Mechanism-based intervention in Non-Alcoholic Fatty Liver Disease: mapping the muscle-liver axis and exploring the effect of L-carnitine on liver fat, insulin sensitivity and mitochondrial energy kinetics

The rising prevalence of non-communicable disease reflects a global healthcare burden that is the product of overnutrition and physical inactivity. As these deleterious lifestyles persist, their metabolic consequences have reached epidemic proportions. In this context, and with its corollaries of obesity and insulin resistance, non-alcoholic fatty liver disease (NAFLD) presents a growing public health and economic challenge. With no approved pharmacological treatment currently available, NAFLD will dominate the landscape of hepatology for the foreseeable future. Its prevalence, associated complications and escalation of disease burden in coming decades together underscore a time-critical unmet need for disease-modifying therapy which is broadly applicable, safe and cost-effective. The research presented in this thesis sought to explore mechanisms underpinning insulin resistance in a NAFLD population, and in particular to determine the role of intramyocellular lipids and muscle insulin resistance in the pathogenesis of NAFLD. Further, the role of chronic L-carnitine supplementation on liver fat, whole-body insulin sensitivity and hepatic energy metabolism in individuals with NAFLD was quantitatively evaluated in a placebo-controlled randomised trial. The data presented herein argues that chronic substrate overabundance, combined with defects in mitochondrial lipid oxidation, give rise to the ‘perfect storm’ of ectopic lipid accumulation and subsequent interruption of insulin signalling pathways in target tissues. This, together with oxidative stress, creates an insulin resistant, pro-fibrogenic milieu at the level of the liver. Leveraging a two-step euglycaemic hyperinsulinaemic clamp technique together with sophisticated precision imaging, I have demonstrated that NAFLD is associated with excess intramyocellular lipid accumulation and whole-body insulin resistance. Further, I have shown that even at an early stage in its natural history, NAFLD is associated with impaired hepatic mitochondrial energetics and defective oxidative phosphorylation. A plausible explanation linking these findings is that local muscle insulin resistance alters the pattern of energy storage, favouring diversion of carbohydrate substrate to the liver and, through de novo lipogenesis, exacerbates intrahepatic lipid accumulation. In a systematic review and meta-analysis, I have synthesised existing data detailing the effects of dietary carnitine loading on liver enzymes, insulin resistance profiles and liver fat in NAFLD. As a naturally occurring quaternary amine, L-carnitine has a well-established safety profile. In its dual role as an essential cofactor for mitochondrial fatty acid β-oxidation, and as a facilitator of muscle glycogen storage, L-carnitine stands at the nexus of glucose and lipid homeostasis. These unique properties render it an intriguing candidate therapy for NAFLD. Results from five randomised trials are presented, arguing that dietary L-carnitine supplementation may be an effective tool in the treatment of NAFLD, through lowering liver lipid, improving biomarkers of liver injury and improving metabolic phenotype. To further explore this hypothesis, I conducted a placebo-controlled randomised trial in young, non-diabetic individuals with NAFLD comparing 24 weeks of twice daily L-carnitine therapy (plus Slimfast supplementation to provide an insulinogenic stimulus for muscle carnitine uptake) versus maltodextrin placebo (plus Slimfast) on intrahepatic triglyceride concentration (IHTG). I have shown that mean IHTG decreased in the L-carnitine group, while it increased in the placebo group (-3.5% [-5.9; -1.5] versus 6.0% [1.5; 11.0], p = 0.002). Further, L-carnitine treatment was associated with a decrease in intramyocellular lipid to extramyocellular lipid (IMCL: EMCL) ratio (0.92 ± 0.62 to 0.42 ± 0.30 in the L-carnitine group versus 1.53 ± 0.99 to 1.47 ± 0.36 in placebo, p < 0.001) and improved leg glucose uptake (p=0.02). Serum alanine aminotransferase (ALT) declined significantly in the L-carnitine group versus placebo (p=0.04). Peripheral (muscle) insulin sensitivity and adipose tissue insulin resistance were not significantly different between groups compared to baseline values (p=0.83 and 0.72, respectively). Finally, we observed an improvement in forward rates of hepatic ATP synthesis with L-carnitine (+ 0.50 mM/s vs - 0.09 Mm/S in placebo group, p=0.025), suggesting that L-carnitine is capable of boosting hepatic mitochondrial energy kinetics. There was no significant change in weight (kg) from baseline within or between groups (p=0.67). In summary, we implicate muscle lipid deposition and insulin resistance as important contributors to NAFLD. We argue that augmenting muscle carnitine content is capable of reducing myocellular lipid and intrahepatic triglyceride. Improved liver mitochondrial energy kinetics are consistent with a reduction in liver fat following carnitine supplementation in a NAFLD population. Whether these effects are indirect, consequent to carnitine loading in muscle, or whether exogenous carnitine exerts a direct effect on hepatocyte lipid metabolism, remains to be established and will require harnessing recent advances in precision imaging to map hepatic and muscle carnitine content non-invasively