Persistent PKA activation redistributes NaV1.5 to the cell surface of adult rat ventricular myocytes

Bernas et al. show that persistent PKA activation increases INa by directing NaV1.5 to the myocyte surface, activating CREB1, upregulating EB1, and reorganizing microtubules. Their study suggests that persistent PKA activation may protect impulse propagation in a chronically stressed heart.

3) Time courses of 8-CPT-cAMP/okadaic acid incubation: the authors convincingly demonstrate in Figure 3 that EB1 expression is maximally increased around 14 hours.However, later figures have different incubation timeframes, such as Figures 5 (12 hours) and Figure 10 (4 hours).The authors frequently use the phrase "Persistent PKA activation" in figure legends and text, which covers a wide time frame.The authors should carefully edit the manuscript such that time frames are defined every time "persistent PKA expression" is stated.
4) The authors should discuss why cell-attached patch clamp was used to compare current magnitudes between cell populations instead of whole-cell.While the authors that show tip resistances were similar, the uncontrollable variable is channel density under the patch.This may explain the variability in maximal current, which is large given the sample sizes.Whole-cell, normalized to call capacitance (size), may have negated this issue and should have generated currents with smoother decay functions.Whole-cell would've also offered the opportunity to more thoroughly investigate channel kinetics, such as inactivation and recovery, which could be implicated if NaV1.5 is phosphorylated.Whole-cell would've increased confidence that cell surface NaV1.5 expression was increasing if macroscopic currents were larger and there were no changes to macroscopic kinetics.These experiments would have provided significant support to increased cell surface expression in myocytes as biotinylation experiments were unsuccessful.a.It's also recommended that the authors choose representative traces for their electrophysiology.Mean peak current is roughly 100 pA in PKA treated but traces appear to be closer to 200 pA.5) Figure 2A/B: it is difficult to determine if these images correspond to the cell surface without a surface marker.Where possible, the authors should include images with cell surface markers (WGA) overlayed to help show reader what is the surface.6) In figure 8B, the authors successfully IP GFP with a molecular weight above 250, suggestive of NaV1.5-GFP.Although it didn't IP with EB1-mRFP, this would suggest the protein was successfully extracted with 1% Triton, correct?If so, the parsimonious conclusion would be these two proteins (GFP-NaV1.5 and EB1-mRFP) do not co-IP in myocytes.Unfortunately, proximity ligation assays and FRAP are useful in showing similar cellular locations, but precise interactions are not described in these assays.Thus, at present, there isn't evidence that EB1 and NaV1.5 physically interact.If the authors wish to test this further they could try different detergents, including those which successfully extracted NaVs for cryo-EM, or use site directed mutagenesis of NaV1.5 in expression systems to better validate the interaction.Otherwise, the authors should temper language in the study discussing a physical interaction.
7) The authors should explain leaky myocytes further in the context of Figure 11A.This result would be stronger in cardiomyocytes, as it was demonstrated in Figure 8 that EB1/NaV1.5 may not interact in cardiomyocytes. Minor: 1) Define PKA in abstract.Consider using full name in title.
2) Abstract: Na should be Na+ 3) Unneeded space between 'aging' and '1'.4) Ca current should be "Ca2+".All ions should be labelled with their charges.5) 8cpt-cAMP should be 8-CPT-cAMP 6) Methods table for antibodies: add another column called "Knockout verified" and state yes or no for each antibody used.7) Grammar: "Fig.2C summaries surface Nav1.5 cluster data".Summary?8) Figure 3A, con, 14 hrs: this myocyte lacks vivid striations and is losing its boxy shape.In my experience, I would consider this an unhealthy myocyte.Do the authors have a better image they could replace it with?Also, the authors should add images for 40 hours, as this is shown in 3B.Reviewer #2 (Comments to the Authors ): Summary: The manuscript by Bernas et al. is a well written account of the effects of persistent PKA activation on Nav1.5's trafficking.The group shows that chronic PKA activation causes a significant increase in peak Nav1.5 amplitude.However, the number of Nav1.5 channels was unchanged while increased trafficking and clustering of the channel at the myocyte surface was observed.The mechanism of altered trafficking is shown to be connected to upregulated EB1 which leads to microtubule reorganization in ventricular myocytes.Specifically, the PKA activation strengthens the interactions between Nav1.5/EB1 and Nav1.5/β-Tub at the intercalated discs of myocytes.The manuscript provides clear evidence to support the conclusion that persistent PKA activation upregulates EB1, which in turn increases Nav1.5 trafficking to the myocyte surface.There are some missing details regarding the pathophysiology