PIP2 modulates TRPC3 activity via TRP helix and S4-S5 linker

The transient receptor potential canonical type 3 (TRPC3) channel plays a pivotal role in regulating neuronal excitability in the brain via its constitutive activity. The channel is intricately regulated by lipids and has previously been demonstrated to be positively modulated by PIP2. Using molecular dynamics simulations and patch clamp techniques, we reveal that PIP2 predominantly interacts with TRPC3 at the L3 lipid binding site, located at the intersection of pre-S1 and S1 helices. We demonstrate that PIP2 sensing involves a multistep mechanism that propagates from L3 to the pore domain via a salt bridge between the TRP helix and S4-S5 linker. Notably, we find that both stimulated and constitutive TRPC3 activity require PIP2. These structural insights into the function of TRPC3 are invaluable for understanding the role of the TRPC subfamily in health and disease, in particular for cardiovascular diseases, in which TRPC3 channels play a major role.

3) It would strengthen the argument that mutagenesis of the residues involved in coordinating PIP2 in the identified novel site decrease affinity for PIP2.The PH domain experiments are a good control but inside-out macropatches and concentration-response experiments of each of the single mutants of the four residues involved would provide strong evidence of their individual contributions to the overall PIP2 affinity.
Minor points 4) Page 4, lines 133-137: "Moreover, PIP2 rapidly binds to TRPC3, and at higher concentrations …which showed that Chol typically binds to TRPC3 at regions inaccessible to phospholipids20".Please show in the supplementary figures the modeling data to substantiate the claims that 10% PIP2 displaces PC, PE and PS. 5) Page 5, lines 154-157: "Moreover, in the presence of the re-entrant loop, the four…enhances it".It would make the argument stronger if the authors provided as a Supp.6) Figure 5e and f, I found confusing referring to the Double Mutant (I would indicated as "DM") as Control.Please replace "Control" with "DM".7) Page 4, line 130: change "can" to "could" in "…, during which PIP2 could laterally diffuse…" 8) In Figure 2h, I, it would help the readers if there was a key (e.g., a line schematic) to put the residue numbers (350…700) in the context of the structurally important domains (e.g., pre-S1, S1, S4-S5 linker, pore domain, re-entrant loop, TRP helix, etc.) 9) Page 6, line 208-209, delete "at" and correct typo "farther" rather than "further" in "…bind here, the majority bind at a site close to the re-entrant loop.A smaller portion binds at a second discrete site slightly farther from the re-entrant loop".10) Page 7, line 230: correct the typo from "R374A" to "R380A" in "…directly interacted with PIP2: R374A, K377A, R380A, and K385A (Fig. 3)".11) Page 8, line 260; Page 9, line 302; Page 10, line 361, treat "data" as a plural word "these data" and accordingly the verb tense in the sentence.
12) What the authors refer to as "beige" in the figures looks "orange" to my eyes.
13) In general, I would be "more" humble in statements that imply that we now understand how this novel site controls gating.We do appreciate that this is likely an important determinant of activity but we still do not understand how.Please tone down the way lines 472-479 are phrased to stress that the results are consistent with the hypothesis that the distant PIP2 site couples to key elements involved in channel gating to control the activity of this channel.
The study by Clarke et al. integrates in-silico prediction with experimental validation to identify a significant PIP2 binding site at the pre-S1/S1 nexus of the TRPC3 channel, termed the L3 binding site.This research demonstrates that PIP2 positively modulates TRPC3 function through its interaction at the L3 site.While the study suggests the possibility of dual regulation of TRPC3 by PIP2 at both L3 and L1 sites, similar to the dual role of PIP2 in TRPV1, the simulation data indicate a lower affinity for the L1 site.A key insight provided by this study is the elucidation of a step-by-step mechanism showing how PIP2 binding at the L3 site induces conformational changes that affect the gating structures in the pore domain.The research provides strong evidence of a coordinated movement involving the TRP helix and the S4-S5 linker in this process.Additionally, it reveals the critical role of residue K385, not only as a lipid binding site but also in transmitting the PIP2 binding signal to the TRP helix.Single-channel experiments further support these findings, showing a significant reduction in open channel probability under PIP2 depleted conditions and in the presence of a K385A mutation.Molecular simulations are rigorously carried out appropriately analyzed.The authors' conclusions are well supported by the data and the manuscript is overall clear and detailed enough to ensure reproducibility of the results.I recommend publication of the manuscript after a minor revision: • The description of the data shown in Supplementary Figure 2 is not sufficiently detailed.The results are relevant as they support the notion that the majority of PIP2 molecules bind to L3.However, the clustering approach used to obtain this result is practically not described.A paragraph with a long and detailed description of these results would improve the clarity of the paper.
Reviewer #3 (Remarks to the Author): As an essential component of biomembrane, PIP2 is critical for regulation of function, expression, and subcellular localization of diverse membrane proteins.Clarke et al. study the structural biological basis of the modulatory action of PIP2 on the transient receptor potential canonical type 3 (TRPC3) channel, which is recognized to play pivotal roles in regulating neuronal excitability in the brain and pathogenesis of cardiovascular disorders.By employing molecular dynamics simulations, site-directed mutagenesis and patch clamp techniques (both single-channel and whole-cell), authors beautifully identified structural components required for the modulatory action of PIP2 on TRPC3 protein.The experiments are well executed and discussion is sound.My comments are as follows.
1) The results firstly reveal the L3 lipid binding site as a predominant interact site of PIP2 in TRPC3.This finding is comparable to the previous identification of the pre-S1 shoulder as the PIP2 interaction site in the closely related homologue TRPC6 through exhaustive screening cytoplasmic positively charged amino acid residue using mutant analysis with whole-cell patch clamp recording and voltage-dependent phosphatase that hydrolyzes PIP2 in situ (Mori, MX et al. Sci. Rep. 12:10766 (2022)).It can be viewed that the previous work nicely provides the present work with sort of a proof of concept for the molecular dynamics approach taken.An achievement that should be truly appreciated is the identification of multistep propagation of PIP2-induced structural modification from L3 to the pore domain via a salt bridge between the TRP helix and S4-S5 linker.This is only achievable by the authors' molecular dynamics approach.In this sense, authors may consider further discussion that attempts a feedback of the knowledge to interpretation of the phenotypes of the TRPC6 mutants reported in the above previous work, which greatly enhances but never devalues the significance of the present work.
2) Given that TRPC3 plays major roles in certain cardiovascular disease, it is better that this is also mentioned in the Abstract.
3) Page 3, lines 83-87: It is curious whether the influence of masking of negative charges of PIP2, which may concentrate cations nearby channels, by the PH domain (of what?) on TRPC3 currents need to be considered.4) Page 11, lines 393-400.When the single mutations of R572E and E684R, both of which significantly impairs the activity and modulation of TRPC3, are combined as the double mutation R572E/E684R, the mutant showed restored function.This exciting data should be corroborated by structural analysis using molecular dynamics and the obtained data can be presented as a figure in the revised manuscript.
3) It would strengthen the argument that mutagenesis of the residues involved in coordinating PIP 2 in the identified novel site decrease affinity for PIP 2 .The PH domain experiments are a good control but inside-out macropatches and concentration response experiments of each of the single mutants of the four residues involved would provide strong evidence of their individual contributions to the overall PIP 2 affinity.

