- Email: [email protected]
Possible contribution of adipocytokines on diabetic neuropathy
Diabetes Research and Clinical Practice 66 (2004) S121–S123
Possible contribution of adipocytokines on diabetic neuropathy M. Matsuda∗ , F. Kawasaki, H. Inoue, Y. Kanda, K. Yamada, Y. Harada, M. Saito, M. Eto, M. Matsuki, K. Kaku Diabetes and Endocrine Division, Department of Medicine, Kawasaki Medical School, 577 Matsushima, Kurashiki-shi, Okayama-ken 701 0192, Japan
Abstract Neuropathy is one of the typical features of chronic complications of diabetes mellitus. Recent analyses indicate that subjects with impaired glucose tolerance (IGT) already have disturbance of peripheral nerve function. To test the role of adipocytokines, that tend to be abnormal in IGT subjects, on diabetic neuropathy, we analyzed the relationship between plasma adipocytokine levels (TNF␣, adiponectin, and leptin) and nerve conduction velocity in 105 type 2 diabetic subjects (M/F = 66/39, age = 60.8 ± 11.8 years, BMI = 24.7 ± 5.0 kg/m2 ). Adipocytokines were measured by ELISA, and motor conduction velocity (MCV) and sensory conduction velocity (SCV) in median, ulnar, and tibial nerve were measured by electrical stimulation. Motor conduction velocity and SCV were corrected by age to be 1.0 as the normal value, and the average of three nerves were used to be the representative value. Relationship between corrected MCV or corrected SCV as a dependent variable and the duration of diabetes, HbA1C , BMI, TNF␣, adiponectin, and leptin concentrations as independent variables were analyzed by multiple regression. Duration of diabetes and HbA1C were highly related with both corrected MCV (P < 0.02 and P < 0.001) and SCV (P < 0.02 and P < 0.05) by this analysis. Only corrected SCV was related significantly with TNF␣ (P < 0.05), and close to significantly with leptin (P = 0.059) concentrations. These results indicate that increased plasma glucose levels and duration of diabetes are the major factors that modulate diabetic neuropathy. However, nerve function may be affected by plasma cytokine levels like TNF␣, and this effect was more significant on sensory nerves than motor nerves. The present results suggest that adipocytokines may play a role not only on angiopathy but also on neuropathy in diabetics. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Adipocytokine; Diabetic neuropathy; TNF␣
1. Introduction Neuropathy is one of typical features of chronic complications of diabetes mellitus. Recent analyses indicate that subjects with impaired glucose tolerance ∗ Corresponding author. Tel.: +81 86 462 1111; fax: +81 86 462 1199. E-mail address: [email protected] (M. Matsuda).
(IGT) already have disturbance of peripheral nerve function [1]. In this early phase of disturbance of glucose homeostasis, in addition to slight increase of glucose concentrations, accumulation of fat tissue is a feature of some subjects with impaired glucose tolerance who have insulin resistance [2], and known as insulin resistance syndrome [3]. Naturally, fully developed type 2 diabetic subjects are insulin resistance. Fat tissue secretes many kinds of substances including TNF␣, adiponectin, and leptin which are called
0168-8227/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2004.05.010
S122
M. Matsuda et al. / Diabetes Research and Clinical Practice 66 (2004) S121–S123
adipocytokines in addition to free fatty acid [4]. Since adipocytokines have biological activities, it is possible that adipocytokines may contribute to the development of peripheral nerve dysfunction. TNF␣ has been demonstrated to have impact on peripheral nervous system in vitro perfusion environment [5]. In this report, we tested the role of adipocytokines, which tend to be abnormal in IGT subjects, on diabetic neuropathy in type 2 diabetic subjects.