of persistent PKA activation in both the introduction and discussion.More information regarding the physiological levels and duration of PKA activation are warranted.Specific Critiques: 1.There are too many figures crowding the main body of the manuscript.Figures 6,7,10,and 12 are not critical for understanding and could be moved to the supplementary information.2. How long and at what levels does PKA activation actually last in a physiological setting?Is persistent activation for 6 -15 hours within the physiological range for humans?Are the levels probed consistent with increased beta-adrenergic tone? 3. The introduction is extremely brief and does not do an adequate job of explaining the background and rationale for this study.In particular, the significance for human health is not well established and really only justified with two sentences: "Persistent βadrenergic receptor stimulation occurs in chronically stressed hearts, such as heart failure and aging.Activation of β-adrenergic downstream effector networks including cAMP-dependent protein kinase (PKA), in concert with other neurohumoral responses, can induce electrical, structural, and metabolic remodeling in the heart, leading to protective but eventually maladaptive consequences."After these, the manuscript lists off a few roles of acute PKA activation, but further elaboration is essentially absent, particularly regarding the potential maladaptive consequences.Are there any specific conditions that can cause the chronic stress that leads to persistent activation?What specific maladaptive consequences can the initially protective βadrenergic responses lead to?How could those consequences affect long-term patient health and outcomes?4. Aside from the significance for human health, the introduction is also missing background information, some of which is included later in the paper but could have been mentioned earlier.In particular, more background on EB1 and Nav1.5's interactions with microtubules would be helpful.The following description (given on page 10) could probably be moved to the introduction to partially accomplish this: "Nav1.5 channels traffic on microtubules, targeted to and anchored at their destinations by protein-protein interactions.Microtubule end binding protein 1 (EB1) is a major microtubule regulator in cardiac myocytes.Furthermore, it has been suggested that EB1 in conjunction with CLIP-associating protein 2 (CLASP2) is involved in targeting Nav1.5 to intercalated discs (ICDs).Importantly, the promoter regions of EB1 genes in human, mouse and rat contain cAMP response elements."5.The first sentence in Abstract has unclear/confusing grammar: "Persistent PKA activation occurs in chronically stressed hearts, that can produce protective or maladaptive effects."It sounds like the chronically stressed hearts are producing the effects independent of the PKA activation.6.On page seven, at the beginning of the second paragraph of the results section, the sentence refers to "PAK activation" when it should say "PKA activation." 7.In the section where CREB1 is first brought up, a brief explanation of the role of phosphorylated CREB1 should be included.The only explanation currently is "CREB1-S133phos enters nuclei and, together with coactivators, modulates gene expression."At the very least, include a sentence justifying why the presence of CREB1-S133phos "confirms PKA activation." Reviewer #3 (Comments to the Authors ): In this manuscript, the authors examined the effects of persistent PKA activation on the expression, distribution and function of Nav1.5 channels in adult rat ventricular myocytes.The authors report that PKA activation increases Na+ current amplitude and Nav1.5 cell surface expression without affecting Nav1.5 total protein expression.Additionally, they show in HEK293 cells that this increased Nav1.5 cell surface expression is prevented by inhibition of trafficking.They also show an increased expression of EB1 protein, an increased co-immunoprecipitation of EB1 and b-tubulin with Nav1.5, and microtubule reorganization upon PKA activation.This study addressed an interesting question regarding ion channel regulation, and the authors used various methods to serve their arguments.Nevertheless, the results present several weaknesses as follow.Also, the data suggest that the regulation of Nav1.5 channels by PKA activation is not associated with a direct post-translational effect on the channel, but that it is rather due to transcriptional regulation of EB1.This mechanism is therefore of interest and should be strengthened by additional experimental demonstrations.If revealed true, it should then be made explicit in the main conclusions so no misinterpretations could be made.Finally, and of major importance, the overall organization of the manuscript and figures is too dense and difficult to follow.The number of figures is too large.Too many technical details are included outside of the materials and methods section.The presentation of data should therefore be revised thoroughly, which will help the reading and increase the impact of the work by better highlighting the various experimental approaches developed.

