Mitochondrial Glrx2 Knockout Augments Acetaminophen-Induced Hepatotoxicity in Mice

Round 1

Reviewer 1 Report

In the manuscript ID Antioxidants-183253 entitled “Mitochondrial Glrx2 knockout augments acetaminophen-induced hepatotoxicity in mice”, the authors reported the importance of Grx2 in APAP-induced hepatotoxicity.　
While some of the observations are interesting, this work is preliminary and the manuscript remains very descriptive.

There are some concerns about the experiments:

 

Major point

1. Does APAP treatment increase the expression of Glrx2? It is necessary to show the expression of Glrx2 induced by APAP treatment.

2. Are mitochondria-localized Grx2 and cytoplasmic Grx1 affected each other? Is there a complementary response such as overexpression of Grx1?

3. Grx2 locates in mitochondria, are there any changes in mitochondria by Glrx2 knockout?

4. There is an assessment of GSH, but is actual oxidative stress elevated? For example, is there an increase in ROS?

 

Minor points

1. What does "arrow" mean in Fig. 1D?

2. Are there two bands in Glrx2-/- APAP+ lane in Fig2C? Why?

3. In Fig4, which band is the target one? Need to quantify the bands by densitometry in Western blotting figure.

4. In Fig5, the localization of AIF is not clear. Higher quality and magnified photographs are needed. It is necessary to distinguish between nuclear and mitochondrial localization with more magnified images. (AIF locates in mitochondria under normal condition?) Proof of nucleoprotein translocation by western blotting is expected if possible. Or with other apoptosis markers?

5. In Fig7, where does the cytoplasmic ROS come from? Is mitochondrial ROS coming from mitochondria? Which is predominant, mitochondrial ROS or cytoplasmic ROS? Grx1 should be described in the figure.

Author Response

In the manuscript ID Antioxidants-183253 entitled “Mitochondrial Glrx2 knockout augments acetaminophen-induced hepatotoxicity in mice”, the authors reported the importance of Grx2 in APAP-induced hepatotoxicity.

While some of the observations are interesting, this work is preliminary and the manuscript remains very descriptive.

 

>>>Thank the reviewer for this comment. Acetaminophen-induced hepatotoxicity has been well studied and known to be related to the reaction with GSH. But the role of other components in the glutathione system are not well understood. Here we showed mitochondrial Grx2 is key player in APAP-induced hepatotoxicity. Since Grx2 is a disulfide/dithiol reductase, we focused on the investigation of cellular thiol regulation change and related events.  The results demonstrated that extensive oxidative stress occurred in the deficiecy of Grx2.

 

 

There are some concerns about the experiments:

Major point

 

1. Does APAP treatment increase the expression of Glrx2? It is necessary to show the expression of Glrx2 induced by APAP treatment.

>>>Thank the reviewer for pointing out this important concern. We have added the protein level of Grx2 after APAP treatment for 24 h in Fig. S1B. APAP treatment caused an increase of the expression of Grx2. This sentence has been added in the page 7, second paragraph, line 292-293.

 

2. Are mitochondria-localized Grx2 and cytoplasmic Grx1 affected each other? Is there a complementary response such as overexpression of Grx1?

>>>Thank the reviewer for this concern. The study of Mailloux and colleagues showed that Grx2 deficiency did not induce a compensatory overexpression of Grx1 in liver. (Julia Chalker et al. Redox Biology, 2018). In this study, we didn’t investigate the interaction between mitochondrial Grx2 and cytosolic Grx1. We addressed this description in the discussion part. (page 12, line 439-441)

 

3. Grx2 locates in mitochondria, are there any changes in mitochondria by Glrx2 knockout?

>>> Previous study displayed morphological and functional alters in Glrx2 deficient liver mitochondria. Glrx2 deficient liver mitochondria showed a changed morphology with the stretched outer membrane and curving and inner membrane forming structures like a “c” or donut shape. Also, a more electron dense mitochondrial matrix was found in Glrx2 deficient liver. Functionally, Glrx2 deficient liver mitochondria showed decreased mitochondrial activity characterized by reduced membrane potential and oxygen consumption rates, but an elevated sensitivity to mitochondrial permeability transition. (Valeria Scalcon et al. Redox Biology, 2022). This has been added in the discussion part (page 12, line 445-446). Since this concern has been well investigated before, it was not explored in this study.

