Coordinating activation of endo‐lysosomal two‐pore channels and TRP mucolipins

Two‐pore channels and TRP mucolipins are ubiquitous endo‐lysosomal cation channels of pathophysiological relevance. Both are Ca2+‐permeable and regulated by phosphoinositides, principally PI(3,5)P2. Accumulating evidence has uncovered synergistic channel activation by PI(3,5)P2 and endogenous metabolites such as the Ca2+ mobilizing messenger NAADP, synthetic agonists including approved drugs and physical cues such as voltage and osmotic pressure. Here, we provide an overview of this coordination.

Recent work has shown that both endogenous and pharmacological agonists of TPC2 and TRPML1, when combined, synergistically mediate cognate channel activation.A similar mode of activation is also seen in other isoforms of the family when agonists are combined with physical stimuli such as voltage and osmotic pressure.Here, we discuss such synergism, its molecular basis and potential physiological and pathophysiological significance.
The availability of cell-permeable TPC2 agonists affords novel opportunities to probe NAADP and PI(3,5)P 2 action in live cells.The effects of co-activating TPC2 were examined given the likely co-occurrence of NAADP and PI(3,5)P 2 under physiological conditions.Strikingly, sequential or co-addition of TPC2-A1-N and TPC2-A1-P caused a dramatic potentiation of Ca 2+ fluxes (from cell imaging) or currents (from patch-clamp recordings) compared with the response seen with either of the agonists alone (Yuan et al., 2022).In marked contrast, Na + fluxes or currents were unchanged (Yuan et al., 2022).Combined NAADP and PI(3,5)P 2 similarly exerted a synergistic increase in TPC2-mediated Ca 2+ but not Na + currents (Yuan et al., 2022).TPC2 therefore selectively favours Ca 2+ permeability during co-activation by its endogenous agonists or their mimetics despite radically different effects of the agonists alone on the channel (Fig. 2).
Although TPC2 can evidently change its ion selectivity, reflected by markedly different permeability ratios of Ca 2+ over Na + , this cannot account for the selective synergism.This is because the permeability ratio for co-stimulated TPC2 was either intermediate in the case of the natural agonist combination or dominated by TPC2-A1-N for the synthetic agonists (Gerndt et al., 2020;Yuan et al., 2022).Rather, a biased increase in channel open probability might explain such effects.Wang and Zhu have likewise noted the synergism between NAADP and PI(3,5)P 2 (Wang & Zhu, 2022).
A recent flurry of studies have provided structural insight into the regulation of TRPMLs by their agonists.Both the binding sites for PI(3,5)P 2 and ML-SA1 (Fine et al., 2018;Schmiege et al., 2017) have been identified in TRPML1.PI(3,5)P 2 binds the N-terminus at the cytosolic end of S1 whereas MLSA1 binds within the pore involving residues in S5, the first pore helix and S6 between subunits.An open structure with both ligands bound has provided insight into synergism through the identification of a cation-pi bond between residues in S3 and S4 (Fine et al., 2018).This interaction is evident in PI(3,5)P 2 -bound but not ML-SA1-bound structures.It appears to be stabilized in the double-liganded structure causing S3-S4 to move away from the pore centre which in turn is coupled to movement of S5 and S6 helices.Activity of channels with a stabilizing mutation at the putative S3/S4 interface was increased by ML-SA1 but decreased by PI(3,5)P 2 , consistent with a central role for this region in coordinating agonist action.Interestingly, ML-SA1 does not activate drosophila TRPML but it is still able to synergize with PI(3,5)P 2 to cause channel activation (Feng et al., 2014).
