In Silico Analyses Suggest That Exercise-Induced Irisin-Mediated Neuroprotection Supports Non-Pharmacological Preventive Strategies for Alzheimer’s Disease in Public Health

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

I am attaching my suggestions for improving the article.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

 

We thank you for the time and attention dedicated to evaluating our manuscript, as well as for the suggestions provided, which have significantly contributed to improving the work. We conducted a thorough revision of the manuscript, seeking to incorporate, in the most appropriate manner possible, all the considerations made by the reviewers.

 

In general, the main changes made were:

• Comprehensive textual revision, correcting inconsistencies, improving the writing, and enhancing clarity in the description of the tables and methodology;
• Inclusion, in the methodology section, of the justification for the waiver of ethics committee submission, considering the exclusive use of public and anonymized data, in accordance with current regulations;
• Complete reformulation of the discussion section, aiming to more consistently align the study findings, clearly distinguishing between evidence, hypotheses, and implications.

 

Regarding your specific comments:

 

1. Provide more detailed methodological descriptions for computer analyses (software, parameters, validation procedures). 

- The versions of the software used were included, as well as the threshold criteria adopted at each analytical step, aiming to ensure greater transparency and reproducibility. We clarify that the validation of the structural models followed widely consolidated parameters in the literature for this type of approach, including metrics such as pLDDT, GDT-HA, RMSD, MolProbity score, clashscore, and Ramachandran plot analysis, as described in the methodology.

 

2. Moderate conclusions drawn from in silico predictions. 

- The discussion section was thoroughly revised with the purpose of improving the coherence between the results obtained and their interpretations, establishing a clearer distinction between directly observed evidence and hypothetical inferences.

 

3. Clarify how epidemiological findings mechanistically relate to the irisin pathway.

- The discussion was also adjusted to more precisely explain the role of the interaction analyses, particularly in the context of molecular dynamics simulations. We sought to highlight that the results obtained support the stability of the complex over time, which is consistent with the possibility that the interaction is maintained for a sufficient period to trigger signal transduction events associated with the neuroprotective effects identified in the network analyses. This approach was particularly relevant given the absence of the RGD motif in the irisin structure, a typical feature of integrin ligands. In this regard, the results are consistent with evidence from the literature indicating that irisin interacts with the αV/β5 receptor at an alternative site to the classical RGD binding pocket.

 

4. Reinforce the section on limitations, with particular reference to the observational nature of the epidemiological data and the lack of experimental validation.

- We sought to more rigorously articulate the study limitations, as well as to align the interpretations with the specific capabilities of each methodological approach employed. Considerations were included, for example, regarding the absence of resolved experimental structures of the αV/β5 integrin complex, in addition to the presence of potential confounding factors in the epidemiological analysis.

 

5. The figure captions should be more clearly separated from the main text and formatted in a smaller font size to improve readability and to allow readers to clearly distinguish between the figure legends and the main manuscript text. 

- We clarify that the manuscript was prepared according to the LaTeX template provided by the journal. We acknowledge, however, that there were occasional difficulties in standardizing certain elements and we have sought to improve these aspects in the present revised version.

We remain at your disposal to address any remaining questions, as well as to make any improvements you deem necessary.

Reviewer 2 Report

Comments and Suggestions for Authors

1. Line 45, please revise “0,05” as “0.05” by using the dot before decimals.
2. Table 1, please specify that the proteins were represented by gene names in the table. Please add the explanation to the table title. Please adjust the format of table 1 to make it more organized.
3. In line 274 “US$ 650 million”, the US and $ is kind of redundancy, please consider deleting one of them. The same for “US$ 11 million” in line 275.
4. Did authors find the difference of irisin’ correlation with AD between men and women, and different APOE gene types?
5. What’s the suggestion of authors for people to prevent AD? E.g. doing exercise to produce more irisin to avoid AD. If yes, what’s the potential concentration of irisin would be helpful for preventing AD?
6. In Table 2, are there any units for the parameters? If this, please add the units.
7. Do other people from different countries show the same correlation of irisin and AD? E.g. North America, Asian etc.
8. The conclusion section could be revised to be more concise and to focus on the key results of the manuscript.

