Electrophysiological and Structural Remodeling of the Atria in a Mouse Model of Troponin-I Mutation Linked Hypertrophic Cardiomyopathy: Implications for Atrial Fibrillation

Round 1

Reviewer 1 Report

In this experimental study, Dr. Lim and colleagues studied the atrial electrophysiological and structural consequences of HCM, that was represented by a transgenic mouse model with Tn-I mutation. They concluded that there were both EP and structural remodeling in the left atria of HCM mice, which could predispose to the initiation of AF. Such remodeling is even more prominent in aged mice. Overall, this is a nicely done study with a potentially insightful findings. However, I have some questions, remarks and comments regarding the study and the manuscript:

• I am surprised by the fact that the ERP of the HCM model did not correspond to the shortening of atrial APD, which is not commonly observed. In young mice, it did not change while in old mice, it prolonged rather than shortened. Please explain why this could happen and add the explanation in the discussion.
• Regarding this sentence "we did not observe statistical differences between the age groups in any of the atrial structural parameters assessed", could this indicate that the electrical remodeling can develop even though the structural remodeling does not progress? In the common situation, electrical remodeling usually denotes the acute phase while the structural indicates the chronic phase. But this observation indicates differently. Please comment on this interesting observation. 
• Previous study (PMID: 32762493) has linked AF with IL-1b and NLRP3 inflammasome. Did the authors also observe an elevation of those inflammatory cytokines in this study, either in circulation or in the tissue? Please perform additional experiments if needed.
• In the method, the authors said "Circulating CRP and IL-6 measurements across groups were consistently higher and lower than the limits of detection respectively and were excluded from the analyses", why did this happen? Please speculate on the cause. 
• Introduction: the authors said that HCM is frequently undiagnosed. Is this true? Please provide the basis of this argument. 
• Introduction and section 3.4: Regarding the translatability and clinical application / relevance of the study, "Identification of the progression of the atrial substrate in HCM may allow for the early prevention of atrial myopathy and AF development." How to do so? Please elaborate, because I think it will be challenging to extract the atrial tissue from living patients to perform those cellular and histological experiments. 
• Introduction: please clarify the functional status of the Gly203Ser cardiac troponin-I transgenic mouse model. Is it loss-of-function or gain-of-function mutation? 
• In the introduction (last paragraph), the authors said that they demonstrated slowed AV conduction conserved with ageing. Which part of the results investigated the AV conduction? Please clarify.
• There is no such thing called "intracellular APD". APD is a cellular parameter, not intracellular. Please replace all of the "intracellular APD" in the text with "cellular APD". 
• Section 2.4: "HCM mice atria demonstrates increased cardiomyocyte hypertrophy"  I think it would be better to specify whether this is eccentric or concentric hypertrophy because it would inform the origin of the remodeling (volume or pressure overload). Please also adapt the rest of the text when explaining this observation (e.g., discussion point 2). 
• In discussion point number 1 and 2, please clarify whether those observations were present in young, old or both models. 
• Regarding this statement "Localized conduction abnormalities have also been observed implicating AF in the setting of atrial ischemia [24,25], obesity [26,27], obstructive sleep apnea [28], hypertension [29,30], cardiomyopathy [31,32], and mitral regurgitation [33].", please also add binge drinking into the list and add this study to the reference (PMID: 32710981) since they also showed similar observations. 
• The authors said that inflammation (inflammatory infiltration) is an important histological finding, but no discussion about this factor. Please discuss briefly the link between inflammation and AF. The authors can benefit from this study by Heijman et al. (PMID: 32762493) and Harada et al. (PMID: 25746525), or other relevant publications. 
• Figure 2D: please add the label and unit of the x- and y axes, also the description of the color bar. 
• Table 1: what is "LA wet weight"? Please clarify or correct it.
• The title needs to include mouse for clarity. I would suggest to add "...in a mouse model of ..."
• Abstract: "Young HCM mice demonstrated significantly shortened atrial action potential duration (APD)"
• Introduction: I think "inherited" is better than "heritable"? 
• Please add the unit of the PCL in Figure 1. It should be (ms).
• Figure 4C, remove the .154 from the bar. Also, Figure 5A, please remove 0.130 from the bar. It is too far from 0.05 (the agreed p-value in this study) so it is not statistically significant, whatever the reason is. The authors can also add more samples to reduce the sample variability (please calculate the required number of samples). 
• Therefore, please refrain from claiming that the circulating TGFb1 was increased in old HCM mice before adding more samples to reduce the sample heterogeneity. Please adapt the statement in point 4 of the discussion and also in Figure 6.
• Please add line numbers to assist the review process.
• Please revise the reference list. The style is different, not using numbering system.
• Please check the writing of the gene vs protein. Gene is always italicized (e.g., MYH7), while protein is not (e.g., MYH7). Please adapt the text necessarily. 