Major comments:
1-In the introduction, the authors discuss the previously published effects of PKA activation on the cardiac Na+ current, mentioning an increased peak Na+ current by PKA activation.It seems, however, that the effects of PKA activation on the cardiac Nav1.5 channels are not that consistent from one experiment/laboratory to another, and that no consensual conclusions can currently be made.This shall therefore be discussed accordingly and with more caution.Additionally, a lot of literature is available on the potential molecular mechanisms mediating PKA regulation and phosphorylation of Nav1.5, which shall also be discussed in the context of this study.2-Although the effect of PKA activation on CREB1 phosphorylation is clear (Figure 1A), the number of repeated experiments should be given, and data from all repeated experiments should be quantified and illustrated as bar/individual data points graphs.3-The authors mention about previously reported subcellular (cell center compared to cell end) differences in amplitude and gating properties of Na+ currents in ventricular myocytes.Although differences in peak amplitudes were not observed here, do the authors observe differences in voltage-dependence of steady-state inactivation as previously reported in several publications (PMID: 21767519 and 29222390)?Whatever the findings, it would be interesting to show these data as well as the other Na+ current gating (activation, inactivation) properties in a table.Additionally, in Figure 1C, it would be preferable to show, on the top of bars, individual data points in order to show data point distribution in each condition.4-The duration of PKA activation (from 6 to 15 hours) is large and variable from experiments to experiments, and also sometimes within each single set of experiments, which is somehow worrisome as a lot of different (transcriptional and/or posttranslational) effects on channel expression and/or gating could occur during short-, mid-or long-time frames.It would therefore be interesting to distinguish the effects obtained after different durations, at least after short and long duration.This will give the reader a better idea on the time frame of the different mechanisms involved in the regulation of interest.5-The use of adenoviruses expressing GFP-Nav1.5 under the control of the CMV promoter is worrisome.It is indeed wellrecognized that the CMV promoter contains cAMP response elements (CRE, PMID: 25612069) that are responsive to CREB, and therefore to PKA activation.Therefore, one could always wonder whether the observed regulation of Na+ channel is the consequence of CMV promoter activation?Information concerning whether this construct has been used in the experiments or whether the experiments were performed on endogenous cardiac channels (which is preferable in this context) is not always clearly defined and should therefore be made clearer in each figure.Alternatively, or additionally, this could also be defined in the methods section.Additionally, the use of this adenovirus refers to reference #8.Annotation of this reference is not correct.This paper was published in Nature Communications, not in eLife.Also, although this adenovirus has been used in this later study, details concerning its construction and content cannot be found in the publication, which requires that the authors report it here in the present manuscript.6-In Figures 1D and 4A, the authors should show only one representative figure (images of repeated experiments will be found in the Source Data file available from JGP).In addition, the bar graph should show bars and individual data points for each condition, which allow evaluation of variability in each condition, rather than a ratio between the two conditions.7-The authors demonstrate that the PKA regulation of Nav1.5 expression depends on an increased EB1 protein expression.They also suggest that this increased expression may be associated with an increased binding of CREB1 to the EB1 gene promoter region.To further the argument that the mechanism involved is indeed a transcriptional regulation, and not a posttranslational regulation of the channel per se, the authors should strengthen this argument with additional experimental demonstrations.Notably, the authors should at least analyze the expression levels of EB1 (and Nav1.5) transcripts in CON and PKA myocytes.If revealed to be true, this finding should be emphasized as it is the mechanism of interest involved in the regulation of Nav1.5.This will avoid confounding the reader into misinterpretation.8-In Figure 4B, data should be quantified and illustrated.Also, images of b-Tub and deY a-Tub should be merged to show changes in microtubule organization.Also, no images of CON myocytes are shown in Figure 4D.Are there any microtubule loops observed in CON myocytes?These data should also be quantified and illustrated.9-In the calculation of degrees of enrichment on lateral surface and at ICD (Figures 6 and S2), why do the authors divide % pixel contents by % area, rather than simply divide pixel contents by area?10-In paragraph 3 of the Results section, the authors say that Nav1.5 and EB1, but no b-Tub, were more enriched along the lateral surface, and that is consistent with image data presented in Figures 1E, 2 and 4B.However, illustration and especially quantification of these data are not provided in these related Figures.11-In the experiments presented in Figure 8A, the authors conclude an increased interaction of EB1 with Nav1.5 upon PKA activation.However, Nav1.5 expression and therefore immunoprecipitation are increased upon PKA activation, which will simply allow greater co-immunoprecipitation of EB1.Also, importantly, this quantification shall be made from the same blot and blot exposure, which is unclear in the presented figure because of the line drowned in between the CON and PKA conditions.12-In Figure 10, images of Nav1.5 immunoprecipitation, as well as of EB1 immunoprecipitation are required to make sure the IPs worked.Also, it would be more convincing to show a representative co-immunoprecipitation in which the expression level of b-tubulin is similar in CON and PKA conditions.Again, in Figure 10B, the authors should show the plots in both CON and PKA conditions, rather than showing a PKA/CON ratio, so the variability in each condition is illustrated.13-In Figure 11A, because of the fact that PKA activation can increase total expression of GFP-Nav1.5 through transcriptional regulation (CMV promoter activation), expression of both total and cell surface expression of GFP-Nav1.5 should be shown in both representative blot images and in the graph.This is in fact the case in each condition (with and without chloroquine).Again, values and data points in each CON and PKA condition should be shown in the graph.Also, in order to demonstrate an effect of PKA activation on Nav1.5 trafficking, the authors should use other tools (such as brefeldin A for example...) in addition to chloroquine which not only impacts trafficking but most certainly many other cellular mechanisms, such lysosomal degradation, protein recycling, etc...The authors shall also pay attention to the vocabulary used as both trafficking and recycling are used in the text.Each mechanism is specific, involves distinct cellular and molecular pathways, and should therefore not be mixed up.14-Too much speculation on mechanisms are provided from experiments performed in Figures 11B and 12.I would recommend removing these data from the manuscript.

