101 (2000) 841–843. O. Morel, B. Hugel, L. Jesel, et al. Sustained elevated amounts of circulating procoagulant membrane microparticles and soluble GPV after acute myocardial infarction in diabetes mellitus, Thromb. Haemost. 91 (2004) 345–353. C.P. Wilkinson, F.L. Ferris III, R.E. Klein, et al. Global Diabetic Retinopathy Project Group, Proposed international clinical diabetic retinopathy and diabetic edema disease severity scales, Ophthalmology 110 (2003) 1677–1682. M. Kaplar, J. Kappelmayer, A. Veszpremi, K. Szabo, M. Udvardy, The possible association of in vivo leucocyte-platelet heterophilic aggregate formation and the development of diabetic angiopathy, Platelets 12 (2001) 419–422. S. Nomura, A. Shozu, S. Omoto, M. Nishikawa, T. Iwasaka, Effect of losartan and simvastatin on monocyte-derived microparticles in hypertensive patients with and without type 2 diabetic mellitus, Clin. Appl. Thromb./Hemost. 10 (2004) 133–141. S. Nomura, S. Kanazawa, S. Fukahaea, Effect of eicosapentaenoic acid on platelet activation markers and cell adhesion molecules in hyperlipidemic patients with Type 2 diabetic raellitus, J. Diabetic Complications 17 (2003) 153–159. S. Nomura, S. Kanazawa, S. Fukahaea, Effect of efonipine on platelet activation markers in hypertensive patients with and without type 2 diabetic mellitus, Hum. Hypertens. 16 (2002) 539–547. S. Kado, N. Nagata, Circulating intercellular adhesion molecule-1, vascular cell adhesion molecule-1, and E-selectin in patients with type 2 diabetes mellitus, Diabetes Res. Clin. Pract. 46 (1999) 143–148.
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201 [36] G.R. Barile, S.S. Chang, L.S. Parl, V.S. Reppucci, V.M. Schiff, A.M. Schimidt, Soluble cellular adhesion molecules in proliferative vitreoretinopathy and proliferative diabetic retinopathy, Curr. Eye Res. 19 (1999) 219–227. [37] K. Miyamoto, S. Khosrof, S.E. Bursell, et al. Prevention of leukostasis and vascular leakage in streptozotocin-induced diabetic retinopathy via intercellular adhesion molecule-1 inhibition, Proc. Natl. Acad. Sci. U.S.A. 96 (1999) 10836– 10841. [38] R. Chibber, B.M. Ben-Mahmud, G.E. Mann, J.J. Zhang, E.M. Kohner, Protein kinase Cb2-dependent phosphorylation of core2 GlcNAc-T promotes leukocyte-endothelial cell adhesion. A mechanism underlying capillary occlusion in diabetic retinopathy, Diabetes 52 (2003) 1519–1527. [39] M. Baj-Krzyworzeka, M. Majka, D. Pratico, et al. Plateletderived microparticles stimulate proliferation, survival, adhesion, and chemotaxis of hematopoietic cells, Exp. Hematol. 30 (2002) 450–459. [40] G.A. Limh, J. Hickman-Casey, R.D. Hollifield, A.H. Chignell, Vascular adhesion molecules in vitreous from eyes with proliferative diabetic retinopathy, Invest. Ophthalmol. Vis. Sci. 40 (1999) 2453–2457. [41] E.C. Barouch, K. Miyamoto, J.R. Allport, et al. Integrinmediated neutrophill adhesion and retinal leucostasis in diabetes, Invest. Ophthalmol. Vis. Sci. 41 (2000) 1153–1158. [42] K. Miyamoto, N. Hiroshiba, A. Tsujikawa, Y. Ogura, In vivo demonstration of increased entrapment in retinal microcirculation of diabetic rats, Invest. Ophthalmol. Vis. Sci. 39 (1998) 2190–2194.
201
[43] A.M. Joussen, T. Murata, A. Tsujikawa, B. Kirchhof, S.E. Bursell, A.P. Adamis, Leucocyte-mediated endothelial cell injury and death in diabetic retina, Am. J. Pathol. 158 (2001) 147–152. [44] H.A. van Leiden, J.M. Dekker, A.C. Moll, et al. Risk factors for incident retinopathy in diabetic and nondiabetic population: the Hoorn study, Arch. Ophthalmol. 121 (2003) 245–251. [45] T. Stulc, Z. Kasalova, M. Prazny, M. Vrablik, J. Skrha, R. Ceska, Microvascular reactivity in patients with hypercholesterolemia: effect of lipid lowering treatment, Physiol. Res. 52 (2003) 439–445. [46] P.I. Sims, T. Wiedmer, C.T. Esmon, H.J. Weiss, S.J. Shattli, Assembly of the platelet prothrombinase complex is link to vesiculation of the platelet plasma membrane, J. Biol. Chem. 264 (1989) 17049–17057. [47] A. Shozu, S. Nomura, T. Hayakawa, et al. Effect of sarpogrelate hydrochloride on platelet-derived microparticles and various soluble adhesion molecules in diabetic mellitus, Clin. Appl. Thromb./Hemost. 6 (2000) 139–143. [48] S. Nomura, A. Shozu, S. Omoto, et al. Effect of cliostazol on soluble adhesion molecules and platelet-derived microparticles in patients with diabetes, Thromb. Haemost. 80 (1998) 388–392. [49] A. Shozu, S. Nomura, S. Omoto, T. Hayakawa, M. Nishikawa, T. Iwasaka, Effect of ticlopidine on monocyte-derived microparticles and activated platelet markers in diabetic mellitus, Clin. Appl. Thromb./Hemost. 10 (2004) 167–173.
