NMDA Receptor Regulation by Calmodulin and α-Actinin-1

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This hypothesis-driven review clearly integrates functional studies with recent cryo-EM and NMR structures and proposes an appealing model in which α-actinin stabilizes the active receptor whereas Ca²⁺–calmodulin promotes desensitization. The manuscript is readable and the figures are effective. However, its central “cytoplasmic plug” mechanism is presented more confidently than the evidence permits. Before considering the paper for pubblication the following points should be addressed:

1. In my view the proposed “plug” is a hypothesis, not a structurally demonstrated mechanism. The C-terminal domains are absent or unresolved in experimental NMDAR structures. The arrangement shown in Figure 4 is produced by combining AlphaFold3 modeling, superposition of isolated NMR complexes and energy minimization. If I am not wrong, it therefore does not establish that four CaM molecules organize the C0 helices into a stable ring, nor that this ring directly blocks the pore. Indeed, C0 is connected to the peripheral M4 helix, whereas the principal intracellular gate is associated with the pore-forming region. A direct steric “plug” is therefore topologically possible only after substantial, presently unvalidated rearrangement.

Figure 4 should be considered explicitly as a speculative working model; the sentence “blocks the channel pore” should be replaced with an allosteric formulation; possibly the authors should provide model confidence, alternative solutions, simulation details and coordinates and distinguish clearly between experimental structures, docking and prediction. This might be difficult, but would reinforce the paper presentation.

2. I find that this manuscript is neither purely a review nor a sufficiently documented original modeling study: The AlphaFold3/MD analysis is the principal novelty, and the Author Contributions section even lists methodology, software, validation and formal analysis. Yet there is no Methods section describing inputs, restraints, model selection, membrane environment, simulation duration or reproducibility. The authors should either provide full methods and model validation or reposition the paper as a Perspective/Hypothesis Review, with the models used only as illustrations.
3. The proposed mechanism assumes that, at low Ca²⁺, α-actinin occupies GluN1-C0 and CaM is absent, whereas published electrophysiological work supports a “resident” or pre-associated CaM that primes receptors for Ca²⁺-dependent inactivation. The negative binding results obtained with isolated peptides do not necessarily exclude binding in the intact receptor, where membrane context, multivalency and associated proteins may be important. The authors should reconcile these two models rather than dismiss apo-CaM association. Their alternative explanations for the negative ITC and NMR results—zero enthalpy or binding without spectral perturbation—are themselves speculative.
4. The title emphasizes α-actinin-2, but the structural analysis and Figure 4 use ACTN1. This is a significant issue because ACTN2 is the more directly relevant neuronal isoform. The authors should either: change the title to “α-actinins” or “α-actinin-1”; or,demonstrate that the relevant GluN1-binding surfaces and functional properties are sufficiently conserved between ACTN1 and ACTN2. The current manuscript moves too easily from ACTN1 structure to ACTN2 physiology.
5. The model is essentially based on GluN1/GluN2A receptors, but several statements refer to NMDARs generally. CDD is subunit-dependent, and the conclusions should be limited accordingly. Likewise, the abstract implies that mutations disrupting CDD are linked specifically to Alzheimer’s disease, depression, stroke, epilepsy and schizophrenia. This conflates direct genetic effects on CDD with broader associations involving altered NMDAR signalling. The disease statement should be substantially moderated and supported with precise examples.

 

Minor points

Figure 4 should be labelled prominently as a hypothetical model.

The comparison in Figure 2 includes an apo GluN1/GluN2B receptor alongside GluN1/GluN2A structures; subunit-dependent differences should be acknowledged.

D-serine should be mentioned alongside glycine as an NMDAR co-agonist.

The manuscript should distinguish agonist-bound closed/desensitized structures from the specific CaM-dependent desensitized state.

Reference and typographical errors require correction, including “Suprmolecular,” “A.B..,” and the apparent page-range error in reference 47.

 

In conclusion:

This is a clear and potentially valuable hypothesis-driven review integrating recent structural and functional studies of NMDAR regulation by calmodulin and α-actinin. The proposed displacement mechanism is interesting and could stimulate experimental work. However, the central cytoplasmic “plug” model is currently presented with greater mechanistic certainty than is justified by the evidence. It is derived from AlphaFold3 modeling, superposition of isolated NMR complexes and energy minimization, rather than from an experimentally determined intact receptor complex.

