Investigating Age-Dependent Oxygenation and Blood Perfusion in a Mouse Model of Peripheral Artery Disease (PAD) Using Multispectral Optoacoustic Tomography (MSOT), Laser Speckle Contrast Imaging (LSCI) and Histology^†

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors describe an interesting study comparing age-dependent recovery in a mouse model of PAD using two innovative imaging modalities. The application of the "oxygen gas-breathing challenge" to assess vascular reserve in ischemic limbs is particularly noteworthy and provides more physiological insight than static measurements. While the technological approach is sound, there are several concerns regarding the consistency of the experimental design, the sample sizes, and the clinical interpretation of the histological findings that need to be addressed.

Major Comments:

1. There is a noticeable discrepancy in the number of animals used across different experiments. In Section 2, the authors state that n=16 (8 young, 8 old) were used for MSOT, while n=24(8 young, 16 old) were used for LSCI. Furthermore, the immunofluorescence analysis (CD31 and Ki-67) in Section 3.4.1 appears to be based on only n=4 per group.

・Could the authors please clarify the rationale for using different cohorts for LSCI and MSOT?

・Why was the sample size for the immunofluorescence analysis reduced to half of the MSOT cohort?

・Please clarify if the mice used for histology were a random subset of the imaging cohorts and provide a statistical justification for the smaller n in the IF studies, as this may limit the robustness of the regenerative markers' data.

 

2. The authors acknowledge that the acute cauterization model differs from the chronic atherosclerotic process in humans. However, as clinicians, we often see that the "spontaneous recovery" observed in mice within 14 days is rarely seen in human PAD/CLTI patients without intervention. Please expand the discussion on how the "vascular reserve" measured by ΔsO2​ in this acute model might translate to predicting revascularization outcomes in chronic human patients where the baseline microvascular environment is severely degraded.

3. In the Discussion, the authors suggest that increased adipocyte presence in young mice might play a "supportive role" in tissue repair due to the secretion of cytokines and growth factors. However, in clinical PAD, adipose infiltration and "fatty degeneration" of the muscle are typically regarded as indicators of poor prognosis and advanced disease stage (sarcopenia/myosteatosis). The authors should address this paradox—why adipose infiltration is considered a positive sign for regeneration in this model, whereas it is a pathological marker in humans.

4. The paper effectively describes the phenomenon of impaired recovery in older mice but does not delve deeply into the underlying mechanisms. While the authors mention VEGF and HIF-1α in the Discussion, these were not measured in this study. To strengthen the paper, I recommend adding protein or mRNA expression data (e.g., via Western Blot or qPCR) for these key angiogenic factors to correlate the MSOT/LSCI functional data with molecular changes. If this is not possible, the limitations regarding the purely observational nature of these findings should be stated more explicitly.

 

Minor Comments:

1. In Section 2.2, please provide details on how the depilation (Veet cream) was managed over the 14-day period. Repeated chemical depilation can cause skin irritation/inflammation, which might influence the LSCI skin perfusion signal or the sO2​ readings in the superficial layers.

Author Response

We appreciate the thoughtful review of our manuscript and are pleased to provide a revised submission herewith. We have made modifications to address the issues and respond as follows.

Reviewer 1:

Major Comments:

1. There is a noticeable discrepancy in the number of animals used across different experiments. In Section 2, the authors state that n=16 (8 young, 8 old) were used for MSOT, while n=24(8 young, 16 old) were used for LSCI. Furthermore, the immunofluorescence analysis (CD31 and Ki-67) in Section 3.4.1 appears to be based on only n=4 per group.

・Could the authors please clarify the rationale for using different cohorts for LSCI and MSOT?

・Why was the sample size for the immunofluorescence analysis reduced to half of the MSOT cohort?

・Please clarify if the mice used for histology were a random subset of the imaging cohorts and provide a statistical justification for the smaller n in the IF studies, as this may limit the robustness of the regenerative markers' data.

Answer 1: The results are the product of a multi institutional collaboration, where we are able to undertake LSCI at UT Arlington and MSOT at UT Southwestern. Transporting mice between locations was not practical and we chose to examine separate cohorts. However, Dr. Nguyen, a trained surgeon generated the PAD models in both laboratories to minimize potential differences in procedures. LSCI had been undertaken first during which time we perfected the technique and established a highly consistent mouse model and imaging methodology so that smaller cohorts of mice could be used for subsequent investigations. We should also note that not all mice were used at every timepoint for LCSI, whereas the 16 mice for MSOT were followed sequentially. This is now clarified in the Methods and Results, and we have added a Supplementary Figure S3 to show the high consistency of investigations by LCSI. Immunofluorescence (CD31 and Ki-67) analyses were performed on a subset (n = 4 per group) of the mice used for MSOT. Specifically, the first 4 old and young mice were investigated as now stated in Methods. Immunofluorescence was performed as a supportive histological confirmation of the imaging findings. IF staining results were consistent with imaging observations.

