Diminished arginine bioavailability during hyperglycaemia distinguishes individuals with A-β+ ketosis-prone diabetes from those with type 2 diabetes: implications for the pathogenesis of ketoacidosis in A-β+ ketosis-prone diabetes.

AIMS/HYPOTHESIS: Individuals with A-β+ ketosis-prone diabetes (KPD) resemble those with type 2 diabetes but are prone to developing unprovoked diabetic ketoacidosis (DKA). We previously noted decreased plasma arginine concentrations in clinically stable individuals with KPD. We hypothesised that defective arginine metabolism could distinguish individuals with KPD from those with type 2 diabetes during hyperglycaemia and contribute to their proclivity to develop DKA. Hence, we quantified arginine production and catabolism, bioavailability and relation to insulin secretion under euglycaemic and hyperglycaemic conditions in participants with KPD compared with type 2 diabetes. Non-diabetic control participants were also studied to establish normative ranges. METHODS: In Protocol 1, participants with KPD and type 2 diabetes and non-diabetic control participants were studied under euglycaemic and hyperglycaemic conditions. Kinetics of arginine, citrulline, ornithine and phenylalanine were measured using stable isotope tracer infusions with mass spectrometric analyses, together with quantification of insulin secretion in response to glucose and arginine. In Protocol 2, ten participants with KPD (adults diagnosed after an episode of unprovoked DKA, lacking islet autoantibodies, with substantial beta cell functional reserve and HbA1c <58.5 mmol/mol [7.5%] on metformin alone) underwent a blinded, randomised crossover trial of citrulline (arginine precursor) compared with alanine (placebo) supplementation. The supplements were provided in identical, opaque capsules and allocated by a pharmacist; participants and investigators were masked to the order of supplementation. The studies were performed in a clinical research centre and the outcome measurements were arginine flux (primary outcome), de novo synthesis and plasma concentration, and nitric oxide (NO) synthesis, together with insulin secretion tests. RESULTS: In Protocol 1, under hyperglycaemic conditions, arginine flux decreased in KPD but not in type 2 diabetes. This resulted in 42.5% greater decline of arginine bioavailability between the euglycaemic and hyperglycaemic conditions in KPD (p=0.035), due primarily to less arginine derived from protein breakdown. KPD and type 2 diabetes groups did not differ in arginine de novo synthesis, arginine hydrolysis, NO synthesis or ornithine flux. Following hyperglycaemia, both diabetes groups exhibited blunted insulin response to a glucose bolus, but KPD showed increased insulin response to arginine. In Protocol 2, supplementation with citrulline but not with alanine elicited increased arginine flux (effect size 0.746; 95% CI 0.258, 0.931), arginine synthesis and NO synthesis in the KPD participants, without adverse effects. CONCLUSIONS/INTERPRETATION: Arginine potentiates insulin secretion during sustained hyperglycaemia, hence its diminished bioavailability in individuals with KPD could contribute to the development of ketoacidosis during a hyperglycaemic crisis. Individuals with KPD retain responsiveness to exogenous arginine during hyperglycaemia and normalise endogenous arginine availability with citrulline supplementation, suggesting therapeutic options for ameliorating or preventing their episodes of DKA. TRIAL REGISTRATION: ClinicalTrials.gov NCT03566524 (for Protocol 2). FUNDING: This work was supported by R01-DK101411 from the National Institutes of Health.