Insulin resistance, beta cell dysfunction and visceral adiposity as predictors of incident diabetes: the Insulin Resistance Atherosclerosis Study (IRAS) Family study.

AIMS/HYPOTHESIS: Central obesity, insulin resistance and beta cell dysfunction are independent risk factors for incident type 2 diabetes, although few studies have used detailed measures of these disorders. Our objective was to study the association of directly measured visceral and subcutaneous adipose tissue (VAT, SAT), insulin sensitivity (S (I)) and the acute insulin response (AIR) with incident type 2 diabetes. METHODS: Participants were 1,230 Hispanic-Americans and African-Americans in the Insulin Resistance Atherosclerosis Study (IRAS) Family Study who were free of type 2 diabetes at baseline (2000-2002). S (I) and AIR were determined from frequently sampled IVGTTs with minimal model analysis. VAT and SAT were determined by computed tomography. Impaired fasting glucose and type 2 diabetes were defined according to American Diabetes Association criteria. RESULTS: Incident type 2 diabetes was diagnosed in 90 participants after 5 years. After adjustment for age, sex, ethnicity, centre, impaired fasting glucose, triacylglycerol, HDL-cholesterol and systolic BP, both S(I) and AIR were inversely associated with type 2 diabetes (S (I), OR 0.53, 95% CI 0.39-0.73; AIR, OR 0.22, 95% CI 0.14-0.34 per SD; both p < 0.001), while both VAT and SAT were positively associated with type 2 diabetes (VAT, OR 1.68, 95% CI 1.22-2.33; SAT, OR 1.49, 95% CI 1.13-1.99; both p < 0.01). In a model including all four factors, S (I) and AIR (S (I), OR 0.55, 95% CI 0.37-0.80; AIR, OR 0.21, 95% CI 0.13-0.33; both p < 0.01) were significant predictors of type 2 diabetes, although associations with VAT and SAT were no longer significant. A significant sex x VAT interaction indicated a stronger association of VAT with type 2 diabetes in women than in men. CONCLUSIONS/INTERPRETATION: Insulin resistance, beta cell dysfunction and VAT predicted incident type 2 diabetes, with evidence of a stronger association of VAT with type 2 diabetes among women.

A genome-wide association scan for acute insulin response to glucose in Hispanic-Americans: the Insulin Resistance Atherosclerosis Family Study (IRAS FS).

AIMS/HYPOTHESIS: This study sought to identify genes and regions in the human genome that are associated with the acute insulin response to glucose (AIRg), an important predictor of type 2 diabetes, in Hispanic-American participants from the Insulin Resistance Atherosclerosis Family Study (IRAS FS). METHODS: A two-stage genome-wide association scan (GWAS) was performed in IRAS FS Hispanic-American samples. In the first stage, 317K single nucleotide polymorphisms (SNPs) were assessed in 229 Hispanic-American DNA samples from 34 families from San Antonio, TX, USA. SNPs with the most significant associations with AIRg were genotyped in the entire set of IRAS FS Hispanic-American samples (n = 1,190). In chromosomal regions with evidence of association, additional SNPs were genotyped to capture variation in genes. RESULTS: No individual SNP achieved genome-wide levels of significance (p < 5 x 10(-7)); however, two regions (chromosomes 6p21 and 20p11) had multiple highly ranked SNPs that were associated with AIRg. Additional genotyping in these regions supported the initial evidence of variants contributing to variation in AIRg. One region resides in a gene desert between PXT1 and KCTD20 on 6p21, while the region on 20p11 has several viable candidate genes (ENTPD6, PYGB, GINS1 and RP4-691N24.1). CONCLUSIONS/INTERPRETATION: A GWAS in Hispanic-American samples identified several candidate genes and loci that may be associated with AIRg. These associations explain a small component of variation in AIRg. The genes identified are involved in phosphorylation and ion transport, and provide preliminary evidence that these processes are important in beta cell response.

Glucose-dependent insulin release from genetically engineered K cells.