Author Response

Dear Reviewer,

 

We thank you for the time and attention dedicated to evaluating our manuscript, as well as for the suggestions provided, which have significantly contributed to improving the work. We conducted a thorough revision of the manuscript, seeking to incorporate, in the most appropriate manner possible, all the considerations made by the reviewers.

 

In general, the main changes made were:

• Comprehensive textual revision, correcting inconsistencies, improving the writing, and enhancing clarity in the description of the tables and methodology;
• Inclusion, in the methodology section, of the justification for the waiver of ethics committee submission, considering the exclusive use of public and anonymized data, in accordance with current regulations;
• Complete reformulation of the discussion section, aiming to more consistently align the study findings, clearly distinguishing between evidence, hypotheses, and implications.

 

Regarding your specific comments:

 

1. Line 45, please revise “0,05” as “0.05” by using the dot before decimals.

Corrected.

 

2. Table 1, please specify that the proteins were represented by gene names in the table. Please add the explanation to the table title. Please adjust the format of table 1 to make it more organized.

The specifications were added to the table legend, as requested.

 

3. In line 274 “US$ 650 million”, the US and $ is kind of redundancy, please consider deleting one of them. The same for “US$ 11 million” in line 275.

Corrected.

 

4. Did authors find the difference of irisin’ correlation with AD between men and women, and different APOE gene types?

According to DiCarlo et al. (2024, DOI: https://doi.org/10.1002/ana.26946), reduced irisin levels were observed in women, particularly those with Alzheimer’s disease, who present significantly lower concentrations in cerebrospinal fluid (CSF) compared to both healthy women and men affected by the disease. Additionally, more consistent negative correlations were identified in females between irisin (plasma and CSF) and markers such as t-tau, p-tau, and clinical severity, indicating an association with greater cognitive and functional impairment. This likely occurs due to hormonal influences, such as reduced estrogen levels after menopause, potentially related to neurodegenerative processes. Although relevant, these aspects were not incorporated into the manuscript because they go beyond the scope of the proposed analysis, which did not include stratification by sex. Similarly, although DATASUS data indicate a higher prevalence of excess weight in women, the interpretation of these results is limited by the greater number of consultations in this group, which could introduce analytical biases. Regarding the APOE gene, experimental evidence indicates that irisin may reduce APOE expression and modulate astrocyte reactivity via STAT3 inhibition (Kim, et al., 2023, DOI: https://doi.org/10.1016/j.neuron.2023.08.012). More specifically, the APOE4 variant, the main genetic risk factor for late-onset AD, appears to influence the relationship between irisin and cognitive function: while elevated irisin levels are associated with preservation of language and executive functions in individuals without the allele, this effect is not observed in APOE4 carriers (Kim et al., 2022, DOI: https://doi.org/10.1016/j.psyneuen.2021.105624). Despite their relevance, these interactions were also not included in the manuscript because they are not directly aligned with the scope of the analysis conducted.

 

5. What’s the suggestion of authors for people to prevent AD? E.g. doing exercise to produce more irisin to avoid AD. If yes, what’s the potential concentration of irisin would be helpful for preventing AD?

No reference concentration of irisin specifically associated with the prevention of Alzheimer’s disease was identified in the literature. Although several studies evaluate its levels in different tissues and biological fluids (plasma, muscle, adipose tissue, and brain), there is no consensus regarding ideal values or cutoff points. In this context, the present study emphasizes the functional role of irisin, especially its association with physical activity practice and its neuroprotective effects. Evidence indicates that these effects may be maintained with regular exercise and, in some cases, observed even in the short term, reinforcing the potential of physical activity as a non-pharmacological strategy for the prevention and management of dementias.

