Insulin secretion and insulin sensitivity in diabetic subgroups: studies in the prediabetic and diabetic state

Aims/hypothesis. To evaluate insulin sensitivity and insulin secretion in prediabetic and diabetic subjects with mutations in MODY1 (HNF-4α) and MODY3 (HNF-1α) genes, in subjects with GAD antibodies, latent autoimmune diabetes in adults and in subjects with the common form of Type II (non-insulin-dependent) diabetes mellitus. Methods. Insulin secretion was measured as the incremental 30-min insulin (I30) and insulin glucose ratio (I:G30) during OGTT whereas insulin sensitivity was measured as the insulin sensitivity index during OGTT in 131 carriers of MODY mutations [NGT = 38, IFG/IGT = 21, diabetes mellitus (DM) = 72], in 293 subjects with GADA (NGT = 47, IFG/IGT = 29, DM = 217) and in 2961 subjects with a family history of the common form of Type II diabetes but without MODY mutations or GADA (NGT = 1360, IFG/IGT = 857, DM = 744). A subgroup of the subjects underwent a euglycaemic clamp (n = 210) and intravenous glucose tolerance test (n = 337) for the estimation of insulin sensitivity and first-phase insulin secretion. Results. Non-diabetic subjects with MODY mutations had pronounced impaired insulin secretion (I30, I:G30) compared with the two other groups (p = 0.005). Normal or non-diabetic glucose tolerance was maintained by enhanced insulin sensitivity compared with the other two groups (p < 0.05 and p < 0.005). In contrast to patients with Type II diabetes and with adult latent autoimmune diabetes, MODY patients showed only a modest deterioration in insulin sensitivity at onset of diabetes. The 2-h glucose values inversely correlated with insulin sensitivity in subjects with GADA (r = –0.447, p < 0.001) and subjects from Type II diabetic families (r = –0.426, p < 0.001), whereas no such relation was observed in subjects with MODY mutations (r = 0.151, p = NS). There were no statistically significant differences in insulin secretion or insulin sensitivity between subjects with GADA or subjects with a family history of Type II diabetes, either at the NGT or the IFG/IGT stage. Conclusion/interpretation. Glucose-tolerant carriers of MODY mutations are characterised by a severe impairment in insulin secretion. Enhanced insulin sensitivity is the most likely explanation for the normal glucose tolerance. Whereas subjects with positive GADA or Type II diabetes have impaired insulin sensitivity with increasing glucose concentrations, MODY mutation carriers seem to be protected from the effect of glucose toxicity. [Diabetologia (2000) 43: 1476–1483] Received: 23 March 2000 and in revised form: 29 August 2000     

Risk and mechanism of dexamethasone-induced deterioration of glucose tolerance in non-diabetic first-degree relatives of NIDDM patients

Summary We tested the hypothesis that glucose intolerance develops in genetically prone subjects when exogenous insulin resistance is induced by dexamethasone (dex) and investigated whether the steroid-induced glucose intolerance is due to impairment of beta-cell function alone and/or insulin resistance. Oral glucose tolerance (OGTT) and intravenous glucose tolerance tests with minimal model analysis were performed before and following 5 days of dex treatment (4 mg/day) in 20 relatives of non-insulin-dependent diabetic (NIDDM) patients and in 20 matched control subjects (age: 29.6 ± 1.7 vs 29.6 ± 1.6 years, BMI: 25.1 ± 1.0 vs 25.1 ± 0.9 kg/m²). Before dex, glucose tolerance was similar in both groups (2-h plasma glucose concentration (PG): 5.5 ± 0.2 [range: 3.2–7.0] vs 5.5 ± 0.2 [3.7–7.4] mmol/l). Although insulin sensitivity (Si) was significantly lower in the relatives before dex, insulin sensitivity was reduced to a similar level during dex in both the relatives and control subjects (0.30 ± 0.04 vs 0.34 ± 0.04 10^–4 min^–1 per pmol/l, NS). During dex, the variation in the OGTT 2-h PG was greater in the relatives (8.5 ± 0.7 [3.9–17.0] vs 7.5 ± 0.3 [5.7–9.8] mmol/l, F-test p < 0.05) which, by inspection of the data, was caused by seven relatives with a higher PG than the maximal value seen in the control subjects (9.8 mmol/l). These “hyperglycaemic” relatives had diminished first phase insulin secretion (?1) both before and during dex compared with the “normal” relatives and the control subjects (pre-dex ?1: 12.6 ± 3.6 vs 26.4 ± 4.2 and 24.6 ± 3.6 (p < 0.05), post-dex ?1: 22.2 ± 6.6 vs 48.0 ± 7.2 and 46.2 ± 6.6 respectively (p < 0.05) pmol · l^–1· min^–1 per mg/dl). However, Si was similar in “hyperglycaemic” and “normal” relatives before dex (0.65 ± 0.10 vs 0.54 ± 0.10 10−4 · min^–1 per pmol/l) and suppressed similarly during dex (0.30 ± 0.07 vs 0.30 ± 0.06 10−4 · min^–1 per pmol/l). Multiple regression analysis confirmed the unique importance of low pre-dex beta-cell function to subsequent development of high 2-h post-dex OGTT plasma glucose levels (R ² = 0.56). In conclusion, exogenous induced insulin resistance by dex will induce impaired or diabetic glucose tolerance in those genetic relatives of NIDDM patients who have impaired beta-cell function (retrospectively) prior to dex exposure. These subjects are therefore unable to enhance their beta-cell response in order to match the dex-induced insulin resistant state. [Diabetologia (1997) 40: 1439–1448] Received: 20 January 1997 and in final revised form: 17 July 1997     

