Alterations of Cerebral Extracellular Vesicle microRNA Profiling Potentially Disrupts Brain Homeostasis Following Myocardial Infarction

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper entitled "Alterations of cerebral extracellular vesicle microRNA profiling potentially disrupts brain homeostasis following myocardial infarction" by MD Monowarul Islam et al. investigates changes in brain extracellular vesicle (EV) miRNA profiles after myocardial infarction (MI). The study suggests that altered EV-associated miRNAs may contribute to cognitive impairment by disrupting brain homeostasis, particularly through pathways related to oxidative stress and neuroinflammation. The topic is timely and interesting, and the study provides novel insight into brain–heart axis communication after MI. However, several issues should be addressed.

1. In the Introduction, cognitive impairment is mainly discussed in relation to aging. Chronic inflammation after cardiac disease is also an important contributing factor and should be mentioned.
2. In the Methods, the authors state that “MI animals with ejection fraction (EF%) more than 40 were excluded from further study.” However, the criteria for defining a successful MI/HF model are not clearly described. These criteria should be provided in detail.
3. In Figure 1A–F, exact P values are shown, whereas Figure 1G uses asterisks. The meaning of the asterisks is not defined in the figure legend. The authors should use a consistent reporting style and clearly define all statistical symbols.
4. In Figure 5, some P values between groups are missing. For example, the comparison between Sham and 6 weeks post-MI in Figure 5B is not shown. Similar issues are present in Figure 5C, D, and F for comparisons between Sham and 3 weeks post-MI.
5. In Figure 6, some group comparisons are also missing. For example, the P value between Sham and 6 weeks post-MI in Figure 6C is not shown. In Figure 6D, the difference between Sham and 6 weeks post-MI appears substantial, but only one asterisk is shown, indicating P < 0.05. The authors should recheck the statistical analysis and ensure that the reported significance levels are accurate.  
6. The authors suggest that brain EV miRNAs may be cardiac in origin after MI. However, as MI-induced heart failure progresses, brain function itself may also be altered, potentially leading to changes in brain-derived EVs. Therefore, the authors should discuss whether dysregulated EVs may originate from both the heart and the brain itself.

Author Response

Reviewer #1

We are grateful for the reviewer’s positive feedback and constructive suggestions. We have carefully addressed each point to enhance the clarity of our manuscript as follows:

Point 1: In the Introduction, cognitive impairment is mainly discussed in relation to aging. Chronic inflammation after cardiac disease is also an important contributing factor and should be mentioned.

Answer: We thank the reviewer for this excellent suggestion. We agree that systemic inflammation is a major contributor to cognitive impairment following cardiac injury. Accordingly, we have revised the Introduction to highlight the critical role of systemic inflammation in heart failure-induced cognitive impairment in red (page 2 lines 48-49).

Point 2: In the Methods, the authors state that “MI animals with ejection fraction (EF%) more than 40 were excluded from further study.” However, the criteria for defining a successful MI/HF model are not clearly described. These criteria should be provided in detail.

Answer: We thank the reviewer for pointing this out. To ensure our pre-clinical model mimics human pathophysiology, we targeted Heart Failure with reduced Ejection Fraction (HFrEF). In rats, a normal left ventricular ejection fraction (LVEF) typically ranges from 55% to ~70% (compared with 50% ~ 60% in humans). For our model, a typical heart failure phenotype was defined by an LVEF of less than 45%. Consequently, MI animals with an LVEF greater than 45% were excluded from the study to maintain cohort consistency. We have clarified these criteria and updated the text accordingly (Page 3, lines 87-89; see also Figure 1 and Supplemental Table 1).

Point 3: In Figure 1A–F, exact P values are shown, whereas Figure 1G uses asterisks. The meaning of the asterisks is not defined in the figure legend. The authors should use a consistent reporting style and clearly define all statistical symbols.

Answer: We appreciate the reviewer’s attention to detail. We have updated Figure 1 and its corresponding legend to ensure a consistent statistical reporting style (see updated Figure 1 on page 6, lines 190-194).

Point 4: In Figure 5, some P values between groups are missing. For example, the comparison between Sham and 6 weeks post-MI in Figure 5B is not shown. Similar issues are present in Figure 5C, D, and F for comparisons between Sham and 3 weeks post-MI.

Answer: We thank the reviewer for identifying these missed comparisons. Figure 5 and its legend have been comprehensively updated to include all relevant statistical comparisons (see updated Figure 5 and Figure legend on page 11, line 275).