The manuscript should clearly separate established findings from structural speculation, provide adequate methodological and confidence information for the computational models, and formulate the proposed CaM/C0 assembly as an allosteric working hypothesis rather than a demonstrated pore-blocking structure. The authors should also reconcile their model with evidence for pre-associated apo-calmodulin, justify the assumed four-CaM stoichiometry, and resolve the inconsistency between the emphasis on ACTN2 and the use of ACTN1 in the structural analysis. Finally, receptor-subtype and disease-related claims should be made more precise.

Author Response

Reviewer 1:  This hypothesis-driven review clearly integrates functional studies with recent cryo-EM and NMR structures and proposes an appealing model in which α-actinin stabilizes the active receptor whereas Ca²⁺–calmodulin promotes desensitization. The manuscript is readable and the figures are effective. However, its central “cytoplasmic plug” mechanism is presented more confidently than the evidence permits. Before considering the paper for pubblication the following points should be addressed:

1. In my view the proposed “plug” is a hypothesis, not a structurally demonstrated mechanism. The C-terminal domains are absent or unresolved in experimental NMDAR structures. The arrangement shown in Figure 4 is produced by combining AlphaFold3 modeling, superposition of isolated NMR complexes and energy minimization. If I am not wrong, it therefore does not establish that four CaM molecules organize the C0 helices into a stable ring, nor that this ring directly blocks the pore. Indeed, C0 is connected to the peripheral M4 helix, whereas the principal intracellular gate is associated with the pore-forming region. A direct steric “plug” is therefore topologically possible only after substantial, presently unvalidated rearrangement. Figure 4 should be considered explicitly as a speculative working model; the sentence “blocks the channel pore” should be replaced with an allosteric formulation; possibly the authors should provide model confidence, alternative solutions, simulation details and coordinates and distinguish clearly between experimental structures, docking and prediction. This might be difficult, but would reinforce the paper presentation.

Response: The AlphaFold3 model of the desensitized NMDAR channel tetramer predicts that each GluN1 M4 helix extends continuously into the GluN1-C0 helix, thus forming one long continuous helix that connects M4 and C0 as described in our recent paper (Bej et al 2026, DOI: 10.1016/j.jbc.2026.111131). The NMR structures of CaM/GluN1-C0 were superimposed onto the GluN1-C0 helices in the channel tetramer model. AlphaFold3 predicted the GluN2A-C0 helices in the tetramer model are arranged concentrically with the GluN1-C0 helices but form steric clashes.  To minimize the steric clashes, we manually adjusted the GluN2A-C0 helices in the tetramer, and the NMR structures of CaM C-lobe/GluN2A-C0 were then superimposed onto the GluN2A-C0 helices in the channel tetramer model to produce the concentric arrangement of C0 helices as shown in Fig. 4B (now called Fig. 5B). So, our structural model does show that four CaM molecules organize the C0 helices into a stable ring (see cyan and yellow structures in Fig. 5B) and a detailed description of this modeling was already presented in our recent publication (Bej et al 2026, DOI: 10.1016/j.jbc.2026.111131). We agree that this ring may not block the channel pore. Therefore, we have removed all mention of the cytosolic plug and revised our description of the structural model as follows: “Our model of the desensitized channel state suggests that Ca2+-CaM binds to helical structures in GluN1-C0 and GluN2A-C0 that come together to form a concentric ring-like structure located underneath the channel pore, which may serve to allosterically stabilize the desensitized channel and facilitate closure of the pore (see cyan and yellow region in Fig. 5B).”  We agree that Figure 4 (now called Figure 5) is a speculative model and therefore changed the title of the figure to include the phrase, “Hypothetical structural model”.  Lastly, we have added in the Figure 5 legend a more detailed description of the methods used to generate the structural models in Figures 5A-B.    

 

 

2. I find that this manuscript is neither purely a review nor a sufficiently documented original modeling study: The AlphaFold3/MD analysis is the principal novelty, and the Author Contributions section even lists methodology, software, validation and formal analysis. Yet there is no Methods section describing inputs, restraints, model selection, membrane environment, simulation duration or reproducibility. The authors should either provide full methods and model validation or reposition the paper as a Perspective/Hypothesis Review, with the models used only as illustrations.

Response:  We have added a more detailed description of the methods used to generated the structural models in the Figure 5 legend. More detailed methods of the structural modeling were described recently by ref 60. Our manuscript is an invited review article and was written as a Perspective/Hypothesis paper, and is not intended to be a modeling study, which is why a specific Methods section is not included.