Genetic engineering of non-beta cells to release insulin upon feeding could be a therapeutic modality for patients with diabetes. A tumor-derived K-cell line was induced to produce human insulin by providing the cells with the human insulin gene linked to the 5'-regulatory region of the gene encoding glucose-dependent insulinotropic polypeptide (GIP). Mice expressing this transgene produced human insulin specifically in gut K cells. This insulin protected the mice from developing diabetes and maintained glucose tolerance after destruction of the native insulin-producing beta cells.

Altered distribution of insulin between intravascular and interstitial compartments in STZ-induced diabetic rats.

These studies were undertaken to determine the distribution of insulin between the intravascular and interstitial compartments in 19 nondiabetic and 16 high-dose (60 mg/kg) STZ-induced diabetic rats. Anesthetized male rats underwent cannulation of the thoracic lymph duct with continuous collection of lymph by passive drainage; 90-min hyperinsulinemic (0, 14, 28, or 42 pmol.kg-1 x min-1) glucose clamps were performed with serum glucose held at the preinfusion level. Integrated lymph samples were collected, with serum collection at the midpoint. Before insulin infusion, insulin levels in lymph as a percentage of levels in serum were higher in the diabetic rats (lymph = 103 +/- 8% of 52 matched serum values, mean +/- SE) compared with nondiabetic rats (lymph = 83 +/- 3% of serum; n = 84, P < 0.008). Under steady-state conditions during insulin infusion (66-90 min), lymph insulin as a percentage of serum insulin decreased significantly in nondiabetic animals (range 53-62% in all insulin-infused groups, vs. 94% in the absence of insulin infusion, P < 0.0001). However, in the diabetic animals, the proportion was similar in the presence or absence of infused insulin (range 57-63% in insulin-infused groups vs. 70% in the absence of infused insulin, P = NS). Lymph flow rate was significantly higher in diabetic than nondiabetic rats (37.7 +/- 7.2 vs. 17.9 +/- 1.5 microliters/min [mean +/- SE], P < 0.02). Distribution of [14C]inulin was uniform between the two compartments in all animals.(ABSTRACT TRUNCATED AT 250 WORDS)

Regulation of rat insulin-receptor kinase by glucose in vivo.

The effects of acute and chronic hyperglycemia on the kinase activity of insulin receptors in vivo were studied in the rat. Skeletal muscle-derived insulin receptors were isolated, with preservation of the in vivo phosphorylation state, from nondiabetic rats subjected to hyperinsulinemic clamps at either euglycemia (mean 5.2 mM) or hyperglycemia (14.4 mM) or from streptozocin-induced diabetic rats at euglycemia (5.1 mM) or hyperglycemia (14.2 mM). Kinase activity toward histone of insulin receptors from nondiabetic animals rendered hyperglycemic for 80-90 min was 50% higher than that of receptors from rats clamped at euglycemia (mean +/- SE 4.5 +/- 0.4 vs. 3.0 +/- 0.3 fmol of phosphate into histone, respectively, P less than 0.02), although kinase activity of receptors isolated from animals rendered diabetic for 10-14 days before hyperglycemic or euglycemic clamps showed no such effect. These results suggest that acute hyperglycemia may increase insulin-receptor kinase activity in vivo, possibly augmenting glucose disposal thereby, whereas the chronic hyperglycemia of diabetes mellitus results in metabolic derangements that nullify this effect.

Rat insulin-receptor kinase activity correlates with in vivo insulin action.

Tyrosine kinase activity of skeletal muscle-derived insulin receptors isolated from rats that had undergone euglycemic clamps at various insulin infusion rates was examined. Receptors were isolated under conditions designed to preserve their in vivo phosphorylation state, and their kinase activity toward histone was measured in the absence of in vitro exposure to insulin. Results showed that significant activation of the insulin-receptor kinase occurred after exposure in vivo to mean serum insulin concentrations as low as 34 +/- 3.5 microU/ml and that maximal activation was achieved by insulin levels less than or equal to 2000 microU/ml. There was a highly significant correlation between receptor kinase activity and serum insulin concentration in the physiologic range (r = .92, P less than .0001) and between kinase activity and glucose utilization rate (r = .74, P less than .0001). These findings further support a role for the insulin-receptor kinase in insulin action in vivo, and this model provides a novel method for the study of the effect of factors known to influence insulin action on the insulin-receptor kinase under physiologic conditions.

