Thomsen · The American journal of clinical nutrition 2018 · randomized crossover trial · n=10

Effects of 3-hydroxybutyrate and free fatty acids on muscle protein kinetics and signaling during LPS-induced inflammation in humans: anticatabolic impact of ketone bodies.

Cited 117 times in the scientific literature.

Level 2 - randomized trial

Randomized crossover trial in humans

PubMed 30239561 · doi:10.1093/ajcn/nqy170 · record verified 2026-08-30

What was done

Ten healthy male subjects completed a randomized, 3-way crossover study examining whether infusions of 3-hydroxybutyrate (3OHB), free fatty acids (FFAs via Intralipid), or saline reduce protein catabolism during acute LPS-induced inflammation. All subjects received LPS and Acipimox (to restrict endogenous lipolysis) across three testing sessions during a 5-hour basal period and a 2-hour hyperinsulinemic-euglycemic clamp. Protein kinetics were measured using labeled phenylalanine, tyrosine, and urea tracers, and muscle signaling was evaluated via Western blot on muscle biopsies.

What was found

Infusion of 3OHB increased circulating 3OHB to 3.5 mM (P < 0.0005) and reduced whole-body phenylalanine-to-tyrosine degradation. Both basal and insulin-stimulated net forearm phenylalanine release decreased by >70% (P < 0.005), with profound reductions in both phenylalanine appearance and disappearance. In skeletal muscle, eIF2alpha phosphorylation at Ser51 increased and S6 kinase phosphorylation at Ser235/236 tended to decrease (P = 0.074), while markers of protein breakdown showed no detectable changes. Lipid infusion had no effect on phenylalanine kinetics, and insulin sensitivity was unaffected across interventions.

Why it matters

This study demonstrates that 3-hydroxybutyrate has a direct anticatabolic effect in human skeletal muscle during acute endotoxemia, primarily by suppressing protein breakdown to a degree that overrides concomitant reductions in protein synthesis.

Limits

The trial had a small sample size (n = 10) consisting solely of healthy males in an acute experimental endotoxemia model. The mechanistic muscle biopsy signaling findings did not reflect the tracer-measured suppression of proteolysis, and the abstract lacks exact baseline kinetic rates and confidence intervals.

Cited by