Identification of the gene altered in Berardinelli-Seip congenital lipodystrophy on chromosome 11q13.

Congenital generalized lipodystrophy, or Berardinelli-Seip syndrome (BSCL), is a rare autosomal recessive disease characterized by a near-absence of adipose tissue from birth or early infancy and severe insulin resistance. Other clinical and biological features include acanthosis nigricans, hyperandrogenism, muscular hypertrophy, hepatomegaly, altered glucose tolerance or diabetes mellitus, and hypertriglyceridemia. A locus (BSCL1) has been mapped to 9q34 with evidence of heterogeneity. Here, we report a genome screen of nine BSCL families from two geographical clusters (in Lebanon and Norway). We identified a new disease locus, designated BSCL2, within the 2.5-Mb interval flanked by markers D11S4076 and D11S480 on chromosome 11q13. Analysis of 20 additional families of various ethnic origins led to the identification of 11 families in which the disease cosegregates with the 11q13 locus; the remaining families provide confirmation of linkage to 9q34. Sequence analysis of genes located in the 11q13 interval disclosed mutations in a gene homologous to the murine guanine nucleotide-binding protein (G protein), gamma3-linked gene (Gng3lg) in all BSCL2-linked families. BSCL2 is most highly expressed in brain and testis and encodes a protein (which we have called seipin) of unknown function. Most of the variants are null mutations and probably result in a severe disruption of the protein. These findings are of general importance for understanding the molecular mechanisms underlying regulation of body fat distribution and insulin resistance.

Major histocompatibility class II genes polymorphism in insulin dependent diabetes mellitus with or without associated thyroid autoimmunity.

Insulin dependent diabetes mellitus (IDDM) is sometimes associated with extrapancreatic organ-specific autoimmune diseases, but whether this phenotype results from a peculiar genetic profile is still unclear. The allelic distribution of the major histocompatibility complex (MHC) class II genes (HLA-DRB1, DQA1, DQB1 and TAP) was analysed in 143 patients with IDDM alone by comparison with 82 IDDM patients with autoimmune thyroid disease (IDDM/AITD). The frequency of the DQB1*0301 IDDM-protective phenotype seemed to be lower in IDDM than in IDDM/AITD patients (16.8% vs 30.5% respectively, p = 0.02). By contrast, the frequency of the DRB1*04-DQB1*0302 IDDM-predisposing phenotype was higher in IDDM than in IDDM/AITD patients (91.3% vs 76.1% of DR4-positive patients respectively, p = 0.007), but these differences were not significant after correcting the p values, except in the case of the DRB1*0405-DQB1*0302 combination (21.3% vs 2.4% of DR4-positive patients, Pc = 0.05). Furthermore, all differences disappeared when patients were matched for age at IDDM-onset. Our data do not long give support for a particular role of MHC class II genes in favouring the occurrence of thyroid autoimmunity in IDDM patients, but rather suggest that some class II alleles or residues might determine the rapidity of progression to IDDM in genetically susceptible individuals. The involvement of non-MHC genes and/or environmental factors remains to be determined.

Linkage disequilibrium between HLA class II (DR, DQ, DP) and antigen processing (LMP, TAP, DM) genes of the major histocompatibility complex.

TAP, LMP and DM genes map within the major histocompatibility complex (MHC) class II region between the DQB1 and DPB1 loci, and are involved in the processing of peptides bound to HLA class I or class II molecules. In order to determine the various linkage disequilibria existing between these genes and HLA class II genes, we have analyzed TAP1, TAP2, LMP2, DMA, DMB, DRB1, DQA1, DQB1 and DPB1 polymorphisms in 162 unrelated healthy Caucasian individuals. Many positive or negative associations were observed between alleles at these loci, such as between DR/DQ and TAP2, DM or LMP, between DP and DMB, and between TAP2 and DM, TAP2 and LMP. Conversely, no linkage disequilibrium was detected between some closely related genes (DR/DQ and TAP1, TAP1 and TAP2, LMP2 and DM), in agreement with the existence of recombination hot spots in this region. Other weak linkage disequilibria are likely to exist in this region. These data allow to define some conserved MHC class II haplotypes including HLA class II and TAP, LMP and DM alleles. Furthermore, the knowledge of such linkage disequilibria is of outstanding importance in order to avoid misinterpretation of the data when studying MHC class II associations with autoimmune diseases.