Recently, rapamycin which is best known as an mTOR inhibitor (Ballou & Lin, 2008;Li et al., 2014) has also been found to directly activate TRPML1 (Gan et al., 2022;Zhang, Chen, Gao, et al., 2019).Rapamycin is a naturally occurring macrocycle found in bacteria and an approved drug for immunosuppression.Similar to ML-SA1, PI(3,5)P 2 synergized with rapamycin or temsirolimus, a structurally related analogue to open the channel (Gan et al., 2022;Zhang, Chen, Gao, et al., 2019) (Fig. 2).This cooperation was also captured structurally through TRPML1 bound to both PI(3,5)P 2 and temsirolimus in the open state (Gan et al., 2022).In this study, local conformational changes associated with PI(3,5)P 2 binding appeared to differ from previous structural studies although there was convergence on S4 as a critical determinant.Like ML-SA1, temsirolimus bound at an inter-subunit interface close to the pore.Binding of temsirolimus and widening of the pore associated with opening was proposed to push S4 towards the PI(3,5)P 2 binding site thus facilitating PI(3,5)P 2 binding.Conversely, binding of PI(3,5)P 2 was proposed to pull S4 and expand the pore through the S4-S5 linker, thereby facilitating temsirolimus binding.TRPML1 is therefore allosterically activated through remote sites.
Interestingly, the binding sites for ML-SA1 and temsirolimus overlap.Indeed, temsirolimus competed with ML-SA1 causing dose-dependent inhibition of channel opening (Gan et al., 2022).Temsirolimus is therefore likely a partial agonist of TRPML1.

Coordinating activation of other TPC and TRPML isoforms
TPC1.TPC1 is the other TPC isoform found in humans and like TPC2 is both NAADP and PI(3,5)P 2 -activated (Brailoiu et al., 2009;Wang et al., 2012).Single channel recordings of TPC1 in lipid bilayers revealed a biased increase in Na + permeability of TPC1 in the presence of PI(3,5)P 2 , suggesting that like TPC2, TPC1 can also switch its ion selectivity.However, in contrast to TPC2, there was no mutual influence between NAADP and PI(3,5)P 2 on activating the channel (Pitt et al., 2014).Independent single channel analyses of TPC1 showed that NAADP-mediated TPC1 activity was regulated by voltage (Rybalchenko et al., 2012).Although initial vacuolar patch-clamp analysis using ramp protocols suggested TPC1 was voltage-insensitive when activated by PI(3,5)P 2 (Wang et al., 2012), subsequent analysis using step protocols confirmed its voltage sensitivity (Cang, Bekele, et al., 2014).Under the latter conditions, PI(3,5)P 2 and voltage can be considered co-agonists because neither alone appears to activate the channel (Fig. 3).Structurally, TPC1 binds PI(3,5)P 2 through domain I (She et al., 2018) and like its plant counterpart senses voltage through positively charged residues in S4 of domain II (Guo et al., 2016;Jaślan et al., 2016;Kintzer & Stroud, 2016).The voltage sensor in the apo form of TPC1 is in the 'up' activated position but the gate is closed.Binding of PI(3,5)P 2 straightens S6 in domain I and causes a rotation which is coupled to rotation of S6 in domain II.The latter movement results in opening of the gate but is sterically hindered when the voltage sensor is in the down position.This explains the dual requirement for PI(3,5)P 2 and voltage for channel opening.In other studies, TPC1 is activated by voltage in the absence of exogenous PI(3,5)P 2 (Pitt et al., 2014).Interestingly, agonists that activate TPC2 in a voltage-dependent manner either reverse rectification or inhibit voltage activation of TPC1 under these conditions (Zhang, Chen, Li, et al., 2019).These findings point to a complex interplay between ligand and voltage activation of TPC1.