Increased levels of platelet-derived microparticles in patients with diabetic retinopathy Nahoko Ogataa,*, Masahito Imaizumia, Shosaku Nomurab, Akira Shozuc, Miwa Arichia, Masato Matsuokaa, Miyo Matsumuraa a
Department of Ophthalmology, Kansai Medical University, Fumizono-cho 10-15, Moriguchi, Osaka 570-8507, Japan b First Department of Internal Medicine, Kansai Medical University, Moriguchi, Osaka, Japan c Third Department of Internal Medicine, Kansai Medical University, Moriguchi, Osaka, Japan Received 10 March 2004; received in revised form 24 May 2004; accepted 13 October 2004 Available online 10 December 2004
Abstract Diabetic retinopathy is caused by capillary occlusions. Platelet-derived microparticles (PMPs) stimulate the coagulation cascade and increase leukocyte and endothelial cell adhesions, both of which are key events in the development of diabetic retinopathy. However, the correlation between the levels of PMPs and diabetic retinopathy has not been precisely determined. The PMPs levels and the expression of platelet CD62P and CD63 were measured in 92 diabetic patients. The level of PMPs was significantly correlated with the expression of CD62P (r = 0.76, P < 0.0001) and CD63 (r = 0.71, P < 0.0001). The mean level of PMPs in diabetics (507 15/104 platelets (plt), mean S.E.) was significantly higher than that in normal. The PMPs levels increased with the progression of the diabetic retinopathy; 480 28/104 plt in diabetic patients without retinopathy (n = 25), 504 40/104 plt with mild or moderate non-proliferative diabetic retinopathy (n = 13), 512 29/104 plt with severe nonproliferative diabetic retinopathy (n = 25), and 528 25/104 plt with proliferative diabetic retinopathy (n = 29). The PMPs level in patients with non-perfused retinal areas (582 27/104 plt, n = 24) was significantly higher than patients without non-perfused areas (469 23/104 plt, n = 30; P = 0.0096) and without diabetic retinopathy (P = 0.024). These high correlations indicate that increased levels of PMPs may accelerate diabetic retinopathy. # 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Platelet-derived microparticles (PMPs); Diabetic retinopathy; Platelet; Coagulation; Diabetes
1. Introduction The number of patients with diabetes is increasing dramatically, and diabetic retinopathy resulting in * Corresponding author. Tel.: +81 6 6992 1001; fax: +81 6 6993 2222. E-mail address: [email protected] (N. Ogata).
retinal ischemia [1–3] is a major cause of adult blindness. To prevent the progression of diabetic retinopathy to proliferative diabetic retinopathy (PDR), retinal photocoagulation is performed. Although photocoagulation is the only effective treatment to reduce the retinal ischemia, this treatment destroys retinal cells including photoreceptor cells and can thereby induce severe vision [4]. Therefore, a
0168-8227/$ – see front matter # 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2004.10.010
194
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
search for new and better non-invasive therapy for diabetic retinopathy is going on. Diabetic retinopathy is characterized at its early stage by areas of microvascular damage and capillary non-perfusion. Platelets play an important role in this vascular occlusive process [1,5–7], and patients with diabetes develop hypercoagulability and hyperaggreability of platelets [5,7]. Various platelet abnormalities appear to be involved in the genesis and/or the evolution of diabetic microangiopathy [5,7], as suggested by the presence of platelet aggregates and thrombi in the small vessels of the retina and the kidneys of diabetic patients [6]. Pro-coagulant activity is expressed as markers of activated platelet, e.g., platelet CD62P (P-selectin), CD63, and plateletderived microparticles (PMPs) [8,9]. PMPs were first observed as vesicles released from activated platelets following their adhesion to vessel walls [10]. PMPs contain a platelet-activating factor (PAF), and express high-affinity receptors for factor VIII and platelet/endothelium attachment receptors such as P-selectin [11,12]. In addition, PMPs increase the level of intracellular adhesion molecule (ICAM)-1, stimulate cytokine secretion, and increase tissue factor expression in endothelial cells [13]. P-selectin is believed to be involved in the initial leukocyte– endothelium interactions [14–16], and ICAM-1 plays an important role in the adhesion and transmigration of leukocytes [17,18]. Thus, PMPs activate the coagulation cascade [19], and increase leukocyte and endothelial cell adhesions [12,20], both of which are key events in the development of capillary occlusion in diabetic patients [21–24]. Therefore, the release of PMPs has been investigated and determined to be evidence of cellular activation occurring in diseases associated with diabetes and thrombotic risk [12,21–29]. However, information on the level of PMPs in patients with diabetic retinopathy is limited [21] and the data have not been well determined. The purpose of this study was to determine whether the microangiopathic changes in the retina arise from the increased level of PMPs. To accomplish this, we have measured the level of PMPs and activated platelets by determining the expression of CD62P (platelet P-selectin) and CD63. Understanding the role of PMPs in retinal ischemia may provide clues for non-invasive therapies for diabetic retinopathy.