Absence of primary association between DM gene polymorphism and insulin-dependent diabetes mellitus or celiac disease.

The DMA and DMB genes encode class II-like heterodimetric molecules located in a specialized endocytic compartment, where they facilitate efficient loading of antigenic peptides on HLA class II molecules. Both genes are located within the MHC class II region and present a limited allelic polymorphism. Here we report the distribution of DM alleles in a group of 75 IDDM patients, 72 CD patients, and 162 random controls. We found a pronounced decreased frequency of DMA*0102 in both patient groups relative to controls. This difference was, however, mainly secondary to a strong negative linkage disequilibrium (LD) between this allele and the IDDM and CD-associated DRB1*03 allele. The DMB phenotype frequencies were similar in CD patients and controls. By contrast, we observed a decreased frequency of DMB*0101 and an increased frequency of DMB*0102 and DMB*0104 in IDDM patients. These differences disappeared when matching individuals for DRB1*03 or DRB1*04 alleles, which was in accordance with strong negative LD between DMB*0101 and DRB1*04 or DQB1*0302 alleles, and positive LD between DMB*0104 and DQB1*0201. Our data suggest that the apparent associations of IDDM or CD with given DM alleles are mostly secondary to primary associations with alleles at the DRB and DQB loci.

Cyclosporin A does not delay insulin dependency in asymptomatic IDDM patients.

OBJECTIVE: To measure the effects of cyclosporin A (CyA) with no insulin therapy on glucose tolerance and beta-cell function in the preclinical phase of insulin-dependent diabetes mellitus (IDDM). RESEARCH DESIGN AND METHODS: beta-cell responses to the intravenous glucose tolerance test (IVGTT), hyperglycemic clamp, intravenous arginine, and intravenous glucagon were evaluated before and after a 6-month course of CyA in seven patients (mean age 19.6 years) with asymptomatic IDDM. RESULTS: Initial insulin secretory responses were severely decreased when the patients were compared with eight healthy control subjects: IVGTT (1 + 3 min): 106 +/- 16 vs. 884 +/- 190 pmol/l (P < 0.001); hyperglycemic clamp: 102 +/- 16 vs. 310 +/- 42 pmol/l (P < 0.001); intravenous arginine: 346 +/- 72 vs. 1104 +/- 168 pmol/l (P < 0.01); and intravenous glucagon: 170 +/- 37 vs. 247 +/- 35 pmol/l (NS). The beta-cell responses remained markedly abnormal after 6 months of CyA, although the response to intravenous glucose and oral glucose tolerance tests improved in three subjects. All the patients became insulin-dependent after 5-36 months. CONCLUSIONS: CyA alone is not a suitable treatment for asymptomatic IDDM. Earlier identification of subjects with substantial beta-cell secretory capacity and newer nontoxic intervention strategies are required for the prevention of IDDM.

Excessive glucose production, rather than insulin resistance, accounts for hyperglycaemia in recent-onset streptozotocin-diabetic rats.

Glucose production and utilization and activities of key enzymes involved in liver and muscle glucose metabolism were studied in post-absorptive streptozotocin-diabetic rats after 12 h of severe hyperglycaemia (17.5 +/- 0.5 mmol/l) and insulinopenia (5 +/- 1 microU/ml). Basal glucose production was increased: 36.6 +/- 3.0 mg.kg.min-1, vs 24.4 +/- 2.5 in controls (p < 0.05); liver glycogen concentration was decreased by 40% (p < 0.05); liver phosphoenolpyruvate carboxykinase and glucose-6-phosphatase activities were increased by 375 and 156%, respectively (p < 0.001 and < 0.01). During a euglycaemic clamp at a plasma insulin level of 200 microU/ml, glucose production was totally suppressed in controls, but persisted at 20% of basal in diabetic rats. In these rats, glucose production was suppressed at a plasma insulin level of 2500 microU/ml. Basal whole body glucose utilization rate, 2-deoxy-1-[3H]-D-glucose ([3H]-2DG) uptake by muscles and muscle glycogen concentrations were similar in both groups, as well as total and active forms of pyruvate dehydrogenase and glycogen synthase activities. During the euglycaemic clamp, the total body glucose utilization rates and [3H]-2DG uptake by muscles were similar in control and diabetic rats at a plasma insulin level of 200 microU/ml, but lower in diabetic rats at a plasma insulin level of 2500 microU/ml. We conclude 1) in recent-onset severely insulinopenic rats, an excessive glucose production via gluconeogenesis prevailed, mainly accounting for the concomitant hyperglycaemia. This excess glucose output cannot be attributed to liver insulin resistance: the gluconeogenic pathway is physiologically less sensitive than glycogenolysis to the inhibition by insulin.(ABSTRACT TRUNCATED AT 250 WORDS)

Mefloquine-associated hypoglycaemia in a cachectic AIDS patient.