Response:
We thank the reviewer for this valuable suggestion.Accordingly, we conducted inside-out patch clamp recordings from TRPC3, K377A, R380A, and K385A to compare the behaviour of TRPC3 WT with that of mutant channels during exposure to the PIP 2 scavenger PalPIP2 (results are summarised in Suppl.Figures 7 and 8).TRPC3 WT, as well as K377A and R380A, exhibited a more pronounced decline (run-down) in channel activity (N*Po) in the presence of PalPIP2 (10 µM; Suppl.Figure 8a) as compared to controls (DMSO; Suppl.Figure 8b).Importantly, we observed stable activity (no significant decline in N*Po) for the K385A mutant in the same setting (Suppl.Figure 8a-b).Our whole-cell recordings from TRPC3overexpressing HEK293 cells revealed that PalPIP2 concentrations of >10 µM are required to affect the activity of TRPC3 WT channels (Suppl.Figure 7a).Using this PalPIP2 concentration the inside-out recording configuration, we observed a slightly but insignificantly faster rundown in K377A or R380A mutants as compared to WT (Figure below) This demonstrates the minor if any impact of these individual charges for the affinity of L3 to bind PIP 2 .The L3 site is a highly positively charged pocket that includes several positively charged residues.Consequently, we assume that the overall positive change density in L3 remains sufficiently high for PIP 2 binding.Specifically, residues adjacent to K377 and R380 can compensate.Similar redundancy has been observed in other membrane proteins regulated by PIP 2 , such as the serotonin transporter SERT 1 .Most importantly, our results obtained from the inside-out patch clamp approach clearly confirm that K385A is essential for productive sensing of PIP 2 binding to L3 by the channel (Figure 4 and Suppl.Figure 8).We excluded the R374A residue from the experiments due to its lowest binding probability for PIP 2 according to the results from our MD simulations (Figure 2f). 1 1 1.Buchmayer, F. et al.Amphetamine actions at the serotonin transporter rely on the availability of phosphatidylinositol-4,5-bisphosphate.Proc.Natl.Acad.Sci.U.S. A. 110, 11642-11647 (2013).
In this figure we show a summary of the linear regression plotting of the N*Po rundown from inside-out patches in presence of PalPIP2 (combined data from Suppl. Figure 7c).Data is presented as Mean±SEM.Statistical significance was assessed using ANOVA followed by a two-tailed multiple t test with Bonferroni correction.**P< 0.01, ns = not significant 4) Page 4, lines 133-137: "Moreover, PIP 2 rapidly binds to TRPC3, and at higher concentrations …which showed that Chol typically binds to TRPC3 at regions inaccessible to phospholipids20".Please show in the supplementary figures the modeling data to substantiate the claims that 10% PIP 2 displaces PC, PE and PS.