 Point 5: In Figure 6, some group comparisons are also missing. For example, the P value between Sham and 6 weeks post-MI in Figure 6C is not shown. In Figure 6D, the difference between Sham and 6 weeks post-MI appears substantial, but only one asterisk is shown, indicating P < 0.05. The authors should recheck the statistical analysis and ensure that the reported significance levels are accurate.  

Answer: We thank the reviewer for this observation. We have fully re-analyzed the statistical data for Figure 6 to ensure complete accuracy. Figure 6 and the accompanying figure legend have been corrected to reflect the precise significance levels and missing group comparisons (see updated Figure 6 and figure legend d on page 12, lines 283-284).

Point 6: The authors suggest that brain EV miRNAs may be cardiac in origin after MI. However, as MI-induced heart failure progresses, brain function itself may also be altered, potentially leading to changes in brain-derived EVs. Therefore, the authors should discuss whether dysregulated EVs may originate from both the heart and the brain itself.

Answer: We completely agree with the reviewer’s insightful comment. The progression of heart failure undoubtedly induces cerebral changes that can alter endogenous EVs. This phenomenon complements our hypothesis regarding the cardiac origin of specific EV-miRNAs. First, our previous work using a novel cardiac-membrane GFP+ reporter mouse model directly demonstrated cardiac EV-mediated heart-to-brain communication following MI.  Second, we previously also demonstrated that brain-derived EVs are significantly enriched with cardiac muscle-specific miRNAs (e.g., miR-208, miR-499), indicative of a potential cardiac origin of miRNAs (Circ Res, 2022, 131: 687-700; JACC Basic Transl Sci. 2025, 10(7):101307). Third, the specific miRNAs validated in the current study display a consistent abundance in cardiac EVs, circulating EVs, and cerebral EVs (Fig 5 and Fig 6), strongly supporting a cardiac contribution.   

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript entitled “Alterations of cerebral extracellular vesicle microRNA profiling potentially disrupts brain homeostasis following myocardial infarction” by Isism et al. investigated the role of extracellular vesicles (EVs) and their microRNA (miRNA) cargo in cognitive impairment associated with heart failure. Using a rat model of myocardial infarction (MI), the authors found progressive alterations in brain EV miRNA profiles during heart failure progression, with minimal changes at 3 weeks post-MI but substantial changes at 6 and 12 weeks. Bioinformatic analyses revealed reduced levels of miRNAs involved in protection against oxidative stress and inflammation, alongside increased expression of miRNAs associated with oxidative stress and neuroinflammation. The authors concluded that cardiac-derived EVs may alter brain EV signaling, disrupt brain homeostasis, and contribute to cognitive impairment in heart failure. These findings are very interesting and may provide a fundamental basis for cardiac-brain crosstalk-related cognitive impairment in heart failure. However, addressing the minor concerns listed below would help improve the quality of the manuscript.

1. It is unclear whether age-matched rats were used as controls at each post-MI time point. Since cardiac dysfunction appeared to stabilize by 6 weeks post-MI, this raises the additional concern that the further dysregulation of miRNAs observed at 12 weeks post-MI may be partially attributable to aging rather than progression of heart failure alone.
2. Although consistent changes in miRNA expression were observed between the heart and brain, it is hard to conclude that the miRNAs enriched in brain EVs are of cardiac origin. Are there any established methods for labeling cardiac EVs? If so, such approaches could enable tracking of EV biodistribution, cellular uptake, and biological functions.
3. Potential mediators promoting EV trafficking from heart to brain should be discussed. Additionally, it is unclear whether the number or abundance of cardiac EVs was altered after MI. Are there any differences in miRNA profiles between brain EVs and cardiac EVs in sham rats?
4. How do the authors distinguish cardiac-derived EVs from other possible contributors, such as systemic inflammation, liver-derived EVs, or BBB disruption?
5. Could the authors clarify the rationale for focusing specifically on EVs isolated from the non-infarcted left ventricle? Infarct, border-zone, and remote myocardial regions may exhibit distinct EV miRNA profiles.
6. The rationale for selecting only the typical heart failure phenotype for sequencing is unclear. Does this imply that heart failure severity is a key determinant of EV trafficking? Additionally, were sequencing data compared between typical and non-typical heart failure rats at each post-MI time point?

Author Response

Point 1: It is unclear whether age-matched rats were used as controls at each post-MI time point. Since cardiac dysfunction appeared to stabilize by 6 weeks post-MI, this raises the additional concern that the further dysregulation of miRNAs observed at 12 weeks post-MI may be partially attributable to aging rather than progression of heart failure alone.

