ObjectiveFasting results in major metabolic changes including a switch from glycogenolysis to gluconeogenesis to maintain glucose homeostasis. However, the relationship between the length of fasting and the relative contribution of gluconeogenic substrates remains unclear. We investigated the relative contribution of glycogen, lactate, and glycerol in glucose production of male C57BL/6 J-albino mice after 6, 12, and 18 h of fasting.MethodsWe used non-perturbative infusions of 13C3 lactate, 13C3 glycerol, and 13C6 glucose combined with liquid chromatography mass spectrometry and metabolic flux analysis to study the contribution of substrates in gluconeogenesis (GNG).ResultsDuring infusion studies, both lactate and glycerol significantly label about 60% and 30–50% glucose carbon, respectively, but glucose labels much more lactate (∼90%) than glycerol carbon (∼10%). Our analyses indicate that lactate, but not glycerol is largely recycled during all fasting periods such that lactate is the largest direct contributor to GNG via the Cori cycle but a minor source of new glucose carbon (overall contribution). In contrast, glycerol is not only a significant direct contributor to GNG but also the largest overall contributor to GNG regardless of fasting length. Prolonged fasting decreases both the whole body turnover rate of glucose and lactate but increases that of glycerol, indicating that the usage of glycerol in GNG become more significant with longer fasting.ConclusionCollectively, these findings suggest that glycerol is the dominant overall contributor of net glucose carbon in GNG during both short and prolonged fasting.

Gluconeogenesis (GNG) is de novo production of glucose from endogenous carbon sources. Although it is a commonly studied pathway, particularly in disease, there is a lack of consensus about substrate preference. Moreover, primary hepatocytes are the current gold standard for in vitro liver studies, but no direct comparison of substrate preference at physiological fasting concentrations has been performed. We show that mouse primary hepatocytes prefer glycerol to pyruvate/lactate in glucose production assays and 13C isotope tracing studies at the high concentrations commonly used in the literature, as well as at more relevant fasting, physiological concentrations. In addition, when glycerol, pyruvate/lactate, and glutamine are all present, glycerol is responsible for over 75% of all glucose carbons labeled. We also found that glycerol can induce a rate-limiting enzyme of GNG, glucose-6-phosphatase. Lastly, we suggest that glycerol is a better substrate than pyruvate to test in vivo production of glucose in fasting mice. In conclusion, glycerol is the major carbon source for GNG in vitro and in vivo and should be compared with other substrates when studying GNG in the context of metabolic disease states.

This work was supported by a Research Grant from the American Society of Reproductive Medicine and support from the Charles and Johanna Busch Memorial Fund at Rutgers, the State University of NJ to K.S. and the Foundation for Embryonic Competence, Inc to N.T. The authors declare no conflicts of interest.