Most plants express a single, voltage-sensitive, TPC1-like isoform in their vacuoles.Arabidopsis TPC shares up to 50% sequence similarity with its animal counterpart (Brailoiu et al., 2009;Jaślan et al., 2020Jaślan et al., , 2022;;Rahman et al., 2014) but it is insensitive to PI(3,5)P 2 and NAADP (Boccaccio et al., 2014;Jaślan et al., 2023;Patel et al., 2022).Of relevance here is its co-regulation by voltage and Ca 2+ .Patch-clamp studies have revealed that submillimolar concentrations of cytoplasmic Ca 2+ are required for the voltage-dependent activation of plant TPC1 channel, and that luminal Ca 2+ inhibits the channel (Hedrich & Neher, 1987;Pottosin et al., 2004).Cytosolic Ca 2+ binds to the second of the pair of EF hands in the cytosolic linker between the two domains with the first serving more of a modulatory role (Guo et al., 2016;Schulze et al., 2011).A recent study uncovered another important Ca 2+ binding site in the N-terminus (Ye et al., 2021).The inhibitory vacuolar calcium sensor consists of three Ca 2+ -binding sites formed by negatively charged amino acids located in different vacuole-facing parts of the channel (Dadacz-Narloch et al., 2011;Dickinson et al., 2022;Kintzer & Stroud, 2016).
Recent cryo-EM studies have provided molecular insight into the mechanism of the co-regulation of AtTPC1 by voltage and Ca 2+ .When the channel is closed, the voltage sensing domain in its down state is partly exposed to the cytosol and is stabilized by the first EF hand as well as the Ca 2+ -occupied inhibitory vacuolar Ca 2+ binding sites.The second EF hand, which is important for channel activation, remains in a Ca 2+ -unbound apo state.Dissociation of Ca 2+ from the vacuolar Ca 2+ binding site induces upward movement of the voltage sensor and priming of both the second EF hand and the N-terminal Ca 2+ binding site to bind Ca 2+ .Such Ca 2+ binding causes a repositioning of the first EF hand, resulting in channel opening (Dickinson et al., 2022;Guo et al., 2016;Hedrich et al., 2023;Kintzer & Stroud, 2016;Kintzer et al., 2018;Ye et al., 2021).In this way, Ca 2+ binding and voltage activation are intimately linked.
TPC3.TPC3 completes the ancestral TPC family in deuterostome animals.Functional characterization of recombinant TPC3 as well as TPC1 and TPC2 from echinoderms such as sea urchins (Brailoiu et al., 2010;Ruas et al., 2010) have taken foundational studies of NAADP-mediated Ca 2+ release in egg homogenates (Clapper et al., 1987;Lee & Aarhus, 1995) into the molecular cell biology era.The TPC3 gene is present throughout the chordates but it has undergone lineage-specific loss in mammals with a notable absence in humans (but not other primates) and mice (but not other rodents) (Brailoiu et al., 2010;Cai & Patel, 2010).
Initial electrophysiological studies of zebrafish and frog TPC3 heterologously expressed in mammalian cells indicated that TPC3 was present on both the plasma membrane as well as acidic organelles and functioned as a voltage-gated channel insensitive to PI(3,5)P 2 (Cang, Aranda, et al., 2014;Dickinson et al., 2020).However, other studies of plasma membrane frog TPC3 expressed homologously in oocytes revealed sensitivity to PI(3,5)P 2 and additionally PI(3,4)P 2 (Shimomura & Kubo, 2019) (Fig. 3).Rabbit TPC3 was also recently reported to have a broader phosphoinositide sensitivity than the other TPCs (Feng et al., 2022).