Overexpression of dominant negative mutant hepatocyte nuclear factor (HNF)-1α inhibits arginine-induced insulin secretion in MIN6 cells

Aims/hypothesis. To explain the mechanisms whereby mutations in the HNF-1α gene cause insulin secretory defects. Methods. A truncated mutant HNF-1α (HNF-1α288 t) was overexpressed in hepatoma cells (HepG2) and murine insulinoma cells (MIN6) using a recombinant adenovirus system and expression of the HNF-1α target genes and insulin secretion were examined. Results. Expression of phenylalanine hydroxylase and α1-antitrypsin genes, the target genes of HNF-1α, was suppressed in HepG2 cells by overexpression of HNF-1α288 t. In MIN6 cells, overexpression of HNF-1α288 t did not change insulin secretion stimulated by glucose (5 mmol/l and 25 mmol/l) or leucine (20 mmol/l). Potentiation of insulin secretion by arginine (20 mmol/l, in the presence of 5 mmol/l or 25 mmol/l glucose) was, however, reduced (p < 0.0001 and p = 0.027, respectively). Similarly reduced responses were observed when stimulated with homoarginine. Expression of the cationic amino acid transporter-2 was not reduced and insulin secretory response to membrane depolarization by 50 mmol/l KCl was intact. Conclusion/interpretation. The HNF-1α288 t, which is structurally similar to the mutant HNF-1α expressed from the common MODY3 allele, P291fsinsC, exerts a dominant negative effect. Suppression of HNF-1α in MIN6 cells severely impaired potentiation of insulin secretion by arginine, whereas glucose-stimulated and leucine-stimulated insulin secretion was intact. Our findings delineate the complex nature of beta-cell failure in patients with MODY3. This cell model will be useful for further investigation of the mechanism of insulin secretory defects in these patients. [Diabetologia (1999) 42: 887–891] Received: 30 November 1998 and in revised form: 16 February 1999     

A missense mutation in the hepatocyte nuclear factor 4 alpha gene in a UK pedigree with maturity-onset diabetes of the young

Summary Maturity-onset diabetes of the young (MODY) is a monogenic subgroup of non-insulin dependent diabetes mellitus (NIDDM) characterised by an early age of onset (< 25 years) and an autosomal dominant mode of inheritance. MODY is genetically heterogeneous with three different genes identified to date; hepatocyte nuclear factor 4 alpha (HNF-4α) [MODY1], glucokinase [MODY2] and hepatocyte nuclear factor 1 alpha (HNF-1α) [MODY3]. A nonsense mutation in the HNF-4α gene has recently been shown to cause MODY in a single large North American pedigree (RW). We screened a large UK Caucasian MODY family which showed weak evidence of linkage to the MODY1 locus on chromosome 20q (lod score for ADA 0.68 at θ = 0) for mutations in the coding region of the HNF-4α gene by direct sequencing. A missense mutation resulting in the substitution of glutamine for glutamic acid was identified in exon 7 (E276Q). The mutation was present in all of the diabetic members of the pedigree plus two unaffected subjects and was not detected in 75 normal control subjects or 95 UK Caucasian subjects with late-onset NIDDM. This is the first missense mutation to be described in the HNF-4α gene. [Diabetologia (1997) 40: 859–862] Received: 7 March 1997 and in revised form: 16 April 1997     

High frequency of mutations in MODY and mitochondrial genes in Scandinavian patients with familial early-onset diabetes

Abstract Aims/hypothesis. To investigate the contribution of mutations in maturity-onset diabetes of the young (MODY) and mitochondrial genes to early-onset diabetes with a strong family history of diabetes in a cohort with a high prevalence of Type I (insulin-dependent) diabetes mellitus. Methods. Screening for sequence variants in the hepatocyte nuclear factor (HNF)–4 α (MODY1), glucokinase (MODY2), HNF-1 α (MODY3) genes and mitochondrial DNA was carried out in 115 Finnish and Swedish patients with early-onset ( ≤ 40 years) diabetes using the single strand conformation polymorphism (SSCP) technique and direct sequencing. Allele frequencies were compared with 118 patients with onset of diabetes Type II (non-insulin-dependent) diabetes mellitus after the age of 40 and 92 non–diabetic control subjects without a family history of diabetes. Results. In total 52 sequence variants were found in the HNF-1α, HNF-4α and glucokinase genes, 12 of which were considered as MODY mutations. Three families had the A3243G mutation in the mitochondrial tRNA ^Leu gene, which resulted in an overall prevalence of these mutations of 13 %. Conclusion/interpretation. Among 115 Scandinavian families, mutations in the HNF-1α gene represented the most common cause of familial early-onset ( ≤ 40 years) diabetes: MODY3 (5.2 %) more than MODY2 (3.5 %) more than MIDD (2.6 %) more than MODY1 (1.7 %). [Diabetologia (1999) 42: 1131–1137] Received: 3 March 1999 and in revised form: 14 May 1999     