Bis(maltolato)oxovanadium(IV) attenuates hyperinsulinemia and hypertension in spontaneously hypertensive rats.

We previously reported that bis(maltolato)oxovanadium(IV) (BMOV), an organic vanadium complex, decreased plasma insulin concentrations in nondiabetic rats without affecting plasma glucose levels (McNeill JH, Yuen VG, Hoveyda HR, Orvig C: Bis(maltolato)oxovanadium(IV) is a potent insulin mimic. J Med Chem 35:1489-1491, 1992). In this study, chronic oral BMOV treatment was started in 6-week-old spontaneously hypertensive (SH) rats and their Wistar-Kyoto (WKY) controls, and the effects of the drug on insulin sensitivity, plasma insulin, and blood pressure (BP) were studied. BMOV (0.35-0.45 mmol.kg-1.day-1) caused a sustained reduction in plasma insulin (198 +/- 6 vs. untreated 366 +/- 13.2 pM, P < 0.0001) and systolic BP (149 +/- 3 vs. untreated 184 +/- 3 mmHg, P < 0.0001) in SH rats. No changes were seen in WKY rats (plasma insulin: treated 228 +/- 4.8 vs. untreated 222.6 +/- 3.6 pM, P > 0.05; BP: treated 134 +/- 3 vs. untreated 134 +/- 5 mmHg, P > 0.05). At 13 weeks of age, euglycemic clamps were performed in fasted, conscious, mobile rats. During low-dose insulin infusions (14 pmol.kg-1.min-1) with concomitant somatostatin administration, neither hepatic glucose output nor total body glucose uptake differed between the untreated SH and WKY rats. Insulin sensitivity, expressed as steady-state glucose clearance per unit of plasma insulin, was higher in the untreated SH compared with the untreated WKY rats (2.1 +/- 0.2 vs. 1.2 +/- 0.1 ml.kg-1.h-1.pM-1, P < 0.002). BMOV further enhanced insulin sensitivity in SH rats (3.6 +/- 0.4, P < 0.002 vs. untreated SH rats).(ABSTRACT TRUNCATED AT 250 WORDS)

Tumor necrosis factor-alpha induces hepatic insulin resistance in obese Zucker (fa/fa) rats via interaction of leukocyte antigen-related tyrosine phosphatase with focal adhesion kinase.

The molecular mechanism whereby tumor necrosis factor-alpha (TNF-alpha) induces insulin resistance in obesity is not well understood. Previously, we have shown that inhibition of TNF-alpha improved hepatic insulin sensitivity in obese Zucker rats without altering the tyrosine phosphorylation of liver insulin receptors (IRs), which indicates that the TNF-alpha and insulin-signaling cascades interact distally to the IR. To assess the effects of TNF-alpha on signaling molecules downstream from the IR, we analyzed the tyrosine phosphorylation patterns of liver homogenate proteins from TNF-alpha-neutralized fa/fa rats and showed that focal adhesion kinase (FAK) was consistently hyperphosphorylated (4.5-fold). Moreover, intravenous insulin increased hepatic FAK phosphorylation in a time-dependent manner in Sprague-Dawley rats, which suggests that TNF-alpha may induce hepatic insulin resistance by preventing FAK phosphorylation in response to insulin treatment. To explore the cellular mechanism whereby TNF-alpha regulates phosphorylation of FAK in the liver, we measured c-Src kinase activity and the abundance of 3 major protein tyrosine phosphatases (PTPs) (PTP-1B, leukocyte antigen-related tyrosine phosphatase [LAR], and src homology 2 domain-containing protein-tyrosine phosphatase [SHPTP-2]) in liver homogenates from obese Zucker rats after TNF-alpha blockade. Hepatic c-Src kinase activity was unaltered, but LAR protein was reduced by 75%. In addition, TNF-alpha blockade reduced hepatic PTP activity toward tyrosine phosphorylated FAK by 70%, and this was accounted for by immunodepletion of LAR. Incubation of HepG2 cells with TNF-alpha increased LAR protein levels in a dose-dependent manner. Additionally, pretreatment with TNF-alpha abolished insulin-stimulated tyrosine phosphorylation of FAK in HepG2 cells but had no effect on IR tyrosine phosphorylation or expression. These data suggest that TNF-alpha promotes LAR expression and thus prevents insulin-mediated tyrosine phosphorylation of FAK. This probably represents the interface between TNF-alpha and insulin signaling in the liver.

