- Email: [email protected]
Elevated serum levels of soluble vascular cell adhesion molecule-1 in NIDDM patients with proliferative diabetic retinopathy
Diabetes Research and Clinical Practice 42 (1998) 65 – 70
Elevated serum levels of soluble vascular cell adhesion molecule-1 in NIDDM patients with proliferative diabetic retinopathy Miyako Yoshizawa a,*, Yukihiro Nagai a, Kensou Ohsawa a, Masayoshi Ohta a, Haruhisa Yamashita a, Azusa Hisada a, Ichiro Miyamoto b, Katsuyuki Miura c, Toshinari Takamura a, Ken-ichi Kobayashi a a
First Department of Internal Medicine, School of Medicine, Kanazawa Uni6ersity. 13 -1 Takara-machi, Kanazawa, Ishikawa 920 -8640, Japan b The Department of Internal Medicine, Toyama Rosai Hospital. 992 Rokuroumaru, Uozu, Toyama 937 -0042, Japan c The Department of Public Health, Kanazawa Medical Uni6ersity. 1 -1 Daigaku, Uchinada, Ishikawa 920 -0265, Japan Received 15 August 1997; received in revised form 24 June 1998; accepted 12 August 1998
Abstract We studied 68 Japanese NIDDM patients (38 men and 30 women), aged 56.9 9 1.2 years (range 33 – 75 years), with a BMI of 23.190.5 kg/m2 without hypertension, dyslipidemia, and diabetic macroangiopathy for evaluating the relationship between serum soluble vascular cell adhesion molecule-1 (sVCAM-1) levels and the severity of diabetic retinopathy. Fundus examination was performed by an ophthalmologist using an ophthalmoscope, and the findings were graded as: (1) no signs of diabetic retinopathy (NDR), (2) background diabetic retinopathy (BDR), or (3) proliferative diabetic retinopathy (PDR). Serum sVCAM-1 levels were measured in duplicate by enzyme-linked immunosorbent assay using the soluble VCAM-1 KIT (R&D Systems Ltd., Ablingdon, Oxfordshire, UK). There was no difference in serum sVCAM-1 levels between patients with BDR (n=17) and patients with NDR (n= 40) (1035.3 9104.4 and 978.8 948.9 ng/ml, respectively, P= 0.8), but patients with PDR (n= 11) showed a significant increase of serum sVCAM-1 levels compared with patients with NDR (1281.8 9166.3 and 978.8 9 48.9 ng/ml, respectively, P =0.02). Although serum sVCAM-1 levels were correlated, not only with age but also with the known diabetic duration (r =0.39, P =0.001, and r= 0.40, P= 0.0007, respectively), age-adjusted sVCAM-1 levels were still significantly higher in the PDR group than in the NDR group. In contrast, serum sVCAM-1 levels were not related to the presence of diabetic nephropathy or HbA1c levels. Our results suggest that sVCAM-1 might be implicated in the development of the diabetic retinopathy, and measurement of serum sVCAM-1 levels in NIDDM patients may
* Corresponding author. Tel.: +81 76 2652234; fax: +81 76 2344250. 0168-8227/98/$ - see front matter © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII: S0168-8227(98)00091-6
66
M. Yoshizawa et al. / Diabetes Research and Clinical Practice 42 (1998) 65–70
be clinically useful for assessing the severity and possibly the activity of diabetic retinopathy. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: VCAM-1; Diabetes mellitus; Diabetic retinopathy
1. Introduction
2. Materials and methods
Vascular cell adhesion molecule-1 (VCAM-1), a member of the immunoglobulin superfamily which contains immunoglobulin-like structures, is a 110 kD cell surface glycoprotein, and plays a major role in the adhesion of leukocytes to the endothelium by interaction with its ligand very late antigen-4 (VLA-4), a member of the b1 integrin family, which is expressed on lymphocytes and monocytes [1]. Recently, Schmidt et al. [2] reported that the advanced glycation endproducts (AGEs) interacting with a cellular receptor for AGE (RAGE) resulted in enhanced cellular oxidant stress with consequent changes in multiple parameters, and increased expression of endothelial cell-associated VCAM-1 as well as increased release of soluble VCAM-1 (sVCAM-1) into culture supernatants. Cominacini et al. [3] also suggested that the metabolic control of the diabetic patients may affect the endothelial cell activity since elevated levels of soluble E-selectin were positively correlated with glycated hemoglobin levels. On the other hand, Koch et al. [4] showed that soluble forms of VCAM-1 and E-selectin induced angiogenesis in rat cornea. sVCAM-1 acts on endothelial cells in part through a VLA-4 dependent mechanism and induces chemotaxis of endothelial cells without stimulating proliferation of endothelial cells in vitro [4]. They suggested an intriguing possibility that shed adhesion molecules might stimulate angiogenesis in vivo. If endothelial cell-associated VCAM-1 is up-regulated by AGE – RAGE interaction, leading to increased circulating levels of sVCAM-1 in diabetic patients, and sVCAM-1 has proangiogenic functions, serum sVCAM-1 levels in diabetic patients with proliferative retinopathy might be increased. In this study, we aimed to investigate the relationship between serum sVCAM-1 levels and the severity of diabetic retinopathy in NIDDM patients.