Mutation screening in 18 Caucasian families suggest the existence of other MODY genes

Summary Maturity-onset diabetes of the young (MODY) is a heterogeneous subtype of non-insulin-dependent diabetes mellitus characterised by early onset, autosomal dominant inheritance and a primary defect in insulin secretion. To date five MODY genes have been identified: hepatocyte nuclear factor-4 alpha (HNF-4 α/MODY1/TCF14) on chromosome 20 q, glucokinase (GCK/MODY2) on chromosome 7 p, hepatocyte nuclear factor-1 alpha (HNF-1 α/MODY3/TCF1) on chromosome 12 q, insulin promoter factor-1 (IPF1/MODY4) on chromosome 13 q and hepatocyte nuclear factor-1 beta (HNF-1 β/MODY5/TCF2) on chromosome 17cen-q. We have screened the HNF-4 α, HNF-1 α and HNF-1 β genes in members of 18 MODY kindreds who tested negative for glucokinase mutations. Five missense (G31D, R159W, A161T, R200W, R271W), one substitution at the splice donor site of intron 5 (IVS5nt + 2T→A) and one deletion mutation (P379fsdelT) were found in the HNF-1 α gene, but no MODY-associated mutations were found in the HNF-4 α and HNF-1 β genes. Of 67 French MODY families that we have now studied, 42 (63 %) have mutations in the glucokinase gene, 14 (21 %) have mutations in the HNF-1 α gene, and 11 (16 %) have no mutations in the HNF-4 α, IPF1 and HNF-1 β genes. Eleven families do not have mutations in the five known MODY genes suggesting that there is at least one additionnal locus that can cause MODY. [Diabetologia (1998) 41: 1017–1023] Received: 5 January 1998 and in revised form: 8 April 1998     

Intramuscular triglyceride content is increased in IDDM

Summary Increased lipid oxidation is related to insulin resistance. Some of the enhanced lipid utilization may be derived from intramuscular sources. We studied muscle triglyceride (mTG) concentration and its relationship to insulin sensitivity in 10 healthy men (age 29 ± 2 years, BMI 23.3 ± 0.6 kg/m²) and 17 men with insulin-dependent diabetes mellitus (IDDM) (age 30 ± 2 years, BMI 22.8 ± 0.5 kg/m², diabetes duration 14 ± 2 years, HbA_{1 c} 7.7 ± 0.3 %, insulin dose 48 ± 3 U/day). Insulin sensitivity was measured with a 4 h euglycaemic (5 mmol/l) hyperinsulinaemic (1.5 mU or 9 pmol · kg^–1· min^–1) clamp accompanied by indirect calorimetry before and at the end of the insulin infusion. A percutaneous biopsy was performed from m. vastus lateralis for the determination of mTG. At baseline the IDDM patients had higher glucose (10.2 ± 0.9 vs 5.6 ± 0.1 mmol/l, p < 0.001), insulin (40.3 ± 3.2 vs 23.2 ± 4.2 pmol/l, p < 0.01), HDL cholesterol (1.28 ± 0.06 vs 1.04 ± 0.03 mmol/l, p < 0.01) and mTG (32.9 ± 4.6 vs 13.6 ± 2.7 mmol/kg dry weight, p < 0.01) concentrations than the healthy men, respectively. The IDDM patients had lower insulin stimulated whole body total (–25 %, p < 0.001), oxidative (–18 %, p < 0.01) and non-oxidative glucose disposal rates (–43 %, p < 0.001), whereas lipid oxidation rate was higher in the basal state ( + 44 %, p < 0.01) and during hyperinsulinaemia ( + 283 %, p < 0.05). mTG concentrations did not change significantly during the clamp or correlate with insulin stimulated glucose disposal. In healthy men mTG correlated positively with lipid oxidation rate at the end of hyperinsulinaemia (r = 0.75, p < 0.05). In conclusion: 1) IDDM is associated with increased intramuscular TG content. 2) mTG content does not correlate with insulin sensitivity in healthy subjects or patients with IDDM. [Diabetologia (1998) 41: 111–115] Received: 12 June 1997 and in revised form: 8 September 1997     

Production and metabolic clearance of glucose under basal conditions in Type II (non-insulin-dependent) diabetes mellitus