Author Response

Response to Reviewer 1 Comments

In this experimental study, Dr. Lim and colleagues studied the atrial electrophysiological and structural consequences of HCM, that was represented by a transgenic mouse model with Tn-I mutation. They concluded that there were both EP and structural remodeling in the left atria of HCM mice, which could predispose to the initiation of AF. Such remodeling is even more prominent in aged mice. Overall, this is a nicely done study with a potentially insightful findings. However, I have some questions, remarks and comments regarding the study and the manuscript:

 

1)    I am surprised by the fact that the ERP of the HCM model did not correspond to the shortening of atrial APD, which is not commonly observed. In young mice, it did not change while in old mice, it prolonged rather than shortened. Please explain why this could happen and add the explanation in the discussion.

Reply: The two parameters are not necessarily concordant although it is generally assumed to be as discussed in our original methods paper (PMID: 26771118). APD depends on when the repolarizing currents overcome the residual  depolarising currents (i.e., when I_K exceeds I_Ca (or I_Na)). While ERP is dependent on APD, it reflects the recovery from inactivation of the depolarizing current (which we think is mostly driven by I_Ca).

It has been well documented that in certain circumstances, such as ischemia or heterozygotic Scn5a+/− mouse model of Brugada Syndrome  (PMID: 10995401; PMID: 21779762), changes between APD and ERP are known to be dissociated. Similarly, ageing studies have demonstrated conflicting data in ERP changes although APD is commonly shortened (Reviewed in PMID: 31942483, added to discussion line #252). The inconsistency of the ERP findings may reflect either regional heterogeneity in ERP distribution (differences in pacing and recording sites) and non-linear correlation between APD and ERP. In the current setup, atrial ERP (epicardial) is determined as the shortest diastolic interval achieved across the entire tissue (organ), whereas APD (cellular) assessed via an intracellular microelectrode inserted into the endocardial myocardium at an opposing region from the stimulation site (due to pacing-induced electrical artefacts). AP morphology is known to differ depending on the geometry, boundary, direction of impulse propagation (PMID: 21762703), structural heterogeneity also modulates ERP with little effects on APD (PMID: 25291380) and investigative drugs have also been shown to affect one and not the other (PMID: 22370957, PMCID: PMC7416459). We have expanded the methods to state differences between APD recording and stimulation sites (line 400) and have now discussed this as recommended.  

2)    Regarding this sentence "we did not observe statistical differences between the age groups in any of the atrial structural parameters assessed", could this indicate that the electrical remodeling can develop even though the structural remodeling does not progress? In the common situation, electrical remodeling usually denotes the acute phase while the structural indicates the chronic phase. But this observation indicates differently. Please comment on this interesting observation. 

Reply: We agree with the reviewer that electrical remodeling can predate structural remodeling in certain conditions, as has been observed in AF- or tachycardia-induced atrial cardiomyopathy. In a canine model of HF, Li et al (1999) demonstrated the promotion of sustained AF by pathological substrate (atrial fibrosis and conduction heterogeneity) without changes in ERP and conduction velocity. Similarly, HF patients have been known to demonstrate structural remodeling in the absence of AF (PMID: 12952837). This may perhaps be reflective of the young HCM mouse developing an intrinsic pathological substrate due to sarcomeric mutations rather than due to electrical remodeling progression. Ageing itself contributes to atrial electrophysiological changes by modulating either the substrate (e.g. fibrosis) or electrical triggers (e.g. ion channels or calcium dysregulation) (PMID: 28904308), with the effects of underlying cardiomyopathy complications remaining relatively unknown. Lastly, as we have not studied mouse ages beyond 12 months (C57BL/6 mice are able to survive up to 24 months, but the lifespan of HCM mice have not been assessed), we are unable to address further if structural and/or electrical remodeling progression (and they may not necessarily be linearly correlated) may well be exacerbated in HCM mice if we followed them long enough towards the end of their life-span. 

