Reversible Thiol Oxidation Increases Mitochondrial Electron Transport Complex Enzyme Activity but Not Respiration in Cardiomyocytes from Patients with End-Stage Heart Failure

Round 1

Reviewer 1 Report

The authors studied the impact of reversible thiol oxidation on myocardial mitochondrial function in patients with HFrEF. The authors found that mitochondrial function can be modulated through reversible thiol oxidation, but other components of mitochondrial energy transfer are rate limiting in end-stage HFrEF. The work is well done, however there are some open questions that need to be addressed before acceptance of the manuscript.

1. There is a lack of a full characterization of the HFrEF patients. The authors have to provide a table with all echocardiography and hemodynamic parameters from these patients in the manuscript.

2. Since there is a reversible thiol oxidation in the mitochondria, which mitochondrial proteins are oxidized in HFrEF patients and responsible for mitochondrial dysfunction.

3. Would be even more innovative if the authors can show cysteines that are oxidized in HFrEF patients and could be potentially involved in mitochondrial dysfunction. This can be achieved through the use of mass spectrometry.

4. the authors used DTT to reverse the mitochondrial changes, DTT reduce the disulfide bonds of proteins, but what about other oxidative post-translational modifications in the mitochondrial proteins such as S-glutathionylation or S-nitrosylation? Would these post-translational modifications not be partially involved in the mitochondrial dysfunction observed in HFrEF patients? If so what would be the contribution of the disulfide bonds to the global cardiac function in HFrEF patients compared to other oxidative post-translational modifications?

Author Response

The authors studied the impact of reversible thiol oxidation on myocardial mitochondrial function in patients with HFrEF. The authors found that mitochondrial function can be modulated through reversible thiol oxidation, but other components of mitochondrial energy transfer are rate limiting in end-stage HFrEF. The work is well done, however there are some open questions that need to be addressed before acceptance of the manuscript.

 

1. There is a lack of a full characterization of the HFrEF patients. The authors have to provide a table with all echocardiography and hemodynamic parameters from these patients in the manuscript.

We have included additional clinical echocardiography parameters (right ventricular and diastolic function) that are now detailed in Table 2 as shown below.

 

Table 2. Echocardiography and hemodynamics variables from patients included in the study

	All Patients (n = 9)
EF (%)		17.2 ± 8.3
LVIDd (cm)		8.0 ± 1.2
LVIDs (cm)		7.3 ± 1.3
RVH (yes/no)		(4/5)
RVSP (mmHg)		42.4 ± 14.2
E wave (cm/s)		78.1 ± 25.6
A wave (cm/s)		61.4 ± 37.5
E’ (cm/s)		8.19 ± 2.49
DT (s)		155.2 ± 65.5
E/A		1.59 ± 0.78
E/E’		9.99 ± 3.21
E/DT (cm/s2)		0.62 ± 0.38
Diastolic dysfunction Grade I/II/III (n, %)		1 (11)/2 (22)/1 (11)

EF, ejection fraction; LVIDd, left ventricular internal diameter during diastole; LVIDs, left ventricular internal diameter during systole; RVH, right ventricular hypertrophy determined by clinical assessment as ‘yes’ or ‘no’; RVSP, right ventricular systolic pressure; DT, deceleration time. Diastolic function was considered normal in 2 patients and not assessed in 3 patients due to severe systolic dysfunction.

 

2. Since there is a reversible thiol oxidation in the mitochondria, which mitochondrial proteins are oxidized in HFrEF patients and responsible for mitochondrial dysfunction. Would be even more innovative if the authors can show cysteines that are oxidized in HFrEF patients and could be potentially involved in mitochondrial dysfunction. This can be achieved through the use of mass spectrometry.

 

A previous study has completed redox proteomics of reversible thiol oxidation and identified proteins in mitochondrial ETS complex I and II that are reversibly oxidized in left ventricle biopsies of patients with end-stage HFrEF (Tomin et al Int J Mol Sci 2021, 22, doi:10.3390/ijms22041787). We consider these are the likely candidates and have specified in the introduction and discussion

 

Page 2, para 2: “Redox proteomics in left ventricle biopsies from patients with end-stage HFrEF reveal increases in reversibly oxidized cysteine residues in proteins of complexes I and II of the electron transport system (ETS) in cardiac mitochondria [14]…. Therefore, the mitochondrial limitations in patients with end-stage HFrEF may be caused to oxidative modifications of thiol redox switches, such as those found in complex I and II [14], that impair mitochondrial protein function.”