Minor comments:
1-In the abstract, the authors use the term "recycling" to talk about the trafficking of proteins from intracellular compartments to plasma membrane.This is incorrect.They should use one or the other depending on findings (see above).2-In the abstract, the authors cannot say that Nav1.5 "directly" interacts with EB1 and b-tubulin because they cannot exclude the possibility, using co-immnunoprecipitation experiments, that the observed interactions are mediated through intermediary proteins.
3-The description of the cell-attached patch-clamp recording method is confusing.Indeed, the voltage-clamp values should be expressed using the usual, intracellular versus extracellular potentials convention, rather than the cell-attached invert convention.This will help comprehension by non-electrophysiologist readers.4-At the end of paragraph 2 in the Results section, the authors used "(see below)" to discuss findings described underneath the text.This should be changed to the associated

PKA activation upregulates EB1, reorganizes microtubules and promotes Nav1.5 to cardiomyocyte surface
Tytus Bernas, John Seo, Zachary T. Wilson, Isabelle Deschenes, Christiane Morecock, Jinze Liu, and G.-N. Tseng We thank all 3 Reviewers and the Editor for their careful consideration of our manuscript.We have conducted new experiments and added the following data: 1. Forskolin (activating adenylate cyclases and producing native cAMP) mimicked the effect of PKA activation, and a membrane-permeable PKA peptide inhibitor (PKI-14-22 amide, myristoylated) reduced the effect of PKA activation without affecting CON myocytes (new Fig. 14).2. Wheat germ agglutinin 'WGA' was used as a cell membrane marker to validate cell surface Nav1.5 cluster identification (Fig. S1). 3. The effect of persistent PKA activation on CREB1-S133 P band intensity was quantified with statistical analysis (Fig. 1A).4. The effects of persistent PKA activation on the transcript levels of EB1 and Nav1.5 (as part of an RNA-seq experiment) were quantified (Fig. S2). 5. Co-immunoprecipitation experiments in HEK293 cells presented in Fig. 9A and Fig. 11 are strengthened with new data.
We have also made the following changes: 1. Adding more data or analysis requested by the reviewers (details in our specific response to the reviewers).2. Uploading 2 videos of 3D views of nCadherin, Nav1.5, -Tubulin and EB1 in CON and PKA myocytes.This is the most efficient way to illustrate how persistent PKA activation affects the distribution and interrelationship of these native proteins in myocytes' ICDs.3. In response to the critique by both reviewers 2 and 3 (the manuscript was too crowded with figures and technical details), we have removed technical details from the main text and kept them either in the Methods section (if generally applicable to all experiments) or in figure legends (specific for the presented experiments).This change streamlines the main text and makes it much more readable.4. The manuscript title was shortened, in accordance with JPG guidelines. 5. Two new authors were added (responsible for RNA-seq analysis).6. Per JGP policy, we have uploaded all Source data (full, uncropped immunoblot images) for those presented in the figures.
Below is our point-by-point response to the Reviewers' comments.

Reviewer #1
Major: 1.The authors are encouraged to use forskolin to increase endogenous cAMP...In addition, using a PKA inhibitor to block this response would also be supportive as it would focus the pathway through PKA.
Thank you for the suggestions.We have followed them and the new are presented in Fig. 14.

Time course of 8CPT-cAMP/okadaic acid incubation:
The authors convincingly demonstrate in Fig. 3  We have specified the 8CPT-cAMP/okadaic acid incubation times in all figure legends or related text.The incubation time differed between myocyte and HEK experiments for the following reasons.a.In myocytes, the EB1 peaking time of 14 hr shown in Fig. 3 was based on the time points we chose in this experiment.The EB1 upregulation, as well as changes in Nav1.5 and microtubule distribution, is expected to occur over a time window of 4 to >15 hr.Therefore, the incubation time was 12-15 hr in most myocyte experiments.The only exception is Fig. 14 (incubation time 4 hr), because we observed myocyte deterioration in forskolin medium after 4 hr.b.A pilot experiment in HEK cells showed that cell surface GFP-Nav1.5 was increased when the 8CPT-cAMP/Okadaic acid incubation time was prolonged from 1 to 4 hr, but was not further increased after 6 hr of incubation.Therefore, the incubation time was 4-6 hr in HEK experiments.
3. The authors should discuss why cell-attached patch clamp was used to compare current magnitudes between cell populations instead of whole-cell.I used cell-attached patch clamp to quantify myocyte surface Nav channels for 2 reasons.First, this recording conditions provided much better temporal and spatial control of membrane voltage, thus much higher fidelity of Nav channel recording, than whole myocyte recording.Adult rat ventricular myocytes are large (120-180 pF), and the Nav current amplitudes are high (with [Na] o 10 mM, the maximal I Na peak amplitude could be 10 nA), making it extremely challenging to achieve appropriate spatio-temporal control of membrane voltage.Indeed, under optimal conditions in my hand (1 M pipette tip, 98% R s compensation, [Na] o 10 mM, room temperature), I still saw signs of losing spatial control of whole myocyte membrane voltage (abrupt increase in I Na peak amplitude upon small depolarization above the I Na threshold).
Second, a previous study reported that in adult rat ventricular myocytes, cell attached patch clamp recordings showed higher I Na amplitudes at cell end than at cell center (Lin et al., 2011).Therefore, I was interested in knowing whether PKA's effect on myocyte surface Nav1.5 was region-dependent.
While the authors show that the tip resistances were similar, the uncontrollable variable is channel density under the patch.This may explain the variability in maximal current, which is large given the sample sizes.
This is an important critique.Patch clamp recording took time: on average, I could do ~ 5 patches within a one-hour time limit after myocytes were removed from the incubator and placed in a cell chamber on the microscope stage superfused with room temperature Tyrode's solution.This one-hour time limit was set to minimize the confounding factor of dissipating PKA effects on Nav channel distribution.I began patch clamping CON myocytes roughly 4 hr after myocyte isolation, and alternated recordings between CON and PKA myocytes over 10 -12 hr total patch clamp time.This long patch clamp procedure and the inevitable variations in 8CPT-cAMP/okadaic acid incubation time are expected to contribute to the variations in reported maximal I Na peak amplitude, especially in the PKA group.
Whole-cell would've also offered the opportunity to more thoroughly investigate channel kinetics, such as inactivation and recovery, which could be implicated if Nav1.5 is phosphorylated..I agree.In adult rabbit ventricular myocytes, -adrenergic stimulation with 10 uM isoproterenol increased I Na by 2 mechanisms.One was PKA-mediated and involved a modest acceleration of I Na inactivation ( of inactivation shortened from 3.87 to 2.57 ms) (Matsuda, Lee and Shibata, 1992).The fact that I did not observe any change in  of inactivation of maximal I Na between PKA and CON myocytes (Fig. 1C, lower) is consistent with the notion that the increase in I Na amplitude was not related to PKA phosphorylation of the Nav channel.