Response:
We have now included new modelling data as Suppl.Figure 2 to confirm this claim.Critically, this data shows that only PS is significantly reduced in the presence of 10% PIP 2 .As a result, we have now amended our sentence in the main text of the manuscript to read: "Moreover, PIP 2 rapidly binds to TRPC3, and at higher concentrations (10%, Fig. 1e) it displaces the negatively charged lipid phosphatidylserine (PS) (Suppl.Figure 2).PIP 2 does not appear to displace phosphatidylcholine (PC) or phosphatidylethanolamine (PE) (Suppl.Figure 2).It also does not appear to compete with cholesterol (Chol) for binding to TRPC3".

5)
Page 5, lines 154-157: "Moreover, in the presence of the re-entrant loop, the four…enhances it".It would make the argument stronger if the authors provided as a Supp.

Response:
We have now created a coarse-grained double mutant (D698K/D699K) and ran 5 independent simulations, each 20 µs long simulation to explore the effect of the double mutant on PIP 2 recruitment to the L3 site.This data is now included in Suppl.Figure 9.

Maximum occupancy and contact duration analysis show that the L3 site (composed of the pre-S1, S1, and re-entrant loop) is still associated with the highest maximum occupancy and contact duration values. Density map analysis confirms that PIP 2 accumulates at the L3.
There appear to be minor changes associated with the mutation, such as a slight increase in PIP 2 binding to the S4-S5 linker; however, given that the majority of PIP 2 binds to the L3 as in WT, we believe our data is consistent with the experimental data which shows no change in net current density between WT and mutant.
The new data is included in Suppl.Figure 9 and we have also included the sentence "Moreover, a coarse-grained computational mutant D698K/D699K accumulated PIP 2 at the L3 site at levels comparable to the WT protein (Suppl.Fig. 9d-f)." in our results section.
The study by Clarke et al. integrates in-silico prediction with experimental validation to identify a significant PIP 2 binding site at the pre-S1/S1 nexus of the TRPC3 channel, termed the L3 binding site.This research demonstrates that PIP 2 positively modulates TRPC3 function through its interaction at the L3 site.While the study suggests the possibility of dual regulation of TRPC3 by PIP 2 at both L3 and L1 sites, similar to the dual role of PIP 2 in TRPV1, the simulation data indicate a lower affinity for the L1 site.A key insight provided by this study is the elucidation of a step-by-step mechanism showing how PIP 2 binding at the L3 site induces conformational changes that affect the gating structures in the pore domain.The research provides strong evidence of a coordinated movement involving the TRP helix and the S4-S5 linker in this process.Additionally, it reveals the critical role of residue K385, not only as a lipid binding site but also in transmitting the PIP 2 binding signal to the TRP helix.Single-channel experiments further support these findings, showing a significant reduction in open channel probability under PIP 2 depleted conditions and in the presence of a K385A mutation.Molecular simulations are rigorously carried out appropriately analyzed.The authors' conclusions are well supported by the data and the manuscript is overall clear and detailed enough to ensure reproducibility of the results.I recommend publication of the manuscript after a minor revision: • The description of the data shown in Supplementary Figure 2 is not sufficiently detailed.The results are relevant as they support the notion that the majority of PIP 2 molecules bind to L3.However, the clustering approach used to obtain this result is practically not described.A paragraph with a long and detailed description of these results would improve the clarity of the paper.