The effect of glyburide on beta-cell sensitivity to glucose-dependent insulinotropic polypeptide.

OBJECTIVE: To assess whether treatment with glyburide alters beta-cell sensitivity to GIP in NIDDM patients. RESEARCH DESIGN AND METHODS: We studied 5 untreated NIDDM patients in a meal study (Ensure, 240 ml/M2) and a 2-h hyperglycemic glucose clamp study (glucose 5.4 mM above fasting). From 60 to 120 min of the clamp, GIP was infused in a primed continuous manner at a rate of 2 pmol.kg-1 x min-1. Subjects then were treated with glyburide. After they had been on a stable dose of medication for 1 mo, the meal study and glucose clamp studies were repeated. RESULTS: In response to treatment, a decrease in fasting glucose and an increase in weight was observed (12.8 +/- 1.8 vs. 8.5 +/- 0.8 mM and 74.3 +/- 6.3 vs. 76.1 +/- 6.3 kg, respectively, P < 0.05). In response to the meal study, the AUC for glucose was less, for insulin was increased, and for GIP was unchanged after treatment (16.9 +/- 2.1 vs. 12.6 +/- 6.9 mM, P < 0.05; 161 +/- 47 vs. 242 +/- 60 pM, P < 0.05; and 199 +/- 22 vs. 219 +/- 18 pM, respectively). During the hyperglycemic clamp, steady-state glucose and 90- to 120-min GIP values were equivalent before and after treatment (18.0 +/- 1.3 vs. 18.3 +/- 1.3 mM and 302 +/- 59 vs. 298 +/- 37 pM, respectively). The 90-120 min insulin responses to the hyperglycemic clamp were greater after therapy (123 +/- 37 vs. 283 +/- 80 pM, P < 0.05) reflecting increased beta-cell responses to GIP. CONCLUSIONS: We conclude that glyburide enhances beta-cell sensitivity to GIP.

An in vivo model for elucidation of the mechanism of tumor necrosis factor-alpha (TNF-alpha)-induced insulin resistance: evidence for differential regulation of insulin signaling by TNF-alpha.

Tumor necrosis factor-alpha (TNF-alpha) has been shown to induce insulin resistance in cultured cells as well as in animal models. The aim of this study was to map the in vivo mechanism whereby TNF-alpha contributes to the pathogenesis of impaired insulin signaling, using obese and lean Zucker rats in which TNF-alpha activity was inhibited through adenovirus-mediated gene transfer. We employed a replication-incompetent adenovirus-5 (Ad5) vector to endogenously express a TNF inhibitor (TNFi) gene, which encodes a chimeric protein consisting of the extracellular domain of the human 55-kDa TNF receptor joined to a mouse IgG heavy chain. Control animals consisted of rats infected with the same titer of adenovirus carrying the lac-z complementary DNA, encoding for beta-galactosidase. There was a significant reduction in plasma insulin and free fatty acid levels in TNFi obese rats 2 days following Ad5 administration. The peripheral insulin sensitivity index was 50% greater, whereas hepatic glucose output was completely suppressed during hyperinsulinemic glucose clamps in TNFi obese animals, with no differences observed between the two lean groups. The improvement in peripheral and hepatic sensitivity to insulin seen in the obese animals was independent of insulin receptor (IR) number and insulin binding affinity for IR. However, TNF-alpha neutralization led to a 2.5-fold increase in tyrosine phosphorylation of IR in skeletal muscle, whereas this was unchanged in liver. There was also a 4-fold increase in particulate protein tyrosine phosphatase activity of skeletal muscle in TNFi obese animals vs. beta-galactosidase controls, whereas protein tyrosine phosphatase activity in liver was unchanged. These results suggest that TNF-alpha is a mediator of insulin resistance in obesity and may modulate IR signaling in skeletal muscle and liver through different pathways. TNF-alpha may affect insulin action in the liver either at sites distal to the IR or indirectly, possibly because of increased provision of gluconeogenic substrates or altered counterregulation. In addition, the Ad5-mediated gene delivery system employed here provides an in vivo model that is efficient and economical for exploring mechanisms involved in TNF-alpha-induced insulin resistance in various genetic models of obesity-linked diabetes.