2. Methods 2.1. Subjects Subjects of this study were type 2 diabetic subjects who were admitted to the Kawasaki Medical School Hospital from 1999 to 2002 for the purpose of diabetes education and evaluation of diabetic complications. The determination of diabetes mellitus was based on World Health Organization criteria [6]. Informed consent for all examinations was obtained from the subjects studied on admission. The onset of diabetes mellitus of all subjects was at the age of 40 years or later, and none had anti-GAD antibody. Blood was taken before breakfast of the second day of hospitalization for measurements of plasma glucose, insulin, adiponectin, TNF␣, leptin, total cholesterol, HDL-cholesterol, and serum triglycerides. The blood samples for the measurement of plasma glucose, insulin, HbA1C , and lipid/cholesterol levels were sent to the hospital laboratory for the assay. LDL-cholesterol was calculated by the Friedewald’s equation [7], and none had a TG value that was greater than 400 mg/dl. Blood pressure was measured with a standard mercury sphygmomanometer placed on the arm while subjects were seated at least for 5 min. In this analysis all subjects did not have any significant kidney diseases except for diabetic nephropathy. 2.2. Nerve function test To assess the peripheral nerve function, we measured motor nerve conduction velocity (MCV) and sensory nerve conduction velocity (SCV), by electric stimulation [8]. We assessed the median, ulnar, and tibial nerves. MCV and SCV were corrected by age, and expressed as the ratio of this corrected velocity to
the standard value of the same age of a subject. The average of ratios of three nerves, that we measured, was used as the index of age-corrected MCV or SCV [10]. 2.3. Assays Plasma adiponectin concentration was measured by an ELISA kit developed by Otsuka Assay (Tokushima, Japan). TNF␣ concentration was determined by QuantiGlo human TNF␣ Chemiluminescent Immunoassay (R&D systems, Inc., Minneapolis, MN, USA). Plasma leptin concentration was measured by an ELISA kit developed by Mitsubishi Biochemial Laboratory (Tokyo, Japan). 2.4. Data analysis All data are expressed as means with S.D. Relationship between two variables is analyzed by standard correlation analysis conducted by StatView version 5.0 (SAS, Cary, NC, USA). Multivariate analysis was performed using a multiple variable regression model to determine the standard coefficients to the index of neuropathy. Statistical significance was accepted when P < 0.05.
3. Results and discussion We studied 105 subjects with 66 males and 39 females. The average age was 60.8 years old with the range of 26–84 years. The mean duration of diabetes mellitus was 10.4 years. The mean BMI was 24.7 kg/m2 , and the mean HbA1C on admission was 8.9%. None was under the medication of epalrestat. Motor conduction velocity and SCV were corrected by age to be 1.0 as the normal value, and the average of three nerves were used to be the representative value, and the mean values of the index of corrected MCV and SCV were 0.96 ± 0.07 and 0.95 ± 0.08 individually. While mean values of MCV and SCV were slightly lower than 1.0, the ranges of MCV and SCV were 0.75–1.13 and 0.69–1.25 individually. Mean leptin, TNF␣, and adiponectin concentrations in this study subjects were not different from the standard values published previously, but plasma levels of these adipocytokines varied in each individual in the study subjects.
M. Matsuda et al. / Diabetes Research and Clinical Practice 66 (2004) S121–S123
The simple regression analysis between the index of age-corrected MCV and SCV revealed that MCV and SCV were strongly significantly related (P < 0.05) with duration of diabetes mellitus and HbA1C as expected. Among adipocytokines, leptin and TNF␣ seemed to be related with SCV. The data that we analyzed were observation of cross-sectional features of study subjects. To eliminate un-controlled features in the study subjects, we applied multiple-regression analysis to figure out contribution of variables on nerve disturbances. Again, we detected a significant negative contribution of TNF␣ on sensory conduction velocity with standard coefficient of −0.189 (P < 0.05). There was a negative correlation (r = −0.225, P < 0.05) between plasma TNF␣ levels and the index of age-corrected SCV. Increased plasma glucose levels and duration of diabetes are the major factors that modulate diabetic neuropathy [9]. However, a certain part of subjects with impaired glucose tolerance have a predromal phenomenon of diabetic painful neuropathy [10,11]. The disturbance in autonomic nervous system in subjects with impaired glucose tolerant has been reported [13,14]. Since increment of plasma glucose levels could be negligible, some attribute insulinopenia from the GK-rat study [12]. Meanwhile, nerve function may be affected by plasma cytokine levels like TNF␣, as shown in an experimental condition [5]. This possible mechanism of this adipocytokine on peripheral nerve system would be applied in type 2 diabetic subjects who have a characteristic feature of insulin resistance. In our analysis TNF␣ has been proven to be related with index of sensory nerve function in type 2 diabetic subjects, and this effect was more significant on sensory nerves than motor nerves, and this coincides with the result reported by Sumner et al. [1]. Thus, the precise mechanism that makes progress in peripheral nerve system is still to be clarified, one of the possible mechanisms must be related with increased plasma TNF␣ levels or inflammatory process in the body. We conclude that adipocytokines may play a role not only on angiopathy but also on neuropathy in patients with type 2 diabetes mellitus. Acknowledgements This study was supported by Grant-In-Aid for scientific research (No. 13671205) from the Japan Soci-
S123
ety for the Promotion of Science, and Research Project Grants (Nos. 12-506, 12-507, 13-507) from Kawasaki Medical School.
