Endocrine Journal 2015, 62 (6), 485-492
Original
The association between impaired proinsulin processing and type 2 diabetes mellitus in non-obese Japanese individuals Hidenori Katsuta1), Sachihiko Ozawa1), Kiyoshi Suzuki2), Kazuto Takahashi1), Toshiaki Tanaka1), Yoshikazu Sumitani1), Susumu Nishida1), Takuma Kondo1), Toshio Hosaka1), Kouichi Inukai1) and Hitoshi Ishida1) 1)
Third Department of Internal Medicine, Division of Diabetes, Endocrinology and Metabolism, Kyorin University School of Medicine, Tokyo 181-8611, Japan 2) Division of Clinical Laboratory, Shimada Municipal Hospital, Shizuoka 427-8501, Japan
Abstract. We aimed to examine the association between impaired proinsulin processing in pancreatic beta cells and type 2 diabetes mellitus in non-obese Japanese patients. Participants were divided into groups for normal glucose tolerance, prediabetes, and type 2 diabetes based on the oral glucose tolerance test (OGTT). Activities of prohormone convertase (PC) 1/3 and PC2 in fasting states were estimated. Multiple regression analysis was undertaken to ascertain if alteration of the activities of these enzymes contributes to the development of impaired glucose tolerance by comparison with HOMA-β and the oral disposition index (DIO). Overall, 452 subjects were included. PC1/3 activity tended to decrease in type 2 diabetes compared with normal glucose tolerance. PC2 activity showed no difference among the three groups. Decreased estimated PC1/3 activity was significantly associated with type 2 diabetes after adjustment for sex, age, creatinine, triglycerides, HOMA-β and DIO. Odds ratios (95% CI) of PC1/3, HOMA-β, and DIO were 2.16 (1.12–4.19), 3.44 (1.82–6.52) and 14.60 (7.87–27.11), respectively. Furthermore, decreased PC1/3(≤1.7) combined with decreased HOMA-β (≤30) had a sensitivity of 73% and specificity of 62%. Decreased PC1/3 activity may be a useful measurement of beta-cell function alongside decreased HOMA-β or DIO. A combined decrease in estimated fasting PC1/3 activity and HOMA-β measurement led to suspicion of type 2 diabetes in the non-obese Japanese population studied. Key words: HOMA-β, Non-obese, Prohormone convertase, Proinsulin processing, Type 2 diabetes
The increasing prevalence of diabetes is a major health problem worldwide. Pancreatic beta-cell dysfunction causes type 2 diabetes mellitus as characterised by impaired acute-phase insulin secretion in response to glucose, partly because of higher circulating concentrations of intact and split proinsulin (PI), both in absolute amounts and as a proportion of total insulin secretory products in the fasting state [1–5]. Although the cause of this disproportional hyperproinsulinaemia in type 2 diabetes is not known, it might arise from inefficient PI processing within the beta-cell secretory granule, or from the premature release of PI as a result of increased demand for insulin in insulin-
resistant states [6]. In the past two decades, the high cross-reactivity between insulin, PI and C-peptide assays has permitted few conclusions regarding the role of insulin and hyperinsulinaemia in the development of type 2 diabetes. Several methods of PI-specific radioimmunoassay (RIA) have been used to measure circulating PI levels. Among them, the human PI RIA kit reveals high cross-reactivity with des-31,32-PI, as well as with the intact molecule of PI. However, a specific chemiluminescence assay (CLIA) virtually without cross-reactivity to des-31,32-PI and more specific for intact PI has been developed [7].