Vanadyl sulfate prevents fructose-induced hyperinsulinemia and hypertension in rats.

To determine whether insulin resistance and hyperinsulinemia are causally related to fructose-induced hypertension, we used vanadyl sulfate, a drug that improves insulin sensitivity in rats. Chronic oral vanadyl treatment was initiated in 6-week-old male Sprague-Dawley rats. One week after vanadyl was started, rats were fed either normal rat chow or a fructose-enriched diet. Plasma glucose and insulin levels and systolic blood pressure were measured weekly for 4 weeks. Fructose feeding induced hyperinsulinemia (fructose-fed, 366.6 +/- 8.4 versus control, 276 +/- 10.8 pmol/L, P < .001) and increased blood pressure (fructose-fed, 160 +/- 3.0 versus control, 124 +/- 3.0 mm Hg, P < .001). Vanadyl (0.4 to 0.6 mmol/kg per day) prevented the rise in plasma insulin (treated, 211.2 +/- 6.0 pmol/L, P < .001) and blood pressure (treated, 127 +/- 3.0 mm Hg, P < .001) in the fructose-fed rats without a change in plasma glucose. No change in blood pressure was seen in the control group. After 4 weeks, euglycemic clamps were performed on 20-hour fasted, conscious, mobile rats. Low-dose porcine insulin infusion (14 pmol/kg per minute) with concomitant somatostatin infusion resulted in similar steady-state plasma glucose and insulin levels in the various groups. Hepatic glucose production was suppressed and similar among various groups under clamp conditions. Insulin sensitivity index (micromoles of glucose per kilogram per hour per picomole per liter of insulin) was reduced in the fructose-fed rats compared with controls (fructose-fed, 0.9 +/- 0.4 versus control, 5.4 +/- 1.2, P < .002).(ABSTRACT TRUNCATED AT 250 WORDS)

Plasma renin and prorenin (inactive renin) in diabetes mellitus: effects of intravenous furosemide.

PRA, active renin, and prorenin were measured in 32 normotensive diabetic patients and 14 normal subjects of similar ages before and after iv injection of 40 mg furosemide. The majority of the diabetic patients had normal PRA and active renin levels before and after furosemide, but in 4 normal subjects and 5 diabetic patients PRA did not rise after furosemide treatment to at least 0.25 ng angiotensin I/(L.s) as previously found in 90% of normal subjects. Two thirds of the diabetic patients had higher plasma prorenin levels than the normal subjects. Four of the 5 diabetic patients whose PRA failed to rise to the usual level after furosemide treatment attained a plasma prorenin level higher than any normal subject, suggesting that furosemide stimulated synthesis and secretion of prorenin, but that conversion of prorenin to active renin was impaired. These 5 diabetic patients also had higher plasma creatinine and potassium levels as well as an increased frequency of albuminuria compared with the other diabetic patients. In contrast, the 4 normal subjects whose PRA failed to reach the usual level after furosemide treatment had low unresponsive prorenin levels, indicating that furosemide did not evoke the expected increase in prorenin or renin synthesis. We conclude that the inability of some diabetic patients to increase PRA after furosemide treatment is not dependent on failure of renin synthesis, but reflects an impairment of the normal processing of prorenin, leading to high levels of prorenin in plasma.

Insulin-mediated increase in blood flow is impaired in the elderly.