 

6. In Table 2, are there any units for the parameters? If this, please add the units.

The units of measurement for the analyzed parameters were included, and their descriptions were duly incorporated into the table legends, aiming for greater clarity and standardization of the information presented.

 

7. Do other people from different countries show the same correlation of irisin and AD? E.g. North America, Asian etc.

No direct correlations between irisin levels and Alzheimer’s disease incidence at the population level were identified. However, epidemiological evidence pointing to differences in dementia incidence between countries of different income levels was considered. In high-income countries, factors such as higher education, better nutrition, access to health services, and lifestyle changes are associated with a reduction in dementia incidence, although recent trends such as increased obesity, diabetes, and sedentary behavior represent potential factors for reversing this scenario. In contrast, low- and middle-income countries concentrate approximately two-thirds of dementia cases worldwide, which may reflect both increased life expectancy and a higher prevalence of modifiable risk factors, such as hypertension, obesity, low education, untreated hearing loss, in addition to rising rates of diabetes and smoking (Livingston et al., 2020, DOI: https://doi.org/10.1016/S0140-6736(20)30367-6). On the other hand, the presence of these modifiable factors indicates potential for preventive interventions. Aspects such as a higher frequency of social interaction in these contexts may act as protective factors, in contrast to the more prevalent social isolation in high-income countries. These points were incorporated into the discussion of the manuscript.

 

8. The conclusion section could be revised to be more concise and to focus on the key results of the manuscript.

The conclusion was revised with the aim of strictly limiting it to the findings obtained in the study, avoiding extrapolations beyond the evidence presented. 

 

 

We remain at your disposal to address any remaining questions, as well as to make any improvements you deem necessary.

Reviewer 3 Report

Comments and Suggestions for Authors

In the manuscript entitled “In silico Analyses Suggest That Exercise-Induced Irisin-Mediated Neuroprotection Supports Non-Pharmacological Preventive Strategies for Alzheimer’s Disease as a Public Health Policy” the authors explore the potential neuroprotective role of the exercise-induced myokine irisin in Alzheimer’s disease (AD) using an integrative in silico framework that combines epidemiological analysis, protein-interaction networks, molecular docking, and molecular dynamics simulations. The authors attempt to connect physical activity with molecular mechanisms potentially involved in neuroprotection and prevention strategies for AD.

The topic is relevant because Alzheimer’s disease represents a major global health challenge, and non-pharmacological strategies such as physical activity are increasingly investigated as preventive approaches. The integration of public health data with bioinformatics and structural biology is an interesting conceptual direction.

However, while the manuscript presents potentially valuable ideas, several methodological, conceptual, and structural issues limit the robustness and clarity of the conclusions. Substantial revision is recommended before the manuscript can be considered for publication.

Major Comments

1. Conceptual integration between epidemiology and molecular analyses

The manuscript attempts to link epidemiological observations regarding overweight prevalence with molecular mechanisms related to irisin signaling. However, the connection between these levels of analysis is not sufficiently justified.

For example, the epidemiological section analyzes correlations between overweight prevalence and Alzheimer’s disease hospitalization in Brazil, while the molecular sections focus on irisin-integrin interactions and signaling pathways. The manuscript does not clearly demonstrate how the epidemiological findings mechanistically support the molecular modeling results.

The authors should more clearly explain:

1. The causal or conceptual framework linking population-level observations with molecular docking analyses.
2. Whether the epidemiological component is intended as supportive context or as direct evidence for the proposed biological mechanism.

Without clearer integration, the manuscript risks appearing as a combination of independent analyses rather than a coherent study.

 

 

2. Limitations of the epidemiological analysis

The epidemiological analysis relies on secondary data from the DATASUS platform and uses a time-lagged correlation approach. While this is appropriate for exploratory analysis, the manuscript does not sufficiently discuss its limitations.