Quinine and its isomer quinidine are well-known causes of iatrogenic hypoglycaemia, due to excessive insulin secretion. The situation is less clear regarding other anti-malarial quinine analogues. In particular, this adverse effect has never been described with mefloquine (Lariam). We report a case of hypoglycaemia after mefloquine therapy (1,500 mg over two days) for severe gastrointestinal cryptosporidiasis in a cachectic AIDS patient with protracted diarrhoea. Blood glucose levels, which were normal before treatment, dropped to 2.3 mmol/l within a few hours and were corrected by i.v. glucose infusion. Hypoglycaemia did not recur despite continued treatment. Rat islets of Langerhans exposed to mefloquine in vitro (10(-8) mol/l to 10(-3) mol/l) secreted significantly more insulin than control islets (up to 980 +/- 180 microU/ml/5 islets incubated with mefloquine 10(-3) mol/l, vs 20 +/- 4 microU/ml/5 untreated islets). Mechanisms and triggering factors of hypoglycaemia induced by mefloquine and some other anti-malarial quinine analogues are discussed. Clinicians who manage cachectic patients, particularly those with protracted diarrhoea and/or receiving anti-malarial drugs including mefloquine, should be aware of the risk of severe hypoglycaemia.

Pentamidine-induced derangements of glucose homeostasis. Determinant roles of renal failure and drug accumulation. A study of 128 patients.

OBJECTIVE: To assess the prevalence, presentation, and risk factors of pentamidine-induced dysglycemia. RESEARCH DESIGN AND METHODS: Blood glucose values were screened in 244 consecutive immunocompromised patients with Pneumocystis carinii pneumonia: 116 being treated with cotrimoxazole and 128 others with pentamidine. RESULTS: Two cotrimoxazole patients developed diabetes as a result of necrotizing pancreatitis (1.7%); the others remained euglycemic. Forty-eight pentamidine-treated patients (38.5%) developed severe glucose homeostasis disorders: hypoglycemia in 7, hypoglycemia and then diabetes in 18, and diabetes alone in 23 (P < 0.001 vs. the cotrimoxazole group). Hypoglycemia was early, sudden, often recurrent, and life-threatening, associated with inappropriately high insulin levels in plasma; the B-cell response to stimuli was poor. Of the 41 diabetic patients, 26 required insulin therapy; their plasma C-peptide levels were lower than normal, and the B-cell secretory responses to stimuli were poor. Islet cell antibodies, insulin antibodies, and insulitis were not detected. The pentamidine-treated dysglycemic patients differed from their euglycemic counterparts by higher pentamidine doses (P < 0.001), higher plasma creatinine levels (P < 0.001), and more severe anoxia (P < 0.05) and shock (P < 0.001). Most of them had received pentamidine mesylate parenterally (n = 36; 75%); six others received the isethionate salt and six exclusively pentamidine aerosols. CONCLUSIONS: Pentamidine-induced dysglycemic accidents are primarily due to inappropriate insulin release and toxicity to the islet B-cells. Drug accumulation due to excessive doses, iterative courses, and/or renal impairment is the determining risk factor.

Course of pancreatic beta cell destruction in prediabetic NOD mice: a histomorphometric evaluation.