It's also recommended that the authors choose representative traces for their electrophysiology.
Mean peak current is roughly 100 pA in PKA treated but traces appear to be closer to 200 pA.
The current traces shown in Fig. 1B were chosen because they were obtained on the same day (November 1, 2022), at 5 -6 pm (CON myocytes) and 11 pm -midnight (PKA myocytes).The new experiment presented in Fig. S1 addresses this point.Please note that although wheat germ agglutinin (WGA) staining may appear as a smooth line along cell edges when imaging zplanes inside myocytes, WGA manifests a distinct distribution pattern when viewed on the myocyte surface.This pattern reflects the uneven distribution of cell surface glycosylated proteins.

5.
In figure 8B, the authors successfully IP GFP with a molecular weight above 250, suggestive of NaV1.5-GFP.Although it didn't IP with EB1-mRFP, this would suggest the protein was successfully extracted with 1% Triton, correct?If so, the parsimonious conclusion would be these two proteins (GFP-NaV1.5 and EB1-mRFP) do not co-IP in myocytes.
You are correct that the extremely faint band at > 250 kDa in WCL lanes means we could extract minute amount of GFP-Nav1.5 using 1% Triton.This band was prominent in the IP lanes because we were loading IP samples equivalent to 20 fold WCL protein in these lanes.We believe the low level of GFP-Nav1.5 extraction with 1% Triton prevented its detection in EB1-mRFP immunoprecipitate.This motivated us to use other means to detect GFP-Nav1.5/EB1-mRFPinteractions.
Unfortunately, proximity ligation assays and FRAP are useful in showing similar cellular locations, but precise interactions are not described in these assays... the authors should temper language in the study discussing a physical interaction.

The authors should explain leaky myocytes further in the context of Figure
Leaky myocytes refer to those that are dying or dead in the culture dishes, which inevitably occurs during myocyte isolation and culture.Shown on the right are immunoblot images from a pilot experiment on adult rat ventricular myocytes.The purpose was to see whether we could detect increase in cell surface native Nav1.5 by biotinylation after PKA activation for 15 hr.The -actin image (middle row) clearly showed contamination of cytosolic proteins in the biotinylated fraction.This motivated us to switch to HEK293 cells for the biotinylation experiments.

Minor:
1. Define PKA in abstract.This is done.Consider using full name in title.This cannot be done because the title is limited to 100 characters plus spaces.2. Abstract: Na should be Na+ Done throughout.3. Unneeded space between 'aging ' and '1'. Corrected. 4. Ca current should be "Ca2+".All ions should be labelled with their charges.Done throughout. 5. 8cpt-cAMP should be 8CPT-cAMP Done throughout.6. Methods table for antibodies: add another column called "Knockout verified" and state yes or no for each antibody used.Done.7. Grammar: "Fig.2C summaries surface Nav1.5 cluster data".Corrected. 8. Figure 3A,con,14 hrs: this myocyte lacks vivid striations and is losing its boxy shape.In my experience, I would consider this an unhealthy myocyte.Do the authors have a better image they could replace it with?Also, the authors should add images for 40 hours, as this is shown in 3B.Both are done.9.

Reviewer 2 Specific Critiques:
1.There are too many figures crowding the main body of the manuscript.Figures 6,7,10,and 12 are not critical for understanding and could be moved to the supplementary information.
We have removed technical details from the main text and kept them either in the Methods section (if generally applicable to all experiments) or in figure legends (specific for the presented experiments).This streamlines the main text and makes it much more readable.

How long and at what levels does PKA activation actually last in a physiological setting? Is
persistent activation for 6 -15 hours within the physiological range for humans?Are the levels probed consistent with increased beta-adrenergic tone?
These are important questions.cAMP production and PKA activation following adrenergic stimulation in myocytes are known to be compartmentalized (reviewed in (Agarwal et al., 2022)).
We are taking a reductionist approach to understand how PKA activation for hours can affect Nav1.5 distribution and function in adult ventricular myocytes and to explore potential mechanisms.

The introduction is extremely brief and does not do an adequate job of explaining the background
and rationale for this study.In particular, the significance for human health is not well established.Thank you for these important comments.At the time of writing, we had little information beyond what our experiments had told us.Now we have finished comprehensive transcriptomic and proteomic experiments on myocytes exposed to PKA activation for 15 hr and time control (N=5 each).These data will allow us to speculate protective and potentially maladaptive consequences of PKA activation in myocytes.
4. Aside from the significance for human health, the introduction is also missing background information, some of which is included later in the paper but could have been mentioned earlier.
Thank you for the suggestion.The description of relationship among Nav1.5, microtubules and EB1 is now in the Introduction.

The first sentence in Abstract has unclear/confusing grammar: "Persistent PKA activation occurs in
chronically stressed hearts, that can produce protective or maladaptive effects."It sounds like the chronically stressed hearts are producing the effects independent of the PKA activation.This sentence has be modified to clarify the point: During chronic stress persistent activation of cAMP-dependent protein kinase (PKA) occurs, that can contribute to protective or maladaptive changes in the heart.
6. On page seven, at the beginning of the second paragraph of the results section, the sentence refers to "PAK activation" when it should say "PKA activation."This is corrected.Thank you.