2. Patients and methods 2.1. Patients Ninety-two type 2 diabetic patients (37 men and 55 women, ages 31–83 years, average 60.5 years) were studied. All patients who were diagnosed as having type 2 diabetes and agreed to have additional examinations were investigated. An informed consent was obtained from all patients after an explanation of the purpose and procedures of the study, and the procedures were performed to conform to the tenets of the declaration of Helsinki. Patients with severe nephropathy, under treatment by hemodialysis, or being treated with anti-thrombotic or anti-platelets agents were excluded. 2.2. Examination of diabetic retinopathy The stage of diabetic retinopathy was determined by ophthalmoscopy and fluorescein angiography prior to the examination of the blood samples, and the patients were classified according to the severity scale of diabetic retinopathy [30]. The 92 patients included 25 with no apparent diabetic retinopathy (NDR), 13 with mild or moderate non-proliferative diabetic retinopathy (mild or moderate NPDR), 25 with severe non-proliferative diabetic retinopathy (severe NPDR), and 29 with proliferative diabetic retinopathy (PDR). The 54 patients with severe NPDR and with PDR were subdivided into two groups; those with and those without non-perfused areas as determined on fluorescein angiograms. Diabetic eyes were classified as not having non-perfused areas if the retinopathy had been treated with retinal photocoagulation, and the previously documented non-perfused areas had regressed fully and were not present when the eyes were evaluated for this study. The eyes were considered to have non-perfused areas if they were not treated or even if the retinopathy had been treated with retinal photocoagulation and non-perfused areas were still present. Using these criteria, 30 patients (16 with severe NPDR and 14 with PDR) were classified as not having non-perfused areas, and 24 patients (9 with severe NPDR and 15 with PDR) were classified as having non-perfused areas. The demographics of the patients with diabetic retinopathy are shown in Table 1.
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
195
Table 1 CD62P, CD63, and PMPs in diabetic retinopathy Diabetic retinopathy
Number of patients
CD62P(%) (mean S.E.) normal level <10%
CD63 (%) (mean S.E.) normal level <10%
PMPs/104 plt (mean S.E.) normal level = 263 9
NDR MNPDR SNPDR PDR Total
25 13 25 29 92
19.4 2.0 22.6 1.1 20.7 1.3 23.8 1.0 21.0 0.7
18.7 1.2 22.3 1.3 20.2 1.3 22.8 1.2 20.4 0.6
480 28 504 40 512 29 527 25 507 15
NDR, no apparent retinopathy; MNPDR, mild or moderate non-proliferative diabetic retinopathy; SNPDR, severe non-proliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy.
2.3. Methods 2.3.1. HbA1c The level of HbA1c was determined in all patients by standard laboratory procedures.
2.3.3. Total cholesterol The concentrations of total cholesterol were determined in all patients by standard laboratory procedures. 2.4. Statistical analyses
2.3.2. Analysis of activated platelets and plateletderived micropartities The concentration of activated platelets and PMPs in the blood was determined by flow cytometry as described [20–22]. Briefly, blood samples were collected in tubes containing 3.8% sodium citrate, and washed platelets were prepared by centrifuging platelet-rich plasma. The washed platelets were fixed with 2% paraformaldehyde, rinsed, and re-suspended in stock solution. The platelets were then incubated with a FITC-labeled monoclonal antibody against platelets GPIX, and analyzed with an ortho cytoron absolute analyzer (Ortho Diagnostic Systems, Tokyo, Japan) to measure the level of PMPs. Only cells and particles positive for GPIX were gated to separate the platelets and PMPs from electric noise. To differentiate between platelets and PMPs, the lower limit of the platelet gate was set at the limit of the forwardscatter profile of resting platelets. The number of PMPs released from 104 platelets (plt) was used as a measure of the PMP concentration. As an index of platelet activation, the expressions of CD62P and CD63 were quantified. The washed platelets were fixed with 2% paraformaldehyde, incubated with anti-CD62P antibody (CLB-thromb/ 6, Immunoteck Marseile, France) and anti-CD63 antibody (CLB-gran/12, Immunoteck Marseile, France). They were then analyzed by flow cytometry as previously described [8]. The platelets positive for CD62P and CD63 were expressed as percentages.