 

2. The authors acknowledge that the acute cauterization model differs from the chronic atherosclerotic process in humans. However, as clinicians, we often see that the "spontaneous recovery" observed in mice within 14 days is rarely seen in human PAD/CLTI patients without intervention. Please expand the discussion on how the "vascular reserve" measured by ΔsO2​ in this acute model might translate to predicting revascularization outcomes in chronic human patients where the baseline microvascular environment is severely degraded.

Answer 2: As noted, the acute ischemia caused by cauterization serves as a model of PAD, but cannot completely recreate the impaired microvascular function and chronic atherosclerotic in human PAD. This model has been used by many investigators, but we note a more complex two-stage limb ischemia model has been proposed and could be preferable for future therapeutic investigations. Our current goal was to readily achieve reproducible ischemia and examine the application of MSOT as compared with the more popular LCSI. While LSCI provides an indication of limb perfusion, the measurement are depth limited, whereas ΔsO₂ measured by MSOT can indicate tissue heterogeneity and local ischemia and hypoxia in muscle (Figs 1 & 2).

We have added a comment “A more complex  two-stage mouse model to better simulate human PAD has been reported [57], but was beyond the scope of our current work which sought to demonstrate and compare imaging methods for examining muscle pathophysiology. “ to Discussion.

Tissue oxygenation and vascular reserve may be more effective is assessing tissue damage and recovery with treatment. Global limb perfusions as indicated by superficial measurements using LSCI may not reflect local heterogeneity; restoration of a major blood vessel may not indicate microvascular improvement. As such heterogeneity of deep tissues should be a better predictive indicator of potential tissue recovery. Here we have demonstrated the ability to make measurements in a study using more animals and over a longer time frame than previous reports for MSOT, e.g. Khaw et al, ref 14.

Our study supports the fact that MSOT is a non-invasive modality which can assess oxygenation based on vascular function, rather than doing only structural or perfusion-based measurements.

 

3. In the Discussion, the authors suggest that increased adipocyte presence in young mice might play a "supportive role" in tissue repair due to the secretion of cytokines and growth factors. However, in clinical PAD, adipose infiltration and "fatty degeneration" of the muscle are typically regarded as indicators of poor prognosis and advanced disease stage (sarcopenia/myosteatosis). The authors should address this paradox—why adipose infiltration is considered a positive sign for regeneration in this model, whereas it is a pathological marker in humans.

Answer 3: We agree that adipose infiltration and "fatty degeneration" of the muscle are typically regarded as indicators of poor prognosis and advanced disease stage (sarcopenia/myosteatosis) in clinical PAD. But the presence of adipocytes in young mice in our study is more likely due to the temporary physiological adaptation rather than the chronic intramuscular fat infiltration. The increased adipose tissue indicates perfused and highly vascularized tissues which are involved in vascular reconfiguration and promote angiogenesis through VEGF, adipokines, cytokines via cross talk with endothelial cells. Such remodeling is well documented regarding angiogenic potential of young tissues after ischemia. Previous studies also supported the fact that increased adipose tissues contribute in the vascular remodeling and helps to improve regenerative capacity (Section 4 and Refs. 41-43 in the manuscript).

 

4. The paper effectively describes the phenomenonof impaired recovery in older mice but does not delve deeply into the underlying mechanisms. While the authors mention VEGF and HIF-1α in the Discussion, these were not measured in this study. To strengthen the paper, I recommend adding protein or mRNA expression data (e.g., via Western Blot or qPCR) for these key angiogenic factors to correlate the MSOT/LSCI functional data with molecular changes. If this is not possible, the limitations regarding the purely observational nature of these findings should be stated more explicitly.

 

Answer 4: We understand that VEGF and HIF-1α are significant regulators for angiogenesis, which would provide depth of mechanism. We investigated CD31 (endothelial) and Ki67 (proliferation) markers which were consistent with our primary findings obtained from MSOT and LSCI imaging. VEGF and HIF-1α were not investigated in our study. In our revised Discussion we now clarify that we only examined histological markers (CD31 and Ki67) and did not investigate upstream molecular mechanism of angiogenic signaling pathways.

 

Minor Comments:

1. In Section 2.2, please provide details on how the depilation (Veet cream) was managed over the 14-day period. Repeated chemical depilation can cause skin irritation/inflammation, which might influence the LSCI skin perfusion signal or the sO2​ readings in the superficial layers.

Answer 1. We do recognize that repeated chemical depilation can cause local skin irritation/inflammation that may influence superficial LSCI perfusion and MSOT-derived sO₂ signals. We did apply Veet cream before each imaging session in a consistent manner for all mice followed by rapid gentle cleansing. The skin was carefully monitored during imaging sessions. We did not find any visible sign of irritation, erythema, or injury. We now elaborate in Methods Section 2.2.

