Defining the miRnome of Saphenous Vein Smooth Muscle Cells from Patients with Type 2 Diabetes Mellitus

Round 1

Reviewer 1 Report

Summary

This paper aimed to examine miRNA profiles of saphenous vein smooth muscle cells form people with and without diabetes, to understand how differences could contribute to their dysfunction in T2DM. I found it easy to read and well-presented but have some suggestions to improve.

 Specific Comments

Introduction

This was a beautiful introduction, covering the background of miRNAs and their therapeutic potential, as well as the known defects of SMCs in T2DM. Great work!

Methods

Line 90-93 miRNA validation experiments had tissue from lower legs as well as CAGB. This is mentioned in results but it is unclear if the number of samples for T2DM and ND are matched in tissue type. I’d like this to be mentioned here or in results, and then in discussion in case these source tissues are a contributor to differences between groups.

105- Only males in this, but some females in validation. I’d like to see sex mentioned in discussion.

151- There is no discussion of power calculations done to determine sample size. It would be beneficial to note if the sample size was selected for a specific reason. If this is atypical for this type of study, then please discard my comment.

Results

Figure 1: Does a volcano plot usually highlight in red the ones that were selected (e.g.  above 2 upregulated or below 0.2 downregulated)? It would nice to see if this can be added. The text in the table is small and a bit blurred, can it be clearer?

Line 189 “Table 1. Criteria for selecting candidate miRNAs for further validation”. The table shows the candidates, not the criteria, so please amend the title. Unsure if p values should be truncated to leading digit?

198-200 notes the different sources of tissue, and I wonder if this was matched between T2DM and ND. If not, could it be a contributing factor to differences? Please mention numbers of each here (or in methods).

Figure 3: What is %U6? Please note in legend.

Figure 4: Is there a name for this type of figure? If so, please note it so we readers know what they are. It’s beautiful!  

Discussion

Line 232-241. I found it a little disordered to introduce changes to these cells from the literature, rather than your results, first. I was wondering why this was being discussed here without clearly linking to finding(s). I feel as though you should present your findings regarding cytoskeleton and then say these are likely to be additional causes of the phenotypic abnormalities to miR-145.  

243-246: This is the methods, and covered in results, so not needed in discussion.

Line 254- It would be best to move the statement of MiR being up in T2DM to the start of this paragraph, then explain what this might mean in terms of known/predicted functions.

Please mention sex in discussion, as the initial dataset had no females (although validation did).

I’d also like to see the potential impact of different tissue sources discussed, at least briefly.

Conclusion

325-331 The limitations should be moved to discussion, not remain here in the conclusion

Supp Fig 2: I really like the colour coding here, nice detail for a supplementary figure!

Author Response

Reviewer 1

Summary

This paper aimed to examine miRNA profiles of saphenous vein smooth muscle cells form people with and without diabetes, to understand how differences could contribute to their dysfunction in T2DM. I found it easy to read and well-presented but have some suggestions to improve.

Specific Comments

Introduction:

1. This was a beautiful introduction, covering the background of miRNAs and their therapeutic potential, as well as the known defects of SMCs in T2DM. Great work!

Thank you for providing positive feedback as well as careful and critical feedback – it is very much appreciated and a good example to set for the early career researchers who feature on this paper.

Methods:

2. Line 90-93 miRNA validation experiments had tissue from lower legs as well as CAGB. This is mentioned in results but it is unclear if the number of samples for T2DM and ND are matched in tissue type. I’d like this to be mentioned here or in results, and then in discussion in case these source tissues are a contributor to differences between groups.

All SMC were isolated from saphenous veins, irrespective of the surgical procedure. During CABG, the saphenous vein is removed from the leg and used to create a bypass graft in diseased coronary arteries, with surplus tissue fragments coming to our laboratory. For the lower limb amputation, the leg was amputated from the patient, the saphenous vein dissected out and then transported to our laboratory. Thus, although the procedures are slightly different, the tissue is the same. In the validation experiments, 3/5 ND-SV-SMC samples were from CABG and 2/5 were from amputation. For the T2DM-SV-SMC, 2/5 were from CABG and 3/5 from amputation thus the numbers in each group were comparable. We have added text to clarify this throughout the manuscript in lines 91, 93-95, 209-212, 287 and 291-293.