7.
In the section where CREB1 is first brought up, a brief explanation of the role of phosphorylated CREB1 should be included.
We have added this information to the beginning of the Results section: PKA can phosphorylate serine 133 of cAMP response element binding protein 1 (CREB1).Phosphorylated CREB1 (CREB1-S133 P ) then enters nuclei and regulates gene expression.

Reviewer #3
General comments 1.The data suggest that the regulation of Nav1.5 channels by PKA activation is not associated with a direct post-translational effect on the channel, but that it is rather due to transcriptional regulation of EB1.This mechanism is therefore of interest and should be strengthened by additional experimental demonstrations.If revealed true, it should then be made explicit in the main conclusions so no misinterpretations could be made.
We have followed this reviewer's suggestion in making our findings explicit in the Summary statement.This is based on the following data: a. Newly finished RNA-seq experiment showed that EB1 transcript was markedly elevated in PKA vs CON myocytes (fold change = 4.52, FDR = 4.82E-07, Fig. S2).b.Immunoblot data showed EB1 protein was increased in PKA vs CON myocytes (cytosolic fraction, Fig. 4A).c.There was no change in the time constant ( of inactivation of maximal I Na between PKA and CON myocytes (Fig. 1C).This is different from the shortening of  of inactivation in acute PKA mediated increase in I Na in adult rabbit ventricular myocytes (Matsuda, Lee and Shibata, 1992).
2. Of major importance, the overall organization of the manuscript and figures is too dense and difficult to follow.The number of figures is too large.Too many technical details are included outside of the materials and methods section.The presentation of data should therefore be revised thoroughly, which will help the reading and increase the impact of the work by better highlighting the various experimental approaches developed.Thank you for this important comment.We have removed technical details from the main text and kept them either in the Methods section (if generally applicable to all experiments) or in figure legends (specific for the presented experiments).This streamlines the main text and makes it much more readable.

Major:
1.In the introduction, the authors discuss the previously published effects of PKA activation on the cardiac Na+ current, mentioning an increased peak Na+ current by PKA activation.It seems, however, that the effects of PKA activation on the cardiac Nav1.5 channels are not that consistent from one experiment/laboratory to another, and that no consensual conclusions can currently be made.This shall therefore be discussed accordingly and with more caution.
We have added a whole new section in the beginning of 'Discussion' to address this issue.Its subtitle is '1.Effects of persistent PKA activation are distinctly different from those of acute PKA activation on Nav1.5 in cardiac myocytes'.This new section is highlighted by gray shading for easy detection.
Additionally, a lot of literature is available on the potential molecular mechanisms mediating PKA regulation and phosphorylation of Nav1.5, which shall also be discussed in the context of this study.
We have cited an important and comprehensive proteomics study of phosphorylation sites in the Nav1.5 channel (Lorenzini.. Marionneau, J Gen Physiol 2021;153:1-23).1A), the number of repeated experiments should be given, and data from all repeated experiments should be quantified and illustrated as bar/individual data points graphs.

Although the effect of PKA activation on CREB1 phosphorylation is clear (Figure
These new data are summarized in Fig. 1A.

3.
The authors mention about previously reported subcellular (cell center compared to cell end) differences in amplitude and gating properties of Na+ currents in ventricular myocytes.Although differences in peak amplitudes were not observed here, do the authors observe differences in voltage-dependence of steady-state inactivation as previously reported in several publications (PMID: 21767519 (Lin et al., 2011) and 29222390 (Rivaud et al., 2017))?Whatever the findings, it would be interesting to show these data as well as the other Na+ current gating (activation, inactivation) properties in a table.I do not have any kinetic data on Nav channels in PKA or CON myocytes, other than the lack of change in  of inactivation of maximal I Na (Fig. 1C, lower panel).The issues of patch clamp experiments are discussed in details in response to Reviewer 1, point 3, above.
I would humbly state that this study is about the effects of persistent PKA activation on Nav1.5 distribution in adult ventricular myocytes and possible underlying mechanism.
Additionally, in Figure 1C, it would be preferable to show, on the top of bars, individual data points in order to show data point distribution in each condition.
The box plot in Fig. 1C has been switched to a bar graph with individual data points as requested.
4. The duration of PKA activation (from 6 to 15 hours) is large and variable from experiments to experiments, and also sometimes within each single set of experiments, which is somehow worrisome as a lot of different (transcriptional and/or post-translational) effects on channel expression and/or gating could occur during short-, mid-or long-time frames.
The issue of different PKA activation times was also raised by Reviewer 1. Please see our detailed response above (to major concern 2).
5. The use of adenoviruses expressing GFP-Nav1.5 under the control of the CMV promoter is worrisome.It is indeed well-recognized that the CMV promoter contains cAMP response elements (CRE, PMID: 25612069) that are responsive to CREB, and therefore to PKA activation.Therefore, one could always wonder whether the observed regulation of Na+ channel is the consequence of CMV promoter activation?Information concerning whether this construct has been used in the experiments or whether the experiments were performed on endogenous cardiac channels (which is preferable in this context) is not always clearly defined and should therefore be made clearer in each figure.Alternatively, or additionally, this could also be defined in the methods section.
We have added one new paragraph in the 'Technical consideration' of Discussion: The cytomegalovirus (CMV) promoter in the Adv-GFP-Nav1.5 construct has three cAMP response elements.Incubation with 8CPT-cAMP will increase GFP-Nav1.5 protein level, interfering with experiments designed to study Nav1.5 distribution and function.Therefore in all myocyte experiments we focused on native Nav1.5, except those presented in the following figures where an increase in GFP-Nav1.5 protein level was not expected to affect the conclusions: Fig. 9B Additionally, the use of this adenovirus refers to reference #8.Annotation of this reference is not correct.This paper was published in Nature Communications, not in eLife.Also, although this adenovirus has been used in this later study, details concerning its construction and content cannot be found in the publication, which requires that the authors report it here in the present manuscript.
The GFP-Nav1.5 construct originated from Dr. Abriel's lab, and this has been properly noted with permission from Dr. Abriel.It has GFP directly fused to the N-terminus of the Nav1.5 isoform, hH1a (Uniprot ID Q14524).There was no other modification made to hH1a.This information has been added to the Materials and Methods section.Reference #8 was removed.1D and 4A, the authors should show only one representative figure (images of repeated experiments will be found in the Source Data file available from JGP).In addition, the bar graph should show bars and individual data points for each condition, which allow evaluation of variability in each condition, rather than a ratio between the two conditions.