 

3. The proposed mechanism assumes that, at low Ca²⁺, α-actinin occupies GluN1-C0 and CaM is absent, whereas published electrophysiological work supports a “resident” or pre-associated CaM that primes receptors for Ca²⁺-dependent inactivation. The negative binding results obtained with isolated peptides do not necessarily exclude binding in the intact receptor, where membrane context, multivalency and associated proteins may be important. The authors should reconcile these two models rather than dismiss apo-CaM association. Their alternative explanations for the negative ITC and NMR results—zero enthalpy or binding without spectral perturbation—are themselves speculative.

Response:  The previous electrophysiological work that suggested a “resident” or pre-associated CaM that primes receptors for Ca2+-dependent inactivation does not necessarily indicate that apo-CaM is pre-associated with NMDAR.  It is possible that a half-calcified form of CaM (Ca2+ bound to EF3 and EF4 and not bound to EF1 and EF2) could be pre-associated with NMDAR under resting Ca2+ conditions (100 nM free Ca2+ concentration). It is important to point out that the Ca2+-bound CaM C-lobe binds to GluN2A-C0 with a dissociation constant of ~100 nM. This implies that the CaM C-lobe bound to the C0 peptide will have an apparent Ca2+ affinity in the nanomolar range, which indicates that a significant fraction of the CaM C-lobe bound to NMDAR will have Ca2+ bound under resting conditions ([Ca2+] = 100 nM). This is quite similar to the recent finding that L-type voltage gated channels (CaV1.2) are likely pre-associated with half-calcified CaM rather than apo-CaM (see Bartels et al 2022 DOI: 10.1016/j.jbc.2022.102701). Now, it is true that fluorescence polarization binding studies have shown that apoCaM can bind to a fluorescently labeled GluN1-C0 peptide.  However, this binding might be an artifact of the attached fluorescent tag, because both ITC and NMR studies have shown that apo-CaM does not bind to GluN1-C0 peptide without a fluorescent tag. In this review, we would prefer to not litigate this issue here and therefore we purposefully chose to not speculate about a possible role of apo-CaM binding to NMDAR. Instead, we suggest that future studies should be performed using the EF-hand mutants (CaM1234 etc) that disable Ca2+ binding and find out if the CaM1234 mutant (that exists only as apoCaM) binds to NMDAR tetramer in live cells under physiological conditions as well as to evaluate whether CaM1234 binding to NMDAR might modulate channel opening probability. Also, if half-calcified CaM is pre-associated with NMDAR, then the CaM34 mutant should abolish CaM pre-association but the CaM12 mutant should not. If apo-CaM is pre-associated with NMDAR, then the CaM1234 mutant should remain pre-associated etc. This is all work that needs to be done in the future and it would be pre-mature to discuss this now in our review.

 

4. The title emphasizes α-actinin-2, but the structural analysis and Figure 4 use ACTN1. This is a significant issue because ACTN2 is the more directly relevant neuronal isoform. The authors should either: change the title to “α-actinins” or “α-actinin-1”; or,demonstrate that the relevant GluN1-binding surfaces and functional properties are sufficiently conserved between ACTN1 and ACTN2. The current manuscript moves too easily from ACTN1 structure to ACTN2 physiology.

Response:  The title has been changed to “a-actinin-1”. We chose to model the structure of ACTN1 because previous NMR structural studies were performed on ACTN1 EF-hand domain bound to GluN-C0 peptide and much more is known about the structure of ACTN1 versus ACTN2. We believe that ACTN1 and ACTN2 are sufficiently conserved based on the high sequence similarity for the residues that contact GluN1-C0. However, we agree that our manuscript moves too easily from ACTN1 structure to ACTN2 physiology. We have added the following text on page 7 that makes it clear that much more is known about ACTN2 physiology and that future studies are needed to elucidate and verify the physiological role of ACTN1: “We believe that ACTN1 and ACTN2 are structurally similar based on the high sequence similarity for the residues that contact GluN1-C0. However, it is important to point out that the physiology of NMDAR regulation is better known for ACTN2 rather than ACTN1, and it is possible that ACTN1 and ACTN2 may have different roles in regulating channel function. Future electrophysiology studies are needed to better characterize NMDAR channel regulation by ACTN1.”