3)    Previous study (PMID: 32762493) has linked AF with IL-1b and NLRP3 inflammasome. Did the authors also observe an elevation of those inflammatory cytokines in this study, either in circulation or in the tissue? Please perform additional experiments if needed.

Reply: We thank the reviewer for his/her comment on other inflammatory cytokines which have been upregulated in AF, and indeed suggested to be linked to HCM patients (PMID: 22447464). We have not assessed these mentioned cytokines due to: (1) limited selection of the cytokine panel available based on the customized multiplex ELISA array and (2) limited blood volume in mice to perform multiple ELISAs individually. Therefore, having to prioritize our customized cytokine panel assessment.

4)    In the method, the authors said "Circulating CRP and IL-6 measurements across groups were consistently higher and lower than the limits of detection respectively and were excluded from the analyses", why did this happen? Please speculate on the cause. 

Reply: We do not know the exact reason(s) for this. However, we note that the Quantibody ELISA array (sandwich-based, glass slide antibody array) was customized to analyze proteins of our interests and may perhaps reflect incompatibility of CRP and IL-6 to be used together with the rest of the custom array, which is a common limitation of multiplex arrays.

5)    Introduction: the authors said that HCM is frequently undiagnosed. Is this true? Please provide the basis of this argument. 

Reply: We thank the reviewer for pointing this out and have amended the sentence with updated references (line 48). Despite increased HCM diagnosis by clinical imaging and cascade genetic testing, only a fraction (~100,000) of the estimated HCM population (~700,000) in the US has been clinically diagnosed (PMID: 27006153), suggesting that the most persons remain undiagnosed throughout their lifetime. 

6)    Introduction and section 3.4: Regarding the translatability and clinical application / relevance of the study, "Identification of the progression of the atrial substrate in HCM may allow for the early prevention of atrial myopathy and AF development." How to do so? Please elaborate, because I think it will be challenging to extract the atrial tissue from living patients to perform those cellular and histological experiments. 

Reply: We thank the reviewer for his query. We acknowledge that translational impact may have been overstated by the claim on preventing HCM-induced AF and have been amended to better reflect our views (introduction and conclusions). 

Early recognition of AF is important and treatment to improve quality of life and clinical outcomes in HCM patients. We propose that it is of interest to investigate the atrial substrates in HCM patients AF to determine if similar atrial electrophysiological and structural remodeling substrates contribute to AF clinically. It may be possible to obtain endomyocardial biopsies from the right atrial septum to study the atrial histological substrate (PMID: 31808221), atrial fibrosis burden can be identified by LGE-MRI (PMID:30898240) and electrophysiological studies and electroanatomical mapping can be safely conducted in living patients with HCM and/or AF, which we believe remains to be studied collectively. AF ablation in HCM patients has been recently shown to be effective for long-term rhythm control (PMID: 33455428). Furthermore, certain atrial substrates such as atrial fibrosis (particularly interstitial fibrosis) may be reversible with drug therapy (PMID: 16141258), although this remains to be studied in HCM models. We propose that the Gly203Ser TnI mutant mouse (and other genetic models of HCM) may serve as investigative models for drug therapies in reversing atrial substrates that may eventuate into better clinical outcomes for HCM patients. 

7)    Introduction: please clarify the functional status of the Gly203Ser cardiac troponin-I transgenic mouse model. Is it loss-of-function or gain-of-function mutation? 

Reply: The model over-expresses the human disease-causing Gly203Ser mutation and has been described to result in disruptions to the normal troponin complex interaction in the mice, producing characteristic features of HCM (PMID: 16950368). This description has been amended into the introduction to justify the model (line 70).

8)    In the introduction (last paragraph), the authors said that they demonstrated slowed AV conduction conserved with ageing. Which part of the results investigated the AV conduction? Please clarify.

Reply: Our prior investigations in surface ECG characteristics and heart rate variability in this mouse model of HCM demonstrated slowed atrial and atrioventricular conduction and depressed HRV (PMID: 26444142). We have amended the sentence for clarity.