The results were analyzed using a one-way ANOVA and Fisher test. The correlation between PMPs and the different factors was examined by Pearson’s product moment coefficient of correlation (r). The Spearman’s rank correlation coefficient was also calculated. P-values less than 0.05 were accepted as significant.
3. Results 3.1. Correlations between PMPs, and CD62P and CD63 The level of PMPs was strongly correlated with the level of CD62P (r = 0.76, P < 0.0001, Pearson’s product moment coefficient of correlation, Fig. 1A). The Spearman’s rank correlation coefficient was also significant (P < 0.0001). The level of PMPs was significantly higher in patients with higher concentration of CD63 (r = 0.71 and P < 0.0001, Fig. 1B). These results demonstrated that the platelets were indeed activated, and in all likelihood, accounted for the higher levels of PMPs. 3.2. Correlation between PMPs and HbA1c There was a significantly higher level of PMPs in patients with higher levels of HbA1c. Calculations
196
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
Fig. 1. (A) Correlation between the level of PMPs and CD62P. (B) Correlation between the level of PMPs and CD63. The levels of CD62P and CD63 increase as the level of PMPs increases. The coefficient of correlation (r) between PMP and CD62P is r = 0.76 (P < 0.0001), and CD63 is r = 0.71 (P < 0.0001).
showed that the correlation between PMPs and HbA1c was significant (r = 0.28, P = 0.011, Fig. 2). 3.3. Correlation between PMPs and the stage of diabetic retinopathy The concentration of PMPs in all patients with diabetes was 507 15/104 plts (mean standard error of the mean, n = 92). The level of PMPs in diabetic patients was significantly higher than that in non-diabetic control subjects (263 9/104 plts, based on our previous data [21]). The level of PMPs
increased as the stage of diabetic retinopathy advanced; 480 28/104 plt in patients with NDR (n = 25), 504 40/104 plt with mild or moderate NPDR (n = 13), 512 29/104 plt with severe NPDR (n = 25), and 527 25/104 plt with PDR (n = 29). However, the differences in the PMPs levels at the different stages of diabetic retinopathy were not significant (Table 1, Fig. 3A). 3.4. PMPs levels in patients with non-perfused areas The mean level of PMPs in patients with nonperfused area was 582 27/104 plts (9 with severe NPDR and 15 with PDR), which was significantly higher than 469 23/104plt in patients without nonperfused areas (16 with severe NPDR and 14 with PDR, P = 0.0096), and also in patients with NDR (P = 0.024, Fig. 3B). 3.5. Correlation between the stage of diabetic retinopathy, and CD62P and CD63
Fig. 2. Correlation between PMPs and HbA1c. The level of HbA1c increases as the level of PMPs increases (r = 0.28, P = 0.011).
The mean level of CD62P in patients with diabetes was 21.0 0.7%, which was higher than that in nondiabetic control subjects (normal level < 10% [21]). The mean level of CD62P was 19.4 2.0% in patients with NDR (n = 25), 22.6 1.1% with mild or moderate NPDR (n = 13), 20.7 1.3% with severe
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
197
Fig. 3. The levels of PMPs in patients with diabetic retinopathy. (A) The level of PMPs increases as the stage of diabetic retinopathy gets worse, but the differences in the PMPs levels at the different stages of diabetic retinopathy are not significant. (B) The level of PMPs in patients with nonperfused areas is significantly higher than that in patients without non-perfused areas (P = 0.0096) and patients with NDR (P = 0.024). NDR, no apparent retinopathy; MNPDR, mild or moderate non-proliferative diabetic retinopathy; SNPDR, severe non-proliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy; non-PFA, retinas with non-perfused areas. Bars represent S.E. and the asterisks (*) indicate P < 0.05.