 

5. The model is essentially based on GluN1/GluN2A receptors, but several statements refer to NMDARs generally. CDD is subunit-dependent, and the conclusions should be limited accordingly. Likewise, the abstract implies that mutations disrupting CDD are linked specifically to Alzheimer’s disease, depression, stroke, epilepsy and schizophrenia. This conflates direct genetic effects on CDD with broader associations involving altered NMDAR signalling. The disease statement should be substantially moderated and supported with precise examples.

Response: We added a new section (called Section 3) on pages 5-6 that describes differences in the GluN2 subtypes and their effects on CDD. We moderated the statement in the abstract as follows: “Genetic mutations of NMDARs that disrupt CDD affect channel function and may be linked to Alzheimer’s disease, depression, stroke, epilepsy, and schizophrenia.”

 

Minor points

Figure 4 should be labelled prominently as a hypothetical model.

Response: The phrase “Hypothetical structural model” has been added to the title of Fig. 4 (now called Fig. 5).

The comparison in Figure 2 includes an apo GluN1/GluN2B receptor alongside GluN1/GluN2A structures; subunit-dependent differences should be acknowledged.

Response:  The subunit differences are indicated by labels in the figure and mentioned in the figure legend.

D-serine should be mentioned alongside glycine as an NMDAR co-agonist.

Response: D-serine now mentioned as a co-agonist.

The manuscript should distinguish agonist-bound closed/desensitized structures from the specific CaM-dependent desensitized state.

Response: Our hypothesis is that the desensitized channel state (agonist-bound and pore closed) has a similar structure (in the ATD, LBD and TMD) whether CaM is bound or not. Our hypothesis is that CaM binds selectively to the desensitized channel state and stabilizes it relative to the open state (agonist-bound and pore open).

Reference and typographical errors require correction, including “Suprmolecular,” “A.B..,” and the apparent page-range error in reference 47.

Response:  The typographical errors are now corrected.

 

In conclusion:

This is a clear and potentially valuable hypothesis-driven review integrating recent structural and functional studies of NMDAR regulation by calmodulin and α-actinin. The proposed displacement mechanism is interesting and could stimulate experimental work. However, the central cytoplasmic “plug” model is currently presented with greater mechanistic certainty than is justified by the evidence. It is derived from AlphaFold3 modeling, superposition of isolated NMR complexes and energy minimization, rather than from an experimentally determined intact receptor complex.

Response:  We removed all mention of the cytoplasmic “plug” model and replaced it with the following sentences:  “Our model of the desensitized channel state suggests that Ca2+-CaM binds to helical structures in GluN1-C0 and GluN2A-C0 that come together to form a concentric ring-like structure located underneath the channel pore, which may serve to allosterically stabilize the desensitized channel and facilitate closure of the pore (see cyan and yellow region in Fig. 5B).”  We now make it clear that our structural model in Figure 5 is a hypothesis and added the phrase “Hypothetical structural model” in the figure title.

The manuscript should clearly separate established findings from structural speculation, provide adequate methodological and confidence information for the computational models, and formulate the proposed CaM/C0 assembly as an allosteric working hypothesis rather than a demonstrated pore-blocking structure. The authors should also reconcile their model with evidence for pre-associated apo-calmodulin, justify the assumed four-CaM stoichiometry, and resolve the inconsistency between the emphasis on ACTN2 and the use of ACTN1 in the structural analysis. Finally, receptor-subtype and disease-related claims should be made more precise.

Response:  the revised manuscript now more clearly indicates that our structural model is a hypothesis. A more detailed description of the modeling methodology has been added to the Figure 5 legend. We have chosen to not speculate about the functional role of apo-CaM as detailed above. The CaM binding stoichiometry of 4 CaM bound per tetramer is based on previous binding studies (Bej et al 2026, DOI: 10.1016/j.jbc.2026.111131) that determined each C0 peptide binds to one CaM molecule. This suggests that 4 CaM bind per tetramer because each tetramer (GluN1/2A) has four C0 sites. We now provide a clearer rationale for focusing on ACTN1 and added the following sentence: “We believe that ACTN1 and ACTN2 are structurally similar based on the high sequence similarity for the residues that contact GluN1-C0. However, it is important to point out that much more is known about the physiology of ACTN2 regulation of NMDARs and it is possible that ACTN1 and ACTN2 may have different roles in regulating channel function. Future electrophysiology studies are needed to better characterize NMDAR channel regulation by ACTN1”.  Lastly, we added Section 3 that now describes the receptor-subtype differences.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript reviews the regulation of NMDA receptors (NMDARs) by calmodulin (CaM) and α-actinin (ACTN), integrating recent structural findings and proposing a "shift" mechanism whereby ACTN1 would favor the open state of the channel, while Ca²⁺-CaM would promote calcium-dependent desensitization (CDD). The review is clear, well-written, and addresses a relevant biological question. Particularly valuable is the incorporation of recent NMR structures of CaM bound to GluN1-C0 and GluN2A-C0 along with cryo-EM structures of NMDARs. However, I have a few comments:
1. The authors should clearly differentiate between experimental evidence and hypotheses.