The course of pancreatic beta cell destruction during the prediabetic period of autoimmune diabetic syndromes is not precisely known. We have analyzed the course of insulitis (n = 140) and the beta cell and lymphocytic volume densities by morphometric methods (n = 80) in NOD mice aged 6 to 45 weeks. In the absence of diabetes the mean beta cell density was only slightly reduced with age: 0.64 +/- 0.04% glandular tissue (mean SEM; n = 12) in 6 wk-old mice, 0.52 +/- 0.06% in 45 wk-old mice (n = 12; ns). However, a minority of pancreases were free of insulitis or showed isolated periinsulitis at the end of the 45-wk follow-up period. Invasive insulitis (i.e. mononuclear cells invading the islet area) was detected in 60-85% of mice from the 12th week on. In non overtly diabetic mice, beta cell density was reduced only when insulitis was invasive in more than 40% of islets: 0.30 +/- 0.03%, (n = 11) vs 0.59 +/- 0.04% (n = 34) in moderately invasive insulitis (p < 0.05). These mice had significant metabolic abnormalities. In diabetic mice, the beta cell density was markedly decreased: 0.02 +/- 0.01% (n = 7; p < 0.001). On the whole, a lymphocytic infiltrate affecting less than 50% of the islet volume was compatible with a normal beta cell density. Beyond this 50% lymphocytic infiltration threshold, beta cell density was tightly and negatively related to the lymphocytic volume density.(ABSTRACT TRUNCATED AT 250 WORDS)

FK 506/fluconazole interaction enhances FK 506 nephrotoxicity.

In a diabetic liver transplant recipient, immunosuppressed with FK 506, a severe digestive moniliasis required treatment with the antifungal imidazole derivative fluconazole. This was followed by a brisk rise in serum creatinine levels, which returned promptly to baseline when this combination was stopped. Serial measurements of FK 506 concentrations in plasma showed excessive areas under the 24 h time-concentration curve and trough levels during the concomitant intake of imidazole antifungals and FK 506. Eighteen months later, this patient presents with normal renal function. Antifungal imidazole derivatives inhibit the P 450 cytochrome enzyme system and may thus decrease the catabolism of FK 506. Meticulous monitoring of FK 506 trough levels and dosage adaptation are compulsory when concomitant treatment with such a drug is unavoidable. The same is true for cyclosporine in auto-immune diabetic patients receiving cyclosporine A.

The kidney in cyclosporin A-treated diabetic patients: a long-term clinicopathological study.

This was an analysis of the renal investigations performed in 248 cyclosporin A (CyA)-treated patients who had recent-onset type I insulin-dependent diabetes mellitus (IDDM) to assess the clinicopathological relationships, risk factors and predictive indices of CyA nephrotoxicity, and renal function observed with different CyA treatment regimens. There were four different protocols, using initial CyA dosages ranging from 7.5 to 10 mg/kg/day, with dose modification according to serum creatinine concentration, which was measured regularly in some patients for up to 9 years after starting treatment. Kidney biopsies were obtained from 125 patients (74 adults and 51 children) who had received only CyA for an average duration of 13 months before biopsy and had no other sources of renal injury at this stage of IDDM. Of these patients, 58% showed normal or minimal changes on biopsy, 26% showed slight abnormalities, and 16% showed medium-grade (grade III nephropathy) abnormalities. Lesion severity was related to the degree of interstitial fibrosis and tubular atrophy which, in turn, was related to the use of high maximum CyA dosages. Patients' age, and excessive CyA dose and blood trough levels were the main risk factors, and serum creatinine increase was the best predictive factor of CyA-induced nephropathy. However, CyA-induced renal dysfunction was essentially reversible on dosage reduction, and morphological changes were not followed by progressive renal insufficiency when CyA doses were low and adjusted according to serum creatinine levels. We conclude that, at present, it is recommended that low-dose CyA in combination with other non-nephrotoxic immunosuppressive strategies be used in patients with IDDM.

[Is a gene therapy for diabetic syndromes foreseeable?]. / Une thérapie génique des syndromes diabétiques est-elle envisageable?

The concepts and methods of gene therapy are summarized in order to assess a possible implication in the treatment of diabetes mellitus. Gene therapy requires identification of the critical genetic defect and then the preparation and introduction of the therapeutic transgene, with an appropriate targeting and a strong regulated expression. The bases of the different human diabetic syndromes are reviewed in their present state of knowledge: they are mostly clarified in the case of MODY, extreme insulin resistance syndromes, and some mitochondrial diabetic syndromes; but still obscure in the case of Type 2 and Type 1 diabetic syndromes. Substantial contributions to the understanding of the pathophysiology of diabetes have been brought by transgenic animal models. Gene therapy of human diabetic syndromes may become available, in an undetermined future, particularly under the forms of insulin secreting transgenic "organoïds". Such treatments should be proportionate to the intrinsic severity of the candidate diseases and carefully screened for safety.