In frog TPC3, PI(3,4)P 2 in particular shifted the voltage-dependence of the channel to more hyperpolarized potentials.Structurally, this was ascribed to electrostatic interactions in the S6 helices of domains I and II.Cystine accessibility analyses and voltage-clamp fluorometry of residues either side of S4 in domain II confirmed a remote sensitizing effect of PI(3,4)P 2 on voltage-sensor movement (Hirazawa et al., 2021).Intriguingly, fluorescence-voltage relationships from voltage-clamp fluorometry were biphasic such that whereas the major fluorescence changes occurred at positive potentials concomitant with channel opening, smaller inverse changes were detectable at negative potentials where the channel is largely closed (Hirazawa J Physiol 602.8 et al., 2021;Shimomura et al., 2023).The latter signal was interpreted as a transition from an intermediate to a fully activated conformation associated with channel opening.Through a series of elegant mutagenesis experiments, the authors found that a forced interaction between S4 and S2 in domain II allowed channel activation at hyperpolarized potentials (Shimomura et al., 2023).Intriguingly, one set of charge reversal mutations eliminated the structural changes at positive potentials producing a channel that was substantially less voltage-sensitive and almost entirely PI(3,4)P 2 sensitive.This likely reflects the putative intermediate state specialized for phosphoinositide-gating but which is normally hidden in TPC3.Indeed, in TPC2, which is normally voltage-insensitive, stabilizing the corresponding S2/S4 interface was found to inhibit voltage-sensitive currents unmasked by desipramine (Shimomura et al., 2023).Taken together, these studies suggest conformational changes in domain II couple phosphoinositide-and voltage-gating.Intriguingly, the TPC blocker naringenin which inhibits PI(3,5)P 2 activation of TPC2 (Pafumi et al., 2017) potentiated currents by desipramine (Shimomura et al., 2023).This suggests that naringenin and perhaps other pharmacological modulators act by biasing TPCs in different activation modes.
TRPML2 and TRPML3.TRPML2 and TRPML3 are the other two, less well-studied members of the TRPML family.They show a more restricted tissue distribution in mammals than TRPML1 (reviewed in Cheng et al., 2010).Interestingly, TRPML2 has emerged as a mechanosensitive channel (Chen et al., 2020).Recent studies have identified agonists selective for each of the TRPML isoforms (Plesch et al., 2018;Spix et al., 2022).TRPML2 activity in response to the TRPML2-selective agonist ML2-SA1 (Plesch et al., 2018) is enhanced under hypotonic conditions (Chen et al., 2020) but hypotonic pressure alone only very modestly activated TRPML2.Such synergism was also evident for activation of TRPML2 by PI(3,5)P 2 under hypotonic conditions (Fig. 3).Mechanosensitivity required a residue close to the PI(3,5)P 2 binding site.Mutation abolished sensitivity to pressure but spared agonist responsiveness.TRPML1 and TRPML3 were not mechanosensitive.TRPML3 is a selective target of PI(3)P (Kim et al., 2022;Xu et al., 2023).PI(3)P binds to both cytosolic N-terminus and the large luminal loop.Interestingly, TRPML3 is activated by cytosolic but not luminal PI(3)P, but when combined, robust currents are observed (Kim et al., 2022) (Fig. 3).

Synergism: what is it good for?
Given that PI(3,5)P 2 regulates both TPCs and TRPMLs, it is likely central for channel regulation in cells.A complete loss of PI(3,5)P 2 is fatal (Ikonomov et al., 2011) and cellular levels of PI(3,5)P 2 can be enhanced by stimuli, e.g.amino acids, insulin (Bridges et al., 2012) and hyperosmotic shock (Sbrissa & Shisheva, 2005).Its level relative to other stimuli will therefore determine the degree of channel activation.What can be expected from such synergism?
In the case of TPC2, the ratio of PI(3,5)P 2 to NAADP will be the deciding factor governing activity.NAADP is synthesized upon cellular stimulation, e.g. during G-protein coupled receptor signalling (Aley et al., 2013;Gul et al., 2016;Kinnear et al., 2004;Pandey et al., 2009;Soares et al., 2007).It is thought to trigger local Ca 2+ events from the lysosome which are then amplified by the endoplasmic reticulum (ER) into global Ca 2+ signals.The large Ca 2+ signals in response to combined TPC2-A1-N and TPC2-A1-P (Yuan et al., 2022) likely reflects this local to global transition (Favia et al., 2014;Notomi et al., 2017) involving lysosome-ER Ca 2+ crosstalk (Kilpatrick et al., 2013;Patel, 2019).Indeed, recent work using Ca 2+ chelators has dissected these signals into local and global components and found that both depend on the coupling between lysosomal TPC2 and ER IP 3 receptors (Yuan et al., 2024).Such coupling is likely via Ca 2+ -induced Ca 2+ release given that TPC2-A1-N but not TPC2-A1-P sensitized IP 3 receptors (Yuan et al., 2024).Alongside Ca 2+ , lysosomal pH and motility were also regulated in a synergistic manner by the TPC2 agonists (Gerndt et al., 2020;Yuan et al., 2022).The largely reduced motility might aid lysosome-ER communication by 'freezing' lysosomes in proximity to the ER (Jaślan et al., 2023;Yuan et al., 2022).