In Figures
The immunoblot experiments summarized in Figs.1A, 1D and 4A were run on different days because they were independent experiments.In each of the experiments, we controlled the amount of protein loading to be equal between CON and PKA samples, all the lanes in the membrane were exposed to the same 1st and 2nd Abs solutions, and subject to ECL simultaneously.However, densitometry readouts would vary among these independent experiments and cannot be directly pooled.To pool these data for statistical analysis, we normalized band intensity in the PKA lane vs that in the CON lane of the same experiment.Therefore, the meaningful data are the PKA : CON ratio.Plotting data as separate CON and PKA conditions will produce CON = 1, and PKA = PKA:CON ratio.
7. The authors demonstrate that the PKA regulation of Nav1.5 expression depends on an increased EB1 protein expression.They also suggest that this increased expression may be associated with an increased binding of CREB1 to the EB1 gene promoter region.To further the argument that the mechanism involved is indeed a transcriptional regulation, and not a post-translational regulation of the channel per se, the authors should strengthen this argument with additional experimental demonstrations.Notably, the authors should at least analyze the expression levels of EB1 (and Nav1.5) transcripts in CON and PKA myocytes.If revealed to be true, this finding should be emphasized as it is the mechanism of interest involved in the regulation of Nav1.5.This will avoid confounding the reader into misinterpretation.
We have finished an RNA-seq experiment on CON and PKA myocytes (n=5 per group) by the VCU Genomics core, and the data have been analyzed by Christiane Morecock and Jinze Liu (Bioinformatics).The transcript of EB1 was markedly elevated while that of Nav1.5 was modestly reduced in PKA myocytes.These new data are presented in Fig. S2.
The observation that the Nav1.5 protein level was not altered in PKA myocytes despite the modest decrease in the mRNA could be explained by the stability of Nav1.5 protein in myocytes: in canine ventricular myocytes the Nav1.5 protein has a 'functional half-life' of ~ 35 hr (Maltsev et al., 2008).8.In Figure 4B, data should be quantified and illustrated.Also, images of b-Tub and deY a-Tub should be merged to show changes in microtubule organization.Also, no images of CON myocytes are shown in Figure 4D.Are there any microtubule loops observed in CON myocytes?These data should also be quantified and illustrated.Microtubule looping is rarely seen in CON myocytes but is a frequent observation in PKA myocytes.Fig. 4D now includes both CON and PKA myocytes to illustrate this point.
I am not aware of any program suitable for quantifying microtubule looping.9.In the calculation of degrees of enrichment on lateral surface and at ICD (Figures 6 and S2), why do the authors divide % pixel contents by % area, rather than simply divide pixel contents by area?If we simply calculate % of pixel contents at ICD and along the lateral surface, the results will be sensitive to the ICD volume and the cell width, respectively.Normalizing % pixel contents by % of cellular area at ICD or cell boundary avoids this issue.10.In paragraph 3 of the Results section, the authors say that Nav1.5 and EB1, but no b-Tub, were more enriched along the lateral surface, and that is consistent with image data presented in Figures  8A, the authors conclude an increased interaction of EB1 with Nav1.5 upon PKA activation.However, Nav1.5 expression and therefore immunoprecipitation are increased upon PKA activation, which will simply allow greater coimmunoprecipitation of EB1.Also, importantly, this quantification shall be made from the same blot and blot exposure, which is unclear in the presented figure because of the line drowned in between the CON and PKA conditions.
The data presented in Fig. 8A (now Fig. 9A) were from the same blot and blot exposure.Full uncropped blot images are uploaded as shown below: 12.In Figure 10, images of Nav1.5 immunoprecipitation, as well as of EB1 immunoprecipitation are required to make sure the IPs worked.Also, it would be more convincing to show a representative co-immunoprecipitation in which the expression level of b-tubulin is similar in CON and PKA conditions.Again, in Figure 10B, the authors should show the plots in both CON and PKA conditions, rather than showing a PKA/CON ratio, so the variability in each condition is illustrated.New data are added to Fig 11 (original Fig. 10) to show that Nav1.5 and EB1 immunoprecipitation were both specific and efficient, and the expression of -tubulin was even between CON and PKA myocytes.New graphs of [+]IP/WCL of both CON and PKA myocytes are included as requested.
13.In Figure 11A, because of the fact that PKA activation can increase total expression of GFP Nav1.5 through transcriptional regulation (CMV promoter activation), expression of both total and cell surface expression of GFP-Nav1.5 should be shown in both representative blot images and in the graph.This is in fact the case in each condition (with and without chloroquine).Again, values and data points in each CON and PKA condition should be shown in the graph.These data are now included as requested by this reviewer in Fig. 13A (original Fig. 11A).
Also, in order to demonstrate an effect of PKA activation on Nav1.5 trafficking, the authors should use other tools (such as brefeldin A for example...) in addition to chloroquine which not only impacts trafficking but most certainly many other cellular mechanisms, such lysosomal degradation, protein recycling, etc... Thank you for the suggestion.We do plan to pursue the pathways of Nav1.5 trafficking in myocytes under control conditions and after PKA activation in future experiments.
The authors shall also pay attention to the vocabulary used as both trafficking and recycling are used in the text.Each mechanism is specific, involves distinct cellular and molecular pathways, and should therefore not be mixed up.
Thank you for the comment.We have revised all 'recycling' to 'trafficking'.
14. Too much speculation on mechanisms are provided from experiments performed in Figures 11B and 12.I would recommend removing these data from the manuscript.We respectfully disagree, and have retained them in the manuscript (now Fig. 12B and Fig. 13).

