The missing ‘link’? β‐Catenin's role in skeletal muscle glucose uptake during exercise and contraction

The complex signalling mechanisms whereby insulin and contraction both mediate GLUT4 translocation to the plasma membrane of skeletal muscle cells to facilitate glucose uptake have been a point of intense research for decades. Both pathways share a common outcome of increased GLUT4 translocation and glucose uptake; however, their upstream signals are independent from one another (Richter & Hargreaves, 2013). This has led to a focus on the convergence of distal signals to better understand how glucose uptake is regulated in a variety of states such as during exercise or disease. Insulin resistance, as is seen in type 2 diabetes, results in an inability to stimulate glucose uptake via insulin. Paradoxically a diabetic individual who undergoes an exercise bout displays increased GLUT4 translocation and glucose uptake, and improved insulin sensitivity during exercise that persist for hours post-exercise (Bird & Hawley, 2017). To reconcile this, studies involving knockout, overexpression or pharmacological intervention at various points of each respective signalling cascade have found several proteins that respond to both insulin and exercise/contraction signals to translocate GLUT4. A protein of interest where this convergence may occur is Rac1, an insulin- and contraction-responsive protein that is implicated in actin cytoskeleton remodelling (Richter & Hargreaves, 2013). Actin cytoskeleton remodelling is the process of polymerization and reorganization of actin filaments to provide a physical structure for signal transduction and, more importantly in this context, intracellular trafficking (Masson et al. 2020; Richter & Hargreaves, 2013). In response to insulin this process is critical for the delivery of GLUT4-containing vesicles to the plasma membrane, subsequent docking and glucose uptake in skeletal muscle as determined in numerous studies involving actin cytoskeleton disruption (Richter & Hargreaves, 2013). Speculatively, exercise/contraction may elicit a similar outcome via Rac1, an actin cytoskeleton-regulating GTPase, as it has been shown to respond to both insulin and contraction signals and is required for maximal glucose uptake (Richter & Hargreaves, 2013). Recently the structural protein β-catenin, a downstream target of Rac1, was identified as a vital component of actin cytoskeleton remodelling (Masson et al. 2020). Phosphorylation on Ser552 of β-catenin by insulin has been found to be essential for insulin-induced GLUT4 translocation (Masson et al. 2020). Interestingly, it has been shown that both Rac1 and β-catenin are exercise-responsive proteins (Richter & Hargreaves 2013; Masson et al. 2021); thus, a link in both insulin- and contraction-mediated GLUT4 trafficking is possible at β-catenin. A recent article by Masson et al. (2021) published in The Journal of Physiology aimed at investigating the role of β-catenin in exercise/contraction-stimulated skeletal muscle glucose uptake, and identifying a putative mechanism by which it may occur. The strongest aspect of this study was the models employed, as an extensive series of in vivo and ex vivo approaches was conducted involving inducible muscle knockouts in mice and cell line cultures. An effort was made to validate their results through exercise, ex vivo electrical pulse stimulation (EPS) and pharmacological interventions. The diversity of experiments and models used allowed the authors to confirm their hypothesis that β-catenin is required for optimal glucose uptake in response to exercise/contraction through a mechanism of actin remodelling. In addition, it was found that muscle contraction/exercise results in phosphorylation at Ser675 of β-catenin, and not the insulin-sensitive Ser552 site, to promote actin cytoskeleton remodelling. Initially, Masson et al. (2021) used the well documented Cre/Lox method to generate β-catenin-inducible muscle knockout mice (BCAT-imKO; 40–60% reduction in β-catenin protein), with induction of Cre expression via tamoxifen treatment. Littermates not expressing the Cre gene were used as control (WT) and received tamoxifen as well. Mice were fasted, received an intraperitoneal injection of radiolabelled 2-deoxyglucose, and exercised at moderate intensity on a treadmill (or rested) prior to having tibialis anterior and gastrocnemius muscles removed for in vivo analysis of glucose uptake. In another subset of BCAT-imKO and WT mice, extensor digitorum longus (EDL) and soleus were isolated for ex vivo analysis using EPS that followed a well-established glucose uptake assay. Regardless of contraction protocol and fibre-type, BCAT-imKO resulted in a significant reduction in glucose uptake in contracting muscle, demonstrating a requirement for β-catenin in maximal glucose uptake. The reduction during in vivo glucose uptake in BCAT-imKO mice could be associated with differences in running capacity, duration and relative exercise intensity, as these are known to affect glucose uptake (Richter & Hargreaves, 2013). However, compared to WT and suggesting a role for β-catenin in glucose uptake, during moderate intensity and run-to-exhaustion exercise protocols no difference was reported in distance run, energy expenditure, fuel utilization, blood lactate and the marker of intracellular energy charge, AMP-activated protein kinase (AMPK) phosphorylation. Comparing in vivo exercise-induced and ex vivo EPS-induced glucose uptake, the reduction during the in vivo trials for BCAT-imKO was ∼40% and ∼25% that of the WT response in tibialis anterior and gastrocnemius, respectively, but the reduction due to loss of β-catenin during ex vivo EPS was nearly to the basal level in both EDL and soleus. Differences in the two absolute reductions between the protocols, as described by the authors, could be attributed to β-catenin having greater importance for the intrinsic ability of the muscle to enhance glucose uptake, rather than an extrinsic factor such as blood flow