On the other hand, there was also a synergistic increase in the frequency of local Ca 2+ events evoked by TPC2 (termed 'Tuffs') when the agonists were co-applied compared with alone (Yuan et al., 2022).The unchanged Na + currents by combined NAADP and PI(3,5)P 2 (or their mimetics) (Yuan et al., 2022) suggests Na + -dependent TPC2 activity is intact during co-activation.Thus, processes such as osmotic balance within the endo-lysosomal system which are thought to be driven by PI(3,5)P 2 -activated TPCs (Freeman et al., 2020) are proposed to be unperturbed.Interestingly, TPC1-A1-P but not TPC2-A1-N stimulates exocytosis (Gerndt et al., 2020).This is curious given exocytosis is Ca 2+ -regulated, and there are several reports suggesting that NAADP induces exocytosis (Arndt et al., 2014;Davis et al., 2012).Thus, although TPC2-A1-N increase lysosomal pH similar to NAADP (Gerndt et al., 2020;Morgan & Galione, 2007;Yuan et al., 2022), it may not faithfully mimic all the functions of NAADP.Indeed, recent work has shown that activation of TPC2 by TPC2-A1-N does not require the NAADP-binding proteins that are essential for NAADP activation (Saito et al., 2023).Thus, despite marked convergence on Ca 2+ permeability and synergism with their cognate ligand, the two agonists activate TPC2 through fundamentally different means with a likely direct effect of TPC2-A1-N on the channel.Therefore, it is possible that NAADP-binding proteins may have additional TPC2-independent functional roles that are bypassed by TPC2-A1-N.
Dual regulation of TPC1 and TPC3 by voltage and phosphoinositides likely serves as a mechanism to couple electrical and chemical signalling.TPC1 renders endolysosomes and vacuoles electrically excitable which could potentially have major effects on ionic flux through other ion channels and transporters.The strict co-requirement for channel activation by voltage and PI(3,5)P 2 in animals and Ca 2+ in plants might thus serve as a safeguard for regulating TPC1 activation.Activation of TPC3 at the plasma membrane with a depolarizing stimulus can generate ultra-long action potentials reminiscent of those produced endogenously at fertilization in frog oocytes (Baud et al., 1982;Cang, Aranda, et al., 2014;Cross & Elinson, 1980;Grey et al., 1982).An intriguing feature of both recombinant and native responses is so called 'induction' whereby pre-pulse depolarization markedly increases the currents (Cang, Aranda, et al., 2014;Shimomura & Kubo, 2019).Interestingly, PI(3,4)P 2 , which regulates TPC3 is demonstrably produced during induction (Shimomura & Kubo, 2019).Additionally, an increase in PI-3-kinase activity that produces PI(3,4)P 2 has been noted during oocyte maturation (Hehl et al., 2001;Liu et al., 1995) prior to fertilization.Therefore, TPC3 might contribute to electrical changes initiated by sperm through synergistic interplay between PI(3,4)P 2 and voltage at least in high Na + environments.Of potential relevance here is the functional role of TPC3 in fertilization of starfish oocytes (Ramos et al., 2014).

Synergism: friend or foe in disease?