References [1] C.J. Sumner, S. Sheth, J.W. Griffin, D.R. Cornblath, M. Polydefkis, The spectrum of neuropathy in diabetes and impaired glucose tolerance, Neurology 60 (2003) 108– 111. [2] W.Y. Fujimoto, R.W. Bergstrom, D.L. Leonetti, L.L. NewellMorris, W.P. Shuman, P.W. Wahl, Metabolic and adipose risk factors for NIDDM and coronary disease in third-generation Japanese-American men and women with impaired glucose tolerance, Diabetologia 37 (1994) 524–532. [3] J.B. Meigs, R.B. D’Agostino Sr, P.W. Wilson, L.A. Cupples, D.M. Nathan, D.E. Singer, The Framingham Offspring Study. Risk variable clustering in the insulin resistance syndrome, Diabetes 46 (1997) 1594–1600. [4] I. Shimomura, T. Funahashi, Y. Matsuzawa, Significance of adipocytokine, fat-derived hormones, in metabolic syndrome, Tanpakushitsu Kakusan Koso 47 (2002) 1896–1903. [5] V.E. Drory, D. Lev, G.B. Groozman, M. Gutmann, J.M. Klausner, Neurotoxicity of isolated limb perfusion with tumor necrosis factor, J. Neurol. Sci. 158 (1998) 1–4. [6] World Health Organization. Report of a WHO Consultation. Part 1.: Diagnosis and Classification of Diabetes Mellutis, Geneva, 1999, WHO/NCD/NCS/99.2. [7] W.T. Friedewald, R.I. Levy, D.S. Fredrickson, Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge, Clin. Chem. 18 (1972) 499–502. [8] G. Gregersen, Diabetic neuropathy: influence of age, sex, metabolic control, and duration of diabetes on motor conduction velocity, Neurology 17 (1967) 972–980. [9] M.A. Pfeifer, M.P. Schumer, Clinical trials of diabetic neuropathy: past, present, and future, Diabetes 44 (1995) 1355–1361. [10] J.R. Singleton, A.G. Smith, M.B. Bromberg, Painful sensory polyneuropathy associated with impaired glucose tolerance, Muscle Nerve 24 (2001) 1225–1228. [11] J.R. Singleton, A.G. Smith, M.B. Bromberg, Increased prevalence of impaired glucose tolerance in patients with painful sensory neuropathy, Diabetes Care 24 (2001) 1448– 1453. [12] Y. Murakawa, W. Zhang, C.R. Pierson, T. Brismar, C.G. Ostenson, S. Efendic, et al., Impaired glucose tolerance and insulinopenia in the GK-rat causes peripheral neuropathy, Diabetes Metab. Res. Rev. 18 (2002) 473–483. [13] K. Suzuki, Neurological disorders associated with impaired glucose tolerance, Nippon Rinsho 54 (1996) 2704–2708. [14] K.F. Rezende, A. Melo, J. Pousada, Z.F. Rezende, N.L. Santos, I. Gomes, Autonomic neuropathy in patients with impaired glucose tolerance, Arq. Neuropsiquiatr. 55 (1997) 703–711.