We studied 68 NIDDM patients (38 men and 30 women) without diabetic symptomatic macroangiopathy, aged 56.99 1.2 years (range 33–75 years), with a BMI of 23.19 0.5 kg/m2. The known diabetic duration was calculated as the duration from the initial diagnosis of NIDDM as defined by World Health Organization criteria [5]. Diabetic symptomatic macroangiopathy was defined as the clinically symptomatic history of ischemic heart disease, cerebral infarction, or arteriosclerosis obliterans. Twenty-one patients were treated by insulin, 29 with oral hypoglycemic agents and 18 with diet alone. Since soluble forms of cell adhesion molecules were reported to increase in some metabolic disorders including hypertension [6] and dyslipidemia [7], patients with hypertension and dyslipidemia were excluded; hypertension was defined by either prior diagnosis, current treatment with antihypertensive agent, or blood pressure greater than 160/ 95 mmHg. Dyslipidemia was considered to be present when patients had serum total cholesterol \ 240 mg/dl [8], and/or triglyceride \ 250 mg/dl [9] and/or HDL cholesterol B 40 mg/dl in women and B 35 mg/dl in men. In all patients, renal insufficiency, liver diseases, malignancies, and chronic inflammatory diseases and pathological processes known to increase adhesion molecules were excluded. The study was approved by our local ethics committee, and informed consent was obtained from each subject. Venous blood was collected by standard venipuncture using vacutainer tubes at 8:00 a.m. after a 12-h fast from all subjects. Serum was extracted following 10-min centrifugation in a bench centrifuge at 2500 rev/min and stored at − 20°C until assayed. HbA1c was measured by affinity chromatography (HA-8131, Kyoto Daiichi Kagaku, Kyoto, Japan, normal range 4.3– 5.8%). Serum creatinine, total cholesterol, and
M. Yoshizawa et al. / Diabetes Research and Clinical Practice 42 (1998) 65–70
67
Table 1 Serum sVCAM-1 levels and clinical characteristics of NIDDM patients
Number (male/female) Age (years) BMI (kg/m2) Known DM duration (years) HbA1c (%) Serum creatinine (mg/dl) Total cholesterol (mg/dl) HDL-cholesterol (mg/dl) LDL-cholesterol (mg/dl) Triglyceride (mg/dl) sVCAM-1 (ng/ml) sVCAM-1 (age adjusted) sVCAM-1 (known DM duration adjusted)
No retinopathy
Background retinopathy
Proliferative retinopathy
40 (24/16) 54.99 1.7 23.29 0.7 3.2 9 0.5 7.49 0.3 0.89 0.02 199.59 4.3 49.89 2.1 126.99 4.6 113.59 10.2 978.89 48.9 1010.99 57.1 1093.09 70.9
17 (7/10) 61.4 91.5* 23.2 90.6 9.2 9 1.2* 7.29 0.4 0.9 90.1 196.4 9 5.2 57.4 93.8 116.2 94.6 113.6 916.6 1035.3 9 104.4 959.7 989.1 946.4 9 94.6
11 (7/4) 56.9 92.6 22.5 91.1 14.8 9 1.3* 8.3 9 0.7 0.9 9 0.1 181.4 9 15.5 50.9+4.4 106.1 912.3 121.5 917.6 1281.8 9166.3* 1281.4 9 107.6* 1003.8 9 148.6
Values are expressed as n or means9 S.E.M. *PB0.05, vs no retinopathy group, by analysis of variance.
triglyceride concentrations were measured by standard laboratory methods. HDL cholesterol was isolated by sequential ultracentrifugation in NaBr solutions, and LDL cholesterol was calculated with the equation of Friedewald et al. [10]. Serum sVCAM-1 levels were measured in duplicate by enzyme-linked immunosorbent assay using the soluble VCAM-1 KIT (R&D Systems Ltd., Ablingdon, Oxfordshire, UK), with both inter- and intra-assay coefficients of variations are 9.2 and 5.9%, respectively. There is a consensus that the screening of microalbuminuria can easily be done by measuring urine albumin-creatinine ratio (A/C ratio) [11], so that we measured A/C ratios in early morning urine (Olympus AU 800, Tokyo, Japan, normal range B18.0 mg/gCr) to evaluate the presence of diabetic nephropathy. The patients with an A/C ratio \30 mg/gCr were defined to have diabetic nephropathy. The study subjects were divided into three groups according to the severity of diabetic retinopathy. After mydriasis, fundus examination was performed by an ophthalmologist using an ophthalmoscope. The findings were graded as: (1) no signs of diabetic retinopathy (NDR); (2) background diabetic retinopathy (BDR); or (3) proliferative diabetic retinopathy (PDR). Statistical differences between three groups
were evaluated by analysis of variance, using Bonferroni’s test for subsequent comparisons. Values are expressed as means9 S.E.M. A 2-tailed Pvalue less than 0.05 was considered to be statistically significant. Statistical analyses were performed using the Statistical Analysis System (SAS Institute Inc., Cary, North Carolina, USA). 3. Results Table 1 shows the mean values of the anthropometric characteristics, glycated hemoglobin, serum total cholesterol, HDL cholesterol, LDL cholesterol, triglyceride, and creatinine in NIDDM patients divided into three groups based on the severity of diabetic retinopathy. There were no significant differences in BMI, serum creatinine, total cholesterol, HDL cholesterol, LDL cholesterol, or triglyceride levels among the three groups. Age of the BDR group showed relatively older than that of the NDR group, but was not different from that of the PDR group significantly. As expected, the PDR group had a significantly longer diabetic duration than both groups of NDR and BDR. As shown in Table 1, there was no difference in serum sVCAM-1 levels between BDR and NDR groups (P= 0.6), but the PDR group showed a significant increase of serum sVCAM-1 levels
68
M. Yoshizawa et al. / Diabetes Research and Clinical Practice 42 (1998) 65–70
compared with the NDR group (P = 0.02). The serum sVCAM-1 levels in all NIDDM patients significantly correlated, not only with age but also with the known diabetic duration (r = 0.39, P = 0.01 and r =0.40, P =0.0007, respectively), thus the mean values of serum sVCAM-1 levels, which were adjusted for age and the known diabetic duration, were calculated using analysis of covariance. The age-adjusted sVCAM-1 levels of the PDR group were still significantly higher than those of the NDR group, but the sVCAM-1 levels adjusted for the known diabetic duration showed no difference between three groups (Table 1). In contrast, the mean values of serum sVCAM1 levels in patients with diabetic nephropathy (n =24) were not different than those in patients without nephropathy (n = 44) (1056.2989.1 and 1034.19 56.9 ng/ml, respectively, P= 0.8). In addition, serum sVCAM-1 levels were not correlated with HbA1c levels (r =0.02, P =0.9).