Important issues that should be addressed include:

1. Potential confounding variables such as aging population structure, socioeconomic factors, or changes in diagnostic criteria.
2. The ecological nature of the analysis, which does not allow inference about individual risk.
3. The relatively short time window analyzed (2000–2020).

A stronger discussion of these limitations is necessary to avoid overstating the conclusions.

3. Interpretation of molecular docking and molecular dynamics

The manuscript presents docking and molecular dynamics analyses suggesting interaction between irisin and the αV/β5 integrin receptor complex. While these analyses are interesting, the conclusions appear stronger than the computational evidence supports.

In particular:

1. Docking scores and MD stability alone do not demonstrate biological relevance.
2. No experimental validation or comparison with known ligands of the receptor is provided.
3. The biological consequences of the interaction remain speculative.

The authors should present these results as hypothesis-generating rather than as evidence of functional signaling mechanisms.

4. Network and pathway analysis

The protein-interaction network analysis identifies pathways such as PI3K-Akt, FoxO, and neurotrophin signaling. These pathways are already widely associated with neuroprotection and exercise biology.

Therefore, the novelty of the network analysis is somewhat limited. The manuscript would benefit from:

1. A clearer explanation of what new insights emerge from this analysis.
2. A comparison with previously reported signaling mechanisms of irisin.

 

5. Interpretation of results

In several sections, the discussion extrapolates beyond what the data can support. For instance, the manuscript suggests that the findings reinforce the use of exercise as a preventive strategy for Alzheimer’s disease at the level of public health policy.

While exercise is clearly beneficial, the presented analyses alone cannot justify policy recommendations. The authors should moderate these claims and clearly distinguish between evidence, hypotheses, and implications.

Minor Comments

1. Clarity and language
   Some sections contain long and complex sentences that reduce readability. Careful language editing would improve clarity.
2. Structure of the Introduction
   The introduction is informative but somewhat lengthy. A clearer progression from epidemiology to molecular mechanisms would strengthen the narrative.
3. Figures
   The schematic representations of signaling pathways and molecular complexes are useful but should be accompanied by more detailed explanations in the figure legends.
4. References
   Several statements regarding the neuroprotective role of irisin would benefit from additional citations from recent literature.
5. Public health relevance section
   The section highlighting public health implications is interesting but could be shortened and better aligned with the actual findings of the study.

Recommendation

Major Revision

The manuscript addresses a relevant topic and presents interesting computational analyses. However, significant revisions are necessary to clarify the conceptual framework, moderate the interpretation of results, and strengthen the methodological discussion. Addressing these issues would considerably improve the scientific rigor and impact of the work.

 

 

 

Author Response

Dear Reviewer,

 

We thank you for the time and attention dedicated to evaluating our manuscript, as well as for the suggestions provided, which have significantly contributed to improving the work. We conducted a thorough revision of the manuscript, seeking to incorporate, in the most appropriate manner possible, all the considerations made by the reviewers.

 

In general, the main changes made were:

• Comprehensive textual revision, correcting inconsistencies, improving the writing, and enhancing clarity in the description of the tables and methodology;
• Inclusion, in the methodology section, of the justification for the waiver of ethics committee submission, considering the exclusive use of public and anonymized data, in accordance with current regulations;
• Complete reformulation of the discussion section, aiming to more consistently align the study findings, clearly distinguishing between evidence, hypotheses, and implications.

 

Regarding your specific comments:

 

1. Conceptual integration between epidemiology and molecular analyses

The epidemiological analyses aimed to highlight the high prevalence of excess weight in the Brazilian population in the early 2000s, recognized as a risk factor for chronic diseases and dementias, including Alzheimer’s disease. The correlation analysis with DATASUS data was conducted in an exploratory manner, seeking to evaluate the biological plausibility of the outcome and reinforce the relevance of AD as a public health problem in the country. Considering that physical activity is a well-established protective factor, structural analyses were incorporated with the purpose of supporting, at a mechanistic level, the hypothesis that irisin, induced by exercise, may interact with the αV/β5 receptor and trigger intracellular pathways associated with neuroprotection, as suggested by the network analysis. In the revised version, this chain of reasoning was explained more clearly, including the discussion that even in the absence of the RGD motif typical of integrin ligands, the interaction can occur stably, which is consistent with the possibility of signal transduction in the nervous system.