3. 105- Only males in this, but some females in validation. I’d like to see sex mentioned in discussion.

Thank you for this suggestion. In our validation experiments, there was one female ND donor and one female T2DM donor. We have amended Figure 3 to show which donors these were; SV-SMC from female donors are now open symbols with male SV-SMC donors in closed symbols. This has also been clarified in the figure legend.

One of our most robust measures of differential phenotype between ND and T2DM donors is the increase in spread cell area as demonstrated in our previous publications [1-3]. Whilst female donors are few, the mean spread cell area of SV-SMC from female donors fits well within the range of male donors (see new supplementary figure S3). From this, we are confident that gender does not impart significant differences on the T2DM SV-SMC phenotype. We have added this supplementary figure and discussion to this effect on line 245.

4. 151- There is no discussion of power calculations done to determine sample size. It would be beneficial to note if the sample size was selected for a specific reason. If this is atypical for this type of study, then please discard my comment.

This was an omission on our part, for which we apologise. We performed a power calculation using the mean and standard deviation of miR-145 in human ND-SV-SMC from our previous publication [2], and input a desired change size of 2-fold. The type I/II error rate alpha was set as 0.05. This revealed that n=5 in each group would be 95% powered to reveal a two-fold statistically significant difference. This information has now been added to the methods in lines 106-109.

Results:

5. Figure 1: Does a volcano plot usually highlight in red the ones that were selected (e.g. above 2 upregulated or below 0.2 downregulated)? It would nice to see if this can be added. The text in the table is small and a bit blurred, can it be clearer?

Thank you for this suggestion. We have coloured the top 30 differentially expressed miRNAs that proceeded to functional analysis in red on the volcano plot. We have also enlarged the table and rearranged the panels to improve the clarity.

6. Line 189 “Table 1. Criteria for selecting candidate miRNAs for further validation”. The table shows the candidates, not the criteria, so please amend the title. Unsure if p values should be truncated to leading digit?

We have truncated the title to ‘Candidate miRNAs selected for further validation’ and clarified in the table legend that the table shows the reason why these miRNAs were selected. We respectfully request to keep the format of the p values as they are to show the difference between them clearly.

7. 198-200 notes the different sources of tissue, and I wonder if this was matched between T2DM and ND. If not, could it be a contributing factor to differences? Please mention numbers of each here (or in methods).

We have addressed this in the response to the previous point regarding lines 90-93.

8. Figure 3: What is %U6? Please note in legend.

U6 is a small nucleolar RNA that is widely used as a housekeeper when analysing miR expression. We have defined this in the legend.

9. Figure 4: Is there a name for this type of figure? If so, please note it so we readers know what they are. It’s beautiful!

Thank you for the positive feedback. As far as we are aware there is no name for this type of figure, it was simply the most accessible way that we thought that the data could be presented. We are very happy that it was appreciated.

Discussion:

10. Line 232-241. I found it a little disordered to introduce changes to these cells from the literature, rather than your results, first. I was wondering why this was being discussed here without clearly linking to finding(s). I feel as though you should present your findings regarding cytoskeleton and then say these are likely to be additional causes of the phenotypic abnormalities to miR-145.

We have added a new sentence towards the beginning of the Discussion briefly highlighting our new findings (lines 243-245). We have kept the original text of lines 232-241 immediately after this as we believe it is important to remind the reader of the features of the persistently aberrant phenotype that are evident in the literature, as the miRNAs that we have identified are clearly linked to all of these features.

11. 243-246: This is the methods, and covered in results, so not needed in discussion.

Thank you, we have removed this text.

12. Line 254- It would be best to move the statement of MiR being up in T2DM to the start of this paragraph, then explain what this might mean in terms of known/predicted functions.

We have moved the sentence highlighting the observed change in the current study to the start of the paragraph as requested, and have repeated this action for miRNAs -30, -130b and -491.

13. Please mention sex in discussion, as the initial dataset had no females (although validation did).

We have addressed this in the response to the previous point regarding line 105.