4. Discussion In this study, we demonstrated the significant elevation of serum sVCAM-1 levels in NIDDM patients with PDR, even after adjustment for age. In contrast, serum sVCAM-1 levels were not related to the presence of diabetic nephropathy or HbA1c levels. Intraocular neovascularization occurs in ischemic retinal disorders, especially diabetic retinopathy, and often results in vitreous hemorrhage, retinal detachment, and neovascular glaucoma, with subsequent visual loss. Recently, it has been suggested that oxidative stress may play an important role in the development of diabetic retinopathy [12], promoting intraocular neovascularization characterized in proliferative diabetic retinopathy. Endothelium is likely to play a major role in angiogenesis, thus the investigation of the basic mechanisms by which endothelial cells are stimulated to undergo angiogenesis is attractive to identify endogenous angiogenetic inhibitors and to use these molecules to control angiogenesis [13]. The numerous growth factors have been known to stimulate angiogenesis in vivo, including vascular endothelial growth factor, fibroblast
growth factor, and insulin like growth factor-1 [14]. Recently soluble endothelial adhesion molecules, sVCAM-1 and E-selectin, were reported to promote angiogenesis [4]. If sVCAM-1 has proangiogenic function in humans, serum levels of sVCAM-1 might be increased in the diabetic patients with active neovascularization on retina. Interestingly, Olson et al. very recently reported that the serum sVCAM-1 and sE-selectin levels in IDDM patients with diabetic retinopathy were elevated and that retinal endothelial cell migration-inducing (ECMI) activity in vitro was increased, which could be blocked by antibodies to sVCAM-1 and sE-selectin in a subset of patients with diabetic retinopathy [15]. They evaluated the severity of diabetic retinopathy in detail, based on the presence of microaneurysms and haemorrhages, hard exudates, cotton wool spots, venous abnormalities, intraretinal microvascular abnormalities and new vessels by using the modified Airlie House technique [16], and pointed out the possibility that each soluble adhesion molecule might act on subtly different states of diabetic retinopathy, i.e. sE-selectin for preproliferative state and sVCAM-1 for proliferative state. In this study, we evaluated the diabetic retinopathy only three main stages, that is, NDR, BDR and PDR, and demonstrated that the mean levels of serum sVCAM-1 in diabetic patients with PDR were significantly increased. However it is likely that the PDR group might have included the diabetic patients with inactive stage of PDR. Although we could not divide the PDR group into two stages, active and inactive, our results indicated the possibility of sVCAM-1 having a pathogenetic role in the induction of angiogenesis on retina in consideration of Olson’s results [15]. One of the most intriguing concerns is a pathogenic role of increased sVCAM-1. Schmidt et al. [17] reported that sVCAM-1 from microalbuminuric diabetic plasma did not inhibit the binding of Molt-4 cells to tumor necrosis factor-a stimulated endothelial cells. Thus sVCAM-1 are not probably protective with respect to inhibiting the interaction of circulating mononuclear cells with activated endothelium. Although the potential functions of sVCAM-1 can be only the subject of speculation at present, sVCAM-1 is possibly
M. Yoshizawa et al. / Diabetes Research and Clinical Practice 42 (1998) 65–70
derived from membrane fragments of damaged endothelium or released from activated endothelium, thus serum sVCAM-1 levels are likely to reflect endothelial injury or activation. However, if increased levels of sVCAM-1 were only the results of vascular endothelial damage, sVCAM- 1 levels might be also increased in patients with diabetic nephropathy. Our results showed that there were no differences in serum sVCAM-1 levels between patients with or without nephropathy, while Schmidt et al. [17] reported the elevated levels of sVCAM-1, possibly due to vascular endothelial damage, in microalbuminuric diabetic patients. In this study, we used urine A/C ratios in the first voided morning urine for screening the microalbuminuria. It may be more reliable to use urinary albumin excretion rate by three-timed overnight urine collections. Therefore, such difference of the procedure for detection of microalbuminuria might partly explain this discrepancy of the results. Our results demonstrated that serum sVCAM-1 levels did not correlate with HbA1c levels. Since glycated hemoglobin is known to simply reflect the current glycemic control, sVCAM-1 levels need not to correlate with HbA1c levels. In agreement with Fascing et al. [18], appearance of sVCAM-1 may need additional metabolic aberration which contributes to vascular endothelial damage. Otsuki et al. has demonstrated the positive correlation between serum sVCAM-1 levels and thickness of the intimal plus medial complex (IMT) of the carotid arteries only in the 101 Japanese NIDDM patients but not in the nondiabetic subjects, and they indicated the circulating sVCAM-1 could be a marker of atherosclerotic lesions in NIDDM patients even with asymptomatic atherosclerosis [19]. Since atherosclerosis is characterized by endothelial cell injury and the adhesion molecules including VCAM-1 have been demonstrated to be expressed in atherosclerotic lesions, it is very important to diagnose atherosclerosis early. However the clinical exception of atherosclerosis is difficult in fact. We excluded the NIDDM patients only with symptomatic diabetic macroangiopathy but not those with asymptomatic macroangiopathy, thus sVCAM-1 levels in the NIDDM patients with
69
asymptomatic macroangiopathy might give some difficulty in the interpretation of our results. Actually, serum sVCAM levels in the NDR group in this study were higher compared with previous reports [15,19]. These differences may be due to the possibility that some patients with asymptomatic macroangiopathy or occult cancer were included. In conclusion, our study showed the elevation of serum sVCAM-1 levels in NIDDM patients with PDR. Although the functional significance remains unclear now, measurement of serum sVCAM-1 levels may be of potential benefit in the assessment of the process of diabetic proliferative retinopathy.