Reviewer 2 Report

Comments and Suggestions for Authors

 

This manuscript presents a preclinical study investigating age-dependent vascular responses in a mouse model of peripheral artery disease (PAD) using multispectral optoacoustic tomography (MSOT), laser speckle contrast imaging (LSCI), and histology. The authors report that younger mice exhibit faster recovery of oxygenation and perfusion after femoral artery cauterization, supported by imaging and histological markers of angiogenesis and proliferation.

The topic is relevant and timely, particularly given the increasing interest in non-invasive imaging modalities for vascular disease. The multimodal approach is a strength. However, several methodological, statistical, and interpretative issues need to be addressed before the manuscript is suitable for publication.

 

1. The manuscript presents a technically solid and well-executed study, but its central contribution needs to be more clearly articulated. While the use of multispectral optoacoustic tomography alongside laser speckle imaging is a strength, the biological finding, that younger mice recover more effectively from ischemia than older ones, is already well established in the literature. As a result, the real novelty should lie in demonstrating what additional insight MSOT provides compared to existing methods, yet this is not fully developed. The paper would benefit from a clearer argument about whether MSOT offers superior sensitivity, depth resolution, or longitudinal monitoring advantages that meaningfully advance the field.
2. There are also important concerns about the experimental design that weaken the strength of the conclusions. Sample sizes are inconsistent across modalities, with some groups underpowered and others imbalanced, particularly in the LSCI experiments. In addition, the exclusive use of female mice is not justified, despite known sex differences in vascular biology. More fundamentally, the model itself represents acute ischemia rather than the chronic progression seen in human peripheral artery disease, which limits the translational relevance of the findings and requires a more cautious interpretation.
3. The statistical analysis further raises questions about robustness. The reliance on multiple t-tests and relatively permissive post hoc testing without clear correction for multiple comparisons increases the risk of false-positive findings. At the same time, inconsistent reporting of variability measures (SEM versus SD) and lack of detail on assumptions such as normality make it difficult to fully assess the validity of the results. Strengthening the statistical framework would significantly improve confidence in the conclusions.
4. While the multimodal approach is a major strength, the study stops short of fully integrating its datasets. Imaging findings, perfusion measurements, and histological markers are presented in parallel but not quantitatively linked, and mechanistic interpretations, such as the involvement of VEGF signaling or endothelial progenitor cells, are introduced without direct evidence. A more rigorous correlation between modalities and a more restrained interpretation of underlying mechanisms would make the study more coherent and scientifically convincing.

Author Response

1. The manuscript presents a technically solid and well-executed study, but its central contribution needs to be more clearly articulated. While the use of multispectral optoacoustic tomography alongside laser speckle imaging is a strength, the biological finding, that younger mice recover more effectively from ischemia than older ones, is already well established in the literature. As a result, the real novelty should lie in demonstrating what additional insight MSOT provides compared to existing methods, yet this is not fully developed. The paper would benefit from a clearer argument about whether MSOT offers superior sensitivity, depth resolution, or longitudinal monitoring advantages that meaningfully advance the field.

We appreciate the thoughtful review of our manuscript and are pleased to provide a revised submission herewith. We have made modifications to address the issues and respond as follows.

Answer 1: As shown in Figures 1 and 2 MSOT reveals heterogeneity of specific tissues of interest. It also reveals both baseline hemoglobin saturation and the response (vascular reserve ΔsO₂) in response to oxygen gas breathing challenge. It remains to be seen whether these parameters will be more useful than LSCI as prognostic indicators of pathophysiological sequalae, treatment response and functional response to therapeutic interventions. The goal here was to demonstrate proof of principle and indicate comparative results by LSCI and histology. We also recognize that MSOT could provide measurements of total hemoglobin and potentially assay materials such as delivery scaffolds for therapeutic agents. We believe there are several novel aspects to our study: there have been few applications of MSOT to PAD previously, with only 3 studies of mice. We successfully used the iThera system as opposed to LAZR and applied 7 wavelengths in place of 2. We examined transaxial sections of thigh as opposed to longitudinal allowing us to compare injured leg with contralateral control leg and spine muscle. We also examined chronic ischemia as opposed to acute ligation. We examined BALB/c mice as opposed to nude or black mice. These points are presented in the Discussion.

 

2. There are also important concerns about the experimental design that weaken the strength of the conclusions. Sample sizes are inconsistent across modalities, with some groups underpowered and others imbalanced, particularly in the LSCI experiments. In addition, the exclusive use of female mice is not justified, despite known sex differences in vascular biology. More fundamentally, the model itself represents acute ischemia rather than the chronic progression seen in human peripheral artery disease, which limits the translational relevance of the findings and requires a more cautious interpretation.

