HSL and ATGL: the movers and shakers of muscle lipolysis

Intramuscular triglyceride (IMTG) accounts for only a fraction of the total lipid store in the human body. However, given its significant contribution as a substrate for ATP synthesis during exercise, and its proposed role in the development of skeletal muscle insulin resistance, IMTG metabolism has proved to be an area of great scientific interest over the last 20 years. Studies investigating IMTG utilisation during exercise remained equivocal, until methods using immunofluorescence microscopy enabled the muscle fibre type-specific assessment of IMTG content without the interference of extramyocellular adipocytes. Studies using this method convincingly demonstrated that in trained individuals 60–70% of IMTG can be depleted in type I muscle fibres during prolonged moderate-intensity exercise, and can account for up to 50% of total lipid oxidation (van Loon, 2004). Of more clinical relevance were studies highlighting the potential role of intramuscular lipid accumulation in the development of skeletal muscle insulin resistance. This hypothesis has been refined over the last decade, as highly insulin-sensitive, endurance-trained athletes exhibit higher levels of total IMTG and diacylglycerol (DAG) (Amati et al. 2011). It is now hypothesised that the accumulation of ceramides and certain species of membrane-bound DAG are responsible for the inhibition of insulin-mediated glucose uptake.

The regulation of intramuscular lipolysis is an important area, given its role in the regulation of fatty acid supply to the mitochondria during exercise and in the determination of skeletal muscle lipid content and composition. However, the precise mechanisms regulating muscle lipolysis remain poorly understood. Hormone-sensitive lipase (HSL) was long considered to be the only enzyme involved in muscle lipolysis, until adipose triglyceride lipase (ATGL) was discovered and found to be expressed in skeletal muscle. As the activities of ATGL and HSL are highly specific for triglyceride (TG) and DAG, respectively, ATGL is now considered to contribute to the initial step in TG hydrolysis, with HSL primarily responsible for the breakdown of DAG.

Important and intriguing new data have recently been presented by Alsted et al. (2013), which question the importance of HSL in contraction-mediated lipolysis in skeletal muscle. The authors hypothesised that lipases other than HSL are responsible for IMTG lipolysis during muscle contraction. In a well-designed series of experiments, the authors used a combination of pharmacological HSL inhibition (using a small-molecule inhibitor of HSL; 76–0079, Novo Nordisk, Bagsvaerd, Denmark) and HSL knockout (KO) mice, with an ex vivo electrical stimulation approach, to probe the effects of muscle contraction on IMTG lipolysis in the absence of HSL activity. In addition, they used histochemical methods to measure contraction-induced changes in IMTG content as a surrogate of IMTG lipolysis.

First, the study demonstrated that IMTG content decreased following muscle contraction. This is in agreement with the findings of various human studies assessing IMTG utilisation during exercise using the neutral lipid dye oil red O (van Loon, 2004). IMTG content is determined by the balance between lipolysis and esterification of intracellular fatty acids at a given time; therefore, it is clear that lipolytic rates outweigh rates of esterification during muscle contraction and dynamic exercise. The degree of IMTG utilisation (∼72–79% following 20 min of contraction) was large and, in contrast with what is generally observed in human studies, not fibre type specific. Studies in humans show that IMTG utilisation of a similar magnitude occurs only in type I muscle fibres after prolonged (2–3 h), moderate-intensity exercise in trained athletes (van Loon, 2004). These differences can probably be attributed to the ex vivo electrical stimulation approach employed in the study, which maximally recruits both type I and type II muscle fibres. Furthermore, IMTG net breakdown is likely to be high as contractions occur in the absence of extracellular fatty acids, meaning that esterification rates will be limited, whereas the rate of lipolysis and oxidation of IMTG-derived fatty acids will be elevated.

The key finding presented in the article showed that the breakdown of IMTG during ex vivo contraction of rat muscle was unaffected by acute inhibition of HSL activity and was no different between the muscle of HSL KO and wild-type mice. Therefore, this study shows that IMTG breakdown during ex vivo electrically stimulated muscle contraction is not affected by the absence of HSL activity. As the authors also demonstrated that HSL and ATGL account for ∼98% of all TG lipase activity, at least in resting muscle, they concluded that ATGL plays an important role in IMTG lipolysis during muscle contraction.