 

Page 10 (last para) – Page 11 (1^st para): “Several processes within the mitochondria are regulated by redox switches which have been recently reviewed [12,13]. Specifically, several studies have investigated the role of thiol oxidation on mitochondrial function, predominately investigating the role of glutathionylation of cysteine residues [34-36]. Cysteines are commonly observed near active sites of enzymes because of their ability to stabilize Fe-S clusters. Two cysteines on the 75 kDa subunit of complex I (NDUSFI) (Cys⁵³¹ and Cys⁷⁰⁴) are particularly sensitive to redox reactions that can impair enzyme function [37-39]. While specific cysteines were not investigated in the current study, the NDUSFI subunit experienced significant oxidative shifts in myocardial biopsies from patients with HFrEF compared to controls in a previous study [14]. It follows that redox reactions on these residues may relate to impaired function that can be targeted with a reducing agent.”

 

4. the authors used DTT to reverse the mitochondrial changes, DTT reduce the disulfide bonds of proteins, but what about other oxidative post-translational modifications in the mitochondrial proteins such as S-glutathionylation or S-nitrosylation? Would these post-translational modifications not be partially involved in the mitochondrial dysfunction observed in HFrEF patients? If so what would be the contribution of the disulfide bonds to the global cardiac function in HFrEF patients compared to other oxidative post-translational modifications?

 

This is an interesting point. It’s worth noting that DTT reduces all reversible thiol oxidations and not only disulfide bonds. Other irreversible oxidative post-translational modifications may certainly be involved in the mitochondrial dysfunction that contributes to cardiomyopathy in HFrEF. We have modified the text to acknowledge this point.

 

Page 13 (Limitations): “It is not possible to distinguish the impact of other oxidative modifications, which are not acutely reversed by DTT, on mitochondrial function. These would require chronic treatment with antioxidants in vivo.”

Reviewer 2 Report

The work presents too few data to sustain the final hypothesis. It will be better to compensate for some aspects the Authors highlighted in the "limitations" section. The organization of the different sections of the paper (introduction, materials and methods, results and discussion) must be improved.

Author Response

The work presents too few data to sustain the final hypothesis. It will be better to compensate for some aspects the Authors highlighted in the "limitations" section. The organization of the different sections of the paper (introduction, materials and methods, results and discussion) must be improved.

 

We agree that the number of patients included in this study is relatively small in the context of a clinical trial. But this is not a clinical trial. We completed the study to test a basic science concept in human samples. Hence, there is inherently few data to support the hypothesis. However, the data that we have, being collected in left ventricle transmural biopsies, is highly significant to resolve the role of reversible thiol oxidation on mitochondrial respiration dysfunction in patients with end-stage HFrEF.

 

We revised the manuscript and sections to add new information on echocardiography parameters and comparisons between diabetic and non-diabetic patients. We are unable to provide a point-by-point response regarding the organization because the reviewer did not provide specific comments and suggestions for improvement.

Reviewer 3 Report

How can be explained the difference between the percent of CAD - 5 pts - and the ischemic aetiology of the cardiomiopathy - 1 pts-?

Is any different analysis regarding the diabetic vs non-diabetic patients?

Author Response

How can be explained the difference between the percent of CAD - 5 pts - and the ischemic aetiology of the cardiomiopathy - 1 pts-?

 

Coronary artery disease that has not progressed to the point of plaque formation and arterial blockage can occur without causing an ischemic event (infarction), whereas heart failure still results from non-ischemic origins.

 

Is any different analysis regarding the diabetic vs non-diabetic patients?

 

We have included additional analysis on differences between diabetic and non-diabetic patients in this revised version of the manuscript (supplemental figure 1). There were no observable differences for the effects of DTT between patients with and without diabetes mellitus.