Activation of TRPML1 and TPCs by approved drugs raises the possibility that lysosomal ion channel activation may contribute to efficacy in the target disorders.TCAs activate TPCs in the tens of micromolar range (Zhang, Chen, Li, et al., 2019).Such concentrations at the clinical doses are unlikely to be achieved in vivo.However, given that TCAs synergize with PI(3,5)P 2 , it is possible that they may activate TPC2 at low concentrations.Similarly, rapamycin activates TRPML1 at relatively high concentrations compared with inhibition of mTOR (Zhang, Chen, Gao, et al., 2019).Again, given synergistic activation of TRPML1 by rapamycin and PI(3,5)P 2 , this might well happen at concentrations achieved in vivo.
Recent work has shown that activating TPC2 with TPC2-A1-P can revert lysosomal defects in MLIV and other lysosomal storage disorders (Prat Castro et al., 2022;Scotto Rosato et al., 2022).This was ascribed to enhanced lysosomal exocytosis because TPC2-A1-P but not TPC2-A1-N had previously been shown to boost lysosomal exocytosis (Gerndt et al., 2020).This is an attractive mechanism as lysosomal exocytosis may serve to clear undigested lysosomal constituents.But the selective effects of TPC2 agonists on lysosomal exocytosis is perplexing because, as mentioned, lysosomal exocytosis is Ca 2+ -dependent but it is TPC2-A1-P (predominantly a 'Na + agonist') which is more efficacious than TPC2-A1-N ('Ca 2+ agonist').One possibility to explain this apparent paradox is that TPC2-A1-P may synergize with endogenous NAADP to boost Ca 2+ signalling and lysosomal exocytosis.Thus, synergistic activation of TPC2 might be beneficial in correcting lysosomal defects in disease.Alternatively, one might postulate that the local Ca 2+ response elicited by TPC2-A1-P application might be sufficient to enhance lysosomal exocytosis.
Conversely, synergistic activation of TPC2 might exacerbate lysosomal defects.TPC2 has been implicated in familial Parkinson's disease caused by mutation of the LRRK2 gene (Gómez-Suaga et al., 2012).Lysosomes are malformed in patient fibroblasts and this phenotype can be reversed by knockdown of TPC2 and antagonizing NAADP action (Hockey et al., 2015).NAADP-evoked Ca 2+ signals are exaggerated in diseased cells suggesting a gain of function but the underlying mechanism is unknown.Given synergism between NAADP and PI(3,5)P 2 , it is possible that the gain of function could manifest through either (or both) ligands.Indeed, endogenous NAADP-evoked Ca 2+ signals are potentiated by overexpression of PIKfyve and inhibited by the PIKfyve blocker, YM201636 (Jha et al., 2014).Consistent with a role for PI(3,5)P 2 in LRRK2-mediated defects is the reversal of the phenotype by the latter although this could be due to a direct blockade of TPC2 by YM201636 (Du et al., 2022).Inhibition of PIKfyve is being pursued in a number of diseases (Burke et al., 2023).As PI(3,5)P 2 effectors, both TPCs and TRPMLs could be considered as possible targets in the actions of PIKfyve inhibitors.

Outlook
TPC2 is highly unusual in switching its permeability in an agonist-dependent manner (Gerndt et al., 2020).Synergism between PI(3,5)P 2 and NAADP and their mimetics manifested as an increase in Ca 2+ currents and fluxes (Yuan et al., 2022).This was selective because Na + currents and fluxes were not changed.Synergism between PI(3,5)P 2 and TCAs manifested as an increase in Na + currents (Zhang, Chen, Li, et al., 2019), but Ca 2+ currents were not measured.Given the peculiar permeability properties of co-liganded TPC2, Ca 2+ measurements are warranted.Indeed, most biophysical studies of TRPML1 also rely on Na + as the permeant ion despite the established Ca 2+ -dependent role in cells.
