OR20-6 Ketones and Sepsis-Induced Muscle Weakness: Signal or Fuel for Protection?

Abstract

Background: Septic patients often develop muscle weakness, contributing to ICU morbidity and mortality. We recently observed that obese septic mice and patients do not develop such weakness, explained by increased lipolysis and hepatic conversion of fatty acids into the ketone bodies. Furthermore, in lean septic mice, supplementing nutrition with 3-hydroxybutyrate (3HB) reproduced the protection against muscle weakness seen with obesity. The aim of the current study was to investigate underlying mechanisms of the 3HB-protection. Methods: We performed a mouse study (n=49) in a centrally-catheterized fluid-resuscitated model of prolonged (5 days) abdominal sepsis (cecal ligation and puncture). Lean septic mice received standard parenteral nutrition (PN) supplemented with either D,L-3HB (5 mg/kg/day) or isocaloric glucose (gluc). Healthy pair-fed mice served as controls. The use of 3HB as energy substrate in the muscle was assessed by quantifiying protein, triglyceride and glycogen content and markers of muscle ketolysis, proteolysis and fatty acid oxidation (FAO). Signaling function of 3HB was assessed by markers of muscle autophagy activation, inflammation and regeneration. Results: Although 3HB infusion abated sepsis-induced upregulation of ubiquitin-proteasome system marker Fxbo32 (p=0.002), loss of muscle mass was not prevented (p=0.8). Sepsis reduced triglyceride content (p≤0.01) equally in PN+gluc and PN+3HB mice (p=0.9), but did not affect markers of muscle FAO (Cd36, Cpt1b, Acadl, Hadha mRNA) in PN+gluc and PN+3HB mice (p≥0.2). Muscle glycogen content was unaltered by sepsis (p=0.1), and the illness-induced decrease in plasma glucose concentrations at day 5 (p≤0.004) was similar in PN+gluc and PN+3HB mice (p=1). Gene expression of the ketolysis enzyme Oxct1 was lower in PN+3HB compared to PN+gluc septic mice and healthy controls (p≤0.05). Also, ketone body transporter Mct1 mRNA was elevated by sepsis (p≤0.01) but attenuated by 3HB (p=0.05) compared to glucose. Sepsis-induced elevation of inflammation markers (Tnfα, IL-1β, Nlrp3 mRNA levels; p≥0.4) and autophagy (Atg5, Atg7 mRNA; p≥0.4; LC3 II/I, p-ULK1/ULK1 protein; p≥0.4) were not countered by 3HB infusion compared to glucose. However, PN+3HB septic mice did show enhanced markers of muscle regeneration (Myod1, Myf5 and Myog mRNA; p≤0.04) compared to PN+gluc septic mice and healthy controls. Also, sepsis-induced elevation of expression of Hdac4, an inhibitor of myoblast differentiation stimulator Mef2c, was reduced by 3HB (p=0.03) and resulted in increased Mef2c mRNA levels (p=0.01) compared to glucose. Conclusion: In conclusion, the observed protection against sepsis-induced muscle weakness with 3HB infusion coincided with enhanced markers of muscle regeneration, but not with changes in substrate handling. Supplemented 3HB during sepsis thus acted as a signaling molecule rather than as an energy source.