 

4. There is an assessment of GSH, but is actual oxidative stress elevated? For example, is there an increase in ROS?

>>>Both total GSH and GSH/GSSG ratio were assessed in this manuscript. Under physiological condition, >99% of the total glutathione pool is in the reduced form (GSH). When oxidative stress occurs, GSSG accumulates and the ratio of GSSG to GSH increases. GSH/GSSG ratio has been widely accepted to be an indicator of oxidative stress. Therefore, we used total GSH, and GSH/GSSG ratio and Trx oxidation state as the markers of cellular redox state.  Of course, it will be better if we provide the result of increased ROS level. But we do not have efficient and useful approaches to detect ROS level in the animal so far.

 

Minor points

 

1. What does "arrow" mean in Fig. 1D?

>>> “Arrows point to necrotic hepatocytes” has been added in Fig.1 legend.

2. Are there two bands in Glrx2-/- APAP+ lane in Fig2C? Why?

>>> Proteins expression were detected after APAP insult, Trx2 showed two bands. This is very interesting, we do not have an exact explanation for this observation. This may indicate that APAP exposure of Glrx2^-/- mice caused highly oxidization of Trx2 so that structure of part of the protein had been changed and could not be completely reduced into initial reduced form by DTT in loading buffer. For example, part of Trx2 had some type of posttranslational modification in the highly oxidative mitochondrial environment. We added this explanation in Page 7, line 301-306.

 

3. In Fig4, which band is the target one? Need to quantify the bands by densitometry in Western blotting figure.

>>> Fig4 showed the redox state of Trx1/2. There are free thiols and oxidized disulfide (Trx-S-S-protein) contained in the structure of Trx1/2, so we used iodoacetamide (IAM) to react with free thiols avoiding Trx oxidation in the process of sample treatment. In the non-reducing gels, the thiols in reduced form of Trx1/2 were alkylated by IAM, which made them appear in the low molecular weight positions (~12 KDa). When Trx1/2 were oxidized, they would form disulfides with other proteins or themselves, appearing in the high molecular weight positions or as fuzzy bands. This explanation has been added in page 8-9, line 337-343. We also have added arrows which pointed to typical oxidized form of Trx1/2. And the bands of oxidized and reduced Trx1/2 have been quantified by Image J and corresponding result was provided in fig S2.

 

4. In Fig5, the localization of AIF is not clear. Higher quality and magnified photographs are needed. It is necessary to distinguish between nuclear and mitochondrial localization with more magnified images. (AIF locates in mitochondria under normal condition?) Proof of nucleoprotein translocation by western blotting is expected if possible. Or with other apoptosis markers?

>>> Thank the reviewer for this suggestion. An inset with higher magnification has been included in fig 5 to show the difference among these groups. Under physiological conditions, AIF has an N-terminal mitochondrial localization signal that is cleaved off to generate a mature protein of 57 kDa after import into the mitochondrial intermembrane space. After some stimuli, AIF is released from mitochondria and translocated to the nucleus. Relative description has been added in page 9, line 362-366. Here, we could compare the level of AIF in nucleus from current figures and find that more AIF were translocated to nucleus in APAP treated Glrx2^-/- mice with decreased fluorescence signal in the intracellular section beyond nucleus. But we did not have western blotting the data about the translocation of AIF and other apoptosis markers. This is a limitation of this study, further investigation about the exact role of Grx2, Grx1 and Trx2 in the regulation of AIF translocation and the relationship with apoptosis should be very intriguing, probably via cell experiments.

 

5. In Fig7, where does the cytoplasmic ROS come from? Is mitochondrial ROS coming from mitochondria? Which is predominant, mitochondrial ROS or cytoplasmic ROS? Grx1 should be described in the figure.