Possible contribution of adipocytokines on diabetic neuropathy M. Matsuda∗ , F. Kawasaki, H. Inoue, Y. Kanda, K. Yamada, Y. Harada, M. Saito, M. Eto, M. Matsuki, K. Kaku Diabetes and Endocrine Division, Department of Medicine, Kawasaki Medical School, 577 Matsushima, Kurashiki-shi, Okayama-ken 701 0192, Japan
Abstract Neuropathy is one of the typical features of chronic complications of diabetes mellitus. Recent analyses indicate that subjects with impaired glucose tolerance (IGT) already have disturbance of peripheral nerve function. To test the role of adipocytokines, that tend to be abnormal in IGT subjects, on diabetic neuropathy, we analyzed the relationship between plasma adipocytokine levels (TNF␣, adiponectin, and leptin) and nerve conduction velocity in 105 type 2 diabetic subjects (M/F = 66/39, age = 60.8 ± 11.8 years, BMI = 24.7 ± 5.0 kg/m2 ). Adipocytokines were measured by ELISA, and motor conduction velocity (MCV) and sensory conduction velocity (SCV) in median, ulnar, and tibial nerve were measured by electrical stimulation. Motor conduction velocity and SCV were corrected by age to be 1.0 as the normal value, and the average of three nerves were used to be the representative value. Relationship between corrected MCV or corrected SCV as a dependent variable and the duration of diabetes, HbA1C , BMI, TNF␣, adiponectin, and leptin concentrations as independent variables were analyzed by multiple regression. Duration of diabetes and HbA1C were highly related with both corrected MCV (P < 0.02 and P < 0.001) and SCV (P < 0.02 and P < 0.05) by this analysis. Only corrected SCV was related significantly with TNF␣ (P < 0.05), and close to significantly with leptin (P = 0.059) concentrations. These results indicate that increased plasma glucose levels and duration of diabetes are the major factors that modulate diabetic neuropathy. However, nerve function may be affected by plasma cytokine levels like TNF␣, and this effect was more significant on sensory nerves than motor nerves. The present results suggest that adipocytokines may play a role not only on angiopathy but also on neuropathy in diabetics. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Adipocytokine; Diabetic neuropathy; TNF␣
1. Introduction Neuropathy is one of typical features of chronic complications of diabetes mellitus. Recent analyses indicate that subjects with impaired glucose tolerance ∗ Corresponding author. Tel.: +81 86 462 1111; fax: +81 86 462 1199. E-mail address: [email protected] (M. Matsuda).
(IGT) already have disturbance of peripheral nerve function [1]. In this early phase of disturbance of glucose homeostasis, in addition to slight increase of glucose concentrations, accumulation of fat tissue is a feature of some subjects with impaired glucose tolerance who have insulin resistance [2], and known as insulin resistance syndrome [3]. Naturally, fully developed type 2 diabetic subjects are insulin resistance. Fat tissue secretes many kinds of substances including TNF␣, adiponectin, and leptin which are called
0168-8227/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2004.05.010
S122
M. Matsuda et al. / Diabetes Research and Clinical Practice 66 (2004) S121–S123
adipocytokines in addition to free fatty acid [4]. Since adipocytokines have biological activities, it is possible that adipocytokines may contribute to the development of peripheral nerve dysfunction. TNF␣ has been demonstrated to have impact on peripheral nervous system in vitro perfusion environment [5]. In this report, we tested the role of adipocytokines, which tend to be abnormal in IGT subjects, on diabetic neuropathy in type 2 diabetic subjects.
