‘Insulin resistance’ (IR) is a widely used clinical term. It is usually defined as a state characterised by reduced glucose-lowering activity of insulin, but it is also sometimes used as a shorthand label for a clinical syndrome encompassing major pathologies such as type 2 diabetes, polycystic ovary syndrome, fatty liver disease and atherosclerosis. Nevertheless, the precise cellular origins of IR, the causal links among these phenomena and the mechanisms underlying them remain poorly understood or contentious. Prevalent IR usually results from a genetic predisposition interacting with acquired obesity; however, even in some lean individuals, very severe degrees of IR can be observed. It is important to identify these people as they often harbour identifiable single-gene defects and they may benefit from molecular diagnosis, genetic counselling and sometimes tailored therapies. Observation of people with known single-gene defects also offers the opportunity to make inferences about the mechanistic links between IR and common pathologies. Herein, we summarise the currently known monogenic forms of severe IR, with an emphasis on the practical aspects of their recognition, diagnosis and management. In particular, we draw distinctions among the biochemical subphenotypes of IR that arise from primary adipose tissue dysfunction or from primary insulin signalling defects and discuss the implications of this dichotomy for management.
Victoria E R Parker and Robert K Semple
Anna McDonald, Rachel M Williams, Fiona M Regan, Robert K Semple and David B Dunger
Severe insulin resistance resulting from known or putative genetic defects affecting the insulin receptor or post-insulin receptor signalling represents a clinical spectrum ranging from Donohue’s and Rabson–Mendenhall syndrome, where the genetic defect is identified, through to the milder phenotype of type A insulin resistance, where a genetic defect can only be detected in around 10% of cases. Paradoxically, subjects with these conditions may present with hypoglycaemia due to mismatch of post-prandial glucose excursion and compensatory hyperinsulinaemia. Ultimately, treatment with insulin and insulin sensitisers will be unsuccessful and subjects may succumb to diabetes or its complications. Recombinant human IGF-I alone or combined with its binding protein (IGFBP-3) provides an alternative therapy as IGF-I receptor shares structural and functional homology with the insulin receptor and recombinant human insulin-like growth factor I (rhIGF-I) therapy could improve glucose disposal by signalling through the IGF-I receptor, whilst reducing the adverse effects of high insulin concentrations. There are also data which indicate that IGF-I signalling through the IGF-I receptor on the pancreatic β-cell may be important in maintaining insulin secretion. Pilot studies confirmed that rhIGF-I could reduce glucose and insulin levels in subjects with type A insulin resistance and those with Rabson–Mendenhall syndrome with sustained beneficial effects on HbA1c. Continued study has confirmed efficacy of rhIGF-I when combined with IGFBP-3 in the treatment of Donohue’s and type A insulin resistance subjects. Observations that IGF-I treatment can improve C-peptide levels in these subjects may indicate that it might be more valuable as a first line intervention to preserve β-cell function, rather than its current use as a medication of last resort in subjects where all other therapies have failed.
Sarah M Leiter, Victoria E R Parker, Alena Welters, Rachel Knox, Nuno Rocha, Graeme Clark, Felicity Payne, Luca Lotta, Julie Harris, Julio Guerrero-Fernández, Isabel González-Casado, Sixto García-Miñaur, Gema Gordo, Nick Wareham, Víctor Martínez-Glez, Michael Allison, Stephen O’Rahilly, Inês Barroso, Thomas Meissner, Susan Davies, Khalid Hussain, Karen Temple, Ana-Coral Barreda-Bonis, Sebastian Kummer and Robert K Semple
Genetic activation of the insulin signal-transducing kinase AKT2 causes syndromic hypoketotic hypoglycaemia without elevated insulin. Mosaic activating mutations in class 1A phospatidylinositol-3-kinase (PI3K), upstream from AKT2 in insulin signalling, are known to cause segmental overgrowth, but the metabolic consequences have not been systematically reported. We assess the metabolic phenotype of 22 patients with mosaic activating mutations affecting PI3K, thereby providing new insight into the metabolic function of this complex node in insulin signal transduction.
Three patients with megalencephaly, diffuse asymmetric overgrowth, hypoketotic, hypoinsulinaemic hypoglycaemia and no AKT2 mutation underwent further genetic, clinical and metabolic investigation. Signalling in dermal fibroblasts from one patient and efficacy of the mTOR inhibitor Sirolimus on pathway activation were examined. Finally, the metabolic profile of a cohort of 19 further patients with mosaic activating mutations in PI3K was assessed.
In the first three patients, mosaic mutations in PIK3CA (p.Gly118Asp or p.Glu726Lys) or PIK3R2 (p.Gly373Arg) were found. In different tissue samples available from one patient, the PIK3CA p.Glu726Lys mutation was present at burdens from 24% to 42%, with the highest level in the liver. Dermal fibroblasts showed increased basal AKT phosphorylation which was potently suppressed by Sirolimus. Nineteen further patients with mosaic mutations in PIK3CA had neither clinical nor biochemical evidence of hypoglycaemia.
Mosaic mutations activating class 1A PI3K cause severe non-ketotic hypoglycaemia in a subset of patients, with the metabolic phenotype presumably related to the extent of mosaicism within the liver. mTOR or PI3K inhibitors offer the prospect for future therapy.