 

Supplementary Figure 1 Comparisons between patients with and without diabetes mellitus. ΔDTT (DTT-treated samples – untreated samples) for mitochondrial parameters assessed in isolated mitochondria (A); and permeabilzed cardiomyocyte bundles (B-D). There were no observable differences for the effects of DTT between patients with and without diabetes for (A) mitochondrial ETS enzyme activities, (B) OXPHOS Conductance, (C) maximal respiration, or (D) ADP sensitivity. Groups compared made via independent t-test.

Reviewer 4 Report

The authors present interesting original ex-vivo data on the involvement of cardiac mitochondrial respiratory function in patients with Heart Failure with reduced Ejection Fraction, and how it can be modulated by using a reducing agent.

One of the major limitations, as the authors themeselves declare, is the complete absence of a control group. On one hand, this could be easily solved, by using specimens from patients undergoing surgery for other reasons and therefore not having HFrEF, nor any major stroctural heart disease. On the other hand, the authors might easily use a cellular (line) or animal model to compare HF with normal LV function.

A better characterization of the patients included is needed. In fact, nothing is reported regardin diastolic function, valve function, pulmonary hypertension, right ventricular function, drug treatment, and so on...

Even though the study requires extensive lab work for each patient, nine subjects are really few, comparing the known heterogeneity of HF patients.

Did the author try to perform the same experiments on fresh biopsies? Do they have a side by side comparison between fresh and frozen samples? This is key to understand how much results on froze samples reflect those eventulaly obtained on fresh biopsies.

 

Author Response

The authors present interesting original ex-vivo data on the involvement of cardiac mitochondrial respiratory function in patients with Heart Failure with reduced Ejection Fraction, and how it can be modulated by using a reducing agent.

 

Thank you for your positive and constructive criticism

 

One of the major limitations, as the authors themselves declare, is the complete absence of a control group. On one hand, this could be easily solved, by using specimens from patients undergoing surgery for other reasons and therefore not having HFrEF, nor any major structural heart disease. On the other hand, the authors might easily use a cellular (line) or animal model to compare HF with normal LV function.

 

We agree with the reviewer that this is unfortunately one of the major limitations of the study. Control ventricular biopsies could, in theory, only be obtained from healthy organ donors. The challenge is that we completed experiments with transmural biopsies from the left ventricle apex. Such biopsy would only be possible from hearts of brain-dead healthy individuals who do not qualify to serve as donor for a heart transplant. The probability of securing such sample is very low. Then, acquiring the biopsy fresh and having age and sex-matching for true controls is even more difficult.

 

The point of using cell lines or animal models as ‘controls’ is well taken. An animal model would be the most reasonable for our purposes. However, previous studies have resolved the effects of DTT in healthy hearts. Thus, we would have difficulty justifying the experiments with the IACUC. We have rephrased the discussion (limitations) to emphasize the DTT effects or lack thereof in hearts from healthy rodents.

 

Page 13 (Limitations): “In healthy rodent hearts, DTT treatment had no effect [15] or even decreased [16] myocardial mitochondrial function, suggesting that any effects of DTT in our study would arise from reversing the oxidized thiol shift seen with end-stage HFrEF [25].”

 

A better characterization of the patients included is needed. In fact, nothing is reported regarding diastolic function, valve function, pulmonary hypertension, right ventricular function, drug treatment, and so on...

 

We have included additional information on echocardiography and hemodynamics variables in Table 2 to address this point.

 

Table 2. Echocardiography and hemodynamic variables from patients included in the study

	All Patients (n = 9)
EF (%)		17.2 ± 8.3
LVIDd (cm)		8.0 ± 1.2
LVIDs (cm)		7.3 ± 1.3
RVH (yes/no)		(4/5)
RVSP (mmHg)		42.4 ± 14.2
E wave (cm/s)		78.1 ± 25.6
A wave (cm/s)		61.4 ± 37.5
E’ (cm/s)		8.19 ± 2.49
DT (s)		155.2 ± 65.5
E/A		1.59 ± 0.78
E/E’		9.99 ± 3.21
E/DT (cm/s2)		0.62 ± 0.38
Diastolic dysfunction Grade I/II/III (n, %)		1 (11)/2 (22)/1 (11)

EF, ejection fraction; LVIDd, left ventricular internal diameter during diastole; LVIDs, left ventricular internal diameter during systole; RVH, right ventricular hypertrophy determined by clinical assessment as ‘yes’ or ‘no’; RVSP, right ventricular systolic pressure; DT, deceleration time. Diastolic function was considered normal in 2 patients and not assessed in 3 patients due to severe systolic dysfunction.