References [1] A.J.H. Gearing, I. Hemingway, R. Pigott, J. Hughes, A.J. Rees, S.J. Cashman, Soluble forms of vascular adhesion molecules, E-Selectin, ICAM-1, and VCAM-1, Ann. N.Y. Acad. Sci. 667 (1992) 324 – 331. [2] A.M. Schmidt, O. Hori, J.C. Chen, et al., Advanced glycation endproducts interacting with their endothelial receptor induce expression of vascular cell adhesion molecule-1 (VCAM-1) in cultured human endothelial cells and in mice, J. Clin. Invest. 96 (1995) 1395 – 1403. [3] L. Cominacini, P.A. Fratta, U. Garbin, et al., Elevated levels of soluble E-selectin in patients with IDDM and NIDDM, Diabetologia 38 (1995) 1122 – 1124. [4] A.E. Koch, M.M. Halloran, C.J. Haskel, M.R. Shah, P.J. Polverini, Angiogenesis mediated by soluble forms of E-selectin and vascular cell adhesion molecule-1, Nature 376 (1995) 517 – 519. [5] World Health Organization. (1985) Diabetes Mellitus: Report of a WHO study group. Geneva, World Health Org. 1985 (Tech. Rep. Ser. no. 727), 1 – 113. [6] A.D. Blann, W. Tse, S.J.R. Maxwell, M.A. Waite, Increased levels of the soluble adhesion molecule E-selectin in essential hypertension, J. Hypertens. 12 (1994) 925 – 928. [7] A. Hackman, Y. Abe, W.J. Insull, et al., Levels of soluble cell adhesion molecules in patients with dyslipidemia, Circulation 93 (1996) 1334 – 1338. [8] S.M. Grundy, D. Bilheimer, A. Chait, et al., Expert panel on detection, evaluation, and treatment of high blood cholesterol in adults: summary of the second report of the National Cholesterol Education Program (NCEP), J. Am. Med. Assoc. 269 (1993) 3015 – 3023. [9] S.M. Grundy, E. Barrett-Connor, E.L. Bierman, et al., Consensus Conference: Treatment of hypertriglyceridemia, J. Am. Med. Assoc. 251 (1984) 1196 – 1201.
70
M. Yoshizawa et al. / Diabetes Research and Clinical Practice 42 (1998) 65–70
[10] W.T. Friedewald, R.I. Levy, D.S. Fredrickson, Estimation of the concentration of low-density lipoprotein cholesterol in plasma without use of preparative ultracentrifuge, Clin. Chem. 18 (1972) 499–502. [11] C.E. Mogensen, W.F. Keane, P.H. Benett, et al., Prevention of diabetic renal disease with special references to microalbuminuria, Lancet 346 (1995) 1080–1084. [12] D. Giugliano, G. Paolliso, A. Ceriello, Oxidative stress and diabetic vascular complications, Diabetes Care 19 (1996) 257 – 267. [13] J. Bischoff, Cell adhesion and angiogenesis, J. Clin. Invest. 99 (1997) 373 –376. [14] P.S. Sharp, The role of growth factors in the developement of diabetic retinopathy, Metabolism 44 (Suppl 4) (1995) 72 – 75. [15] J.A. Oslon, C.M. Whitelaw, K.C. Mc Hardy, D.W.M. Peason, J.V. Forrester, Soluble leucocyte a and adhesion
[16]
[17]
[18]
[19]
.
molecules in diabetic retinopathy stimulate retinal capillary endothelial cell migration, Diabetologia 40 (1997) 1166 – 1171. B.E.K. Klein, M.D. Davis, P. Segal, et al., Diabetic retinopathy: Assessment of severity and progression, Ophthalmology 91 (1984) 10 – 17. A.M. Schmidt, J. Crandall, O. Hori, R. Cao, E. Lakatta, Elevated plasma levels of vascular cell adhesion molecule1 (VCAM-1) in diabetic patients with microalbuminuria, Br. J. Hematol. 92 (1996) 747 – 750. P. Fasching, Elevated circulating adhesion molecules in NIDDM-potential mediators in diabetic macroangiopathy, Diabetologia 39 (1996) 1242 – 1244. M. Otsuki, K. Hashimoto, Y. Morimoto, T. Kishimoto, S. Kasayama, Circulating vascular cell adhesion molecule-1 (VCAM-1) in atherosclerotic NIDDM patients, Diabetes 46 (1997) 2096 – 2101.