9)    There is no such thing called "intracellular APD". APD is a cellular parameter, not intracellular. Please replace all of the "intracellular APD" in the text with "cellular APD". 

Reply: We acknowledge the reviewer’s comment and have removed the term throughout.

10)  Section 2.4: "HCM mice atria demonstrates increased cardiomyocyte hypertrophy"  I think it would be better to specify whether this is eccentric or concentric hypertrophy because it would inform the origin of the remodeling (volume or pressure overload). Please also adapt the rest of the text when explaining this observation (e.g., discussion point 2). 

Reply: We thank the reviewer for his/her comment for a better description of atria hypertrophy in response to pressure/volume overload. Unlike the ventricular myoarchitecture where myocytes are well-organized in uniformed directions with thicker wall structural, it is more challenging to determine eccentric or concentric hypertrophy in the atria where atrial myocytes are less compact with thin walls, especially so in the mouse (where the atrial appendage occupies the majority of the atrial chamber; PMID: 25928887). Concentric and eccentric hypertrophy is more apt to be used when describing the ventricular myocardium. Atrial dilatation/enlargement are often used when describing the atria (organ) by echocardiography. In the current paper, we described increased atrial mass in HCM mice demonstrating organ enlargement (discussion point 1) and atrial-specific cardiomyocyte hypertrophy (by histology) which we feel is a legitimate observation for describing atrial, albeit cellular, structural changes in HCM.

11)  In discussion point number 1 and 2, please clarify whether those observations were present in young, old or both models. 

Reply: We have amended the discussion points in accordance with the reviewer's suggestion.

12)  Regarding this statement "Localized conduction abnormalities have also been observed implicating AF in the setting of atrial ischemia [24,25], obesity [26,27], obstructive sleep apnea [28], hypertension [29,30], cardiomyopathy [31,32], and mitral regurgitation [33].", please also add binge drinking into the list and add this study to the reference (PMID: 32710981) since they also showed similar observations. 

Reply:  We have added the reference as suggested.

13)  The authors said that inflammation (inflammatory infiltration) is an important histological finding, but no discussion about this factor. Please discuss briefly the link between inflammation and AF. The authors can benefit from this study by Heijman et al. (PMID: 32762493) and Harada et al. (PMID: 25746525), or other relevant publications. 

Reply:  We thank the reviewer for his suggestion and have addressed this in the discussion (line 302).

14)  Figure 2D: please add the label and unit of the x- and y axes, also the description of the color bar. 

Reply:  We thank the reviewer for pointing out our oversight. This has been rectified in Figure 2 and the corresponding figure legend.

15) Table 1: what is "LA wet weight"? Please clarify or correct it.

Reply:   This has been corrected to LA/RA mass (table 1).

16)  The title needs to include mouse for clarity. I would suggest to add "...in a mouse model of ..."

Reply: We have added this according to reviewer’s suggestion. 

17)  Abstract: "Young HCM mice demonstrated significantly shortened atrial action potential duration (APD)"

Reply:  We have amended this accordingly. 

18)  Introduction: I think "inherited" is better than "heritable"? 

Reply:  We have amended the text accordingly.

19)  Please add the unit of the PCL in Figure 1. It should be (ms).

Reply:  We have amended this accordingly.

20)  Figure 4C, remove the .154 from the bar. Also, Figure 5A, please remove 0.130 from the bar. It is too far from 0.05 (the agreed p-value in this study) so it is not statistically significant, whatever the reason is. The authors can also add more samples to reduce the sample variability (please calculate the required number of samples). Therefore, please refrain from claiming that the circulating TGFb1 was increased in old HCM mice before adding more samples to reduce the sample heterogeneity. Please adapt the statement in point 4 of the discussion and also in Figure 6.

Reply:  We acknowledge the reviewer’s critique and have removed non-significant values from the relevant figures accordingly. We have amended the statement and figure as advised.

21)  Please add line numbers to assist the review process.

Reply: We note that the line numbering system was in place prior and erroneously lost upon uploading the IJMS word doc template onto the submission system. This has now been rectified.

22)  Please revise the reference list. The style is different, not using numbering system.

Reply: We note that the reference numbering system was in place prior and erroneously lost during uploading of the word doc template onto the submission system. We have rectified the referencing style.

23)  Please check the writing of the gene vs protein. Gene is always italicized (e.g., MYH7), while protein is not (e.g., MYH7). Please adapt the text necessarily. 