>>> Thank the reviewer for this suggestion. Fig 7 has been redrawn which included two figures of separate pathways related to the wild-type and the knockout mice, and showed the difference between the two groups. Grx1 has been included in the figure. Trx and GSH systems in cytosol and mitochondria acts as important antioxidant systems to scavenge ROS and keep redox balance in cytosol and mitochondria respectively. Thus, when TrxR activity was inhibited and GSH amount was reduced, the balance between ROS and antioxidants was disrupted and ROS were accumulated. Cytoplasmic ROS may be generated from NADPH oxidases, or some flavoproteins, etc. Mitochondria is also a main site to produce ROS. In this study we did not clarify the predominance of mitochondrial and cytoplasmic ROS. Here we focus on the ROS contributed by mitochondria due to the Grx2 deficiency. The enhance of ROS production in mitochondria induced by APAP  thus could increase the total cellular ROS level, including in cytosol.

 

Reviewer 2 Report

The paper “Mitochondrial Glrx2 knockout augments acetaminophen-induced hepatotoxicity in mice” presents an elegant work aimed to clarify the oxidative stress/ antioxidant mechanisms involved in the development of hepatotoxicity caused by overdosing of APAP.  The text is in general well written, experiments are related results are convincing. Only some minor points need attention, as detailed below, making the paper to need a minor revision.

 

Line 146, please check “several hours”, and clarify if blood was allowed to cloth, and thus if separation of surnatant, serum or plasma depending on the use of anticoagulants, was obtained.

 

Lines 94-95, thioredoxin reductase (TrxR), abbreviaton is already given above, lines 49-50

Line 36: “..explore the reactivity of WT and Glrx2-/- mice to APAP…,” is likely better to state: “..explore the acute toxicity of WT and Glrx2-/- mice to APAP…”, please revise

 

Figure 5 Details in pictures in the AIF column are poorly distinguishable, if possible contrast in the picture is to be improved. Also, a description of changes in details and an inset showing a significant part of the pictures at higher magnification could help the reader to better appreciate the differences in labelling;

the values of the bars cannot be read, please improve or add to the legend.  

 

 

Author Response

Comments and Suggestions for Authors

The paper “Mitochondrial Glrx2 knockout augments acetaminophen-induced hepatotoxicity in mice” presents an elegant work aimed to clarify the oxidative stress/ antioxidant mechanisms involved in the development of hepatotoxicity caused by overdosing of APAP.  The text is in general well written, experiments are related results are convincing. Only some minor points need attention, as detailed below, making the paper to need a minor revision.

 

>>> Thank the reviewer very much for these comments and appreciation.

 

 

Line 146, please check “several hours”, and clarify if blood was allowed to cloth, and thus if separation of supernatant, serum or plasma depending on the use of anticoagulants, was obtained.

>>> Thank the reviewer for this comment. The statement “several hours” has been changed to “1 to 2 hours”. There was no anticoagulant used in this process. Blood was collected in clear tubes without anticoagulant, after 1-2 hours stilling, supernatant (serum) was obtained by centrifugating at 2500×g for 20 min.

 

Lines 94-95, thioredoxin reductase (TrxR), abbreviaton is already given above, lines 49-50

>>> “thioredoxin reductase (TrxR)” has been changed to “TrxR”.

Line 36: “..explore the reactivity of WT and Glrx2^-/- mice to APAP…,” is likely better to state: “..explore the acute toxicity of WT and Glrx2^-/- mice to APAP…”, please revise

>>> Thank the reviewer for this suggestion. This sentence has been changed to “To explore the acute toxicity of APAP in WT and Glrx2^-/- mice…”

 

Figure 5 Details in pictures in the AIF column are poorly distinguishable, if possible contrast in the picture is to be improved. Also, a description of changes in details and an inset showing a significant part of the pictures at higher magnification could help the reader to better appreciate the differences in labelling; the values of the bars cannot be read, please improve or add to the legend. 

>>> Thank the reviewer for this suggestion. An inset with higher magnification has been included in fig 5 to show the difference among these groups. A description of changes has been added in line 378-381. Also, the values (50 μm) of the bars and the arrows meaning (AIF in nucleus) have been added in the legend of fig 5.

