The experiments presented in this work provide novel evidence regarding the effect of BTP2 on skeletal muscle fibres. We show that very low µM concentrations of BTP2 block a t-system Ca2+ leak channel, likely to be Orai1. However, at concentrations above 5 µM, RyR Ca2+ leak and release was impaired by BTP2. The inhibitory effect of BTP2 on the RyR was observed following either extracellular or intracellular application of the agent. The sensitivity of low [BTP2] to a t-system Ca2+ leak channel has implications for understanding sarcolemmal Ca2+ leak, which is important in conditions such as muscular dystrophy (Turner et al 1991) and malignant hyperthermia (Eltit et al 2013; Cully et al 2018). Additionally, the promiscuity of BTP2 at commonly used doses highlight a need for re-evaluation of some conclusions regarding SOCE function in skeletal muscle that have been based under the premise of a specific action of BTP2 on Orai1.
BTP2 mechanism of action
The lack of effect of 1 mM BTP2 perfused into Jurkat cells cytoplasm in contrast to the inhibitory effect observed on CRAC channel when applied extracellularly suggests that, regardless of being a membrane permeant molecule, 1 mM BTP2 exerts its action through extracellular interaction with this channel (Zitt et al. 2003). In accordance with Zitt et al, the administration of cytosolic 1 mM BTP2 in this work did not exert a significant effect on Ca2+ movements, whereas higher concentrations such as 10 mM (where the RyR becomes affected) were needed to see an effect, coinciding with what is typically used in the skeletal muscle field (Zhao et al. 2005; Thornton et al. 2011; Wei-Lapierre et al. 2013). Experiments shown on figure 6 where 10 mM BTP2 was applied extracellularly strongly suggest that regardless of the nature of exposure, either cytosolic or extracellular administration of 10 mM BTP2 impairs the release of Ca2+ through the RyR.
Being a cell permeant molecule, it would not be surprising that 10 mM BTP2 could interact with both extracellular or cytosolic targets; importantly, the two possibilities are not mutually exclusive. In regards of putative BTP2 intracellular effectors, a chemico-genetic analysis revealed the F-actin binding cytosolic protein Drebrin as a direct ligand of BTP2 (Mercer et al. 2010). Intriguingly, Drebrin knock down leads to inhibition of SOCE at comparable levels to BTP2 treatment and the authors did not observe synergic effect when Drebrin knock down cells were treated with BTP2, suggesting that the SOCE inhibitory effect of BTP2 occurs through Drebrin inhibition. Anecdotally, when Zitt et al. perfused 1 mM BTP2 into Jurkat cells, they reported a drastic change in cell morphology that was not explored in detail. This observation goes on the same line of BTP2 as a modulator of the actin cytoskeleton, a well-known determinant of cell morphology.
The role of cytoskeleton on SOCE remains controversial. While some reports suggest that actin filaments participate in STIM1 rearrangement and association with SOC channels (Galán et al. 2011), previous ones did not observe an effect of actin cytoskeleton modulating SOCE (Ribeiro, et al. 1997). Interestingly, Rahman et al. observed no effect of BTP2 in STIM1 aggregation after SR Ca2+ depletion (Rahman and Rahman 2017), suggesting that BTP2 could be impairing Orai1 function independently of STIM1 modulation. In line with this, evidence of the modulation of actin cytoskeleton to TRP channels has been described for TRPC1, TRPC4 and TRPC5 (Rosado, et al. 2000; Tang et al. 2000). Therefore, one possibility is that BTP2 mechanism of action on Orai1 involves, as has been previously suggested, Drebrin inhibition, leading to the impairment of actin filaments interacting with SOC channels and subsequently the dysregulation of them. Structural studies of the interplay between actin filaments and SOC channels would improve our knowledge about how these components cooperate.