Aims/hypothesis: The pathogenesis of fasting hyperglycaemia in Type II (non-insulin-dependent) diabetes mellitus has yet to be clarified. Rates of glucose production (R _a), utilization and metabolic clearance rate were therefore measured during an extended fast, in control subjects and in Type II diabetic patients. Methods: Nine subjects with newly-diagnosed or diet-treated diabetes and seven control subjects matched for age and weight (BMI 36.0 ± 2.4 and 35.3 ± 3.1 kg/m² respectively) underwent an overnight fast followed by a 10-h unprimed infusion of [6-³H]glucose. Plasma tracer concentrations were fitted by a single-compartment model. Results: The metabolic clearance rate was near-constant [61.7 + 2.4 ml/(min-m²)] in diabetic patients and [75.5 ± 3.3 ml/(min-m²)] in control subjects (p < 0.05). It was correlated to the glucose concentrations both at t = 0 (r = –0.752, p = 0.0008) and t = 10 h (r = –0.675, p = 0.004). The calculated volume of distribution was 17.3 ± 1.4 l (18.2 % weight, diabetes), 19.6 ± 2.4 l (18.4 % weight, control). Glycaemia fell from 10.7 ± 0.8 mmol/l to 6.5 ± 0.3 mmol/l by 10 h (p < 0.05) in diabetes and from 5.6 ± 0.6 to 4.8 ± 0.1 mmol/l in control subjects (p < 0.05). The rate of glucose production decreased in parallel, from 563 ± 33 to 363 ± 23 μmol/(min-m²) (p < 0.05) in diabetes from 419 ± 20 to 347 ± 32 μmol/(min-m²) in control subjects. Initial R _a was higher in diabetic patients than in control subjects (p < 0.05) and was highly correlated to glycaemia (r = 0.836, p = 0.0001). By 10 h, R _a had converged in diabetic patients and control subjects and all correlation with glycaemia was lost (r = 0.0017, p = 0.95). Conclusions/interpretation: In relatively early diabetes, the more “labile” portion of fasting hyperglycaemia, which subsequently decreased, was closely related to the simultaneously decreasing R _a. The 25 % increase in glucose concentrations which persisted as stabilized R _a, resulted from about a 20 % lower metabolic clearance rate. [Diabetologia (2001) 44: 983–991] Received: 28 February 2001 and in revised form: 17 April 2001     

Inappropriate suppression of glucagon during OGTT but not during isoglycaemic i.v. glucose infusion contributes to the reduced incretin effect in type 2 diabetes mellitus

Aims/hypothesis We investigated glucagon responses during OGTT and isoglycaemic i.v. glucose infusion, respectively, to further elucidate the mechanisms behind the glucose intolerance in patients with type 2 diabetes. Materials and methods Ten patients (eight men) with type 2 diabetes (age: 64 [51–80] years; BMI: 23 [21–26] kg/m²; HbA_1c: 6.9 [6.2–8.7]%, values mean [range]) and ten control subjects matched for sex, age and BMI were studied. Blood was sampled on two separate days following a 4-h 50-g OGTT and an isoglycaemic i.v. glucose infusion, respectively. Results Isoglycaemia during the 2 days was obtained in both groups. In the control subjects no difference in glucagon suppression during the first 45 min of OGTT and isoglycaemic i.v. glucose infusion (−36 ± 12 vs −64 ± 23 mmol/l × 45 min; p = NS) was observed, whereas in the group of patients with type 2 diabetes significant glucagon suppression only occurred following isoglycaemic i.v. glucose infusion (−63 ± 21 vs 10 ± 16 mmol/l × 45 min; p = 0.002). The incretin effect was significantly reduced in patients with type 2 diabetes compared with control subjects, but no significant differences in the secretion of glucagon-like peptide-1 or glucose-dependent insulinotropic polypeptide between the two groups during OGTT or isoglycaemic i.v. glucose infusion, respectively, could explain this. Conclusions/interpretation Attenuated and delayed glucagon suppression in patients with type 2 diabetes occurs after oral ingestion of glucose, while isoglycaemic i.v. administration of glucose results in normal suppression of glucagon. We suggest that this phenomenon contributes both to the glucose intolerance and to the reduced incretin effect observed in patients with type 2 diabetes.     

Hyperlipidaemia is associated with increased insulin-mediated glucose metabolism, reduced fatty acid metabolism and normal blood pressure in transgenic mice overexpressing human apolipoprotein C1