Submitted Dec. 19, 2014; Accepted Mar. 9, 2015 as EJ14-0611 Released online in J-STAGE as advance publication Apr.16, 2015
Abbreviations: CLIA, chemiluminescence assay; CPE, carboxypeptidase E; DIO, oral disposition index; ER, endoplasmic reticulum; FPG, fasting plasma glucose; IFG: impaired fasting glucose; IGT, impaired glucose tolerance; iPI, intact-PI; IRI, immunoreactive insulin; NGT, normal glucose tolerance; OGTT, oral glucose tolerance test; PC, prohormone convertase; PI, proinsulin; TC, total cholesterol; TG, triglycerides; tPI, total-PI
Correspondence to: Hitoshi Ishida, Third Department of Internal Medicine, Division of Diabetes, Endocrinology and Metabolism, Kyorin University School of Medicine, 6-20-2 Shinkawa, Mitaka, Tokyo 181-8611, Japan. E-mail: [email protected] ©The Japan Endocrine Society
Katsuta et al.
486
It is well known that the predominant processing pathway in beta cells first converts PI to des-31,32-PI by prohormone convertase (PC) 1/3, and subsequently to insulin and C-peptide by PC2 [8]. Another intermediate, des-64,65-PI, is difficult to detect in patients with type 2 diabetes, and is thought to not contribute to increased levels of PI-related molecules [9, 10]. In the present study, we aimed to investigate any association between impaired PI processing and type 2 diabetes in a non-obese Japanese population. We measured circulating blood levels of fasting plasma glucose (FPG), immunoreactive insulin (IRI), totalPI (tPI), and intact-PI (iPI) simultaneously in subjects classified as having normal glucose tolerance (NGT), prediabetes, including impaired fasting glucose (IFG and/or impaired glucose tolerance (IGT)), or diabetes. We then estimated PC1/3 and PC2 activity, taking into consideration cross-reactivities of PI measurement systems, to investigate the utility of a newly proposed clinical surrogate marker of pancreatic beta-cell function.
Materials and Methods A cohort of 798 individuals who had a routine health check between 1998 and 2002 at the Health Management Center in our affiliated hospital (Shimada Municipal Hospital, Shizuoka, Japan) was used in this study. A total of 346 subjects were excluded for obesity (body mass index (BMI) ≥25 kg/m2) or for unavailable data on fasting insulin, tPI, or iPI. The final study population for the analysis was 452. All subjects were admitted to the hospital the day before starting the study and ate an adequate dinner (hospital diet). They received a 75-g oral glucose tolerance test (OGTT) after a 10-h overnight fast, and were divided into three groups: NGT, prediabetes (IFG and/or IGT), or type 2 diabetes. Basal blood samples were drawn for measurement of various PI-related molecules. Samples for glucose determination were collected in tubes with fluoride and heparin; those for insulin, tPI, and iPI were collected in tubes with EDTA with 1000 KIE aprotinin/ mL whole blood. Glucose samples were sent immediately to the laboratory for measurements. Other samples were frozen and kept at −70°C until assay. Plasma glucose levels were determined by the glucose oxidase method (Beckman Instruments, Fullerton, CA, USA). Serum insulin level was measured by the monoclonal insulin ELISA method using an AxSYM insulin kit (Dinabot, Tokyo, Japan) [11]. iPI was
assessed using an iPI CLIA kit (Molecular Light Technology Research Ltd., Cardiff, UK). tPI was assayed using a human PI RIA kit (LINCO Research, Inc., St Charles, MO, USA) [7]. Minimum detectable sensitivities of the insulin, C-peptide, iPI, and tPI assays were 1.0, 3.3, 0.6, and 2.0 pmol/L; inter-assay coefficient of variations (CVs) were 5.2, 5.8, 8.6, and 8.6%; and intra-assay CVs were 3.6, 5.1,5.2, and 5.2%, respectively. Cross-reactivities for des-31,32-PI were 1.4% and 95% on assays of intact and total PI, respectively. Neither of these PI measurement systems indicated cross-reactivity with C-peptide or insulin. Assay characteristics are reported from manufacturer instructions for each assay. Upon consideration of cross-reactivities, levels of circulating des-31,32-PI were estimated. Activities of PC1/3 and PC2 were defined using the following formulae [12]: des-31,32-PI = (tPI − iPI)/0.95 PC1/3 activity = des-31,32-PI/iPI PC2 activity = insulin/des-31,32-PI According to 2003 ADA criteria [13], subjects were categorised by 75-g OGTT as NGT (fasting plasma glucose [FPG]