NPDR (n = 25), and 23.8 1.0% with PDR (n = 29, Table 1). All of these levels were significantly higher than that of controls, and the difference between in patients with NDR and PDR was significant (P = 0.029). The mean level of CD62P was 23.6 1.9% (n = 24) in patients with non-perfused areas and 20.7 1.1% (n = 30) in patients without non-perfused areas. This difference was not significant (P = 0.12) but the mean CD62P level in patients with nonperfused areas and the level in patients with NDR was significant (P = 0.039; Table 1, Fig. 4A). The mean level of CD63 in patients with diabetes was 20.4 0.6%, which was higher than that in controls (<10% [21]). The levels of CD63 were 18.7 1.2% in patients with NDR (n = 25), 22.3 1.3% with mild or moderate NPDR (n = 13), 20.2 1.3% with severe NPDR (n = 25), and 22.8 1.2% with PDR (n = 29, Table 1). The differences in the CD63 levels at the different stages of diabetic retinopathy were not significant. However, the mean CD63 level in patients with non-perfused areas (23.9 1.3%, n = 24) was sig-
nificantly higher than that in NDR patients (P = 0.006) and in patients without non-perfused areas (19.7 1.2%, n = 30, P = 0.023, Fig. 4B). 3.6. Correlation between the stage of diabetic retinopathy and total cholesterol The mean level of cholesterol in patients with diabetes was 203.7 5.7 mg/dl which is slightly high but still in the normal range (136–220 mg/dl). The mean level of cholesterol was 203.3 7.5 mg/dl in patients with NDR (n = 25), 204.1 15.4 mg/dl with mild or moderate NPDR (n = 13), 192.9 14.8 mg/dl with severe NPDR (n = 25), and 220.0 17.3 mg/dl with PDR (n = 29). The differences in the cholesterol levels at the different stages of diabetic retinopathy were not significant. The mean level of cholesterol was 195.4 10.8 mg/dl (n = 24) in patients with nonperfused areas and 218.0 15.0 mg/dl (n = 30) in patients without non-perfused areas, and this difference was not significant. In addition there was no correlations between the cholesterol levels and PMPS (r = 0.07).
198
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
Fig. 4. The Levels of CD62P and CD63 in patients with diabetic retinopathy. (A) The mean CD62P level in patients with non-perfused areas is higher than that in patients without non-perfused areas, but the difference is not significant (P = 0.08). (B) The mean CD63 level in patients with non-perfused areas is significantly higher than that in patients without diabetic retinopathy (P = 0.006) and without non-perfused areas (P = 0.023). NDR, no apparent retinopathy; MNPDR, mild or moderate non-proliferative diabetic retinopathy; non-PFA, retinas with nonperfused areas. Bars represent S.E. and the asterisks (*) indicate P < 0.05.
4. Discussion Diabetic retinopathy is caused by capillary occlusions, and capillary occlusions result from microvascular thrombi in which erythrocytes, platelets, and leukocytes play a role [1,31]. The role of platelets in the pathogenesis of diabetic retinopathy has been pointed out [5] and diabetic patients have activated platelets [5,6,31], but the role of activated platelets in diabetic retinopathy has not been determined. Activated platelets express adhesive molecules, e.g., CD62P (platelet P-selectin) and CD63 [9,23,24]. PMPs are also released from activated platelets, and the PMPs contain inner granules, membranous microvesicles, membranous fragments produced by mechanical disruptions, and have coagulative activity [9,12,19]. Earlier, we found that the PMPs levels are correlated with presence of hypertension, nephropathy, atherosclerosis, and hyperlipidemia [12,22,23, 32–34]. There have been reported that PMPs are higher in type 2 diabetic patients [12,21–23], and we have reported on the potential role of PMPs in the
complication of diabetes such as neuropathy, nephropathy, retinopathy, and atherosclerosis [21–24]. However, the relationship between PMPs and diabetic retinopathy has not been clearly assessed. The level of PMPs in our patients with type 2 diabetes was significantly higher than that of control subjects confirming our previous reports [12,21–24]. The higher levels of PMPs most likely resulted from the activation of platelets as shown by its high correlation to CD62P and CD63, both markers of platelet activation. Interestingly, a recent study reported a significant elevation of PMPs in type l diabetes, and no significant increase in type 2 diabetic patients and the age-matched control subjects [25]. This discrepancy in the level of PMPs in type 2 diabetic patients may arise from differences in the patient and control populations (Japanese and Caucasian) and/or from the techniques used to measure the level of the PMPs. The significant correlation between the levels of PMPs and HbA1c in our type 2 diabetic patients suggests that diabetic patients with poor control of glucose levels have higher levels of PMPs.