A question that is raised from our work is how BTP2 impairs RyR function. Following the evidence of actin cytoskeleton as a target of BTP2, it would not be surprising that the cytoskeletal negative modulation could lead to the impairment of the RyR. In a study performed on the neuroblastoma cell line NG115-401L, Bose et al found that actin disruption using cytochalasin D impairs the RyR-mediated ER Ca2+ release (Bose and Thomas, 2009). In the case of skeletal muscle, γ-actin has been found attached to the SR and its disruption leads to impairment of SR Ca2+ release, suggesting that actin cytoskeleton plays a modulatory role on RyR function in skeletal muscle (Gokhin and Fowler 2011). However, we cannot rule out a direct action of BTP2 on RyR function.
BTP2, Ca2+ leak and SOCE across the t-system membrane
In resting myotubes, Eltit et al (2013) found a similar inhibition of the influx of Ca2+ in the presence of 5 µM BTP2 or with the overexpression of dominant negative Orai1E190Q. This similarity and the well-established effect of BTP2 on Orai1 suggests the BTP2-sensitive channel in the resting cell is likely to be at least partially through Orai1. Additionally, BTP2 may inhibit a resting Ca2+ conductance through TRP channels (Eltit et al 2013). Our results and those of Eltit et al (2013) indicate a basal conductance of Ca2+ through Orai1/TRP channels in the absence of STIM1 activation. The level of Orai1 Ca2+ flux in muscle is amplified from its basal level by the dissociation of Ca2+ from STIM1 as the RyR Ca2+ conductance increases during action potential-induced Ca2+ release, probably to remain proportional the activation of Ca2+ efflux pathways across the t-system with increasing [Ca2+]JS (Cully et al 2018; Koenig et al 2018; Azimi et al 2020).
The higher [BTP2] required to inhibit SOCE in skinned fibres (Fig. 1) than the t-system Ca2+ leak (Fig. 2) likely reflects the increased activation of the STIM1/Orai1 pathway as caffeine thoroughly depletes the SR of Ca2+. Additionally, we note that the selective Orai1 enhancer, IA65, increased the t-system Ca2+ flux in the presence of low Mg2+ and caffeine in skinned fibres, confirming the molecular identity of the t-system SOCE channel as Orai1 (Azimi et al 2020). However, it remains possible that other channels also conduct SOCE in the muscle.
Pathophysiological Ca2+ entry in muscular dystrophy and malignant hyperthermia are both likely to have components that are dependent and independent of RyR Ca2+ leak (Turner et al 1991; Eltit et al 2013; Cully et al 2018). BTP2 and, for example, tetracaine, provide complimentary tools for deciphering the modes of excessive Ca2+ entry in muscle. Pathways for Ca2+ leak into the muscle that are blockable by GsMTx-4 and heavy metals also exist (Eltit et al 2013), indicating that the t-system is a compartment very leaky to Ca2+. The t-system is dependent on the function of the SR in sequestering and concentrating cytoplasmic Ca2+ into the junctional space via RyR Ca2+ leak, where the local increase [Ca2+] at the t-system increases plasma membrane Ca2+ ATPase activity to maintain the steep t-system Ca2+ gradient (Cully et al 2018). Thus, the pathophysiological excessive Ca2+ influx under muscular dystrophies and RyR-related myopathies should be considered conditions of altered basal Ca2+ handling at the SR and t-system.
It is possible BTP2 also effects other RyR isoforms. In the sinoatrial node, Liu et al. showed a negative effect of BTP2 on spontaneous and caffeine-induced SR Ca2+ release (Liu et al. 2015). This observation made on intact cells may be attributed to a direct effect of BTP2 on RyR2 Ca2+ efflux or an effect on Orai1 Ca2+ current. However, the our demonstration of an effect of BTP2 on RyR1 it would seem that the results of Liu et al are at least in part due to a slowing of Ca2+ efflux through RyR2 after exposure to the drug. An evaluation of the effect of BTP2 on RyR2 is warranted.
