Deficits in Mitochondrial Spare Respiratory Capacity Contribute to the Neuropsychological Changes of Alzheimer’s Disease

Round 1

Reviewer 1 Report

This is a very  interesting paper in which generates data that has considerable potential in the the early diagnosis of AD and for developing targeted therapies, an issue that is obviously of huge clinical relevance. I am very much in support of the authors aims.

My concern with the paper is that the authors make huge interpretative demands of a sample that is very small for the type of analysis they undertake-namely regression analysis. I did not find evidence of Power calculations for the study. I am pleased that they undertook adjustments for the significance levels.

Throughout, the authors present data they refer to as correlation data but report r-squared rather than r-values. They also do not make it clear whether they are presenting adjusted r-squared values.  In addition Figure 3 graphs have what I would regard as regression lines-this is not appropriate for correlation analysis.  I would state that the authors need to clearly distinguish when correlation or regression analysis is being undertaken. I am satisfied that the authors undertake covariate adjustment, but this is a very risky strategy for samples that are as small as these and should be undertaken in a very limited manner.

The authors should not state "no correlation" but "no significant correlation"

The data presented is very interesting, researching AD in any of its forms is very challenging, and large samples are often difficult to obtain for studies such as this. I would ask that the authors report this as a proof of concept or pilot study or stress in more detail their limitations 

Author Response

Response to Reviewer 1

We thank the Reviewer for their comments and have listed below how we have addressed each point made.

My concern with the paper is that the authors make huge interpretative demands of a sample that is very small for the type of analysis they undertake-namely regression analysis. I did not find evidence of Power calculations for the study. I am pleased that they undertook adjustments for the significance levels.

We thank the Reviewer for their comment. To address the point they make we have referred to the study as a proof-of-concept study throughout the manuscript. This is to highlight the preliminary phase we are at in characterising the “mitochondrial phenotype” of AD and how this links to clinical variables. Studies like this which show positive findings in spite of a relatively small sample size are essential to encourage the design of larger studies which require more resources in terms of funding and commitment. On this note, we recognise that it is important to interpret our findings accordingly. For this reason, we have removed any direct or indirect reference to the application of regression models (see also response to the subsequent comment). In the limitations paragraph of the discussion (starts at line 370) we have also highlighted the need for both bigger cohorts of patients and replication of these findings by other groups before exploring fibroblast metabolism as an Alzheimer’s disease biomarker. This is also the reason we had not added a power calculation.

In conclusion, we agree with the Reviewer on the need to tone down our interpretative remarks and to achieve this, in this updated version of the manuscript we have referred to the study as a proof of concept investigation, which highlights the “preliminary” nature of the findings and contribution to demonstrating “feasibility” of this approach.

Throughout, the authors present data they refer to as correlation data but report r-squared rather than r-values. They also do not make it clear whether they are presenting adjusted r-squared values.  In addition Figure 3 graphs have what I would regard as regression lines-this is not appropriate for correlation analysis.  I would state that the authors need to clearly distinguish when correlation or regression analysis is being undertaken. I am satisfied that the authors undertake covariate adjustment, but this is a very risky strategy for samples that are as small as these and should be undertaken in a very limited manner.

1. We thank the Reviewer for their comments about the clarity of which statistical test has been used. Overall, we agree with the idea that it is important to be consistent throughout the paper with the use (and the description) of the statistical methods. We have now removed all references to regression analyses. The confusion was due to the addition of the line (together with the error curves) in Figure 3, which in our view served to highlight the association between the variables on axes x and y, to the benefit of the viewer. In this updated version of the manuscript, however, this has been removed, to avoid any misunderstanding. Moreover, the choice of performing exclusively correlation analyses was also not to over-interpret the association between the neuropsychological tests and metabolic parameters, as in correlation models x and y are mathematically interchangeable and the model does not imply any statistical directionality (as a regression would do). Moreover, R-squared values have been removed and replaced with r-values. Changes are highlighted with track changes between lines 255 and 277 in the manuscript.

The authors should not state "no correlation" but "no significant correlation"

1. We thank the Reviewer for their comment and have changed the wording as suggested on line 297, 333, 340, 349, and 362 of the revised manuscript.

The data presented is very interesting, researching AD in any of its forms is very challenging, and large samples are often difficult to obtain for studies such as this.

1. We thank the Reviewer for their observation.

I would ask that the authors report this as a proof of concept or pilot study or stress in more detail their limitations 

1. We are happy to make this change and have referred to the study as a” proof-of-concept report” on line 123, and 301 of the revised manuscript.

Author Response File: Author Response.docx

Reviewer 2 Report

My life work has been on Gly-His-Lys:Cu 2+, a human plasma constituent that declines sharply with age. It appears to strongly control or influence about 31% of the human genes. GHK-Cu does increase carbon dioxide production in live mice and isolated mouse liver slices. Gene expression studies also strongly suggest that it increases mitochondrial activity. The gene expression data is from the Broad Institute at MIT and Harvard, Cambridge, MA, USA,

GHK-Cu is used by physicians in the USA for the healing of spinal-cervical damage when injected near the injury at 0.25 mgs per day.

Russian studies report that GHK-Cu's most potent effects in rats as reductions in anxiety and pain-induced aggression.

A recent paper reported that GHK-Cu improves mental function in aged mice. 

The potential of GHK as an anti-aging peptide Yan Dou, Amanda Lee, Lida Zhu, John Morton, Warren Ladiges Aging Pathobiology and Therapeutics 2020; 2(1):58-61 58    Recent papers on GHK-Cu are  Pickart, L., & Margolina, A. (2018). Regenerative and protective actions of the GHK-Cu peptide in the light of the new gene data. International journal of molecular sciences, 19(7), 1987.   Pickart, L., Vasquez-Soltero, J. M., & Margolina, A. (2017). The effect of the human peptide GHK on gene expression relevant to nervous system function and cognitive decline. Brain sciences, 7(2), 20.   My main advice to the authors is to buy some of the sublingual GHK-Cu, 5 mgs and give it to the patients. Direct-Peptides sells it in the UK. GHK-Cu is probably safer than water.  

Author Response

Response to Reviewer 2

We thank the reviewer for the positive judgement of the paper and the authors will investigate and carefully consider GHK as a potential therapeutic strategy in our future work. As the reviewer mentions, studies have shown that GHK, by increasing gene expression, has an effect on improving mitochondrial activity and influencing cognitive decline (Pickart et al 2017). We have now mentioned in the introduction that there is evidence that GHK can influence mitochondrial activity and cognitive decline and referenced the suggested paper. This change can be found from line 113 to line 115 of the revised manuscript

Reference

Pickart, L., Vasquez-Soltero, J. M., & Margolina, A. (2017). The effect of the human peptide GHK on gene expression relevant to nervous system function and cognitive decline. Brain sciences, 7(2), 20.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have addressed all my previous concerns