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201
It has been reported that PMPs trigger different types of biological responses after interacting with their target cells. For example, they increase the level of ICAM-1, and ICAM-1 is known to mediate leukocyte adhesion and transmigration. Thus, ICAM-1 may be associated with the circulatory stasis observed in diabetic retinopathy [35–38]. In addition, PMPs may activate endothelial cells and monocytes, or may stimulate cytokine secretion and tissue factor expression in endothelial cells [12,13,24,39]. PMPs express several platelet/endothelial attachment receptors on their surface, such as glycoprotein Ilb/IIIa (aIIb b3), factor VIII, GPIb, and GPIa/IIa, and CD62P (P-selectin) [9,11,12]. P-selectin is believed to be involved in the initial stages of leukocyte–endothelium interactions [14–16]. Thus, these properties suggest that PMPs themselves increase leukocyte–endothelial cell adhesion. As in atherogenesis, increased leukocyte–endothelial cell adhesion is a key early event in the development of capillary occlusion in diabetic retinopathy [31,36–38,40,41]. Leukocytes from diabetic patients have been shown to be more adherent to endothelial cells [42], and in experimental diabetes, they tend to be easily entrapped in retinal capillaries which then leads to the areas of capillary nonperfusion and endothelial cell damage [43]. These properties may also be modulated by PMPs. Significantly higher levels of CD62P and CD63 were found in our diabetic patients, and especially in those with non-perfused areas. In addition, the levels of PMPs increased with the advancement in the stage of diabetic retinopathy and especially in diabetic retinopathy with non-perfused areas. Some risk factors for the development of retinopathy in diabetic patients have been reported, i.e., hypertension, hypercholesterolemia, and HbA1c levels [44,45]. These factors are also potent stimulators for the release of PMPs [12,33,34]. However in our patients, we did not find a significant correlation between the levels of total cholesterol and the stage of diabetic retinopathy and the level of PMPs. Thus, these findings together with our results suggest that the level of PMPs is a sensitive indicator of microvascular occlusion in diabetic retinopathy. PMPs may be involved in the normal hemostatic response to vascular injury because they have prothrombinase activity [46]. At present, the procoagulant activity of the PMPs in the circulation of
199
diabetic patients has not been studied. PMPs are reported to have both pro- and anti-coagulant activity [12,19], and many studies have reported that PMPs are rich in membrane receptors for coagulation factor Va and provide a catalytic surface for the assembly of the prothrombinase complex [9,11,12]. Thus, PMPs produced by circulating blood cells could play a role in the dissemination of potential pro-coagulants. We found relatively higher levels of PMPs, CD62P, and CD63 in patients with mild or moderate NPDR than in patients with NDR or with diabetic retinopathy without non-perfused areas. There were some patients with mild or moderate NPDR who showed high levels of these factors. Because PMPs bind to fibrin and act as pro-coagulants [19], PMPs activate hypercoagulation and induce platelet hyperaggreability which leads to premature atherosclerosis in diabetic patients [1,20,22]. Then these patients, who showed high levels of these factors, may have poorer prognosis for retinopathy, although further study will be necessary. It has been reported that the platelet aggregation inhibitors, and treatment of hypertension and hyperlipidemia are useful for preventing PMP-dependent vascular damages in type 2 diabetic patients [33,34,47–49]. Such non-invasive therapy may be also effective for diabetic retinopathy. In conclusion, our findings demonstrated that there was a tendency for an increase in the PMPs levels according to the stage of diabetic retinopathy, and the significantly higher levels of PMPs were detected in diabetic retinopathy with non-perfused areas. This would suggest that the high levels of PMPs may accelerate the progression of diabetic retinopathy and may be a prognostic factor for the progression of diabetic retinopathy.
Acknowledgement This study was supported in part by a Grant-in Aid for Scientific Research from the Ministry of Education in Japan.
References [1] S. Schroder, W. Palinski, G.W. Schmid-Schobein, Activated monocytes and granulocytes, capillary non-perfusion and neo-
200
[2] [3]
[4]
[5] [6]
[7]
[8]
[9]
[10] [11]
[12] [13]
[14]
[15]
[16]
[17] [18]
[19]
[20]