2. I suggest reducing overly conclusive statements about the cytosolic "plug."

3. Discuss the limitations of AlphaFold3 and structural modeling.

4. Incorporate a broader view of CTD regulation of NMDARs. 5. Refine the claims related to apo-CaM.

6. Explicitly identify structural figures as predictive models.

Author Response

Reviewer 2:  This manuscript reviews the regulation of NMDA receptors (NMDARs) by calmodulin (CaM) and α-actinin (ACTN), integrating recent structural findings and proposing a "shift" mechanism whereby ACTN1 would favor the open state of the channel, while Ca²⁺-CaM would promote calcium-dependent desensitization (CDD). The review is clear, well-written, and addresses a relevant biological question. Particularly valuable is the incorporation of recent NMR structures of CaM bound to GluN1-C0 and GluN2A-C0 along with cryo-EM structures of NMDARs. However, I have a few comments:

1. The authors should clearly differentiate between experimental evidence and hypotheses.

Response: We added the phrase “Hypothetical structural model” to the title of Fig. 5 to indicate that our structural model is a hypothesis.

2. I suggest reducing overly conclusive statements about the cytosolic "plug."

Response: All statements about the cytosolic “plug” have been removed and replaced with the statement: “Our model of the desensitized channel state suggests that Ca2+-CaM binds to helical structures in GluN1-C0 and GluN2A-C0 that come together to form a concentric ring-like structure located underneath the channel pore, which may serve to allosterically stabilize the desensitized channel and facilitate closure of the pore (see cyan and yellow region in Fig. 5B).”

3. Discuss the limitations of AlphaFold3 and structural modeling.

Response: The AlphaFold3 model of the desensitized NMDAR channel tetramer predicts that each GluN1 M4 helix extends continuously into the C0 helix, thus forming one long continuous helix that connects M4 and C0 in the tetramer.  The NMR structures of CaM/GluN1-C0 were superimposed onto the GluN1-C0 helices in the channel tetramer model.  One limitation to the AlphaFold3 model is that the predicted GluN2A-C0 helices in the tetramer model form steric clashes when it adopts a concentric arrangement with the GluN1-C0 helices. To minimize the steric clashes, we manually adjusted the the GluN2A-C0 helices within the concentric structure, and NMR structures of CaM C-lobe/GluN2A-C0 were then superimposed onto the GluN2A-C0 helices in the channel tetramer model to produce the structure as shown in Fig. 4B (now called Fig. 5B). A more detailed description of the modeling procedure was added to the Figure 5 legend.  

4. Incorporate a broader view of CTD regulation of NMDARs.

Response:  We added a new section (called Section 3 and Figure 4) that describes differences in the CTD of GluN2 subtypes and their effects on CDD.

5. Refine the claims related to apo-CaM.

Response:  Previous electrophysiological work that suggested a “resident” or pre-associated CaM that primes receptors for Ca2+-dependent inactivation does not necessarily indicate that apo-CaM is pre-associated with NMDAR.  It is possible that a half-calcified form of CaM (Ca2+ bound to EF3 and EF4 and not bound to EF1 and EF2) could be pre-associated with NMDAR under resting Ca2+ conditions (100 nM free Ca2+ concentration), similar to the recent finding that L-type voltage gated channels (CaV1.2) are likely pre-associated with half-calcified CaM rather than apo-CaM (see Bartels et al 2022 DOI: 10.1016/j.jbc.2022.102701).  It is true that fluorescence polarization binding studies have shown that apoCaM can bind to fluorescently labeled GluN1-C0 peptide.  However, this binding might be an artifact of the attached fluorescent tag, because both ITC and NMR studies have shown that apo-CaM does not bind to GluN1-C0 peptide without a fluorescent tag.  In this review, we would prefer to not speculate about a possible role of apo-CaM binding to NMDAR. Instead, we suggest that future studies should be performed using the EF-hand mutants (CaM1234 etc) that disable Ca2+ binding and find out if the CaM1234 mutant (exists only as apoCaM) binds to NMDAR tetramer in live cells under physiological conditions as well as to evaluate whether CaM1234 binding to NMDAR might cause an increase in channel opening probability. Also, if half-calcified CaM is pre-associated with NMDAR, then the CaM34 mutant should abolish CaM pre-association but the CaM12 mutant should not. If apo-CaM is pre-associated with NMDAR, then the CaM1234 mutant should remain pre-associated etc. This is all work that needs to be done in the future and it would be pre-mature to discuss this now in our review.