14. I’d also like to see the potential impact of different tissue sources discussed, at least briefly.

We have addressed this in the response to the previous point regarding lines 90-93.

Conclusion:

15. 325-331 The limitations should be moved to discussion, not remain here in the conclusion

Thank you for the suggestion, we have moved the text as requested.

16. Supp Fig 2: I really like the colour coding here, nice detail for a supplementary figure!

Thank you again for your combination of positive and critical feedback.

Reviewer 2 Report

Dear Editor,

The article by Alisah Hussain et al. titled "Defining the miRname of Saphenous Vein Smooth Muscle Cells from Patients with Type 2 Diabetes Mellitus" presents an intriguing experimental design, but there are several limitations that warrant clarification:

1. The sample size of the analyzed population is small and should be increased to enhance the statistical power. A larger and more diverse sample would allow for a more comprehensive understanding of miRNA expression patterns in saphenous vein smooth muscle cells of patients with type 2 diabetes mellitus (T2DM).

2. The characteristics of the diabetic patients included in the study should be thoroughly reported and discussed. Given the known heterogeneity within the diabetic population, it is crucial to assess and discuss any significant differences in miRNA expression profiles among diabetic patients. Factors such as disease duration, glycemic control, and comorbidities may influence miRNA expression and should be carefully considered.

3. It would be valuable to explore whether the expression of the investigated miRNAs can be modulated following specific treatments, such as insulin or hyperglycemia, for instance. 

4. To further elucidate the functional role of the identified miRNAs, specific target proteins expression regulated by these miRNAs is suggested to be evaluated. Validating the predicted target proteins and assessing their expression levels could help confirm the proposed functional roles of the miRNAs discussed in the article.

Addressing these limitations would enhance the scientific rigor and clinical relevance of the study, contributing to a more comprehensive understanding of the role of miRNAs in T2DM pathophysiology.

Author Response

Reviewer 2

The article by Alisah Hussain et al. titled "Defining the miRnome of Saphenous Vein Smooth Muscle Cells from Patients with Type 2 Diabetes Mellitus" presents an intriguing experimental design, but there are several limitations that warrant clarification:

1. The sample size of the analyzed population is small and should be increased to enhance the statistical power. A larger and more diverse sample would allow for a more comprehensive understanding of miRNA expression patterns in saphenous vein smooth muscle cells of patients with type 2 diabetes mellitus (T2DM).

We agree that increasing our n numbers would provide a more comprehensive understanding, however our sample size was adequate to detect significant differences between the two groups. We had performed a power calculation using the mean and standard deviation of miR-145 expression in human ND-SV-SMC, and input a desired change size of 2-fold. The type I/II error rate alpha was set as 0.05. This revealed that n=5 in each group would be 95% powered to reveal a statistically significant difference. We had missed this information initially from the manuscript, but have now added it to the methods in lines 106-109. We also emphasise in lines 335-338 that given high variability that is inherent in tissue and cells from different donors, higher n numbers would further strengthen the novel findings that we present here.

2. The characteristics of the diabetic patients included in the study should be thoroughly reported and discussed. Given the known heterogeneity within the diabetic population, it is crucial to assess and discuss any significant differences in miRNA expression profiles among diabetic patients. Factors such as disease duration, glycemic control, and comorbidities may influence miRNA expression and should be carefully considered.

Thank you for this suggestion. The limited data we have available to us refers to the presence or absence of cardiovascular disease, and some data on medications.

The RNA samples for the array dataset were taken from SV-SMC from patients undergoing CABG at Leeds General Infirmary. Their clinical characteristics were captured within Table 2 of our previous publication [2]. All patients within this cohort had overt cardiovascular disease and were undergoing CABG surgery. There was no difference in circulating LDL-C or creatinine between non-diabetic and T2DM donors. The frequency of prescriptions for statins, ACE inhibitors or angiotensin II receptor blockers, beta blockers, antiplatelet / anticoagulants and diuretics was similar between the two groups. For those with T2DM, the majority were on metformin therapy with 21.4% being supplemented with sulfonylureas, 7.1% supplemented with gliptins and 30.8% having concurrent insulin therapy. One of our most robust and reliable measures of the T2DM-SV-SMC phenotype is enlarged spread cell area. From this previous dataset which had much higher n numbers, we were able to stratify spread cell area according to treatment for glycaemic control. There was no significant difference in spread cell area for patients on diet therapy or those taking oral medications or insulin (see new supplementary figure S3). It seems that the only discriminator for increased spread cell area is a diagnosis of T2DM and we allude to this in lines 245-246.

