Identification of circulating sphingosine kinase-related metabolites for prediction of type 2 diabetes

Background: Sphingosine Kinase (SphK) that catalyzes sphingosine (Sph) to sphingosine 1-phosphate (S1P), plays a key role in both sphingolipid metabolism and cellular signaling. While SphK has been implicated in type 2 diabetes mellitus (T2DM), it is unexplored in humans. Herein, we investigated whether circulating SphK-related metabolites are associated with T2DM incidence in an established prospective cohort. Methods: Levels of SphK-related sphingolipid metabolites, including Sph, S1P, dihydrosphingosine (dhSph) and dihydro-S1P (dhS1P) in serum were measured by targeted-lipidomic analyses. By accessing to an established prospective cohort that involves a total of 2486 non-diabetic adults at baseline, 100 subjects who developed T2DM after a mean follow-up of 4.2-years, along with 100 control subjects matched strictly with age, sex, BMI and fasting glucose, were randomly enrolled for the present study. Results: Comparison with the control group, medians of serum dhS1P and dhS1P/dhSph ratio at baseline were elevated significantly prior to the onset of T2DM. Each SD increment of dhS1P and dhS1P/dhSph ratio was associated with 53.5% and 54.1% increased risk of incident diabetes, respectively. The predictive effect of circulating dhS1P and dhS1P/dhSph ratio on T2DM incidence was independent of conventional risk factors in multivariate regression models. Furthermore, combination of serum dhS1P and dhS1P/dhSph ratio with conventional clinical indices significantly improved the accuracy of T2DM prediction (AUROC, 0.726), especially for normoglycemic subjects (AUROC, 0.859). Conclusion: Circulating levels of dhS1P and dhS1P/dhSph ratio are strongly associated with increased risk of T2DM, and could serve as a useful biomarker for prediction of incident T2DM in normoglycemic populations.

Automated summary

Circulating levels of dhS1P and dhS1P/dhSph ratio are strongly associated with increased risk of T2DM, and could serve as a useful biomarker for predicting incident T2DM in normoglycemic populations.