Minor:
1.In the abstract, the authors use the term "recycling" to talk about the trafficking of proteins from intracellular compartments to plasma membrane.This is incorrect.They should use one or the other depending on findings (see above).Done and tone down.
2. In the abstract, the authors cannot say that Nav1.5 "directly" interacts with EB1 and b-tubulin because they cannot exclude the possibility, using co-immunoprecipitation experiments, that the observed interactions are mediated through intermediary proteins.
We agree that co-immunoprecipitation may occur through intermediate proteins, and thus have toned down the language.
3. The description of the cell-attached patch-clamp recording method is confusing.Indeed, the voltage-clamp values should be expressed using the usual, intracellular versus extracellular potentials convention, rather than the cell-attached invert convention.This will help comprehension by non-electrophysiologist readers.
Since during cell-attached patch recordings the myocytes were superfused with Tyrode's solution that contained 4 mM [K + ], I did not have data on the true membrane potential.4. At the end of paragraph 2 in the Results section, the authors used "(see below)" to discuss findings described underneath the text.This should be changed to the associated Figure.
We apologize for not understanding that this means.
5. The increased % ICD area occupied by clusters and cluster size shall be better illustrated in Figure 6A.In other words, images in Figure 6A are not representative of the quantified data in Figure 6B.We respectfully disagree.The cluster density and size for all proteins (Nav1.5, EB1 and -Tub) in the ICD areas were clearly higher in PKA than CON myocytes, shown in the representative cases of Fig. 7A (original Fig. 6A).

Figures 8C and 8D are not essential in the main Figures and shall be included as Supplemental
Figures.
We kept these two panels because they are important for the points we want to make.
9) Figure 5A: authors encouraged to move Fig.S2 into Fig 5. It's necessary for interpretation of lateral surface and intercalated disk.10) 'We sought 2 alternative strategies'.
Figure.5-The increased % ICD area occupied by clusters and cluster size shall be better illustrated in Figure 6A.In other words, images in Figure 6A are not representative of the quantified data in Figure 6B.6-Figures 8C and 8D are not essential in the main Figures and shall be included as Supplemental Figures.7-In Figure 8E, is it really useful to double the Nav1.5 and EB1 graphs showing fluorescence intensity (end combined with center; and center and end separated)?8-Figure 11C should stand by itself.
Figure 2A/B: it is difficult to determine if these images correspond to the cell surface without a surface marker.Where possible, the authors should include images with cell surface markers (WGA) overlaid to help show reader what is the surface.
Figure 5A: authors encouraged to move Fig.S2 into Fig 5. It's necessary for interpretation of lateral surface and intercalated disk.Done.10. 'We sought 2 alternative strategies'.Corrected.
Fig. 4B now includes merged -tubulin and deY -tubulin signals as requested.

7.
In Figure8E, is it really useful to double the Nav1.5 and EB1 graphs showing fluorescence intensity (end combined with center; and center and end separated)?In Fig.9E(original Fig.8E), the upper row shows superimposed cell end and cell center time courses from the specific experiment depicted above.The lower row shows averaged time courses from multiple experiments, with SE bars and superimposed curve fitting.If I combined the cell end and cell center average time courses, it would be difficult to highlight the time point of 2 min after photobleaching (as a measure of FRAP rates used in the upper right panel).8.Figure 11C should stand by itself.
which covers a wide time frame.The authors should carefully edit the manuscript such that time frames are defined every time "persistent PKA expression" is stated.
that EB1 expression is maximally increased around 14 hours.However, later figures have different incubation timeframes, such as Figures 5 (12 hours [on myocytes]) and Figure 10 (4 hours [on HEK cells]).The authors frequently use the phrase "Persistent PKA activation" in figure legends and text, 1E, 2 and 4B.However, illustration and especially quantification of these data are not provided in these related Figures.Image data presented in Figures1E, 2and 4B illustrate the point, and summary quantification is presented in Fig.6.11.In the experiments presented in Figure