 

 

2. Limitations of the epidemiological analysis

We acknowledge that the epidemiological analysis has limitations inherent to the ecological design, which were not fully explored in the initial version of the manuscript. In response to the suggestion, we expanded the discussion of these aspects, including potential confounding factors and methodological constraints, aiming to provide greater transparency in the interpretation of the results without compromising the exploratory value of the analysis.

 

3. Interpretation of molecular docking and molecular dynamics

The absence of resolved experimental structures of the αV/β5 integrin complex, particularly in association with different ligands except irisin (Mu et al., 2023, DOI: https://doi.org/10.1016/j.molcel.2023.05.008), limits direct comparisons involving affinity, interaction residues, and conformational dynamics. Therefore, we chose to restrict the discussion to structural patterns and conformational movements already described for other integrins, acknowledging their possible applicability to the studied system. Additionally, we adjusted the interpretation of the structural results to more rigorously reflect the theoretical nature of the approaches employed. The parameters obtained were used to indicate the feasibility of the interaction, without inferring direct evidence, maintaining the focus on generating biologically plausible hypotheses.

 

4. Network and pathway analysis

- The main contribution of the network analysis lies in integrating pathways already described into a unified functional context centered on irisin. Considering that part of the interactions is not yet fully annotated in databases, we recognize that the presented network may not fully reflect the complexity of the system. Nevertheless, we sought to interpret the results in light of the literature, articulating these pathways with the structural and epidemiological findings. In this context, we propose that the interaction of irisin with the αV/β5 receptor may act as an initial event capable of triggering multiple intracellular cascades associated with neuroprotection and neuronal plasticity, supporting the hypothesis that regular physical activity may contribute to the prevention of Alzheimer’s disease.

 

5. Interpretation of results

The language of the manuscript was comprehensively revised with the aim of aligning interpretations with the level of evidence presented. We acknowledge that the initial version could lead to an overly conclusive interpretation; therefore, we primarily reformulated the discussion section to more clearly distinguish between evidence, hypotheses, and implications. We emphasize that the results are not confirmatory in nature but indicate consistency across different levels of analysis, suggesting that irisin, produced during physical exercise, may be associated with neuroprotective effects mediated by interaction with the αV/β5 receptor and by the activation of networks related to the maintenance of nervous tissue integrity.

 

Minor Comments

The manuscript underwent thorough textual revision, with restructuring of long sentences, terminological improvement, and adjustments to cohesion and objectivity, aiming for greater clarity and precision in scientific communication. Some references aimed at improving the clarity of the discussion were added, and the legend of Figure 1 was better described. Other minor changes were made in an attempt to address these suggestions.

 

We remain at your disposal to address any remaining questions, as well as to make any improvements you deem necessary.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript has been significantly improved, and my concerns have been satisfactorily addressed. I consider it suitable for publication in its current form.

Reviewer 3 Report

Comments and Suggestions for Authors

The revised manuscript demonstrates clear and substantial improvement. The authors have adequately addressed previous concerns, particularly by strengthening the epidemiological interpretation, explicitly acknowledging limitations, and avoiding causal overstatements. The inclusion of confounding factors and the improved discussion significantly enhance the scientific rigor of the work.

The addition of a public health perspective and clearer methodological descriptions further improves the manuscript’s relevance and transparency. The integration of epidemiological, network, and structural analyses is coherent and well articulated, and the mechanistic discussion of the irisin–αV/β5 interaction adds originality.

Minor language issues remain but do not affect the overall quality of the study.

Recommendation:
I recommend acceptance with minor editorial revisions.