Regulation and Distribution of Endogenous Erythritol, a Biomarker of Cardiometabolic Disease

Abstract

Serum erythritol is a novel predictive biomarker of chronic disease development and complications. In cohort studies, elevated serum erythritol at baseline was predictive of central adiposity gain, T2DM, and cardiovascular disease development. Higher erythritol has also been associated with complications of these chronic diseases such as diabetic retinopathy and nephropathy. Erythritol is a 4-carbon polyol that has been well characterized as a non-nutritive sweetener. More recently, erythritol was discovered to be synthesized endogenously in humans from glucose through the pentose phosphate pathway (PPP). The PPP generates NADPH for anabolic reactions and to combat oxidative stress. We identified two human enzymes that catalyze the conversion of erythrose-4-phosphate to erythritol: sorbitol dehydrogenase (SORD) and alcohol dehydrogenase 1 (ADH1). SORD knockdown significantly decreases erythritol synthesis in vitro. The regulation of erythritol synthesis in mammals is not characterized. The purpose of this study was to assess stimuli that modulate intracellular erythritol levels and the distribution of erythritol in key metabolic tissues. We hypothesize that erythritol synthesis will be induced by stimuli that increase PPP activity, and that erythritol will be present in tissues with high SORD expression. First, we fed 8-week-old C57BL/6J mice a defined diet (AIN93G) for two weeks. After two weeks, plasma and tissue erythritol including liver, kidney, white adipose, and skeletal muscle, were measured by GC-MS. To further evaluate the cellular regulation of erythritol synthesis, we used A549 cells due to their high PPP activity. Using siRNA, we knocked down SORD and glucose-6-phosphate dehydrogenase (G6PD) under normal (5mM) and high (25mM) glucose conditions, then measured intracellular erythritol. We also exposed A549’s to hydrogen peroxide to assess if erythritol synthesis is elevated by oxidative stress. Plasma erythritol ranged from 0.44 uM to 0.73 uM in young, healthy mice. Erythritol was significantly different based on tissue type (p < 0.0001, one-way ANOVA). The liver and kidney had the highest relative erythritol, followed by quadriceps, whereas white adipose tissue contained the least erythritol. Interestingly, SORD knockdown significantly reduced erythritol synthesis under high glucose, but not normal glucose conditions (p < 0.01, Sidak's multiple comparisons test). Knockdown of G6PD did not impact erythritol synthesis at either glucose concentration. Finally, treatment with hydrogen peroxide significantly increased intracellular erythritol (p < 0.0001, student's t-test). In mice, erythritol was highest in tissues known to have high SORD expression. Unexpectedly, erythritol was lowest in adipose, which is known to have high PPP activity. This finding is supported by the reduction of intracellular erythritol induced by SORD, but not G6PD knockdown. Erythritol synthesis was also elevated by two factors known to stimulate the PPP: high glucose and oxidative stress. Together, these data suggest that erythritol synthesis is linked to PPP flux, but is regulated downstream of the oxidative phase of the PPP.