Structural Basis of Cold and Menthol Sensing by TRPM8

  1. Hyuk-Joon Lee
  2. Cheon-Gyu Park
  3. Justin Gerald Fedor
  4. Wyatt A. Peele
  5. Mario J. Borgnia
  6. Seok-Yong Lee

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

The transient receptor potential melastatin member 8 (TRPM8) is a polymodal ion channel that senses cold and menthol in mammals. Despite prior structural studies, the mechanisms by which cold and menthol activate TRPM8 remain unresolved. Here, we present cryo-EM structures representing the cold and menthol-dependent activation trajectories, combined with extensive functional analyses. We captured snapshots of cooling-dependent pore opening, which involves dramatic pore rearrangement, suggesting a mechanism for cold sensing. Moreover, menthol binds dynamically to induce channel activation, which may underlie menthol specificity for TRPM8. Finally, we show how TRPM8 integrates multiple modalities (cold and menthol) through overlapping but non-identical pathways, revealing the temperature-specific “cold spot”. These findings enhance our understanding of the molecular basis of physically and chemically induced cool sensation in mammals.

Article activity feed

  1. PREreview

    This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/17960492.

    Comments:

    TRPM8 is a thermosensitive ion channel which responds to multiple stimuli such as chemical and thermal stimuli in a multimodal manner. This ion channel has been studied extensively and is of immense importance in understanding the behavioral adaptation of animals across phyla in response to external thermal stimuli. There is a great deal of interest in the pharmacological field also regarding TRPM8, owing to its implication in pain sensation and therapeutics. Initial structural investigation of TRMP8 involved solving the cryo-EM structures of avian TRPM8, but due to species-specific differences in TRPM8-mediated responses to cold stimuli, recent structural investigation pursued mammalian TRPM8 in complex with various agonists, antagonists, and lipids in various conformational states (PMIDs: 29217583; 30733385; 31488702; 36227998; 37857704) All those studies provided important insights into the function of TRMP8. The structural mechanism of cold-induced open state has remained elusive for a long while, and only recently cryo-EM structures of cold-evoked opening of the TRPM8 channel has been reported (PMID: 40661591). The authors in the current study have done extensive cryo-EM analysis of a series of conformational states of TRPM8 and reported the open state structure induced by cold conditions and augmented by type I and II cold agonists. In addition, the authors have attempted extensive functional characterization to delineate the mechanism of polymodal regulation of TRPM8 channels in response to combined action of thermal and chemical stimuli. The experiments carried out in the work is meticulously designed and well executed, with results supporting most of the claims.  The manuscript overall is well written and clear. I have some questions, additional comments and suggestions for improvement/clarity.

    Major:

    1.    The main question posed in the manuscript was investigating the cold-induced open state, potentiated through cooling agents. An I846V mutant was instrumental in obtaining the open state. I am curious if the authors have considered looking into the difference in channel function behavior of this mutant under given ligand conditions at 20 and 4-8 0C? Since the enhancement of menthol induced activation seems crucial in obtaining the open state, a few lines mentioning the molecular insights into how this mutation potentiates the binding will be helpful, given the structure is already there.

    2.    The dynamic mode of menthol binding is interesting. Please consider adding a panel with all the binding poses of menthol superimposed, which can better guide the sequential rotation/movement through different states. Also, as depicted in Figure 1F, only the C2M state has a near unambiguous pose. Since the associated maps are not yet released, it is hard to follow the dynamic changes in the orientation of menthol in other maps with only the density features shown as of now. 

    3.    AITC, a type II cold agonist for TRPM8 was used in the sample prep throughout, at sub-activating concentration of 0.5 mM and additional 250 µM added during grid vitrification. Can the authors comment on the actual activating concentration and if using higher concentration of AITC could possibly have destabilized the closed state more from the beginning and consequently stabilized the cold induced open state more?

    4.    The authors attributed and have shown Phe979 is not involved in gating and have tried to functionally probe its role, but the Phe979Cys lacks channel activity. Have the authors considered examining the surface expression of this mutant to assess if they are being trafficked to membrane normally?

    Minor:

    1.    Although I am not an expert in electrophysiology, it seems the ephys experiments have been meticulously carried out to validate the ligand binding pocket and gating residues in the pore. I have one query though, in the figure legends, biological replicates are mentioned, but I couldn't find any details in methods as to what the biological replicates refer to, is it the number of independent experiments (starting from different passage of cells, transfection etc.)?It would be helpful for the authors to clarify this in the text.

    2.    It would be helpful if authors can mention the temperature range for activation of TRMP8 in mammals, to help the readers draw a parallel between the physiological temperatures at which TRPM8 functions and the temperature range used in the experiments.    

    3.    Please mention at what cryo-EM map level the density feature for the ligand menthol is contoured. This is also suggested for all other panels in which the cryo-EM density features for ligands/residues are shown to improve clarity.