Answer: We thank the reviewer for raising this important consideration regarding aging as a potential confounding factor. To minimize confounding by age-related variables, all animals used in this study were strictly age- and weight-matched at baseline. Control animals in the sham group were euthanized between 6 and 12 weeks post-surgery sequentially to serve as appropriate controls across time points. As detailed in Supplemental Table 1, hemodynamic and echocardiographic parameters that had stabilized by week 6 remained severely impaired compared with the sham group. Thus, we feel it is highly unlikely the progressive miRNA alterations observed between 6 and 12 weeks are due to aging per se.

Point 2: Although consistent changes in miRNA expression were observed between the heart and brain, it is hard to conclude that the miRNAs enriched in brain EVs are of cardiac origin. Are there any established methods for labeling cardiac EVs? If so, such approaches could enable tracking of EV biodistribution, cellular uptake, and biological functions.

Answer: We completely agree that definitive tracking of EVs is a major challenge. In our previous publications, we established and utilized a novel cardiac-membrane GFP⁺ reporter mouse model to explicitly track cardiac EV biodistribution, cellular uptake, and downstream biological functions in the brain following MI. Furthermore, our study demonstrated that brain-derived EVs are enriched for cardiac muscle-specific miRNAs (e.g., miR-208, miR-499), indicating a potential cardiac origin (Circ Res, 2022, 131: 687-700; JACC Basic Transl Sci. 2025, 10(7):101307). In addition, the fact that the specific miRNAs validated in this study are consistently abundant across cardiac, circulating, and cerebral EV compartments (Fig 5 and Fig 6) further supporting their potential cardiac origin.  

 Point 3: Potential mediators promoting EV trafficking from heart to brain should be discussed. Additionally, it is unclear whether the number or abundance of cardiac EVs was altered after MI. Are there any differences in miRNA profiles between brain EVs and cardiac EVs in sham rats?

Answer: We appreciate these insightful suggestions. In accordance with the reviewer’s comment, we have added a dedicated paragraph discussing the potential mechanisms promoting cardiac EV trafficking to the brain (page 14, lines 357-362). Moreover, our prior findings and emerging literature indicate that the abundance of cardiac-derived EVs increase substantially following MI (Circ Res, 2022, 131: 687-700; Circ Res 2025, 136 (11):1513-1515). This information has been integrated into the Discussion (page 13, lines 347-349). In addition, we did not compare the miRNA profiles between brain EVs and cardiac EVs in sham rats. While this represents a compelling question, we feel it may be outside the scope of the current study.

Point 4: How do the authors distinguish cardiac-derived EVs from other possible contributors, such as systemic inflammation, liver-derived EVs, or BBB disruption?

Answer: We cannot exclude the contributions of other possible mechanisms (page 2 line 48-53) however our goal was to highlight cardiac-derived EVs as a novel, direct mechanism contributing to cognitive impairment following MI.

Point 5: Could the authors clarify the rationale for focusing specifically on EVs isolated from the non-infarcted left ventricle? Infarct, border-zone, and remote myocardial regions may exhibit distinct EV miRNA profiles.

Answer: We are grateful for the opportunity to clarify our rationale. Following myocardial infarction, the infarcted scar tissue rapidly transitions into a dense, permanent fibrotic area devoid of functional myocardium and normal extracellular space. To optimize EV yield, the fibrotic scar tissue was physically removed. EVs were subsequently isolated from the remaining functional left ventricular (LV) tissue, which encompasses both the border-zone and the remote myocardial regions of the LV. We have clarified this methodology in the revised Methods section (page 3, line 109-110).

 Point 6: The rationale for selecting only the typical heart failure phenotype for sequencing is unclear. Does this imply that heart failure severity is a key determinant of EV trafficking? Additionally, were sequencing data compared between typical and non-typical heart failure rats at each post-MI time point?

Answer: In this study, we focused on HFrEF (Heart Failure with reduced Ejection Fraction) to establish a successful preclinical model that reliably mimics the human pathophysiology of heart failure. Based on the longitudinal miRNA expression profiles captured across our time points following the progression of heart failure (Figures 1 and 3), we believe that heart failure severity dictates the specific molecular cargo of EVs, rather than driving EV trafficking itself, which occurs even under baseline physiological conditions. We did not compare typical and non-typical heart failure cohorts at each time point. While it is an interesting topic and is promising for biomarker screening, it remains beyond the scope of our current study.