Reviewer 3 Report

APAP exhibited higher hepatotoxicity in Glrx2(-/-) than the wild type. The authors investigated the mechanisms which could be useful for a better understanding of APAP-induced hepatotoxicity. Reduction of GSH and TrXR by APAP were enhanced in Glrx2(-/-) mice. In addition, the nuclear translocation of AIF increased in APAP-treated Glrx2(-/-) mice. Authors speculated that Grx2 deficiency deteriorated APAP-induced hepatotoxicity by interrupting thiol-redox compensatory response, enhancing the AIF pathway-mediated oxidative damage. The data are descriptive but show the difference in Glrx2(-/-).

 

Points that should be addressed:

1. It is difficult to read Figure 4. Please provide the densitometry results as well as a further explanation.

 

2. In Figure 3 legend, delete ‘**p<0.01;***<0.001’

 

3. Figure 7 is not much helpful for readers. Please provide two figures of pathways corresponding to the wild-type and the knockout mice, so that readers could understand the difference.

Author Response

Comments and Suggestions for Authors

APAP exhibited higher hepatotoxicity in Glrx2(-/-) than the wild type. The authors investigated the mechanisms which could be useful for a better understanding of APAP-induced hepatotoxicity. Reduction of GSH and TrXR by APAP were enhanced in Glrx2(-/-) mice. In addition, the nuclear translocation of AIF increased in APAP-treated Glrx2(-/-) mice. Authors speculated that Grx2 deficiency deteriorated APAP-induced hepatotoxicity by interrupting thiol-redox compensatory response, enhancing the AIF pathway-mediated oxidative damage. The data are descriptive but show the difference in Glrx2(-/-).

 

>>> Thank the reviewer very much for these comments and suggestions.

 

 

Points that should be addressed:

 

1. It is difficult to read Figure 4. Please provide the densitometry results as well as a further explanation.

>>> The bands of oxidized and reduced Trx1/2 have been quantified by Image J and densitometry result was provided in fig S2. The statement “Under non-reducing condition, the free thiols in reduced Trx1/2 were alkylated by IAM to avoid oxidizing of Trx1/2 in processing samples, thus we could find the reduced Trx1/2 in predicted molecular weight (~12 KDa). When Trx1/2 was oxidized, they would form disulfides with proteins (including Trx1/2) making them appearing in the high molecular weight positions or as fuzzy bands. DTT was added to open the disulfide bond between Trx1/2 and proteins to validate that the bands in high molecular weight are specific bands of Trx1/2” has been added to further explain this result.

 

2. In Figure 3 legend, delete ‘**p<0.01;***<0.001’

>>> “**p<0.01;***<0.001” has been deleted from the legend of Figure 3.

 

3. Figure 7 is not much helpful for readers. Please provide two figures of pathways corresponding to the wild-type and the knockout mice, so that readers could understand the difference.

>>>Thank the reviewer for this suggestion. Figure 7 has been redrawn which included two figures of pathways corresponding to the wild-type and the knockout mice, and showed the difference between the two groups.

Round 2

Reviewer 1 Report

Thank you for your revision.

All my concerns are cleared.

Author Response

Review 1

 Comments and Suggestions for Authors

Thank you for your revision.

All my concerns are cleared.

>>>Thank you very much for these comments.

Reviewer 2 Report

The paper “Mitochondrial Glrx2 knockout augments acetaminophen-induced hepatotoxicity in mice” has been revised according to most of the remarks. However, two points are still needing attention, as detailed below. Therefore, the paper in this form still needs attention from the Authors. 

- Literature citations at lines 201-202 are wrong: James [31] in the text, in the Literature is [33], and Roberto [32] could apply to Motterlini, as the family name? Please check also adjacent literature citations.

- The revision of Figure 5 is not fully responding to the need of showing the nuclear translocation of AIF.  Apart that both previous and revised versions of the Figure can be found in the text, the differences between the patterns observed for the different kinds of liver conditions should be described, in the legend or in the related text. Insets at higher magnification provided also in the pictures in the “Merge” column could better help the reader to appreciate the differences in labelling.