Elevated serum levels of soluble vascular cell adhesion molecule-1 in NIDDM patients with proliferative diabetic retinopathy Miyako Yoshizawa a,*, Yukihiro Nagai a, Kensou Ohsawa a, Masayoshi Ohta a, Haruhisa Yamashita a, Azusa Hisada a, Ichiro Miyamoto b, Katsuyuki Miura c, Toshinari Takamura a, Ken-ichi Kobayashi a a
First Department of Internal Medicine, School of Medicine, Kanazawa Uni6ersity. 13 -1 Takara-machi, Kanazawa, Ishikawa 920 -8640, Japan b The Department of Internal Medicine, Toyama Rosai Hospital. 992 Rokuroumaru, Uozu, Toyama 937 -0042, Japan c The Department of Public Health, Kanazawa Medical Uni6ersity. 1 -1 Daigaku, Uchinada, Ishikawa 920 -0265, Japan Received 15 August 1997; received in revised form 24 June 1998; accepted 12 August 1998
Abstract We studied 68 Japanese NIDDM patients (38 men and 30 women), aged 56.9 9 1.2 years (range 33 – 75 years), with a BMI of 23.190.5 kg/m2 without hypertension, dyslipidemia, and diabetic macroangiopathy for evaluating the relationship between serum soluble vascular cell adhesion molecule-1 (sVCAM-1) levels and the severity of diabetic retinopathy. Fundus examination was performed by an ophthalmologist using an ophthalmoscope, and the findings were graded as: (1) no signs of diabetic retinopathy (NDR), (2) background diabetic retinopathy (BDR), or (3) proliferative diabetic retinopathy (PDR). Serum sVCAM-1 levels were measured in duplicate by enzyme-linked immunosorbent assay using the soluble VCAM-1 KIT (R&D Systems Ltd., Ablingdon, Oxfordshire, UK). There was no difference in serum sVCAM-1 levels between patients with BDR (n=17) and patients with NDR (n= 40) (1035.3 9104.4 and 978.8 948.9 ng/ml, respectively, P= 0.8), but patients with PDR (n= 11) showed a significant increase of serum sVCAM-1 levels compared with patients with NDR (1281.8 9166.3 and 978.8 9 48.9 ng/ml, respectively, P =0.02). Although serum sVCAM-1 levels were correlated, not only with age but also with the known diabetic duration (r =0.39, P =0.001, and r= 0.40, P= 0.0007, respectively), age-adjusted sVCAM-1 levels were still significantly higher in the PDR group than in the NDR group. In contrast, serum sVCAM-1 levels were not related to the presence of diabetic nephropathy or HbA1c levels. Our results suggest that sVCAM-1 might be implicated in the development of the diabetic retinopathy, and measurement of serum sVCAM-1 levels in NIDDM patients may
* Corresponding author. Tel.: +81 76 2652234; fax: +81 76 2344250. 0168-8227/98/$ - see front matter © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII: S0168-8227(98)00091-6
66
M. Yoshizawa et al. / Diabetes Research and Clinical Practice 42 (1998) 65–70
be clinically useful for assessing the severity and possibly the activity of diabetic retinopathy. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: VCAM-1; Diabetes mellitus; Diabetic retinopathy
1. Introduction
2. Materials and methods
Vascular cell adhesion molecule-1 (VCAM-1), a member of the immunoglobulin superfamily which contains immunoglobulin-like structures, is a 110 kD cell surface glycoprotein, and plays a major role in the adhesion of leukocytes to the endothelium by interaction with its ligand very late antigen-4 (VLA-4), a member of the b1 integrin family, which is expressed on lymphocytes and monocytes [1]. Recently, Schmidt et al. [2] reported that the advanced glycation endproducts (AGEs) interacting with a cellular receptor for AGE (RAGE) resulted in enhanced cellular oxidant stress with consequent changes in multiple parameters, and increased expression of endothelial cell-associated VCAM-1 as well as increased release of soluble VCAM-1 (sVCAM-1) into culture supernatants. Cominacini et al. [3] also suggested that the metabolic control of the diabetic patients may affect the endothelial cell activity since elevated levels of soluble E-selectin were positively correlated with glycated hemoglobin levels. On the other hand, Koch et al. [4] showed that soluble forms of VCAM-1 and E-selectin induced angiogenesis in rat cornea. sVCAM-1 acts on endothelial cells in part through a VLA-4 dependent mechanism and induces chemotaxis of endothelial cells without stimulating proliferation of endothelial cells in vitro [4]. They suggested an intriguing possibility that shed adhesion molecules might stimulate angiogenesis in vivo. If endothelial cell-associated VCAM-1 is up-regulated by AGE – RAGE interaction, leading to increased circulating levels of sVCAM-1 in diabetic patients, and sVCAM-1 has proangiogenic functions, serum sVCAM-1 levels in diabetic patients with proliferative retinopathy might be increased. In this study, we aimed to investigate the relationship between serum sVCAM-1 levels and the severity of diabetic retinopathy in NIDDM patients.