It has recently been recognized that the ability of insulin to augment blood flow is reduced in insulin-resistant conditions such as obesity and noninsulin-dependent diabetes mellitus. Normal aging is characterized by resistance to insulin-mediated glucose uptake. We undertook the following studies with the hypothesis that the resistance to insulin-mediated glucose uptake that occurs with aging is caused in part by a reduction in insulin-mediated blood flow. These experiments were conducted on healthy young (n = 13; age, 24 +/- 1 yr; body mass index: 22.2 +/- 0.6 kg/m2) and old (n = 13; age, 77 +/- 1 yr; body mass index: 24.2 +/- 0.5 kg/m2) subjects. Each subject underwent two studies. In the control study, saline was infused for 4 h. In the euglycemic clamp study, insulin was infused for 4 h at a rate of 40 mU/m2.min in the young subjects and 34 mU/m2.min in the old subjects. Blood samples were taken, and calf blood flow was measured using venous occlusion plethysmography at regular intervals in each study. Basal calf blood flow was lower in the elderly (young subjects: 1.51 +/- .08 mL/100 mL tissue per min; old subjects: 1.15 +/- 0.07 mL/100 mL tissue per min, P < 0.002). During the euglycemic clamp studies, steady-state insulin and glucose values were similar in the two age groups. Glucose disposal rates were significantly higher in the young subjects (P = 0.01 by analysis of variance). Mean arterial pressure values were significantly higher in the elderly (P < 0.001 by analysis of variance) throughout the clamp, but there was no significant change over time in either age group. The mean incremental blood flow rate at steady-state (180-240 min) was significantly higher in the young subjects (0.76 +/- 0.23 mL/100 mL tissue per min) than in the old subjects (0.05 +/- 0.09 mL/100 mL tissue per min, P < 0.01). There was a significant correlation between steady-state glucose disposal rate values and incremental blood flow rates in the young subjects (r = 0.59, P < 0.05) but not in the old subjects (r = 0.21, P = NS). We conclude that normal aging is characterized by an impairment in the ability of insulin to modulate blood flow, which may contribute in part to the insulin resistance of aging.

Increased inactive renin in diabetes mellitus without evidence of nephropathy.

PRA, active renin, and inactive renin (IR; activated by dialysis to pH 3.3 and 7.4) were measured in the plasma of 53 patients with diabetes mellitus and 32 normal volunteers (group 1). Proteinuria was present in 21 diabetics (group 3; nephropathy) and absent in 32 diabetics (group 2). The mean PRA was lower in group 3 than in groups 1 and 2. PRA less than 0.2 ng/ml . h occurred more frequently and at a younger age in uncomplicated diabetics than in normal controls. Despite very low PRA, plasma aldosterone was normal in most of the diabetics. IR was significantly higher than normal in the uncomplicated diabetics and was greatly increased in diabetics with nephropathy. Since the kidneys are a principal source of IR, and since patients with diabetic nephropathy have consistently elevated plasma IR, it is possible that increased plasma IR in patients without proteinuria or reduced renal function might be an early sign of renal involvement. However, as other explanations of increased plasma IR exist, the hypothesis must be tested by longitudinal studies of diabetic patients.

The interaction of a type A retroviral particle and class II human leukocyte antigen susceptibility genes in the pathogenesis of Graves' disease.

We have previously reported that over 85% of patients with Graves' disease have detectable serum antibodies against a human intracisternal type A retroviral particle (HIAP), which are not present in age- and gender-matched controls, suggesting a role for HIAP in triggering the autoimmune process leading to Graves' disease. To investigate the interaction of this viral particle with genetic factors, 35 members of 3 kindreds, selected because of a high family prevalence of Graves' disease (a total of 11 members affected), were examined for clinical signs of thyroid dysfunction, goiter, and opthalmopathy. Thyroid function tests and autoimmune serological profiles were also obtained. In addition, subjects were tested for the presence of antibodies against HIAP by means of immunoblot analysis of their sera, and their human leukocyte antigen (HLA) class II alleles were determined by DNA methodology. Molecular genetic analyses enabled the detection of postulated HLA susceptibility haplotypes in each family. These families had 8, 4, and 5 members, respectively, with such apparent susceptibility genes and 11, 5, and 9 members, respectively, with immunological evidence of retroviral exposure. In the presence of both factors (codetected in a total of 15 members of the 3 kindreds), the incidence of Graves' disease was 100%, 67%, and 80%, respectively. One additional member of family B and 3 in family C with both viral and genetic susceptibility factors were found to have serological abnormalities and/or goiter and ocular signs consistent with evolving or preclinical Graves' disease. In families A and C, tight linkage between HLA haplotypes and Graves' disease was demonstrated in a manner consistent with recessive inheritance. The association between the occurrence of both anti-HIAP-I antibody positivity and HLA susceptibility and the presence of Graves' disease was highly significant (P < 0.001). The pathogenesis of Graves' disease in these families appears to be attributable to the interaction between the immune response to an intracisternal type A retroviral particle and immunogenetic susceptibility, leading to the autoimmune processes that underlie Graves' disease, with subsequent development of the characteristic features of the illness. Data from these families suggest that both of these factors are necessary for final disease expression. These results imply that serological evidence of retroviral exposure together with genetic HLA susceptibility are the two major predisposing factors underlying the pathogenesis of Graves' disease. Further studies will establish whether prospective identification of persons at risk for Graves' disease is possible by this means.