Reply: We have corrected the gene vs. protein nomenclature throughout the manuscript.

Reviewer 2 Report

This is an interesting study describing the effects of Gly203Ser cardiac Troponin-I mutation on atrial function and structure in transgenic mice as they age.  The paper is well written, the experiments are appropriate and I agree with the author's conclusions.  I have a few suggestions for improvement.  The authors sometimes present the data as though the reader knows the story already.  The readers need more help in understanding what is presented.  For example, in the introduction some language should be added to indicate that the mouse mutation mirrors one found in human patients, thus justifying the model.  Because so many abbreviations are used, it is important to make sure the reader can find what they mean easily.  In figure 2D the authors present a figure which has both X and Y axes as well as a third dimension indicated by color.  Neither axis is labeled.  The legend does not explain what the figure is showing, only the conclusion.  Of course one does not want to write the Materials and Methods in a figure legend.  One does want to walk the reader briefly through what is being presented. 

Author Response

Response to Reviewer 2 Comments:

This is an interesting study describing the effects of Gly203Ser cardiac Troponin-I mutation on atrial function and structure in transgenic mice as they age.  The paper is well written, the experiments are appropriate and I agree with the author's conclusions.  I have a few suggestions for improvement.  The authors sometimes present the data as though the reader knows the story already.  The readers need more help in understanding what is presented.  For example, in the introduction some language should be added to indicate that the mouse mutation mirrors one found in human patients, thus justifying the model.  Because so many abbreviations are used, it is important to make sure the reader can find what they mean easily.  In figure 2D the authors present a figure which has both X and Y axes as well as a third dimension indicated by color.  Neither axis is labeled.  The legend does not explain what the figure is showing, only the conclusion.  Of course one does not want to write the Materials and Methods in a figure legend.  One does want to walk the reader briefly through what is being presented.

Reply: We are thankful for the reviewer’s suggestions to improve the readability of the manuscript. 

We have amended the introductory sentence describing the mouse model which mirrors the known human mutation to justify the use of the model (line 70). 

We have included abbreviations in parentheses in the first-mention in the 3 sections as instructed by journal: abstract, main-text, and first figure/table to aid reader navigation. The journal does not appear to have a dedicated section for abbreviations in their ‘instructions to authors’ which we are happy to accommodate if necessary.

We thank the reviewer for pointing out the oversight in Figure 2. We have added the axes titles accordingly in Fig 2D and amended the figure legend to describe the axes for improved understanding accordingly.

Reviewer 3 Report

I did not understand why the ”Material and method” section is placed at the end of the article and not immediately after the Introduction and the ”Results” section is placed at the beginning of the article. Ussualy in most articles, especially the experimental ones, the method is described  before the results are presented. 

The section of Conclusions is missing.

Most bibliographic references are old and very old, very few from the last 5 years. 

Author Response

Response to Reviewer 3 Comments:

I did not understand why the ”Material and method” section is placed at the end of the article and not immediately after the Introduction and the ”Results” section is placed at the beginning of the article. Ussualy in most articles, especially the experimental ones, the method is described  before the results are presented. 

Reply: We thank the reviewer for his comments and share the same opinion as the reviewer on the manuscript sections. However, the section structure has been arranged according to specific journal instructions which we are unable to alter.

The section of Conclusions is missing.

Reply: We have now replaced the “Clinical implications for AF” with “Conclusion” subheading in text (line 311) for this section.

Most bibliographic references are old and very old, very few from the last 5 years. 

Reply: We have accepted the reviewer’s critique, prioritized the list of references and amended references that may seem outdated in the field. However, we also think that prioritizing of references should be balanced for appropriateness rather than recency. This is especially so when discussing data/evidence/models first presented and landmark studies (e.g., ‘AF begets AF’ by Wijffels et al. 1995) that explains well-known theoretical frameworks and concepts that remain relevant in today’s time and space. We have now significantly reduced the number of references >10 years from 47 to only 19 references, which we hope demonstrates that we have taken the reviewer’s critique seriously and further improved the manuscript.

Round 2

Reviewer 1 Report

Thank you very much for addressing my comments. I have no further questions or remarks.

Author Response

We thank the reviewer for his/her helpful comments that have greatly improved the manuscript.