Aims/hypothesis. Insulin resistance for glucose metabolism is associated with hyperlipidaemia and high blood pressure. In this study we investigated the effect of primary hyperlipidaemia on basal and insulin-mediated glucose and on non-esterified fatty acid (NEFA) metabolism and mean arterial pressure in hyperlipidaemic transgenic mice overexpressing apolipoprotein C1 (APOC1). Previous studies have shown that APOC1 transgenic mice develop hyperlipidaemia primarily because of an impaired hepatic uptake of very low density lipoprotein (VLDL). Methods. Basal and hyperinsulinaemic (6 mU · kg^–1· min^–1), euglycaemic (7 mmol/l) clamps with 3-³H-glucose or 9,10-³H-palmitic acid infusions and in situ freeze clamped tissue collection were carried out. Results. The APOC1 mice showed increased basal plasma cholesterol, triglyceride, NEFA and decreased glucose concentrations compared with wild-type mice (7.0 ± 1.2 vs 1.6 ± 0.1, 9.1 ± 2.3 vs 0.6 ± 0.1, 1.9 ± 0.2 vs 0.9 ± 0.1 and 7.0 ± 1.0 vs 10.0 ± 1.1 mmol/l, respectively, p < 0.05). Basal whole body glucose clearance was increased twofold in APOC1 mice compared with wild-type mice (18 ± 2 vs 10 ± 1 ml · kg^–1· min^–1, p < 0.05). Insulin-mediated whole body glucose uptake, glycolysis (generation of ³H₂O) and glucose storage increased in APOC1 mice compared with wild-type mice (339 ± 28 vs 200 ± 11; 183 ± 39 vs 128 ± 17 and 156 ± 44 vs 72 ± 17 μmol · kg^–1· min^–1, p < 0.05, respectively), corresponding with a twofold to threefold increase in skeletal muscle glycogenesis and de novo lipogenesis from 3-³H-glucose in skeletal muscle and adipose tissue (p < 0.05). Basal whole body NEFA clearance was decreased threefold in APOC1 mice compared with wild-type mice (98 ± 21 vs 314 ± 88 ml · kg^–1· min^–1, p < 0.05). Insulin-mediated whole body NEFA uptake, NEFA oxidation (generation of ³H₂O) and NEFA storage were lower in APOC1 mice than in wild-type mice (15 ± 3 vs 33 ± 6; 3 ± 2 vs 11 ± 4 and 12 ± 2 vs 22 ± 4 μmol · kg^–1· min^–1, p < 0.05) in the face of higher plasma NEFA concentrations (1.3 ± 0.3 vs 0.5 ± 0.1 mmol/l, p < 0.05), respectively. Mean arterial pressure and heart rate were similar in APOC1 vs wild-type mice (82 ± 4 vs 85 ± 3 mm Hg and 459 ± 14 vs 484 ± 11 beats · min^–1). Conclusions/interpretation. 1) Hyperlipidaemic APOC1 mice show reduced NEFA and increased glucose metabolism under both basal and insulin-mediated conditions, suggesting an intrinsic defect in NEFA metabolism. Primary hyperlipidaemia alone in APOC1 mice does not lead to insulin resistance for glucose metabolism and high blood pressure. [Diabetologia (2001) 44: 437–443] Received: 14 September 2000 and in revised form: 23 November 2000     

Insulin Deficiency and Increased Plasma Concentration of Intact and 32/33 Split Proinsulin in Subjects with Impaired Glucose Tolerance

In order to determine insulin status and beta cell function during the oral glucose tolerance test (OGTT), in impaired glucose tolerance (IGT), 51 such subjects and matched controls, identified during a population survey for diabetes, underwent a 75 g OGTT. Fasting, 30 min and 2 h insulin and intact proinsulin, and fasting and 2 h 32/33 split proinsulin, were measured by specific two-site immunoradiometric assays. The subjects with IGT had higher fasting (geometric mean ± SD, 5.0 ± 4.0 pmol^?1 vs 2.9 ± 1.7, p < 0.02) and 2 h intact proinsulin (23 ± 14 vs 14 ± 12, p < 0.0001), and fasting (3.2 ± 3 pmol^?1 vs 1.8 ± 1.8, p < 0.0007) and 2 h 32/33 split proinsulin (18.3 ± 19 pmol^?1 vs 6.6 ± 15, p < 0.0001). Despite higher plasma glucose concentrations, the IGT group had similar fasting insulin, lower 30 min insulin (216 ± 124 pmol^?1 vs 278 ± 130, p < 0.02), and a lower 30 min insulin/glucose ratio (23.7 ± 2.1 vs 34.8 ± 2.3, p < 0.002). The percentage of fasting proinsulin-like to total insulin-like molecules was higher in those with IGT (15.3 ± 8% vs 11.6 ± 8, p < 0.04). After 6 months, at repeat OGTT, the same subjects with IGT were classified as ‘persisters’ or ‘reverters’. The persister (24/51 47.1%), at initial OGTT, had a higher 2 h glucose level, a greater BMI and higher systolic blood pressure, but other parameters were similar to the reverters. In the reverters, when baseline variables were compared to those recorded at six month follow-up, there was a reduction in 2 h intact (23.8 ± 13 pmol^?1 vs 19.4 = 10, p < 0.02) and 32/33 split proinsulin (20.4 ± 18 pmol^?1 vs 13.8 ± 13, p < 0.006), and an increase in fasting insulin (41 ± 30 pmol^?1 vs 54 ± 35, p < 0.02), respectively, despite no change in fasting glucose. These findings show that IGT is associated with beta cell dysfunction and reduced early insulin secretion during the OGTT. In some subjects with IGT these abnormalities show improvement in the short term.     