The finding that IMTG breakdown occurs during muscle contraction without HSL activity is somewhat surprising, given that the rise in TG hydrolase activity following electrical stimulation of rodent muscle is completely prevented when HSL activity is blocked (Langfort et al. 2000). Importantly, the latter study measured TG hydrolase activity after contractions using an in vitro assay-based approach and HSL activity was blocked with the addition of a HSL antibody to the muscle lysate. In contrast, Alsted et al. (2013) assessed IMTG breakdown by measuring IMTG content before and after muscle contraction, meaning that IMTG breakdown was stimulated and measured in intact muscle fibres. This approach offers several advantages. First, IMTG content was assessed in both type I and type II muscle fibres and without the interference of triacylglycerol contained in adipocytes located between muscle fibres, which can contaminate homogenisation-based approaches. In addition, this approach maintains lipid droplet (LD) morphology and localisation, meaning that in vivo translocation of lipases and their interaction with the regulatory proteins on the LD surface remain intact. HSL is known to translocate to the LD upon contraction and adrenaline stimulation of skeletal muscle (Prats et al. 2006), and it is becoming evident that muscle lipolysis is positively and negatively regulated by co-regulators on the LD surface [comparative gene identification-58 (CGI-58) and G0/G1 switch gene 2 (G0S2), respectively]. Furthermore, LD-associated proteins, particularly the perilipin family of proteins, have been proposed to regulate both lipase access to the LD and the interaction of lipases with co-regulators at the LD surface (see Zechner et al. 2012 for a comprehensive review on the role of LD-associated proteins in lipolysis). Therefore, the normal lipolytic response to muscle contraction is probably dependent on maintaining the subcellular compartmentalisation of the LD, lipolytic enzymes and associated regulatory proteins, all of which are lost when measuring TG hydrolase activity in vitro.

The experimental approach using ex vivo electrically stimulated contractions allows the effect of muscle contraction to be investigated in isolation, without the external hormonal and substrate influences which are known to affect lipolytic rates during endurance-type exercise and muscle contraction. This approach has previously provided important insights into HSL translocation in skeletal muscle. However, it is important to consider that muscle HSL activity and the rate of IMTG lipolysis are mediated by adrenaline and long-chain fatty acyl-CoA availability, in addition to contraction-mediated mechanisms involving Ca²⁺-mediated activation of protein kinase C and the extracellular signal-regulated kinase pathway. Therefore, the full influence of lipase activity may only be observed with activation by both hormonal- and contraction-mediated mechanisms, as occurs during dynamic exercise.

The current study suggests that ATGL plays a pivotal role in contraction-mediated skeletal muscle lipolysis, and this highlights the need for studies to specifically investigate the role of ATGL in muscle lipolysis. Previous studies using ATGL KO mice have shown that muscle contractile function performed ex vivo is unaffected by ATGL deletion (Huijsman et al. 2009). Nevertheless, combining ATGL KO and/or inhibition studies, in the presence and absence of HSL activity, will be an important step in confirming the critical role of ATGL in contraction-mediated muscle lipolysis. In addition, the mechanisms regulating ATGL and HSL activation in skeletal muscle remain poorly understood. ATGL activity appears to be regulated by phosphorylation, protein–protein interaction with its co-regulators CGI-58 and G0S2, and appropriate localisation to the LD surface. Several ATGL phosphorylation sites have been described, although it is currently unclear whether ATGL phosphorylation is an important regulator of ATGL activation and muscle lipolysis during exercise. In vitro cell-based studies have shown that the LD-associated protein perilipin 5 can bind both ATGL and CGI-58 on the LD surface, and perilipin 5 itself is present on the surface of the majority of LDs in skeletal muscle (Shepherd et al. 2013). Although the precise role of perilipin 5 in muscle lipolysis is unclear, the bulk of the available evidence indicates that it is responsible for limiting basal lipolytic rates. However, on protein kinase A (PKA)-mediated phosphorylation, perilipin 5 dissociates CGI-58, apparently to permit ATGL activation and enhance lipolytic rates. Therefore, perilipin 5 may prove to be central in regulating ATGL activity and skeletal muscle lipolysis.

Exercise training is known to increase the expression of ATGL in skeletal muscle, whereas HSL expression is unaffected. Given the apparent importance of ATGL in contraction-mediated IMTG lipolysis, it is tempting to speculate that an increase in muscle ATGL will facilitate greater muscle lipid utilisation post-training. Such an improvement in IMTG lipolysis and utilisation is proposed to contribute to the insulin-sensitising effect of regular exercise. Of course, full elucidation of the mechanisms underpinning the processes of IMTG lipolysis and IMTG synthesis remains a key area for future research. Accordingly, the evidence presented by Alsted et al. (2013) that ATGL plays an important role in contraction-mediated muscle lipolysis is an important step forward in our understanding of IMTG metabolism.