Physiological significance of SOCE in skeletal muscle
A major hypothesis regarding the physiological role of SOCE in skeletal muscle is that SOCE provides resistance to fatigue, where it is proposed that a loss of Ca2+ from the stimulated muscle needs to be replenished by Ca2+ from outside the fibre via SOCE (eg. Zhao et al 2005; Thornton et al 2011; Wei-Lapierre et al 2013). To support this theory, researchers have followed two different strategies. The first one is related to the use of genetically modified animals where SOCE is impaired in constitutively Orai1 or STIM1 knock-out mice. However, this approach presents developmental defects in the adult mutant muscle (Stiber et al. 2008) (Wei-Lapierre et al 2013) and thus complicates the comparison between wild-type and STIM1/Orai1 knock-out mice as to the action of SOCE during repetitive cycles of EC coupling. In contrast to this result, the only study of an inducible Orai1 knock-out mouse concluded that there is no significant role for acute SOCE in resisting fatigue in repetitive EC coupling cycles (Carrell et al. 2016).
The second strategy is based on the use of multiple pharmacological agents that impair SOCE. Some of the most popular ones have been BTP-2, 2-ABP and SKF-96365. 2-ABP and SKF-96365 have been shown to be unselective for the SOC channels (Launikonis and Ríos 2007; Olivera and Pizarro 2010). In respect of BTP2, the data presented here suggest that [BTP2] should not exceed 5 µM intracellularly to avoid RyR impairment. The BTP2 concentration chosen to block Orai1 in muscle experiments has typically been 10 µM, which in mechanically skinned fibres curtails the release of Ca2+ from the SR and the activation of SOCE. In these experiments the decline of the action potential-induced Ca2+ transients in the presence of BTP2 in intact and skinned fibres show decline (Fig 3; Wei-Lapierre et al 2013). While the patterns of decline where not identical (eg. Small shift in concentration dependence), this may be due to factors such as diffusion rates across the plasma membrane, potency of different batches of BTP2 or other minor differences in handling the agent across different labs. Importantly, the decline of the Ca2+ transient was not dependent on which side of the plasma membrane it was applied, nor was it dependent on number of action potentials stimulating the fibre, as a single direct stimulation of RyR opening was affected by exposure to BTP2 (Figs 5 & 6). It follows that an action of BTP2 on the RyR on intact fibre preparation is likely to be the cause of the Ca2+ transient decline in those experiments as well.
Additionally, the biophysical properties of Ca2+ release and SOCE flux in skeletal muscle do not provide a framework within which it is possible to model SOCE as a “store-refiller” during repetitive cycles of EC coupling. During EC coupling, SOCE is activated very rapidly and briefly following the release of Ca2+ from the SR, which causes the near-membrane depletion of Ca2+ in the SR terminal cisternae to activate the local STIM1 (Koenig et al 2018, 2019). The proportional contribution of the SR and t-system to the Ca2+ entering the cytoplasm during EC coupling is 99:1, making it difficult to argue that SOCE contributes significantly to the Ca2+ in the muscle cytoplasm during repetitive cycles of EC coupling (Koenig et al 2018). The development of fatigue is more likely due to the inhibition of the Ca2+ release mechanism of the muscle by the build-up of metabolites than an inherent loss of fibre calcium (Allen et al. 2008) (Olsson et al 2020). We also point out that the SOCE mechanism and kinetics in muscle is not different between slow- and fast-twitch fibres (Cully et al. 2016), thus making it unlikely that a role of SOCE is based around fatigue-resistance.
Ivarsson et al (2019) and Nelson et al (2019) have recently reported enlightening results around the role of SOCE signalling in muscle. Nelson et al. described a set of conserved phosphorylation events in mouse, rat and human in response to exercise. One protein whose phosphorylation was conserved in response to exercise was STIM1. Moreover, the authors demonstrated that STIM1 phosphorylation in response to exercise negatively regulates SOCE (Nelson et al. 2019). This is contradictory to the idea of SOCE as a SR refiller to prevent fatigue during exercise. Interestingly, SOCE probably acts as a signal for muscle adaptation following exercise. Ivarsson et al (2019) showed increases in RyR Ca2+ leak following endurance exercise in mice to induce store-dependent influx during periods of rest and decreases in muscle STIM1 content as mice became fitter (Ivarsson et al. 2019). Near-membrane depletion of Ca2+ inside the SR due to the leaky RyR will cause Ca2+ dissociation from STIM1, providing physiological activation of SOCE while the SR Ca2+ content remains relatively high in the presence of a fully functional SR Ca2+ pump (Cully et al 2016, 2018).