to working muscle. After a moderate intensity running protocol in a subset of wild-type mice with normally expressed β-catenin, phosphorylation at Ser675 increased ∼33% compared to rest but interestingly not that at Ser552, the documented insulin-sensitive phosphorylation site (Masson et al. 2020). The significance of two independent phosphorylation sites with a mutual outcome is possibly due to the redundancy of the system to allow GLUT4 translocation; as mentioned above, disruption of the actin cytoskeleton impairs glucose uptake, and therefore it would be relevant to have multiple ways to activate and permit translocation. Further, redundancy is a common theme in governing glucose uptake at skeletal muscle, as is seen in instances of reactive oxygen species, AMPK and calcium having the ability to stimulate the process (Richter & Hargreaves, 2013). In an attempt to further characterize β-catenin's role in glucose uptake, Masson et al. (2021) performed EPS and pharmacological activation of AMPK by MK8722 in separate experiments on C2C12 myotubes. EPS elicited Ser675 phosphorylation; however, pharmacological AMPK activation did not. This may suggest that phosphorylation of β-catenin is not dependent on energy charge, but rather another mechanism intrinsic to the muscle, for instance mechanical stress. It has been reported that Rac1 can be activated by stretch (Sylow et al. 2016), and conceivably, this could be a means of activation during exercise/contraction. However, β-catenin's response to stretch via Rac1 signalling and not AMPK is speculative, as it was not confirmed in MK8722 treatment conditions that AMPK was activated, and so this remains to be resolved. An argument for β-catenin's role in response to mechanical stretch can be seen in further C2C12 myotube experiments where pharmacological inhibition of Rac1 during EPS results in no change in β-catenin's Ser675 phosphorylation, and known downstream targets of Rac1 are also unchanged as validation. To highlight β-catenin's upstream signals, Masson et al. (2021) utilized Cre/Lox generated Rac1-imKO mice and compared them to WT littermates after acute moderate intensity exercise; here, β-catenin was not phosphorylated in either group. Data from this set of experiments were equivocal, as the Rac1-imKO used was not confirmed to have Rac1 ablated in skeletal muscle and when further analysing the effects on AMPK Thr172 phosphorylation after exercise, there was no statistical change from rest. Phosphorylation of AMPK Thr172 is an established confirmation for exercise/muscle contraction and has previously been shown to increase significantly in an identical exercise protocol (Sylow et al. 2016), so it remains undetermined as to whether exercise elicited its desired stress. Notwithstanding, the authors interpret the data as β-catenin's phosphorylation being intensity dependent. In a final set of cell culture experiments, Rac1 involvement was confirmed in C2C12 myotubes and L6-G4-myc myoblasts; focus will be placed on the latter as both cell lines share similar findings and the specific property of L6-G4-myc myoblasts of overexpressing GLUT4 makes this an interesting approach. The L6-G4-myc myoblasts were split into either a control or an experimental group with transfected mutant β-catenin (Ser → Ala at Ser675) to analyse GLUT4 trafficking and actin cytoskeleton remodelling. Upon stimulation with the Rac1 activator carbachol, Ser675 phosphorylation was significantly increased with concomitant increase in GLUT4 at the plasma membrane. Surprisingly, the mutated myoblasts responded with a significant increase in GLUT4 at the plasma membrane, albeit significantly lower than the control. To reconcile this, analysis of actin cytoskeleton dynamics showed that nearly an identical total number of actin polymers was associated after Rac1 stimulation. However, a difference in the change in actin polymerization upon stimulation from the basal state was apparent, with the mutant form showing minimal increase in actin polymerization. The effect of β-catenin on actin cytoskeleton remodelling was further validated through known actin cytoskeleton staining with phalloidin. The BCAT-imKO mice, in response to high-intensity exercise, displayed negligible change in actin dynamics, with WT counterparts showing the anticipated remodelling. This insight into β-catenin's role in response to contraction and exercise helps bridge a gap in our understanding of the physical process of GLUT4 vesicle trafficking within skeletal muscle. It is interesting to note that the findings in this paper are fibre type independent despite differences in fast and slow twitch fibres’ reliance on blood glucose for fuel. Future research may focus on β-catenin's role in more specific exercise scenarios where a greater reliance is placed on blood glucose, such as in a glycogen depleted state. Further, it would be interesting to compare Rac1-imKO and BCAT-imKO during matched exercise volume and intensity as the current article does not have consistent exercise protocols throughout, making conclusions more limited. Finally, a future series of experiments to consider is induction of insulin resistance to analyse β-catenin function and the direct targeting of its Ser552 and Ser675 phosphorylation sites to induce GLUT4 translocation and glucose uptake as a potential therapeutic target. In conclusion, the authors executed an array of experiments that shed light on β-catenin's requirement for maximal glucose uptake in skeletal muscle in response to exercise/contraction. Although further research is required to understand the complex regulation of glucose uptake in skeletal muscle, the findings by Masson et al. (2021) help establish a link in the process and represent an important advancement in this field. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. No competing interests declared. Sole author. No funding was received for the completion of this manuscript-journal club submission.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.