First report of recombination between the HLA-DR and HLA-DQ loci within a family.

Although unusual associations of HLA-DR and HLA-DQ alleles seen in ancestral haplotypes have indicated that recombination between these genes occurred in the past, an actual crossover event between DR and DQ has never been shown within a family. In a study of families with Graves' disease we have identified an individual from a three generation family who inherited a maternal haplotype that is the result of a recombinational event between the HLA-DR and the HLA-DQ loci on her chromosomes. Family members were typed for HLA class I by the lymphocyte microcytotoxicity test and for HLA class II by polymerase chain reaction (PCR) with sequence-specific primers or with sequence-specific oligonucleotide probes after PCR. Based on linkage disequilibrium it is likely that the recombinant haplotype is present in the proband rather than his brother. This haplotype was subsequently inherited by one of the proband's sons. The data presented support the conclusion that the recombinant haplotype resulted from a crossover event between the mother's DRB1 and DQA1 genes. Thus, recombination between the HLA-DR and HLA-DQ genes has been demonstrated within this family; a recombination event not previously described.

Amylin-mediated reduction in insulin sensitivity corresponds to reduced insulin receptor kinase activity in the rat in vivo.

Studies were undertaken to elucidate further the mechanism whereby the pancreatic peptide amylin induces insulin resistance. Sixteen male Sprague-Dawley rats underwent hyperinsulinemic (14 pmol/kg/min, 0 to 120 minutes) euglycemic clamps in the presence or absence of amylin (500 pmol/kg/min, 60 to 120 minutes). Amylin induced insulin resistance at both the hepatic level (mean +/- SE: hepatic glucose output [HGO] with amylin 1.4 +/- 0.2 v without amylin -1.9 +/- 0.3 mmol/kg/h, P < .001) and peripheral level (glucose disposal [Rd] with amylin 5.0 +/- 0.2 v without amylin 8.5 +/- 0.6 mmol/kg/h, P < .001). Serum insulin levels were similar in the presence or absence of amylin alone (661 +/- 89 v 636 +/- 50 pmol/L, respectively, P = NS), but were significantly less when somatostatin (SRIF) was simultaneously infused (408 +/- 15 pmol/L, P < .02 v the other two groups). This suggests that endogenous insulin production was not suppressed by amylin under these study conditions. Similar findings were obtained in 18 animals in the absence of exogenous insulin infusion. In vitro kinase activity toward histone of skeletal muscle insulin receptors (IRs) activated by insulin in vivo was reduced in the presence of amylin to 6.0 +/- 0.8 versus 9.1 +/- 1.2 fmol phosphate into histone (insulin-infused) and 3.9 +/- 0.7 versus 6.9 +/- 1.4 (non-insulin-infused; P < .03 by ANOVA). Serum calcium was significantly decreased in amylin-treated animals (1.93 +/- 0.04 v 2.30 +/- 0.05 mmol/L, P < .001).(ABSTRACT TRUNCATED AT 250 WORDS)

Human pancreatic islets transfected to produce an inhibitor of TNF are protected against destruction by human leukocytes.