2. Methods 2.1. Subjects Subjects of this study were type 2 diabetic subjects who were admitted to the Kawasaki Medical School Hospital from 1999 to 2002 for the purpose of diabetes education and evaluation of diabetic complications. The determination of diabetes mellitus was based on World Health Organization criteria [6]. Informed consent for all examinations was obtained from the subjects studied on admission. The onset of diabetes mellitus of all subjects was at the age of 40 years or later, and none had anti-GAD antibody. Blood was taken before breakfast of the second day of hospitalization for measurements of plasma glucose, insulin, adiponectin, TNF␣, leptin, total cholesterol, HDL-cholesterol, and serum triglycerides. The blood samples for the measurement of plasma glucose, insulin, HbA1C , and lipid/cholesterol levels were sent to the hospital laboratory for the assay. LDL-cholesterol was calculated by the Friedewald’s equation [7], and none had a TG value that was greater than 400 mg/dl. Blood pressure was measured with a standard mercury sphygmomanometer placed on the arm while subjects were seated at least for 5 min. In this analysis all subjects did not have any significant kidney diseases except for diabetic nephropathy. 2.2. Nerve function test To assess the peripheral nerve function, we measured motor nerve conduction velocity (MCV) and sensory nerve conduction velocity (SCV), by electric stimulation [8]. We assessed the median, ulnar, and tibial nerves. MCV and SCV were corrected by age, and expressed as the ratio of this corrected velocity to
the standard value of the same age of a subject. The average of ratios of three nerves, that we measured, was used as the index of age-corrected MCV or SCV [10]. 2.3. Assays Plasma adiponectin concentration was measured by an ELISA kit developed by Otsuka Assay (Tokushima, Japan). TNF␣ concentration was determined by QuantiGlo human TNF␣ Chemiluminescent Immunoassay (R&D systems, Inc., Minneapolis, MN, USA). Plasma leptin concentration was measured by an ELISA kit developed by Mitsubishi Biochemial Laboratory (Tokyo, Japan). 2.4. Data analysis All data are expressed as means with S.D. Relationship between two variables is analyzed by standard correlation analysis conducted by StatView version 5.0 (SAS, Cary, NC, USA). Multivariate analysis was performed using a multiple variable regression model to determine the standard coefficients to the index of neuropathy. Statistical significance was accepted when P < 0.05.
3. Results and discussion We studied 105 subjects with 66 males and 39 females. The average age was 60.8 years old with the range of 26–84 years. The mean duration of diabetes mellitus was 10.4 years. The mean BMI was 24.7 kg/m2 , and the mean HbA1C on admission was 8.9%. None was under the medication of epalrestat. Motor conduction velocity and SCV were corrected by age to be 1.0 as the normal value, and the average of three nerves were used to be the representative value, and the mean values of the index of corrected MCV and SCV were 0.96 ± 0.07 and 0.95 ± 0.08 individually. While mean values of MCV and SCV were slightly lower than 1.0, the ranges of MCV and SCV were 0.75–1.13 and 0.69–1.25 individually. Mean leptin, TNF␣, and adiponectin concentrations in this study subjects were not different from the standard values published previously, but plasma levels of these adipocytokines varied in each individual in the study subjects.
M. Matsuda et al. / Diabetes Research and Clinical Practice 66 (2004) S121–S123
The simple regression analysis between the index of age-corrected MCV and SCV revealed that MCV and SCV were strongly significantly related (P < 0.05) with duration of diabetes mellitus and HbA1C as expected. Among adipocytokines, leptin and TNF␣ seemed to be related with SCV. The data that we analyzed were observation of cross-sectional features of study subjects. To eliminate un-controlled features in the study subjects, we applied multiple-regression analysis to figure out contribution of variables on nerve disturbances. Again, we detected a significant negative contribution of TNF␣ on sensory conduction velocity with standard coefficient of −0.189 (P < 0.05). There was a negative correlation (r = −0.225, P < 0.05) between plasma TNF␣ levels and the index of age-corrected SCV. Increased plasma glucose levels and duration of diabetes are the major factors that modulate diabetic neuropathy [9]. However, a certain part of subjects with impaired glucose tolerance have a predromal phenomenon of diabetic painful neuropathy [10,11]. The disturbance in autonomic nervous system in subjects with impaired glucose tolerant has been reported [13,14]. Since increment of plasma glucose levels could be negligible, some attribute insulinopenia from the GK-rat study [12]. Meanwhile, nerve function may be affected by plasma cytokine levels like TNF␣, as shown in an experimental condition [5]. This possible mechanism of this adipocytokine on peripheral nerve system would be applied in type 2 diabetic subjects who have a characteristic feature of insulin resistance. In our analysis TNF␣ has been proven to be related with index of sensory nerve function in type 2 diabetic subjects, and this effect was more significant on sensory nerves than motor nerves, and this coincides with the result reported by Sumner et al. [1]. Thus, the precise mechanism that makes progress in peripheral nerve system is still to be clarified, one of the possible mechanisms must be related with increased plasma TNF␣ levels or inflammatory process in the body. We conclude that adipocytokines may play a role not only on angiopathy but also on neuropathy in patients with type 2 diabetes mellitus. Acknowledgements This study was supported by Grant-In-Aid for scientific research (No. 13671205) from the Japan Soci-
S123
ety for the Promotion of Science, and Research Project Grants (Nos. 12-506, 12-507, 13-507) from Kawasaki Medical School.