Original
The association between impaired proinsulin processing and type 2 diabetes mellitus in non-obese Japanese individuals Hidenori Katsuta1), Sachihiko Ozawa1), Kiyoshi Suzuki2), Kazuto Takahashi1), Toshiaki Tanaka1), Yoshikazu Sumitani1), Susumu Nishida1), Takuma Kondo1), Toshio Hosaka1), Kouichi Inukai1) and Hitoshi Ishida1) 1)
Third Department of Internal Medicine, Division of Diabetes, Endocrinology and Metabolism, Kyorin University School of Medicine, Tokyo 181-8611, Japan 2) Division of Clinical Laboratory, Shimada Municipal Hospital, Shizuoka 427-8501, Japan
Abstract. We aimed to examine the association between impaired proinsulin processing in pancreatic beta cells and type 2 diabetes mellitus in non-obese Japanese patients. Participants were divided into groups for normal glucose tolerance, prediabetes, and type 2 diabetes based on the oral glucose tolerance test (OGTT). Activities of prohormone convertase (PC) 1/3 and PC2 in fasting states were estimated. Multiple regression analysis was undertaken to ascertain if alteration of the activities of these enzymes contributes to the development of impaired glucose tolerance by comparison with HOMA-β and the oral disposition index (DIO). Overall, 452 subjects were included. PC1/3 activity tended to decrease in type 2 diabetes compared with normal glucose tolerance. PC2 activity showed no difference among the three groups. Decreased estimated PC1/3 activity was significantly associated with type 2 diabetes after adjustment for sex, age, creatinine, triglycerides, HOMA-β and DIO. Odds ratios (95% CI) of PC1/3, HOMA-β, and DIO were 2.16 (1.12–4.19), 3.44 (1.82–6.52) and 14.60 (7.87–27.11), respectively. Furthermore, decreased PC1/3(≤1.7) combined with decreased HOMA-β (≤30) had a sensitivity of 73% and specificity of 62%. Decreased PC1/3 activity may be a useful measurement of beta-cell function alongside decreased HOMA-β or DIO. A combined decrease in estimated fasting PC1/3 activity and HOMA-β measurement led to suspicion of type 2 diabetes in the non-obese Japanese population studied. Key words: HOMA-β, Non-obese, Prohormone convertase, Proinsulin processing, Type 2 diabetes
The increasing prevalence of diabetes is a major health problem worldwide. Pancreatic beta-cell dysfunction causes type 2 diabetes mellitus as characterised by impaired acute-phase insulin secretion in response to glucose, partly because of higher circulating concentrations of intact and split proinsulin (PI), both in absolute amounts and as a proportion of total insulin secretory products in the fasting state [1–5]. Although the cause of this disproportional hyperproinsulinaemia in type 2 diabetes is not known, it might arise from inefficient PI processing within the beta-cell secretory granule, or from the premature release of PI as a result of increased demand for insulin in insulin-
resistant states [6]. In the past two decades, the high cross-reactivity between insulin, PI and C-peptide assays has permitted few conclusions regarding the role of insulin and hyperinsulinaemia in the development of type 2 diabetes. Several methods of PI-specific radioimmunoassay (RIA) have been used to measure circulating PI levels. Among them, the human PI RIA kit reveals high cross-reactivity with des-31,32-PI, as well as with the intact molecule of PI. However, a specific chemiluminescence assay (CLIA) virtually without cross-reactivity to des-31,32-PI and more specific for intact PI has been developed [7].
Submitted Dec. 19, 2014; Accepted Mar. 9, 2015 as EJ14-0611 Released online in J-STAGE as advance publication Apr.16, 2015
Abbreviations: CLIA, chemiluminescence assay; CPE, carboxypeptidase E; DIO, oral disposition index; ER, endoplasmic reticulum; FPG, fasting plasma glucose; IFG: impaired fasting glucose; IGT, impaired glucose tolerance; iPI, intact-PI; IRI, immunoreactive insulin; NGT, normal glucose tolerance; OGTT, oral glucose tolerance test; PC, prohormone convertase; PI, proinsulin; TC, total cholesterol; TG, triglycerides; tPI, total-PI
Correspondence to: Hitoshi Ishida, Third Department of Internal Medicine, Division of Diabetes, Endocrinology and Metabolism, Kyorin University School of Medicine, 6-20-2 Shinkawa, Mitaka, Tokyo 181-8611, Japan. E-mail: [email protected] ©The Japan Endocrine Society
Katsuta et al.