Response:
We agree that Suppl.Figure 2 (in the revised version it is Suppl.Figure 3) was poorly explained in the original manuscript.We have now included an expanded explanation in the Figure legend and also included an additional section in the Methods section.We have also amended the figure slightly to improve the clarity.We thank the reviewer for the comments.

Reviewer #3 (Remarks to the Author):
As an essential component of biomembrane, PIP 2 is critical for regulation of function, expression, and subcellular localization of diverse membrane proteins.Clarke et al. study the structural biological basis of the modulatory action of PIP 2 on the transient receptor potential canonical type 3 (TRPC3) channel, which is recognized to play pivotal roles in regulating neuronal excitability in the brain and pathogenesis of cardiovascular disorders.By employing molecular dynamics simulations, site-directed mutagenesis and patch clamp techniques (both single-channel and whole-cell), authors beautifully identified structural components required the modulatory action of PIP 2 on TRPC3 protein.The experiments are well executed and discussion is sound.My comments are as follows.
1) The results firstly reveal the L3 lipid binding site as a predominant interact site of PIP 2 in TRPC3.This finding is comparable to the previous identification of the pre-S1 shoulder as the PIP 2 interaction site in the closely related homologue TRPC6 through exhaustive screening cytoplasmic positively charged amino acid residue using mutant analysis with whole-cell patch clamp recording and voltage-dependent phosphatase that hydrolyzes PIP 2 in situ (Mori, MX et al. Sci. Rep. 12:10766 (2022)).It can be viewed that the previous work nicely provides the present work with sort of a proof of concept for the molecular dynamics approach taken.An achievement that should be truly appreciated is the identification of multistep propagation of PIP 2 -induced structural modification from L3 to the pore domain via a salt bridge between the TRP helix and S4-S5 linker.This is only achievable by the authors' molecular dynamics approach.In this sense, authors may consider further discussion that attempts a feedback of the knowledge to interpretation of the phenotypes of the TRPC6 mutants reported in the above previous work, which greatly enhances but never devalues the significance of the present work.

Response:
We thank the reviewer for the suggestion.The work from Mori et al. nicely complements our present study and we are happy to include a more thorough discussion of this issue: "This localisation of a regulatory PIP 2 binding is consistent with a site recently proposed for the highly homologous TRPC6 channel (Mori et al, 2022).Significantly, in this study, a K442Q mutation was found to result in a decrease of PIP 2 binding affinity about 5to 8-fold as compared to WT TRPC6 (Mori et al, 2022).K442 is identical to the K385 residue found in TRPC3, which we here propose to function both as a PIP 2 binding residue and a transducer of the PIP 2 binding signal.Interestingly, the Mori et al study (2022), also proposes a role for residues of the distal TRP box, such as K781 and K782, in PIP 2 sensing.Due to lack of structural information of this region in the cryo-EM structures, we have been unable to explore whether this region plays a role in PIP 2 sensing by TRPC3.However, with the advent of de novo structure prediction methods such as AlphaFold, this may successfully be addressed by future research." 2) Given that TRPC3 plays major roles in certain cardiovascular disease, it is better that this is also mentioned in the Abstract.
Fig. comparable results to Fig. 2g-i on maximal occupancy and contact duration of the D698K/D699K DM.
Fig. comparable results to Fig. 2g-i on maximal occupancy and contact duration of the D698K/D699K DM.