6. Explicitly identify structural figures as predictive models.

Response:  The structural models in Figure 5 are now emphasized to be predictive models and we make it clear that they do not represent experimental structure determinations.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have substantially improved the manuscript and have satisfactorily addressed the principal concern regarding presentation of the proposed cytoplasmic “plug” as an established mechanism. The revised Figure 5 is now clearly identified as hypothetical, and the mechanism is formulated in more appropriate allosteric terms. The distinction between ACTN1 and ACTN2 and the GluN2 subtype dependence of CDD have also been clarified.

A few issues should nevertheless be corrected before acceptance. The authors should clarify the availability and validation status of the computational models, explicitly identify the four-CaM stoichiometry as a modelling assumption rather than an experimentally demonstrated intact-receptor arrangement, acknowledge pre-associated apo- or partially Ca²⁺-loaded CaM as an unresolved alternative basal state, and further moderate the disease-related statement in the abstract. These changes do not require new experiments and could be handled as a minor revision.

Author Response

Reviewer 1:  The authors have substantially improved the manuscript and have satisfactorily addressed the principal concern regarding presentation of the proposed cytoplasmic “plug” as an established mechanism. The revised Figure 5 is now clearly identified as hypothetical, and the mechanism is formulated in more appropriate allosteric terms. The distinction between ACTN1 and ACTN2 and the GluN2 subtype dependence of CDD have also been clarified.

1. A few issues should nevertheless be corrected before acceptance. The authors should clarify the availability and validation status of the computational models, explicitly identify the four-CaM stoichiometry as a modelling assumption rather than an experimentally demonstrated intact-receptor arrangement, acknowledge pre-associated apo- or partially Ca²⁺-loaded CaM as an unresolved alternative basal state, and further moderate the disease-related statement in the abstract. These changes do not require new experiments and could be handled as a minor revision.

Response:  Atomic coordinates of the computational models will be provided to interested readers upon request as stated in the Figure 5 legend. The CaM binding stoichiometry of 4 CaM bound per tetramer is assumed in the modeling and is based on previous binding studies (Bej et al 2026, DOI: 10.1016/j.jbc.2026.111131) that determined each C0 peptide binds to one CaM molecule, which suggests that 4 CaM bind per tetramer because each tetramer (GluN1/2A) has four C0 sites. This sentence is now added to the Figure 5 legend (highlighted in blue text). With regard to the pre-association of CaM in the basal state, we now add a sentence on page 5 (highlighted in blue): “Therefore, in this review, we prefer to not speculate about a functional role for apo-CaM binding to NMDAR and will not present a model of the basal state.”.  We also added the following sentences about half-calcified CaM on page 5 (highlighted in blue): “Alternatively, a half-calcified form of CaM (Ca2+ bound to EF3 and EF4 and not bound to EF1 and EF2) might pre-associate with NMDAR under resting Ca2+ conditions ([Ca2+] = 100 nM). Since the Ca2+-bound CaM C-lobe binds to the GluN1-C0 peptide with a dissociation constant of 100 nM [12], this implies that the CaM C-lobe bound to the C0 peptide should have an apparent Ca2+ affinity in the nanomolar range, suggesting that a significant fraction of the CaM C-lobe bound to NMDAR may have Ca2+ bound under basal conditions. This is analogous to the recent finding that L-type voltage gated Ca2+ channels (CaV1.2) are likely pre-associated with half-calcified CaM rather than apo-CaM (Ref 48, DOI: 10.1016/j.jbc.2022.102701).”  With regard to the disease-related statement in the abstract, we moderated this sentence as “Defects in the regulation of NMDAR function are associated with a spectrum of neurological diseases.”

Reviewer 2 Report

Comments and Suggestions for Authors

The authors responded to the comments.

Author Response

This reviewer did not have any comments.