    4.    Because the relative spatial positioning of the pore residues defines the various conformational states, the density feature for the corresponding residues (Figure 2A) may be shown, perhaps in supplement.

    5.    Figure 1: The panel in Figure E is too dense, and it is hard to judge the quality difference among the different maps from a quick glance. It will be helpful for interpretation if the GSFSC resolutions are provided as inset in the bottom of individual maps. In Panel F, can a marker in terms of distance between two side chain features be included? This will aid in conveying the pore opening through different conformational states to the readers in general. Also, are all the maps contoured at the same level?

    6.    Figure S3: The workflow is too dense. To help, the authors could consider splitting into two. For clarity, please mention how the reference map (EMD-27892) was used in heterogenous classification.

    In other words, at what resolution reference map was low pass filtered? I am assuming other volumes or bad classes were also used to remove junk particles, what were those volumes? Were any ab-initiomodels generated?

    7.    Figure S4: Please mention the number of micrographs for each of the datasets used in image processing, or mention in an additional row in Table S1 to aid reader interpretation.

    8.    Figure S1G, S6A, S9A: Please mention what atoms were used for superposition/overlay for clarity.

    Competing interests

    The authors declare that they have no competing interests.

    Use of Artificial Intelligence (AI)

    The authors declare that they did not use generative AI to come up with new ideas for their review.

  2. PREreview

    This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/17943297.

    This manuscript by Lee et al. presents an extensive structural and functional investigation of the cold- and menthol-activated ion channel TRPM8, providing a series of cryo-EM structures captured in multiple functional states across diverse ligand and temperature conditions. By integrating these structural analyses with mutagenesis and electrophysiology, the authors propose a detailed activation mechanism and delineate distinct conformational pathways underlying cold and menthol activation, including the identification of a putative "coldspot." In addition, the authors argue that a recently reported "semi-swapped" TRPM8 architecture from another preprint (doi: 10.1101/2025.06.06.658377) is incompatible with the structural and functional requirements for pore dilation and S6-mediated gating, based on the evidence presented here.

    The work is ambitious in scope, technically impressive, and addresses long-standing mechanistic questions in the TRPM8 field. Below is a list of comments that may improve clarity and deepen understanding of the manuscript.

    Major points

    1. The I846V mutant was used to obtain the pre-open and open conformations. However, the rationale for how I846V increases menthol-mediated potency and stabilizes these states is not sufficiently discussed. I would encourage the authors to address the role of this mutated residue in their activation model more in the discussion.

    2. The authors propose that Val976 acts as the S6 gate residue, in contrast to the conclusions of a recent preprint by the Julius group (doi: 10.1101/2025.06.06.658377) that suggested Phe979 as the gate. In Fig. 2C, Val976 does not appear to project toward the pore pathway in the closed or pre-open state, whereas Phe979 does. Could the author explain why they think Val976 is the gate residue instead of Phe979?

    3. All cryo-EM structures in this study were reconstructed under C4 symmetry. The authors may consider processing the maps without symmetry imposition to see if there is asymmetric ligand binding status or different subunit conformation.

    4. Except for the C1A state, all other structures, including the open state, were obtained under co-application of cold and menthol. Meanwhile, the model in Fig. 6 shows separate cold-driven and menthol-driven activation mechanisms. Could the authors revise the figure to more accurately reflect the present structural study?

    5. The authors have utilized multiple ligands and temperature conditions to determine the structures across various functional states. It would be helpful if they could provide a summary table that lists all sample preparation conditions along with their functional states.

    Minor Points

    1. The naming and annotation scheme of receptor structures is difficult to follow, as the same ligand label "M" is used to nearly all states without relevant discussion of its necessity. A shorter naming scheme placed earlier in the text may reduce confusion.

    2. Contour levels are missing for the cryo-EM density maps shown in the figures. Providing the values would be helpful for comparing the ligand densities and key residues across different functional states.

    3. Line 1 of page 6, rapamycin was introduced in some reconstructions, but its use and expected impact on stabilizing channel opening are not clearly justified. The authors may consider explaining the intended rationale of this compound.

    4. The abbreviation "PH" appears on page 8 for the first time without definition. It would improve reader understanding if the authors could define this term earlier with related information.

    5. It would be helpful for reader to understand the rationale if the authors could provide Kd or Ki, and fold-excess of the applied RR (Fig. 3F–G).

    6. There are some typos throughout the text that the authors could correct to improve readability:

    Figure 1: the legends for panels C and D are reversed.

    Page 9: "4.6A" should be "4.6 Å."

    Page 14: "Fig. 3 and 6" should be "Figs. 3 and 6."

    Competing interests

    The authors declare that they have no competing interests.

    Use of Artificial Intelligence (AI)

    The authors declare that they did not use generative AI to come up with new ideas for their review.

  3. Version published to 10.1101/2025.09.09.675254 on bioRxiv