486
It is well known that the predominant processing pathway in beta cells first converts PI to des-31,32-PI by prohormone convertase (PC) 1/3, and subsequently to insulin and C-peptide by PC2 [8]. Another intermediate, des-64,65-PI, is difficult to detect in patients with type 2 diabetes, and is thought to not contribute to increased levels of PI-related molecules [9, 10]. In the present study, we aimed to investigate any association between impaired PI processing and type 2 diabetes in a non-obese Japanese population. We measured circulating blood levels of fasting plasma glucose (FPG), immunoreactive insulin (IRI), totalPI (tPI), and intact-PI (iPI) simultaneously in subjects classified as having normal glucose tolerance (NGT), prediabetes, including impaired fasting glucose (IFG and/or impaired glucose tolerance (IGT)), or diabetes. We then estimated PC1/3 and PC2 activity, taking into consideration cross-reactivities of PI measurement systems, to investigate the utility of a newly proposed clinical surrogate marker of pancreatic beta-cell function.
Materials and Methods A cohort of 798 individuals who had a routine health check between 1998 and 2002 at the Health Management Center in our affiliated hospital (Shimada Municipal Hospital, Shizuoka, Japan) was used in this study. A total of 346 subjects were excluded for obesity (body mass index (BMI) ≥25 kg/m2) or for unavailable data on fasting insulin, tPI, or iPI. The final study population for the analysis was 452. All subjects were admitted to the hospital the day before starting the study and ate an adequate dinner (hospital diet). They received a 75-g oral glucose tolerance test (OGTT) after a 10-h overnight fast, and were divided into three groups: NGT, prediabetes (IFG and/or IGT), or type 2 diabetes. Basal blood samples were drawn for measurement of various PI-related molecules. Samples for glucose determination were collected in tubes with fluoride and heparin; those for insulin, tPI, and iPI were collected in tubes with EDTA with 1000 KIE aprotinin/ mL whole blood. Glucose samples were sent immediately to the laboratory for measurements. Other samples were frozen and kept at −70°C until assay. Plasma glucose levels were determined by the glucose oxidase method (Beckman Instruments, Fullerton, CA, USA). Serum insulin level was measured by the monoclonal insulin ELISA method using an AxSYM insulin kit (Dinabot, Tokyo, Japan) [11]. iPI was
assessed using an iPI CLIA kit (Molecular Light Technology Research Ltd., Cardiff, UK). tPI was assayed using a human PI RIA kit (LINCO Research, Inc., St Charles, MO, USA) [7]. Minimum detectable sensitivities of the insulin, C-peptide, iPI, and tPI assays were 1.0, 3.3, 0.6, and 2.0 pmol/L; inter-assay coefficient of variations (CVs) were 5.2, 5.8, 8.6, and 8.6%; and intra-assay CVs were 3.6, 5.1,5.2, and 5.2%, respectively. Cross-reactivities for des-31,32-PI were 1.4% and 95% on assays of intact and total PI, respectively. Neither of these PI measurement systems indicated cross-reactivity with C-peptide or insulin. Assay characteristics are reported from manufacturer instructions for each assay. Upon consideration of cross-reactivities, levels of circulating des-31,32-PI were estimated. Activities of PC1/3 and PC2 were defined using the following formulae [12]: des-31,32-PI = (tPI − iPI)/0.95 PC1/3 activity = des-31,32-PI/iPI PC2 activity = insulin/des-31,32-PI According to 2003 ADA criteria [13], subjects were categorised by 75-g OGTT as NGT (fasting plasma glucose [FPG]