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201 vascularization in diabetic retinopathy, Am. J. Pathol. 139 (1991) 81–100. M.D. Davis, Diabetic retinopathy: a clinical overview, Diabetes Care 15 (1993) 1844–1873. E.M. Kohner, R. Chibber, Diabetic retinopathy, in: J.E. Tooke (Ed.), Diabetic Angiography, Oxford University Press, Oxford UK, 1997, pp. 233–247. Early Treatment Diabetic Retinopathy Study Research Group, Early photocoagulation for diabetic retinopathy. ETDRS report number 9, Ophthalmology 98 (1999) 766–785. J.F. Mustrard, M.A. Packham, Platelets and diabetes mellitus, N. Engl. J. Med. 311 (1984) 665–666. J.G. Dobbie, H.C. Kwann, J. Colwell, N. Suwanwela, Role of platelets in pathogenesis of diabetic retinopathy, Arch. Ophthalmol. 91 (1974) 107–109. I. Aoki, K. Shimoyama, N. Aoki, et al. Platelet-dependent thrombin generation in patients with diabetes mellitus: effects of glycemic control on coagulability in diabetes, J. Cell Cardiol. 27 (1996) 560–566. T. Murakami, Y. Komiyama, M. Masuda, et al. Flow cytometric analysis of platelet activation markers CD62P and CD63 in patients with coronary artery disease, Eur. J. Clin. Invest. 26 (1996) 996–1003. Y. Miyazaki, S. Nomura, T. Miyake, et al. High shear stress can initiate both platelet aggregation and shedding of procoagulant containing microparticles, Blood 88 (1996) 3456–3464. P. Wolf, The nature and significance of platelet products in human plasma, Br. J. Hematol. 53 (1972) 206–215. E. Gilbert, P.J. Sims, T. Wiedmer, et al. Platelet-derived microparticles express high affinity receptors for factor VIII, J. Biol. Chem. 266 (1991) 17261–17268. S. Nomura, Function and clinical significance of plateletderived microparticles, Int. J. Hematol. 74 (2001) 397–404. O.P. Barry, D. Pratico, R.C. Savani, et al. Modulation of monocyte–endothelial cell interactions by platelet microparticles, J. Clin. Invest. 102 (1998) 136–144. M. Dore’, R.J. Korthuis, D.N. Granger, et al. P-selectin mediates spontaneous leucocyte rolling in vivo, Blood 82 (1993) 1308–1316. M.B. Lawrence, T.A. Springer, Leucocytes roll on selectin at physiologic flow rates: distribution from and prerequisite for adhesion through integrins, Cell 65 (1991) 859–873. T.N. Mayadas, R.C. Johnson, H. Raybum, R.O. Hyner, D.D. Wagner, Leucocyte rolling and extravasation are severely compromised in P-selectin-deficient mice, Cell 74 (1993) 541–554. L. Osborn, Leucocyte adhesion to endothelium in inflammation, Cell 62 (1990) 3–6. J.E. Sligh, C.M. Ballantyne, S.S. Rich, et al. Inflammatory and immune responses are impaired in mice deficient in intercellular adhesion molecule 1, Proc. Natl. Acad. Sci. U.S.A. 90 (1993) 8529–8533. P. Sijander, O. Carpen, R. Lassila, Platelet-derived microparticles associate with fibrin during thrombosis, Blood 87 (1996) 4651–4663. S. Nomura, N.N. Tandon, T. Nakamura, et al. High-shearstress-induced activation of platelets and microparticles
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
enhances expression of cell adhesion molecules in THP-1 and endothelial cells, Atherosclerosis 158 (2001) 277–287. S. Omoto, S. Nomura, A. Shouzu, et al. Detection of monocyte-derived microparticles in patients with type II diabetes mellitus, Diabetologia 45 (2002) 550–555. S. Nomura, A. Suzuki, K. Katsura, et al. Platelet-derived microparticles may influence the development of atherosclerosis in diabetes mellitus, Atherosclerosis 116 (1995) 235–240. S. Omoto, S. Nomura, A. Shozu, et al. Significance of plateletderived microparticles and activated platelet in diabetic nephropathy, Nephron 81 (1999) 271–277. S. Nomura, A. Shouzu, S. Omoto, M. Nishikaw, S. Fukuhara, Significance of chemokines and activated platelets in patients with diabetes, Clin. Exp. Immunol. 121 (2000) 437–443. F. Sabatier, P. Darmon, B. Hugel, et al. Type 1 and type 2 diabetic patients display different patterns of cellular microparticles, Diabetes 51 (2002) 2820–2845. M. Galli, A. Grassi, T. Barbui, Platelet-derived microvesicles in thrombotic thrombocytopenic purpura and hemolytic uremic syndrome, Thromb. Haemost. 75 (1996) 427–431. B. Hugel, G. Socie, T. Vu, et al. Elevated levels of circulating procoagulant microparticles in patients with paroxysmal nocturnal hemoglobinuria and aplastic anemia, Blood 93 (1999) 3451–3456. Z. Mallat, H. Benamer, B. Hugel, et al. Elevated levels of shed membrane microparticles with procoagulant potential in the peripheral circulating blood of patients with acute coronary syndromes, Circulation 101 (2000) 841–843. O. Morel, B. Hugel, L. Jesel, et al. Sustained elevated amounts of circulating procoagulant membrane microparticles and soluble GPV after acute myocardial infarction in diabetes mellitus, Thromb. Haemost. 91 (2004) 345–353. C.P. Wilkinson, F.L. Ferris III, R.E. Klein, et al. Global Diabetic Retinopathy Project Group, Proposed international clinical diabetic retinopathy and diabetic edema disease severity scales, Ophthalmology 110 (2003) 1677–1682. M. Kaplar, J. Kappelmayer, A. Veszpremi, K. Szabo, M. Udvardy, The possible association of in vivo leucocyte-platelet heterophilic aggregate formation and the development of diabetic angiopathy, Platelets 12 (2001) 419–422. S. Nomura, A. Shozu, S. Omoto, M. Nishikawa, T. Iwasaka, Effect of losartan and simvastatin on monocyte-derived microparticles in hypertensive patients with and without type 2 diabetic mellitus, Clin. Appl. Thromb./Hemost. 10 (2004) 133–141. S. Nomura, S. Kanazawa, S. Fukahaea, Effect of eicosapentaenoic acid on platelet activation markers and cell adhesion molecules in hyperlipidemic patients with Type 2 diabetic raellitus, J. Diabetic Complications 17 (2003) 153–159. S. Nomura, S. Kanazawa, S. Fukahaea, Effect of efonipine on platelet activation markers in hypertensive patients with and without type 2 diabetic mellitus, Hum. Hypertens. 16 (2002) 539–547. S. Kado, N. Nagata, Circulating intercellular adhesion molecule-1, vascular cell adhesion molecule-1, and E-selectin in patients with type 2 diabetes mellitus, Diabetes Res. Clin. Pract. 46 (1999) 143–148.