The RNA samples for the validation experiments were taken from SV-SMC from patients undergoing CABG at Castle Hill Hospital or undergoing lower limb amputation at the Bradford Royal Infirmary. Similar to the array cohort, all had evidence of cardiovascular disease, and the prescription of cardiovascular medications was similar between the two groups (see new Table 1). Given the low n number in each group, statistical analysis was not possible. However, given the similarity with the previous larger cohort in the 2014 study, we are confident that cardiovascular medications have not played a role in the variation in miR expression in the current study. We have also added a sentence to the discussion highlighting that disease duration and comorbidities would add to the new knowledge presented in the paper (lines 339-341).

3. It would be valuable to explore whether the expression of the investigated miRNAs can be modulated following specific treatments, such as insulin or hyperglycemia, for instance.

This would be an interesting aspect to study. Indeed, for miR-143 and miR-145 we previously assessed this by culturing cells in hyperglycaemia or hyperinsulinaemia, or in the presence of proinflammatory cytokines interleukin-1alpha and tumour necrosis factor, or transforming growth factor beta (all of which are known to be upregulated in the circulation of T2DM patients) [2]. However, we found that only TGFb had any impact on miR expression and importantly this was transient and the stimulatory effect was lost once the stimulus was removed. One of the most interesting aspects of the T2DM-SV-SMC phenotype is its persistence in culture – enlarged spread cell area and altered miR expression are maintained across multiple passage in culture [3], even when the growth conditions of the cells is identical between those from ND and T2DM donors. We have added discussion of this on lines 353-357.

4. To further elucidate the functional role of the identified miRNAs, specific target proteins expression regulated by these miRNAs is suggested to be evaluated. Validating the predicted target proteins and assessing their expression levels could help confirm the proposed functional roles of the miRNAs discussed in the article. Addressing these limitations would enhance the scientific rigor and clinical relevance of the study, contributing to a more comprehensive understanding of the role of miRNAs in T2DM pathophysiology.

We agree that it would be beneficial to validate in vitro the in silico predicted targets. However, we unfortunately do not have the funds or resources to carry out these extra experiments. We have discussed the cellular evidence of altered epigenetic and transcriptional control, cell cycle and survival, cytoskeletal organisation and changes in cell signalling in the discussion in lines 319-334 which supports the functions highlighted in the predicted target analysis, and have described how the lack of protein data is a weakness in this study in line 341-343 and 356-357.

1. Madi, H.A.; Riches, K.; Warburton, P.; O'Regan, D.J.; Turner, N.A.; Porter, K.E. Inherent differences in morphology, proliferation, and migration in saphenous vein smooth muscle cells cultured from nondiabetic and Type 2 diabetic patients. Am J Physiol Cell Physiol 2009, 297, C1307-1317, doi:10.1152/ajpcell.00608.2008.
2. Riches, K.; Alshanwani, A.R.; Warburton, P.; O'Regan, D.J.; Ball, S.G.; Wood, I.C.; Turner, N.A.; Porter, K.E. Elevated expression levels of miR-143/5 in saphenous vein smooth muscle cells from patients with Type 2 diabetes drive persistent changes in phenotype and function. J Mol Cell Cardiol 2014, 74, 240-250, doi:10.1016/j.yjmcc.2014.05.018.
3. Riches, K.; Warburton, P.; O'Regan, D.J.; Turner, N.A.; Porter, K.E. Type 2 diabetes impairs venous, but not arterial smooth muscle cell function: possible role of differential RhoA activity. Cardiovasc Revasc Med 2014, 15, 141-148, doi:10.1016/j.carrev.2014.02.005.

Round 2

Reviewer 2 Report

Dear Editor,

the English need just minor editing.