We studied 68 NIDDM patients (38 men and 30 women) without diabetic symptomatic macroangiopathy, aged 56.99 1.2 years (range 33–75 years), with a BMI of 23.19 0.5 kg/m2. The known diabetic duration was calculated as the duration from the initial diagnosis of NIDDM as defined by World Health Organization criteria [5]. Diabetic symptomatic macroangiopathy was defined as the clinically symptomatic history of ischemic heart disease, cerebral infarction, or arteriosclerosis obliterans. Twenty-one patients were treated by insulin, 29 with oral hypoglycemic agents and 18 with diet alone. Since soluble forms of cell adhesion molecules were reported to increase in some metabolic disorders including hypertension [6] and dyslipidemia [7], patients with hypertension and dyslipidemia were excluded; hypertension was defined by either prior diagnosis, current treatment with antihypertensive agent, or blood pressure greater than 160/ 95 mmHg. Dyslipidemia was considered to be present when patients had serum total cholesterol \ 240 mg/dl [8], and/or triglyceride \ 250 mg/dl [9] and/or HDL cholesterol B 40 mg/dl in women and B 35 mg/dl in men. In all patients, renal insufficiency, liver diseases, malignancies, and chronic inflammatory diseases and pathological processes known to increase adhesion molecules were excluded. The study was approved by our local ethics committee, and informed consent was obtained from each subject. Venous blood was collected by standard venipuncture using vacutainer tubes at 8:00 a.m. after a 12-h fast from all subjects. Serum was extracted following 10-min centrifugation in a bench centrifuge at 2500 rev/min and stored at − 20°C until assayed. HbA1c was measured by affinity chromatography (HA-8131, Kyoto Daiichi Kagaku, Kyoto, Japan, normal range 4.3– 5.8%). Serum creatinine, total cholesterol, and
M. Yoshizawa et al. / Diabetes Research and Clinical Practice 42 (1998) 65–70
67
Table 1 Serum sVCAM-1 levels and clinical characteristics of NIDDM patients
Number (male/female) Age (years) BMI (kg/m2) Known DM duration (years) HbA1c (%) Serum creatinine (mg/dl) Total cholesterol (mg/dl) HDL-cholesterol (mg/dl) LDL-cholesterol (mg/dl) Triglyceride (mg/dl) sVCAM-1 (ng/ml) sVCAM-1 (age adjusted) sVCAM-1 (known DM duration adjusted)
No retinopathy
Background retinopathy
Proliferative retinopathy
40 (24/16) 54.99 1.7 23.29 0.7 3.2 9 0.5 7.49 0.3 0.89 0.02 199.59 4.3 49.89 2.1 126.99 4.6 113.59 10.2 978.89 48.9 1010.99 57.1 1093.09 70.9
17 (7/10) 61.4 91.5* 23.2 90.6 9.2 9 1.2* 7.29 0.4 0.9 90.1 196.4 9 5.2 57.4 93.8 116.2 94.6 113.6 916.6 1035.3 9 104.4 959.7 989.1 946.4 9 94.6
11 (7/4) 56.9 92.6 22.5 91.1 14.8 9 1.3* 8.3 9 0.7 0.9 9 0.1 181.4 9 15.5 50.9+4.4 106.1 912.3 121.5 917.6 1281.8 9166.3* 1281.4 9 107.6* 1003.8 9 148.6
Values are expressed as n or means9 S.E.M. *PB0.05, vs no retinopathy group, by analysis of variance.
triglyceride concentrations were measured by standard laboratory methods. HDL cholesterol was isolated by sequential ultracentrifugation in NaBr solutions, and LDL cholesterol was calculated with the equation of Friedewald et al. [10]. Serum sVCAM-1 levels were measured in duplicate by enzyme-linked immunosorbent assay using the soluble VCAM-1 KIT (R&D Systems Ltd., Ablingdon, Oxfordshire, UK), with both inter- and intra-assay coefficients of variations are 9.2 and 5.9%, respectively. There is a consensus that the screening of microalbuminuria can easily be done by measuring urine albumin-creatinine ratio (A/C ratio) [11], so that we measured A/C ratios in early morning urine (Olympus AU 800, Tokyo, Japan, normal range B18.0 mg/gCr) to evaluate the presence of diabetic nephropathy. The patients with an A/C ratio \30 mg/gCr were defined to have diabetic nephropathy. The study subjects were divided into three groups according to the severity of diabetic retinopathy. After mydriasis, fundus examination was performed by an ophthalmologist using an ophthalmoscope. The findings were graded as: (1) no signs of diabetic retinopathy (NDR); (2) background diabetic retinopathy (BDR); or (3) proliferative diabetic retinopathy (PDR). Statistical differences between three groups
were evaluated by analysis of variance, using Bonferroni’s test for subsequent comparisons. Values are expressed as means9 S.E.M. A 2-tailed Pvalue less than 0.05 was considered to be statistically significant. Statistical analyses were performed using the Statistical Analysis System (SAS Institute Inc., Cary, North Carolina, USA). 3. Results Table 1 shows the mean values of the anthropometric characteristics, glycated hemoglobin, serum total cholesterol, HDL cholesterol, LDL cholesterol, triglyceride, and creatinine in NIDDM patients divided into three groups based on the severity of diabetic retinopathy. There were no significant differences in BMI, serum creatinine, total cholesterol, HDL cholesterol, LDL cholesterol, or triglyceride levels among the three groups. Age of the BDR group showed relatively older than that of the NDR group, but was not different from that of the PDR group significantly. As expected, the PDR group had a significantly longer diabetic duration than both groups of NDR and BDR. As shown in Table 1, there was no difference in serum sVCAM-1 levels between BDR and NDR groups (P= 0.6), but the PDR group showed a significant increase of serum sVCAM-1 levels
68
M. Yoshizawa et al. / Diabetes Research and Clinical Practice 42 (1998) 65–70
compared with the NDR group (P = 0.02). The serum sVCAM-1 levels in all NIDDM patients significantly correlated, not only with age but also with the known diabetic duration (r = 0.39, P = 0.01 and r =0.40, P =0.0007, respectively), thus the mean values of serum sVCAM-1 levels, which were adjusted for age and the known diabetic duration, were calculated using analysis of covariance. The age-adjusted sVCAM-1 levels of the PDR group were still significantly higher than those of the NDR group, but the sVCAM-1 levels adjusted for the known diabetic duration showed no difference between three groups (Table 1). In contrast, the mean values of serum sVCAM1 levels in patients with diabetic nephropathy (n =24) were not different than those in patients without nephropathy (n = 44) (1056.2989.1 and 1034.19 56.9 ng/ml, respectively, P= 0.8). In addition, serum sVCAM-1 levels were not correlated with HbA1c levels (r =0.02, P =0.9).