We used female mice exclusively since they tend to fight less, thereby avoiding injury, scarring or death. Our goal was to evaluate the application of the imaging methods, and compare young versus aged mice, not examine any sex differences, which would have required a considerably larger study.

As noted in the response to Reviewer 1, comment 1, above: the results are the product of a multi institutional collaboration. LSCI was undertaken first during which time we perfected the technique and established a highly consistent mouse model and imaging methodology so that smaller cohorts of mice could be used for subsequent investigations. We should also note that not all mice were used at every timepoints for LCSI, whereas the 16 mice for MSOT were followed sequentially. This is now clarified in the Methods and Results, and we have added a Supplementary Figure S3 to show the high consistency of investigations by LCSI. Immunofluorescence (CD31 and Ki-67) analyses were performed on a subset (n = 4 per group) of the mice used for MSOT. Group sizes were not rigorously balanced, but we do not believe experiments were underpowered.

As noted in Answer 2  to Reviewer 1, the acute ischemia caused by cauterization serves as a model of PAD, but we recognize that it does not completely recreate the impaired microvascular function and chronic atherosclerotic in human PAD. This model has been used by many investigators, but we note a more complex two-stage limb ischemia model has been proposed and could be preferable for future therapeutic investigations. Our current goal was to readily achieve reproducible ischemia and examine the application of MSOT as compared with the more popular LCSI. While LSCI provides an indication of limb perfusion, the measurements are depth limited, whereas ΔsO₂ measured by MSOT can indicate tissue heterogeneity and local ischemia and hypoxia in muscle (Figs 1 & 2). We have added a comment “A more complex  two-stage mouse model to better simulate human PAD has been reported [57], but was beyond the scope of our current work which sought to demonstrate and compare imaging methods for examining muscle pathophysiology. “ to Discussion.

 

3. The statistical analysis further raises questions about robustness. The reliance on multiple t-tests and relatively permissive post hoc testing without clear correction for multiple comparisons increases the risk of false-positive findings. At the same time, inconsistent reporting of variability measures (SEM versus SD) and lack of detail on assumptions such as normality make it difficult to fully assess the validity of the results. Strengthening the statistical framework would significantly improve confidence in the conclusions.

Answer 3. Thank you for pointing out the dangers of multiple t-tests and our mixed presentation of SEM and SD. We have now replaced the analysis with multivariate ANOVA and modified figures to reflect modified statistical significance. All Figures now show SEM.

4. While the multimodal approach is a major strength, the study stops short of fully integrating its datasets. Imaging findings, perfusion measurements, and histological markers are presented in parallel but not quantitatively linked, and mechanistic interpretations, such as the involvement of VEGF signaling or endothelial progenitor cells, are introduced without direct evidence. A more rigorous correlation between modalities and a more restrained interpretation of underlying mechanisms would make the study more coherent and scientifically convincing.

Answer 4: Direct correlation of the LSCI and MSOT observations in identical mice would be more powerful as noted. Unfortunately, our instruments are located at 2 collaborating institutions 15 miles apart and it was neither practical to move mice nor instruments between laboratories. We have investigated CD31 (endothelial) and Ki67 (proliferation) markers which corroborate our primary findings obtained from MSOT and LSCI imaging. VEGF and HIF-1α were not investigated in our study and we are not currently in a position to add such data. We do now mention this in the revised Discussion.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for your very thorough and thoughtful responses to my previous comments. The revised manuscript is significantly improved, and I deeply appreciate the transparent and scientifically rigorous approach you have taken in this revision.

Specifically, I commend the following improvements:

• Your decision to remove the speculative assertions regarding the "supportive role" of adipose tissue and instead properly frame the adipose infiltration as structural evidence of ischemia-associated remodeling is highly appropriate.

• The addition of a dedicated paragraph discussing the limitations of the acute cauterization model compared to the chronic pathophysiology of human PAD adds great maturity and clinical relevance to your Discussion.

• Explicitly stating that VEGF/HIF-1 activity was not directly examined, while maintaining the literature-based context, ensures the readers can accurately interpret your histological findings.

• The detailed explanations regarding the sample sizes (including the rationale for using a subset for immunofluorescence) and the specific management of the depilation cream have fully resolved my previous methodological concerns.

The multimodal imaging approach (MSOT and LSCI) demonstrated here holds genuine promise for the longitudinal assessment of peripheral vascular disease. Your manuscript now provides a highly balanced, robust, and valuable contribution to the field of cardiovascular imaging and diagnostics.

I have no further comments. Congratulations on a well-executed study and an excellent revision.

Sincerely,

Reviewer

Reviewer 2 Report

Comments and Suggestions for Authors

Manuscript ready for publication.