The objective of this study was to determine whether transfection of human islets with an adenovirus construct encoding an inhibitor of tumor necrosis factor (TNFi) was effective at limiting damage to beta cells induced by human peripheral blood leukocytes (huPBL). Human islets transfected with TNFi or control islets were transplanted under the kidney capsule of NOD-scid mice. After a 15-day engraftment period, half of the mice received injections of activated huPBL and half received buffer injections. Islet graft function was assessed by two different methods, both of which use a species-specific radioimmunoassay to determine human insulin. In some mice, insulin production following intraperitoneal glucose injection was determined in serum. In other mice, total graft insulin content was determined by acid ethanol extraction. Histochemical stains were performed on some kidneys at the termination of the experiment to evaluate graft presence, transgene expression, and huPBL infiltration. In huPBL injected mice, graft performance was maintained in mice whose grafts were transfected with TNFi but declined substantially in control groups with sham transfected or beta-galactosidase transfected islet grafts. Similar results were obtained using either glucose-stimulated insulin release or graft insulin content as a measure of graft survival. There was no significant difference in graft function between control groups receiving buffer injections, regardless of whether the islets had been transfected. Human leukocytes were found in all huPBL groups regardless of islet transfection status. We conclude that transfection of human islets with an adenovirus encoding TNFi protects beta cells from destruction induced by human leukocytes. The local production of TNFi does not prevent graft infiltration by leukocytes, only the destruction of grafts by the infiltrating leukocytes. These results raise the possibility that local expression of an inhibitor of the proinflammatory cytokine TNF-alpha may also prevent graft failure in clinical islet transplantation.

Feedback regulation of glucose-dependent insulinotropic polypeptide (GIP) secretion by insulin in conscious rats.

Feedback regulation of glucose-dependent insulinotropic polypeptide (GIP) was studied in a conscious rat model. Male Wistar rats were preconditioned to partial restraint, the tail artery and vein were cannulated under local anesthesia. All animals received 1 g/kg oral glucose by gavage and were divided into 3 groups: One group ('euglycemic hyperinsulinemic', EH) underwent rapid induction of hyperinsulinemia by i.v. insulin infusion from 5 min prior to oral glucose until 120 min after (mean plasma insulin = 2285 +/- 108 pM +/- S.E.M., n = 5); a second group ('hyperglycemic hyperinsulinemic', HH) underwent rapid induction of both hyperinsulinemia and hyperglycemia (mean serum glucose = 12.9 +/- 0.4 mM +/- S.E.M., mean plasma insulin 3160 +/- 109 pM, n = 5). A third group ('control') underwent saline infusion alone (n = 5). Arterial blood was collected for GIP estimation at -10, 0, 10, 20, 30, 50, 70, 90 and 120 min after oral glucose. In the control group GIP rose by 96% from a mean basal concentration of 114 +/- 12 pM to a peak of 224 +/- 14 pM by 20 min, and returned to baseline within 70 min. In EH, the GIP rise was blunted and the peak (146 +/- 31 pM) occurred at 10 min, while in HH GIP peaked at 192 +/- 32 pM 10 min after oral glucose (a 92% increase over basal). Compared to controls, total area under the curve for GIP was less for EH (598 +/- 112 versus 971 +/- 94 pmol/l/h +/- S.E.M., P < 0.034). GIP response in HH was similar to the control group at 853 +/- 134 pmol/l/h.(ABSTRACT TRUNCATED AT 250 WORDS)

Modified method for the performance of glucose insulin clamp studies in conscious rats.

A modified method for performance of insulin-glucose clamp studies in rats was developed via catheterization of the tail vessels, after preconditioning of animals to limited restraint. The procedure is performed in conscious animals under local anesthesia and employs a specially designed foam rubber jacket which allows the animal mobility of the limbs and forward vision. In addition, a table utilizing a belt system allows easy positioning of the animal in the left and right lateral and supine positions during surgery. After initial development of the procedure, its use in 123 animals is reported. Line placement was successfully achieved in all cases with insignificant blood loss or morbidity and zero mortality. We note that 11% of animals did not complete the subsequent insulin-glucose clamp study due to either one of the vascular cannulae leaving the vessel (one animal), venous rupture (12 animals), or cannula blockage unrelated to surgical technique (one animal). Studies on Wistar Kyoto, Spontaneously Hypertensive, and Sprague-Dawley rats showed a fall in catecholamines after animals were replaced in cages, with stabilization within 30 min. In comparison to traditional techniques, this method is, therefore, proposed as a less traumatic and rapid way of performing infusion studies in conscious rats with a high success rate and minimization of loss of animal life due to procedural problems.