4. Discussion In this study, we demonstrated the significant elevation of serum sVCAM-1 levels in NIDDM patients with PDR, even after adjustment for age. In contrast, serum sVCAM-1 levels were not related to the presence of diabetic nephropathy or HbA1c levels. Intraocular neovascularization occurs in ischemic retinal disorders, especially diabetic retinopathy, and often results in vitreous hemorrhage, retinal detachment, and neovascular glaucoma, with subsequent visual loss. Recently, it has been suggested that oxidative stress may play an important role in the development of diabetic retinopathy [12], promoting intraocular neovascularization characterized in proliferative diabetic retinopathy. Endothelium is likely to play a major role in angiogenesis, thus the investigation of the basic mechanisms by which endothelial cells are stimulated to undergo angiogenesis is attractive to identify endogenous angiogenetic inhibitors and to use these molecules to control angiogenesis [13]. The numerous growth factors have been known to stimulate angiogenesis in vivo, including vascular endothelial growth factor, fibroblast
growth factor, and insulin like growth factor-1 [14]. Recently soluble endothelial adhesion molecules, sVCAM-1 and E-selectin, were reported to promote angiogenesis [4]. If sVCAM-1 has proangiogenic function in humans, serum levels of sVCAM-1 might be increased in the diabetic patients with active neovascularization on retina. Interestingly, Olson et al. very recently reported that the serum sVCAM-1 and sE-selectin levels in IDDM patients with diabetic retinopathy were elevated and that retinal endothelial cell migration-inducing (ECMI) activity in vitro was increased, which could be blocked by antibodies to sVCAM-1 and sE-selectin in a subset of patients with diabetic retinopathy [15]. They evaluated the severity of diabetic retinopathy in detail, based on the presence of microaneurysms and haemorrhages, hard exudates, cotton wool spots, venous abnormalities, intraretinal microvascular abnormalities and new vessels by using the modified Airlie House technique [16], and pointed out the possibility that each soluble adhesion molecule might act on subtly different states of diabetic retinopathy, i.e. sE-selectin for preproliferative state and sVCAM-1 for proliferative state. In this study, we evaluated the diabetic retinopathy only three main stages, that is, NDR, BDR and PDR, and demonstrated that the mean levels of serum sVCAM-1 in diabetic patients with PDR were significantly increased. However it is likely that the PDR group might have included the diabetic patients with inactive stage of PDR. Although we could not divide the PDR group into two stages, active and inactive, our results indicated the possibility of sVCAM-1 having a pathogenetic role in the induction of angiogenesis on retina in consideration of Olson’s results [15]. One of the most intriguing concerns is a pathogenic role of increased sVCAM-1. Schmidt et al. [17] reported that sVCAM-1 from microalbuminuric diabetic plasma did not inhibit the binding of Molt-4 cells to tumor necrosis factor-a stimulated endothelial cells. Thus sVCAM-1 are not probably protective with respect to inhibiting the interaction of circulating mononuclear cells with activated endothelium. Although the potential functions of sVCAM-1 can be only the subject of speculation at present, sVCAM-1 is possibly
M. Yoshizawa et al. / Diabetes Research and Clinical Practice 42 (1998) 65–70
derived from membrane fragments of damaged endothelium or released from activated endothelium, thus serum sVCAM-1 levels are likely to reflect endothelial injury or activation. However, if increased levels of sVCAM-1 were only the results of vascular endothelial damage, sVCAM- 1 levels might be also increased in patients with diabetic nephropathy. Our results showed that there were no differences in serum sVCAM-1 levels between patients with or without nephropathy, while Schmidt et al. [17] reported the elevated levels of sVCAM-1, possibly due to vascular endothelial damage, in microalbuminuric diabetic patients. In this study, we used urine A/C ratios in the first voided morning urine for screening the microalbuminuria. It may be more reliable to use urinary albumin excretion rate by three-timed overnight urine collections. Therefore, such difference of the procedure for detection of microalbuminuria might partly explain this discrepancy of the results. Our results demonstrated that serum sVCAM-1 levels did not correlate with HbA1c levels. Since glycated hemoglobin is known to simply reflect the current glycemic control, sVCAM-1 levels need not to correlate with HbA1c levels. In agreement with Fascing et al. [18], appearance of sVCAM-1 may need additional metabolic aberration which contributes to vascular endothelial damage. Otsuki et al. has demonstrated the positive correlation between serum sVCAM-1 levels and thickness of the intimal plus medial complex (IMT) of the carotid arteries only in the 101 Japanese NIDDM patients but not in the nondiabetic subjects, and they indicated the circulating sVCAM-1 could be a marker of atherosclerotic lesions in NIDDM patients even with asymptomatic atherosclerosis [19]. Since atherosclerosis is characterized by endothelial cell injury and the adhesion molecules including VCAM-1 have been demonstrated to be expressed in atherosclerotic lesions, it is very important to diagnose atherosclerosis early. However the clinical exception of atherosclerosis is difficult in fact. We excluded the NIDDM patients only with symptomatic diabetic macroangiopathy but not those with asymptomatic macroangiopathy, thus sVCAM-1 levels in the NIDDM patients with
69
asymptomatic macroangiopathy might give some difficulty in the interpretation of our results. Actually, serum sVCAM levels in the NDR group in this study were higher compared with previous reports [15,19]. These differences may be due to the possibility that some patients with asymptomatic macroangiopathy or occult cancer were included. In conclusion, our study showed the elevation of serum sVCAM-1 levels in NIDDM patients with PDR. Although the functional significance remains unclear now, measurement of serum sVCAM-1 levels may be of potential benefit in the assessment of the process of diabetic proliferative retinopathy.