References [1] C.J. Sumner, S. Sheth, J.W. Griffin, D.R. Cornblath, M. Polydefkis, The spectrum of neuropathy in diabetes and impaired glucose tolerance, Neurology 60 (2003) 108– 111. [2] W.Y. Fujimoto, R.W. Bergstrom, D.L. Leonetti, L.L. NewellMorris, W.P. Shuman, P.W. Wahl, Metabolic and adipose risk factors for NIDDM and coronary disease in third-generation Japanese-American men and women with impaired glucose tolerance, Diabetologia 37 (1994) 524–532. [3] J.B. Meigs, R.B. D’Agostino Sr, P.W. Wilson, L.A. Cupples, D.M. Nathan, D.E. Singer, The Framingham Offspring Study. Risk variable clustering in the insulin resistance syndrome, Diabetes 46 (1997) 1594–1600. [4] I. Shimomura, T. Funahashi, Y. Matsuzawa, Significance of adipocytokine, fat-derived hormones, in metabolic syndrome, Tanpakushitsu Kakusan Koso 47 (2002) 1896–1903. [5] V.E. Drory, D. Lev, G.B. Groozman, M. Gutmann, J.M. Klausner, Neurotoxicity of isolated limb perfusion with tumor necrosis factor, J. Neurol. Sci. 158 (1998) 1–4. [6] World Health Organization. Report of a WHO Consultation. Part 1.: Diagnosis and Classification of Diabetes Mellutis, Geneva, 1999, WHO/NCD/NCS/99.2. [7] W.T. Friedewald, R.I. Levy, D.S. Fredrickson, Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge, Clin. Chem. 18 (1972) 499–502. [8] G. Gregersen, Diabetic neuropathy: influence of age, sex, metabolic control, and duration of diabetes on motor conduction velocity, Neurology 17 (1967) 972–980. [9] M.A. Pfeifer, M.P. Schumer, Clinical trials of diabetic neuropathy: past, present, and future, Diabetes 44 (1995) 1355–1361. [10] J.R. Singleton, A.G. Smith, M.B. Bromberg, Painful sensory polyneuropathy associated with impaired glucose tolerance, Muscle Nerve 24 (2001) 1225–1228. [11] J.R. Singleton, A.G. Smith, M.B. Bromberg, Increased prevalence of impaired glucose tolerance in patients with painful sensory neuropathy, Diabetes Care 24 (2001) 1448– 1453. [12] Y. Murakawa, W. Zhang, C.R. Pierson, T. Brismar, C.G. Ostenson, S. Efendic, et al., Impaired glucose tolerance and insulinopenia in the GK-rat causes peripheral neuropathy, Diabetes Metab. Res. Rev. 18 (2002) 473–483. [13] K. Suzuki, Neurological disorders associated with impaired glucose tolerance, Nippon Rinsho 54 (1996) 2704–2708. [14] K.F. Rezende, A. Melo, J. Pousada, Z.F. Rezende, N.L. Santos, I. Gomes, Autonomic neuropathy in patients with impaired glucose tolerance, Arq. Neuropsiquiatr. 55 (1997) 703–711.