Relationships between plasma measures of oxidative stress and metabolic control in NIDDM

Summary Diabetes mellitus may be associated with increased lipid peroxidation which may contribute to long-term tissue damage. To test this hypothesis, we measured hydroperoxides (ROOHs) as well as α-tocopherol in plasma from healthy subjects and individuals with non-insulin-dependent diabetes mellitus (NIDDM) (n = 41 and 87, respectively). ROOHs were analysed using the ferrous oxidation with xylenol orange version II (FOX2) assay in conjunction with a specific ROOH reductant, triphenylphosphine. α-Tocopherol was analysed by HPLC with fluorimetric detection. NIDDM patients had lower cholesterol standardised α-tocopherol levels as compared to control subjects (3.3 ± 1.0 vs 5.1 ± 2.3 (μmol/l)/(mmol/l); p < 0.0005, Mann-Whitney test): range (1.5–6.5 vs 1.9–13.0, respectively). Plasma ROOHs were substantially higher in the diabetic subjects compared to those of the control subjects (9.4 ± 3.3 vs 4.1 ± 2.2 μmol/l; p < 0.0005 Mann-Whitney test: range 2.7–16.8 vs 0.4–10.3, respectively). ROOH/cholesterol standardised α-tocopherol ratio was significantly higher in the diabetic patients compared to control subjects (3.2 ± 1.6 vs 0.9 ± 0.6; p < 0.0005, Mann-Whitney test: range 0.7–8.3 and 0.1–2.7, respectively). Plasma levels of ROOHs and α-tocopherol were similar in diabetic patients with or without complications as well as in smokers and non-smokers. The present study confirms previous findings from this laboratory that NIDDM is associated with increased oxidative stress as assessed by plasma ROOHs. Increased oxidative stress in diabetic patients appears to be related to the underlying metabolic abnormalities in diabetes, rather than to the complications of this disease. We therefore suggest that oxidative stress is an early stage in the disease pathology, which may contribute to the development of complications. [Diabetologia (1997) 40: 647–653] Received: 21 October 1996 and in final revised form: 12 February 1997     

Evidence for Impaired Pancreatic Beta Cell Function in South African Indians with Impaired Glucose Tolerance

In a prospective study of South African Indians with impaired glucose tolerance (IGT), the serum insulin response during a 75 g oral glucose tolerance test (OGTT) was examined in 128 subjects who were classified as IGT 1 year previously (year 0) and in 60 matched control subjects. Based on the results at year 1, study subjects were divided into three groups, using World Health Organization criteria for glucose tolerance: IGT (n = 47), diabetes (n = 41), and transient IGT (normal glucose tolerance) (n = 40). When compared with the control group, despite higher plasma glucose concentrations, the IGT group showed similar fasting insulin, but lower 30-min insulin response (57.4 ± 1.9 mUI^?1 vs 86.5 ± 1.8, p<0.001) and lower 30-min insulin/glucose ratio (7.4 ± 5.2 vs 13.3 ± 8.7, p < 0.001). The insulinogenic index was lower in the IGT group than in the control group at 30, 60, 90, and 120 min (p < 0.01, p < 0.001, p < 0.001, p < 0.001, respectively). The 2-h insulin response was higher in the IGT group (106.7 ± 1.9 mUI^?1 vs 59.2 ± 1.9, p < 0.01). The IGT group displayed a delayed pattern of insulin response with maximum levels only at 2-h. Insulin area was similar in the two groups. In the transient IGT group, despite similar plasma glucose levels, the insulin responses at 0, 15, 30, and 60 min (p < 0.01, p < 0.001, p < 0.001, p < 0.001, respectively) were lower than in the control group; the 30-min insulin/glucose ratio (7.1 ± 5.1 vs 13.3 ± 8.7, p < 0.001) and 60-min insulinogenic index (46.9 ± 86.3 vs 123.4 ± 206.3, p < 0.001) were also lower in the transient IGT group. This study has shown that IGT in South African Indians is characterized by a diminished early phase insulin response and delayed (2-h) hyperinsulinaemia during OGTT. Such findings would suggest that in this population group impaired early beta cell function is an important pathophysiological abnormality underlying IGT.     

Assessment of insulin sensitivity and secretion with the labelled intravenous glucose tolerance test: improved modelling analysis