- Also, at line 367 “…some stimuli..” is insufficient, the kinds of stimuli is to be specified.

 

Author Response

Reviewer 2

Comments and Suggestions for Authors

The paper “Mitochondrial Glrx2 knockout augments acetaminophen-induced hepatotoxicity in mice” has been revised according to most of the remarks. However, two points are still needing attention, as detailed below. Therefore, the paper in this form still needs attention from the Authors. 

- Literature citations at lines 201-202 are wrong: James [31] in the text, in the Literature is [33], and Roberto [32] could apply to Motterlini, as the family name? Please check also adjacent literature citations.

>>>Thank the reviewer for pointing out these mistakes. For reference 31, the full name of the first author is James D Adams, so “James [31]” in the text has been changed to “Adams JD, et. al. [31]”. Also “Roberto [32]” has been changed to “Motterlini R, et. al. [32]” in the text.

- The revision of Figure 5 is not fully responding to the need of showing the nuclear translocation of AIF. Apart that both previous and revised versions of the Figure can be found in the text, the differences between the patterns observed for the different kinds of liver conditions should be described, in the legend or in the related text. Insets at higher magnification provided also in the pictures in the “Merge” column could better help the reader to appreciate the differences in labelling.

>>>Thank the reviewer for this suggestion. The differences of various kinds of liver conditions have been described in figure 5 legend. And extra “Magnification” panels have been added in figure 5.

- Also, at line 367 “…some stimuli..” is insufficient, the kinds of stimuli is to be specified.

>>> “After some stimuli” at line 367 has been changed to “Upon apoptotic stimuli such as oxidative stress”.

Reviewer 3 Report

The authors responded to the points raised. But one can not see the meaning of the revised Figure 7, which is supposed to explain ‘possible mechanisms underlying Grx2 deficiency-caused deterioration’, but only shows the lack of Grx2. Figure 7 may be deleted entirely.

Author Response

Reviewer 3

Comments and Suggestions for Authors

 

The authors responded to the points raised. But one can not see the meaning of the revised Figure 7, which is supposed to explain ‘possible mechanisms underlying Grx2 deficiency-caused deterioration’, but only shows the lack of Grx2. Figure 7 may be deleted entirely.

>>>Thank the reviewer for this suggestion. Figure 7 has been deleted entirely.

Round 3

Reviewer 2 Report

 

The paper “Mitochondrial Glrx2 knockout augments acetaminophen-induced hepatotoxicity in mice” has been duly  revised and corrected according to the remarks. Only minor corrections to the text are still suggested, as detailed below.  

- Literature citations at lines 201-202, now corrected.

- The revision of Figure 5 is now really illustrative of the AIF absence or accumulation in the nuclei. At line 378 the sentence: ”… accumulated in nuclei and thus showing the merged color (indicated by red arrows).” Spuld be better as: “accumulated in nuclei as shown by the merged colors (indicated by red arrows).”

- Ex line 367, now line 365, correction is OK, but “figure”, is to be specified with number, or removed.  

 

Author Response

The paper “Mitochondrial Glrx2 knockout augments acetaminophen-induced hepatotoxicity in mice” has been duly  revised and corrected according to the remarks. Only minor corrections to the text are still suggested, as detailed below. 

- Literature citations at lines 201-202, now corrected.

>>>Thank you for this comment.

 

- The revision of Figure 5 is now really illustrative of the AIF absence or accumulation in the nuclei. At line 378 the sentence: ”… accumulated in nuclei and thus showing the merged color (indicated by red arrows).” Spuld be better as: “accumulated in nuclei as shown by the merged colors (indicated by red arrows).”

 

>>>Thank the reviewer for this suggestion. Now the expression has been changed.

 

- Ex line 367, now line 365, correction is OK, but “figure”, is to be specified with number, or removed. 

 

>>>Thank the reviewer for pointing out this mistake. “Figure” has been removed.

Reviewer 3 Report

No further comments.

Author Response

Comments and Suggestions for Authors

No further comments.

>>>Thank you for this comment.