References [1] A.J.H. Gearing, I. Hemingway, R. Pigott, J. Hughes, A.J. Rees, S.J. Cashman, Soluble forms of vascular adhesion molecules, E-Selectin, ICAM-1, and VCAM-1, Ann. N.Y. Acad. Sci. 667 (1992) 324 – 331. [2] A.M. Schmidt, O. Hori, J.C. Chen, et al., Advanced glycation endproducts interacting with their endothelial receptor induce expression of vascular cell adhesion molecule-1 (VCAM-1) in cultured human endothelial cells and in mice, J. Clin. Invest. 96 (1995) 1395 – 1403. [3] L. Cominacini, P.A. Fratta, U. Garbin, et al., Elevated levels of soluble E-selectin in patients with IDDM and NIDDM, Diabetologia 38 (1995) 1122 – 1124. [4] A.E. Koch, M.M. Halloran, C.J. Haskel, M.R. Shah, P.J. Polverini, Angiogenesis mediated by soluble forms of E-selectin and vascular cell adhesion molecule-1, Nature 376 (1995) 517 – 519. [5] World Health Organization. (1985) Diabetes Mellitus: Report of a WHO study group. Geneva, World Health Org. 1985 (Tech. Rep. Ser. no. 727), 1 – 113. [6] A.D. Blann, W. Tse, S.J.R. Maxwell, M.A. Waite, Increased levels of the soluble adhesion molecule E-selectin in essential hypertension, J. Hypertens. 12 (1994) 925 – 928. [7] A. Hackman, Y. Abe, W.J. Insull, et al., Levels of soluble cell adhesion molecules in patients with dyslipidemia, Circulation 93 (1996) 1334 – 1338. [8] S.M. Grundy, D. Bilheimer, A. Chait, et al., Expert panel on detection, evaluation, and treatment of high blood cholesterol in adults: summary of the second report of the National Cholesterol Education Program (NCEP), J. Am. Med. Assoc. 269 (1993) 3015 – 3023. [9] S.M. Grundy, E. Barrett-Connor, E.L. Bierman, et al., Consensus Conference: Treatment of hypertriglyceridemia, J. Am. Med. Assoc. 251 (1984) 1196 – 1201.
70
M. Yoshizawa et al. / Diabetes Research and Clinical Practice 42 (1998) 65–70
[10] W.T. Friedewald, R.I. Levy, D.S. Fredrickson, Estimation of the concentration of low-density lipoprotein cholesterol in plasma without use of preparative ultracentrifuge, Clin. Chem. 18 (1972) 499–502. [11] C.E. Mogensen, W.F. Keane, P.H. Benett, et al., Prevention of diabetic renal disease with special references to microalbuminuria, Lancet 346 (1995) 1080–1084. [12] D. Giugliano, G. Paolliso, A. Ceriello, Oxidative stress and diabetic vascular complications, Diabetes Care 19 (1996) 257 – 267. [13] J. Bischoff, Cell adhesion and angiogenesis, J. Clin. Invest. 99 (1997) 373 –376. [14] P.S. Sharp, The role of growth factors in the developement of diabetic retinopathy, Metabolism 44 (Suppl 4) (1995) 72 – 75. [15] J.A. Oslon, C.M. Whitelaw, K.C. Mc Hardy, D.W.M. Peason, J.V. Forrester, Soluble leucocyte a and adhesion
[16]
[17]
[18]
[19]
.
molecules in diabetic retinopathy stimulate retinal capillary endothelial cell migration, Diabetologia 40 (1997) 1166 – 1171. B.E.K. Klein, M.D. Davis, P. Segal, et al., Diabetic retinopathy: Assessment of severity and progression, Ophthalmology 91 (1984) 10 – 17. A.M. Schmidt, J. Crandall, O. Hori, R. Cao, E. Lakatta, Elevated plasma levels of vascular cell adhesion molecule1 (VCAM-1) in diabetic patients with microalbuminuria, Br. J. Hematol. 92 (1996) 747 – 750. P. Fasching, Elevated circulating adhesion molecules in NIDDM-potential mediators in diabetic macroangiopathy, Diabetologia 39 (1996) 1242 – 1244. M. Otsuki, K. Hashimoto, Y. Morimoto, T. Kishimoto, S. Kasayama, Circulating vascular cell adhesion molecule-1 (VCAM-1) in atherosclerotic NIDDM patients, Diabetes 46 (1997) 2096 – 2101.