Summary A new modelling analysis was developed to assess insulin sensitivity with a tracer-modified intravenous glucose tolerance test (IVGTT). IVGTTs were performed in 5 normal (NGT) and 7 non-insulin-dependent diabetic (NIDDM) subjects. A 300 mg/kg glucose bolus containing [6,6-²H₂]glucose was given at time 0. After 20 min, insulin was infused for 5 min (NGT, 0.03; NIDDM, 0.05 U/kg). Concentrations of tracer, glucose, insulin and C-peptide were measured for 240 min. A circulatory model for glucose kinetics was used. Glucose clearance was assumed to depend linearly on plasma insulin concentration delayed. Model parameters were: basal glucose clearance (Cl_b), glucose clearance at 600 pmol/l insulin concentration (Cl₆₀₀), basal glucose production (P_b), basal insulin sensitivity index (BSI = Cl_b/basal insulin concentration); incremental insulin sensitivity index (ISI = slope of the relationship between insulin concentration and glucose clearance). Insulin secretion was calculated by deconvolution of C-peptide data. Indices of basal pancreatic sensitivity (PSI_b) and first (PSI₁) and second-phase (PSI₂) sensitivity were calculated by normalizing insulin secretion to the prevailing glucose levels. Diabetic subjects were found to be insulin resistant (BSI: 2.3 ± 0.6 vs 0.76 ± 0.18 ml · min^–1· m^–2· pmol/l^–1, p < 0.02; ISI: 0.40 ± 0.06 vs 0.13 ± 0.05 ml · min^–1· m^–2· pmol/l^–1, p < 0.02; Cl₆₀₀: 333 ± 47 vs 137 ± 26 ml · min^–1· m^–2, p < 0.01; NGT vs NIDDM). P_b was not elevated in NIDDM (588 ± 169 vs 606 ± 123 μmol · min^–1· m^–2, NGT vs NIDDM). Hepatic insulin resistance was however present as basal glucose and insulin were higher. PSI₁ was impaired in NIDDM (67 ± 15 vs 12 ± 7 pmol · min^–1· m^–2· mmol/l^–1, p < 0.02; NGT vs NIDDM). In NGT and in a subset of NIDDM subjects (n = 4), PSI_b was inversely correlated with BSI (r = 0.95, p < 0.0001, log transformation). This suggests the existence of a compensatory mechanism that increases pancreatic sensitivity in the presence of insulin resistance, which is normal in some NIDDM subjects and impaired in others. In conclusion, using a simple test the present analysis provides a rich set of parameters characterizing glucose metabolism and insulin secretion, agrees with the literature, and provides some new information on the relationship between insulin sensitivity and secretion. [Diabetologia (1998) 41: 1029–1039] Received: 17 September 1997 and in final revised form: 28 April 1998     

Prevention of Type II diabetes in subjects with impaired glucose tolerance: the Diabetes Prevention Study (DPS) in Finland Study design and 1-year interim report on the feasibility of the lifestyle intervention programme

Aims/hypothesis. The aim of the Diabetes Prevention Study is to assess the efficacy of an intensive diet-exercise programme in preventing or delaying Type II (non-insulin-dependent) diabetes mellitus in subjects with impaired glucose tolerance, to evaluate the effects of the intervention programme on cardiovascular risk factors and to assess the determinants for the progression to diabetes in persons with impaired glucose tolerance. Methods. A total of 523 overweight subjects with impaired glucose tolerance ascertained by two oral glucose tolerance tests were randomised to either a control or intervention group. The control subjects received general information at the start of the trial about the lifestyle changes necessary to prevent diabetes and about annual follow-up visits. The intervention subjects had seven sessions with a nutritionist during the first year and a visit every 3 months thereafter aimed at reducing weight, the intake of saturated fat and increasing the intake of dietary fibre. Intervention subjects were also guided individually to increase their physical activity. Results. During the first year, weight loss in the first 212 study subjects was 4.7 ± 5.5 vs 0.9 ± 4.1 kg in the intervention and control group, respectively (p < 0.001). The plasma glucose concentrations (fasting: 5.9 ± 0.7 vs 6.4 ± 0.8 mmol/l, p < 0.001; and 2-h 7.8 ± 1.8 vs 8.5 ± 2.3 mmol/l, p < 0.05) were significantly lower in the intervention group after the first year of intervention. Favourable changes were also found in blood pressure, serum lipids and anthropometric indices in the intervention group. Conclusion/interpretation. The interim results show the efficacy and feasibility of the lifestyle intervention programme. [Diabetologia (1999) 42: 793–801] Received: 7 December 1998 and in revised form: 23 February 1999     

The set point for maternal glucose homeostasis is lowered during late pregnancy in the rat: the role of the islet beta-cell and liver

Summary The aim of this study was to determine the effects of late pregnancy on the ability of insulin to suppress maternal hepatic glucose production in the rat. Unlike in most previous studies, suppression of hepatic glucose production was measured at levels of glycaemia above the relatively hypoglycaemic basal pregnant level. Glucose kinetics were measured using steady-state tracer methodology in chronically catheterised, conscious virgin control and pregnant rats, firstly, during basal and low-dose hyperinsulinaemic euglycaemic clamp conditions and secondly, during a three-step glucose infusion protocol (glucose infusion rates of 0, 60 and 150 μmol · kg⁻¹· min⁻¹). During the clamps, plasma glucose levels were not different (6.1 ± 0.4 vs 6.5 ± 0.3 mmol/l, pregnant vs virgin; N. S.), but plasma insulin levels were higher in the pregnant rats (242 ± 30 vs 154 ± 18 pmol/l, pregnant vs virgin; p < 0.05) most probably due to stimulated endogenous insulin release in this group. Hepatic glucose production was suppressed from basal levels by 41 % in virgin and 90 % in pregnant rats. During the glucose infusion studies, at matched insulin levels (147 ± 10 vs 152 ± 14 pmol/l), but at plasma glucose levels which were much lower in the pregnant rats (5.5 ± 0.2 vs 8.4 ± 0.6 mmol/l, pregnant vs virgin; p < 0.0001), hepatic glucose production was shown to be suppressed by a similar degree in both groups (41 ± 5 vs 51 ± 5 % from basal, pregnant vs virgin; N. S.). Both the plasma insulin and percentage suppression of hepatic glucose production dose responses to plasma glucose were markedly shifted to the left indicating that the plasma glucose set point is lowered in pregnancy. In conclusion, suppression of hepatic glucose production by insulin is not impaired and the set point for plasma glucose homeostasis is lowered during late pregnancy in the rat. [Diabetologia (1996) 39: 785–792] Received: 2 October 1995 and in final revised form: 1 February 1996     