N. Ogata et al. / Diabetes Research and Clinical Practice 68 (2005) 193–201 [36] G.R. Barile, S.S. Chang, L.S. Parl, V.S. Reppucci, V.M. Schiff, A.M. Schimidt, Soluble cellular adhesion molecules in proliferative vitreoretinopathy and proliferative diabetic retinopathy, Curr. Eye Res. 19 (1999) 219–227. [37] K. Miyamoto, S. Khosrof, S.E. Bursell, et al. Prevention of leukostasis and vascular leakage in streptozotocin-induced diabetic retinopathy via intercellular adhesion molecule-1 inhibition, Proc. Natl. Acad. Sci. U.S.A. 96 (1999) 10836– 10841. [38] R. Chibber, B.M. Ben-Mahmud, G.E. Mann, J.J. Zhang, E.M. Kohner, Protein kinase Cb2-dependent phosphorylation of core2 GlcNAc-T promotes leukocyte-endothelial cell adhesion. A mechanism underlying capillary occlusion in diabetic retinopathy, Diabetes 52 (2003) 1519–1527. [39] M. Baj-Krzyworzeka, M. Majka, D. Pratico, et al. Plateletderived microparticles stimulate proliferation, survival, adhesion, and chemotaxis of hematopoietic cells, Exp. Hematol. 30 (2002) 450–459. [40] G.A. Limh, J. Hickman-Casey, R.D. Hollifield, A.H. Chignell, Vascular adhesion molecules in vitreous from eyes with proliferative diabetic retinopathy, Invest. Ophthalmol. Vis. Sci. 40 (1999) 2453–2457. [41] E.C. Barouch, K. Miyamoto, J.R. Allport, et al. Integrinmediated neutrophill adhesion and retinal leucostasis in diabetes, Invest. Ophthalmol. Vis. Sci. 41 (2000) 1153–1158. [42] K. Miyamoto, N. Hiroshiba, A. Tsujikawa, Y. Ogura, In vivo demonstration of increased entrapment in retinal microcirculation of diabetic rats, Invest. Ophthalmol. Vis. Sci. 39 (1998) 2190–2194.
201
[43] A.M. Joussen, T. Murata, A. Tsujikawa, B. Kirchhof, S.E. Bursell, A.P. Adamis, Leucocyte-mediated endothelial cell injury and death in diabetic retina, Am. J. Pathol. 158 (2001) 147–152. [44] H.A. van Leiden, J.M. Dekker, A.C. Moll, et al. Risk factors for incident retinopathy in diabetic and nondiabetic population: the Hoorn study, Arch. Ophthalmol. 121 (2003) 245–251. [45] T. Stulc, Z. Kasalova, M. Prazny, M. Vrablik, J. Skrha, R. Ceska, Microvascular reactivity in patients with hypercholesterolemia: effect of lipid lowering treatment, Physiol. Res. 52 (2003) 439–445. [46] P.I. Sims, T. Wiedmer, C.T. Esmon, H.J. Weiss, S.J. Shattli, Assembly of the platelet prothrombinase complex is link to vesiculation of the platelet plasma membrane, J. Biol. Chem. 264 (1989) 17049–17057. [47] A. Shozu, S. Nomura, T. Hayakawa, et al. Effect of sarpogrelate hydrochloride on platelet-derived microparticles and various soluble adhesion molecules in diabetic mellitus, Clin. Appl. Thromb./Hemost. 6 (2000) 139–143. [48] S. Nomura, A. Shozu, S. Omoto, et al. Effect of cliostazol on soluble adhesion molecules and platelet-derived microparticles in patients with diabetes, Thromb. Haemost. 80 (1998) 388–392. [49] A. Shozu, S. Nomura, S. Omoto, T. Hayakawa, M. Nishikawa, T. Iwasaka, Effect of ticlopidine on monocyte-derived microparticles and activated platelet markers in diabetic mellitus, Clin. Appl. Thromb./Hemost. 10 (2004) 167–173.