We have a good understanding of how TRPML is synergistically activated by analyses of PI(3,5)P 2 -bound structures together with MLSA1 (Gan et al., 2022) and temsirolimus (Gan et al., 2022;Zhang, Chen, Gao, et al., 2019).However, this is not the case for TPC2 where the only liganded structure available is that for PI(3,5)P 2 alone (She et al., 2019).As mentioned, channel activation by the PI(3,5)P 2 mimetic TPC2-A1-P is partially dependent on a lysine residue that forms the PI(3,5)P 2 binding site (Gerndt et al., 2020), suggesting that there may be overlap in the binding sites of the two.But it is intriguing that TPC2-A1-P is selective for TPC2 over TPC1 despite very similar PI(3,5)P 2 binding sites in the two isoforms (She et al., 2018(She et al., , 2019)).Of further mechanistic significance is the apparent dependence of the PI(3,5)P 2 site for NAADP but not TPC2-A1-N (Saito et al., 2023).This is intriguing given the remote actions of NAADP.Conversely, a mutation in the voltage sensor of the second domain (R557A) blocked activation of TPC2 by TPC2-A1-N but not NAADP.However, this mutation also blocked the action of PI(3,5)P 2 despite PI(3,5)P 2 binding to the first domain of the channel.Co-activation is thus decidedly complex and will require structures of TPC2 in complex with its associated binding proteins as well as its ligands to understand.In the interim, it will be of interest to further probe the combinatorial roles of the various activators, e.g.does TPC2-A1-N synergize with PI(3,5)P 2 and TPC2-A1-P with NAADP?Are there any functional interactions between TCAs and riluzole and with NAADP?
As discussed, there is little doubt that PI(3,5)P 2 can activate both TPCs and TRPMLs in electrophysiological experiments and that it can synergize with other agonists.But does it necessarily follow that it is a 'true' agonist in the context of other ligand-gated ion channels (for example IP 3 activation of IP 3 receptors)?An increase in PI(3,5)P 2 level to cellular/extracellular stimuli has been reported but only in a limited number of studies, for it is not trivial to probe its changes within the cells (Bridges et al., 2012;Sbrissa & Shisheva, 2005).Could it be that PI(3,5)P 2 is instead a modulator that sets the sensitivity of the channel to other agonists?It is worth stressing that this is the case for the isomer, PI(4,5)P 2 at the plasma membrane which modulates the sensitivity of a number of channels.Indeed, TPC1 is voltage-gated but only in the presence of PI(3,5)P 2 , and channel activation of TRPML1 by PI(3,5)P 2 is weak compared with synthetic activators.In order to record PI(3,5)P 2 currents through TPCs and TRPMLs it is necessary to enlarge lysosomes, most commonly with vacuolin-1 or apilimod.Both are PIKfyve inhibitors.This treatment might strip channels of PI(3,5)P 2 , which is normally bound in its native lysosome.This would uncover PI(3,5)P 2 activation in vitro but potentially mask constitutive activity in cells.More work is required to understand the exact role PI(3,5)P 2 plays in vivo.
Related is the question of whether there are additional 'natural' activators of lysosomal channels awaiting discovery.In the case of TRPML1, reactive oxygen species may serve such a role (Zhang et al., 2016), but whether they synergize with PI(3,5)P 2 remains to be established.Intriguingly, the TRPML1 blocker ML-SI3 blocks channel activation by synthetic activators but not PI(3,5)P 2 (Schmiege et al., 2021).Yet it has functional effects on processes such as autophagy (Scotto Rosato  , 2019;Wang et al., 2015).This suggests the existence of endogenous ML-SA1-like molecules.However, the stereochemistry of ML-SI3 was not defined (Kriegler et al., 2022;Leser et al., 2021) and ML-SI1, a distinct inhibitor, has been reported to block both ML-SA1 and PI(3,5)P 2 -mediated TRPML1 currents (Samie et al., 2013).
To conclude, TPCs and TRPMLs can integrate multiple signals (Table 1).We are beginning to understand this structurally and, importantly, in the context of both health and disease.

Figure 4 .
Figure 4. Synergistic activation of lysosomal TPC2 and TRPML1 by rapamycin Schematic depicting TPC2 and TRPML1 and synergism between PI(3,5)P 2 and rapamycin by indirect inhibition of mTOR or direct channel activation.Figs 1-4 were created with BioRender.com