Lowering fatty acids potentiates acute insulin response in first degree relatives of people with Type II diabetes

Summary Studies have shown that a high plasma non-esterified fatty acid concentration may inhibit glucose induced insulin secretion in vitro and in vivo. The effect of lowering the fatty acid concentration on the acute insulin response was investigated in first degree relatives of people with Type II diabetes in a double-blind, randomised, placebo-controlled trial. Fifty first degree relatives of people with Type II diabetes volunteered for the study. Twenty five were given acipimox (250 mg/day, four times daily) and 25 placebo. The group treated with acipimox had a lower 2-h plasma glucose concentration (6.1 ± 0.2 vs 7.7 ± 0.3 vs mmol/l, p < 0.01); better insulin-mediated glucose uptake (35.4 ± 0.5 vs 28.3 ± 0.4 μmol/kg fat free mass per min, p < 0.01), acute insulin response (68 ± 4.4 vs 46 ± 7.3 mU/l, p < 0.01) and respiratory quotient (0.81 ± 0.02 vs 0.77 ± 0.03, p < 0.05); and a rise in the plasma glucagon (164 ± 63 vs 134 ± 72 ng/l, p < 0.05), growth hormone (1.31 ± 0.13 vs 0.97 ± 0.21 μg/l, p < 0.03) and cortisol (325 ± 41 vs 284 ± 139 nmol/l, p < 0.05) concentrations. The difference in the acute insulin response persisted, even after adjustment for the 2-h plasma glucose concentration, insulin-mediated glucose uptake, the fasting plasma glucagon concentration and the growth hormone concentration (p < 0.05). In a subgroup of eight patients acipimox was compared with acipimox plus intralipid. The acute insulin response (44 ± 5.1 vs 71 ± 5.3 mU/l, p < 0.01) and the insulin-mediated glucose uptake (27.4 ± 0.4 vs 36.7 ± 0.5 μmol/kg fat free mass per min, p < 0.003) were lower with acipimox plus intralipid treatment than with acipimox alone. It is concluded that long term acipimox treatment lowers the plasma fasting free fatty acid concentration and improves the acute insulin response and the insulin mediated glucose uptake. [Diabetologia (1998) 41: 1127–1132] Received: 27 January 1998 and in final revised form 29 April 1998     

The relationship of glycaemic level to advanced glycation end-product (AGE) accumulation and retinal pathology in the spontaneous diabetic hamster

Summary To assess the relationship between glucose and advanced glycation end products (AGE) and the relationship between AGE and retinal changes in vivo, we studied the time course of retinopathy over 12 months in trypsin digest preparations and measured glycaemia and retinal AGE in spontaneous diabetic hamsters of mild (MD) and severe (SD) phenotypes. Blood glucose levels were elevated in MD (9.44 ± 0.76 mmol/l) and in SD (3 months: 24.3 ± 1.4 mmol/l; 12 months: 31.7 ± 0.8 mmol/l) over non-diabetic controls (NC: 7.15 ± 0.25 mmol/l; p < 0.05 or less vs MD; p < 0.001 vs SD). Similar relations were found for HbA₁. Retinal AGE in mild diabetes was 405 ± 11.3 arbitrary units (AU) (NC 245 ± 7.7; p < 0.01) after 3 months and remained unchanged. A non-linear increase of AGE over time was found in severe hyperglycaemic hamsters (466 ± 21 AU after 3 months and 758 ± 21 AU after 12 months; p < 0.001 vs MD). Pericyte loss in mild diabetes progressed from –26 % after 3 months to –41 % after 12 months (p < 0.001 vs NC). Whereas the initial pericyte loss in severely diabetic hamsters was identical to the mildly diabetic group, a higher degree of pericyte loss occurred after 12 months (–57 %; p < 0.05 vs MD). Endothelial cell numbers remained unaffected by mild hyperglycaemia, but significantly increased over time in severe diabetes reaching 31.7 % above controls after 12 months (p < 0.001 vs NC and MD). Microaneurysms were absent in all retinae examined. Acellular capillary segments were increased in mild diabetes (3.83 ± 0.31 per mm² of retinal area) and severe diabetes (7.83 ± 0.73) over controls (1.0 ± 0.23). These data suggest that a threshold of glycaemia might exist above which AGE removal systems become saturated. Pericyte loss and acellular capillary formation are associated with mild increases in blood glucose and AGE levels while endothelial cell proliferation requires higher glucose and AGE levels. [Diabetologia (1998) 41: 165–170] Received: 31 July 1997 and in revised form: 29 September 1997