Background: The majority of diabetic cats in remission have abnormal glucose tolerance, and approximately one third relapse within 1 year. Greater understanding of the metabolic characteristics of diabetic cats in remission, and predictors of relapse is required to effectively monitor and manage these cats. Objectives: To identify and compare differences in plasma metabolites between diabetic cats in remission and healthy control cats using a metabolomics approach. Secondly, to assess whether identified metabolites are predictors of diabetic relapse. Animals: Twenty cats in diabetic remission for a median of 101 days, and 22 healthy matched control cats. Methods: Cats were admitted to a clinic, and casual blood glucose was recorded. After a 24 h fast, blood glucose concentration was measured, then a blood sample was taken for metabolomic (GCMS and LCMS) analyses. Three hours later, a simplified intravenous glucose tolerance test (1 g glucose/kg) was performed. Cats were monitored for diabetes relapse for at least 9 months (270 days) after baseline testing. Results: Most cats in remission continued to display impaired glucose tolerance. Concentrations of 16 identified metabolites differed (P ≤ 0.05) between remission and control cats: 10 amino acids and stearic acid (all lower in remission cats), and glucose, glycine, xylitol, urea and carnitine (all higher in remission cats). Moderately close correlations were found between these 16 metabolites and variables assessing glycaemic responses (most |r| = 0.31 to 0.69). Five cats in remission relapsed during the study period. No metabolite was identified as a predictor of relapse. Conclusion and clinical importance: This study shows that cats in diabetic remission have abnormal metabolism.

Genetic variants that are associated with susceptibility to type 2 diabetes (T2D) are important for identification of individuals at risk and can provide insights into the molecular basis of disease. Analysis of T2D in domestic animals provides both the opportunity to improve veterinary management and breeding programs as well as to identify novel T2D risk genes. Australian-bred Burmese (ABB) cats have a 4-fold increased incidence of type 2 diabetes (T2D) compared to Burmese cats bred in the United States. This is likely attributable to a genetic founder effect. We investigated this by performing a genome-wide association scan on ABB cats. Four SNPs were associated with the ABB T2D phenotype with p values <0.005. All exons and splice junctions of candidate genes near significant single-nucleotide polymorphisms (SNPs) were sequenced, including the genes DGKG, IFG2BP2, SLC8A1, E2F6, ETV5, TRA2B and LIPH. Six candidate polymorphisms were followed up in a larger cohort of ABB cats with or without T2D and also in Burmese cats bred in America, which exhibit low T2D incidence. The original SNPs were confirmed in this cohort as associated with the T2D phenotype, although no novel coding SNPs in any of the seven candidate genes showed association with T2D. The identification of genetic markers associated with T2D susceptibility in ABB cats will enable preventative health strategies and guide breeding programs to reduce the prevalence of T2D in these cats.