 

 

Even though the study requires extensive lab work for each patient, nine subjects are really few, comparing the known heterogeneity of HF patients.

 

In general, we agree with the reviewer that our sample size would be small from a ‘clinical’ standpoint considering the heterogeneity of HF patients. However, our population is restricted to severe/end-stage heart failure which makes the population more homogenous from the cardiovascular standpoint. It is also important to consider that we approached the study seeking mechanistic insights into mitochondrial dysfunction. The co-morbidities are relevant but the reality of this population such that our significant findings in enzymatic assays but lack of significance in mitochondrial respiration come across as more robust.

 

Did the author try to perform the same experiments on fresh biopsies? Do they have a side-by-side comparison between fresh and frozen samples? This is key to understand how much results on froze samples reflect those eventually obtained on fresh biopsies.

 

This is a good point. The most important mechanistic insight from our study is the lack of DTT effect on the fresh biopsy.  We opted to perform the enzymatic assays in frozen samples because these are based on a plate reader such that we could perform all samples simultaneously. The assays, albeit seemingly simple, were complex and time-consuming to implement with several troubleshooting steps that made it unfeasible to complete side-by-side with fresh samples. Thus, we decided to withhold the enzymatic analysis until we had a working protocol for the human heart biopsies.  We modified the discussion (limitations) to acknowledge this point.

 

Page 13 (Limitations): “Some of the discrepant findings in oxidative phosphorylation conductance and respiration vs ETS complex enzyme activity could arise from use of fresh vs frozen samples, respectively. We could not perform ETS enzyme activity assays in fresh samples due to technical challenges.”

Round 2

Reviewer 1 Report

The authors have addressed most of my concerns

Author Response

We thank the reviewer for their time and efforts in reviewing our work.

Reviewer 2 Report

Dear Authors,

I am sorry to say that the manuscript remains not convincing, even if you performed very precise and elaborate experiments. Your hypothesis was that the treatment with DTT could reverse the oxidation of thiols in mitochondrial complexes, rescueing the reduced mitochondrial respiration in HFrEF patients. Unfortunately, you found out an increase in complex I and III activity, but not followed by a consequent increase in respiration, in oxphos conductance and in ADP kinetics. Thus, accepting your ideas about the discrepancy of your data from other works (stated at page 12, before the limitation section), this means you are not in the right experimental conditions to demonstrate your hypothesis. Your work, detecting only the increase in Complex I and III activity after DTT treatment, is not sufficient to confirm that modulation of reversible thiol oxidation is important to modulate mitochondrial respiration in HFrEF patients.

Moreover, data are not adequately discussed in the results section, with clear reference to fig. A, B or C and to the meaning of the results; figure 1 and table 3 are discussed in the materials and methods section and almost none in the results section;  I cannot see the CAD data in table 1; Figure 2 is not distinguished in A and B;  the discussion is long and focused on results obtained in other papers with the techniques used in this manuscript, but unfortunately this is not the initial hypothesis to demonstrate; reference to metabolic alterations in HFrEF in the manuscript are out of focus respect the main topic of the manuscript.

Author Response

We thank the reviewer for their thorough evaluation of our work and agree with many of their identified limitations. However, despite these limitations, we believe the results of our study provide valuable information regarding redox modulation of mitochondrial function (effects on individual complex function but no translation to respiration) and lay the groundwork for future investigations to attempt to clarify some of the observed discrepancies. To not publish these results simply because they did not meet our initial hypothesis would be inappropriate and limiting to other scientific groups investigating similar fields.

Our aim of the Results section was to simply present the findings in an unbiased manner. Our aim in the Discussion section was to frame our complicated results in the context of previous studies (redox modulation of mitochondrial function; and energetics in the failing myocardium) which required a lengthier approach. We apologize for the arrangement of figures and tables, these have been adjusted in the revised text.