Serum Branch Chain Amino Acids (BCAAs) Are Elevated Due to Decreased Catabolism in Patients With Ketosis-Prone Diabetes at the Time of Presentation With DKA

Patients with “A-β+” Ketosis-Prone Diabetes (KPD) develop diabetic ketoacidosis (DKA) despite lacking islet autoantibodies and a phenotype of T1D, have good beta cell function and can come off insulin therapy 4–8 weeks after the DKA episode. When near-normoglycemic and stable on metformin, they have accelerated BCAA catabolism which promotes ketogenesis (Patel SG et al, Diabetes 2013). Here we measured BCAAs, their metabolites and acylcarnitine esters (C5,C3) in blood samples obtained from adults with DKA (N=74) compared to those with non-ketotic hyperglycemic crisis (N=21) at the time of acute presentation to the emergency center, and to healthy controls (N=17). Of the DKA patients, 53 were classified as likely A-β+ KPD based on absence of GAD65Ab and C-peptide levels or clinical features, and the 21 patients with non-ketotic hyperglycemia were classified as T2D. Serum concentrations of leucine, isoleucine and valine and their respective branch chain keto acids (BCKA) were higher (p<0.05) in KPD patients compared to T2D and control. The ratio of each BCKA to its precursor BCAA was calculated as an index of its rate of transamination. Serum KIC/Leu, KMV/Ile and KIV/Val were significantly lower (p<0.05) in KPD compared to T2D. The ratio of each acylcarnitine to its precursor BCKA was calculated as an index of its rate of entry and metabolism within mitochondria. Serum C5/KIC, C5/KMV and C5/KIC+KMV were lower (p<0.05) in KPD patients compared to T2D patients. Serum C3/KIV, C3/KMV and C3/KIV+KMV were significantly lower (p<0.05) in KPD patients compared to controls. Since KIC can be converted to acetoacetate and then reduced to β-hydroxybutyrate (BHOB), and KIC and KMV can be metabolized to acetyl CoA, the ratios of KIC+KMV/C2 and KIC/BHOB were calculated as indicators of their relative conversion to acetyl CoA and acetoacetate respectively. KIC+KMV/C2 was significantly lower (p<0.001) in KPD than T2D and control and KIC/BOHB was lower (p<0.001) in KPD than T2D. Acetyl carnitine was markedly elevated in the KPD group, indicating accelerated production of acetyl CoA from free fatty acids. During acute DKA, KPD patients have higher serum BCAAs because their catabolism is decreased, due to slower rate of transamination in the cytosol by BCAA transaminase 1 (BCAT1) and slower rate of entry into mitochondria and metabolism to acetyl CoA and acetoacetate by BCAT2, BCKA dehydrogenase and other catabolic enzymes. This is diametrically opposite to their profile in the stable, near-normoglycemic state, when BCAA catabolism is accelerated. We propose that during acute DKA, accelerated flux of fatty acids to acetyl CoA diminishes carnitine and NAD+ availability for mitochondrial transport and metabolism of BCAA catabolites in KPD patients, whereas in the near-normoglycemic state they have heightened dependence on BCAA catabolism for energy production through acetyl CoA and ketogenesis.