Neutrophils promote the development of reparative macrophages mediated by ROS to orchestrate liver repair

Phagocytes, including neutrophils and macrophages, have been suggested to function in a cooperative way in the initial phase of inflammatory responses, but their interaction and integration in the resolution of inflammation and tissue repair remain unclear. Here we show that neutrophils have crucial functions in liver repair by promoting the phenotypic conversion of pro-inflammatory Ly6ChiCX3CR1lo monocytes/macrophages to pro-resolving Ly6CloCX3CR1hi macrophages. Intriguingly, reactive oxygen species (ROS), expressed predominantly by neutrophils, are important mediators that trigger this phenotypic conversion to promote liver repair. Moreover, this conversion is prevented by the depletion of neutrophils via anti-Ly6G antibody, genetic deficiency of granulocyte colony-stimulating factor, or genetic deficiency of NADPH oxidase 2 (Nox2). By contrast, adoptive transfer of WT rather than Nox2−/− neutrophils rescues the impaired phenotypic conversion of macrophages in neutrophil-depleted mice. Our findings thus identify an intricate cooperation between neutrophils and macrophages that orchestrate resolution of inflammation and tissue repair.

The authors addressed all of the comments of the reviewers and significantly improved the manuscript.
Reviewer #3 (Neutrophil biology, liver inflammation)(Remarks to the Author): This is an interesting manuscript with an interesting relationship between neutrophils and monocytes/macrophage in sterile injury and repair. I have a number of general comments that should be addressed. 1) Not sure that a Ly6C high macrophage is an accepted term. Generally, these are referred to as monocytes. Throughout the manuscript the authors refer to Ly6C high and Ly6C low macrophage. However, immunologists would see this as markers of monocytes and macrophage respectively. While this could just be terminology it could also be a completely alternative interpretation ie., neutrophils help monocytes mature into macrophage. The authors must demonstrate at the beginning of their results that they truly are looking at macrophage and macrophage ie., is F4/80 and/or MerTK and/or other macrophage markers on the surface of both Ly6C high and Ly6C low cells.
2) In the abstract and throughout the manuscript the authors make the following claim verbatim: "We mechanistically demonstrate that neutrophil-derived reactive oxygen species (ROS) trigger the phenotypic switch between macrophages via AMP-activated protein kinase (AMPK)" This is not shown in this manuscript. The authors show that AMPK is oxidized/activated but whether it is responsible in the switch is not examined. The authors only make one sentence statement saying that the knockout has fewer reapir genes. This needs to be better fleshed out. Was there less repair in this knockouts? 3) I am not sure the results of depletion of neutrophils in CCR2-/-mice provide a clear conclusion that neutrophil repair is related to monocytes. is poorly developed in this manuscript; significantly more work needs to be done to convince the reader that neutrophil ROS are responsible for mediating liver repair via macrophages. The data does not support the idea that "Neutrophils license macrophages for orchestrating liver repair by release reactive oxygen species".

Response:
First, we wish to express our sincere thanks for your constructive comments and suggestions. Your questions and suggestions have been thoroughly discussed among all authors and we addressed your concerns in the following.
We have re-organized the manuscript and greatly reduced redundant data to better present the main information in a streamlined logical flow, as you suggested.
Regarding the issue of neutrophils regulating macrophage phenotypic switching, we realized that we did not elaborate on the importance of this main finding. We have discussed the topic in the Discussion Section in the revised manuscript.
We thank you for pointing out the lack of strong evidence for the role of neutrophil-derived ROS. As far as we know, there are two classical approaches to assess the role of specific types of cell-derived mediators under in vivo conditions. The ideal is to create neutrophil-specific deletions using Cre-lox system. The other is to generate bone marrow chimeras by transplanting mixed bone marrow cells from a neutrophil-deficient and a gene knockout mouse into previously lethally irradiated wild type mice. In our previous study, we had considered the use of lineage-specific conditional knockout mice to assess the pro-repair role of neutrophil-derived ROS. We found that three transgenic mice may be used (e.g. MRP8 (S100A8)-Cre 1, 2, 3 , GE (Ela)-Cre 4, 5 and LysM-Cre 6, 7 ) for conditional deletion of target genes in neutrophils. However, it is important to note that high specificity cannot always be obtained when using these neutrophil-specific conditional knockout mice 8 . MRP8-Cre mice have been reported to allow for Cre-mediated deletion of loxP-flanked target genes in granulocytes, monocytes and a fraction of granulocyte-macrophage progenitor 1, 2, 9 . GE-Cre mice can be used for conditional gene targeting in the myeloid lineage, including granulocytes and inflammatory macrophages 4,10 . LysM-Cre mice have been widely used to genetically target myeloid cells, including granulocytes, macrophages, inflammatory monocytes and myeloid-derived dendritic cells 7,11 . Given the limited specificity of these lineage-specific conditional knockout mice, they may not be suitable for us to assess the role of neutrophil-derived ROS in liver repair.
Generation of mixed bone marrow chimera provides an alternative method to investigate the importance of a molecule in certain cell types 8 . This approach is also widely used in many studies 12,13,14 . Thus, we chose bone marrow chimeras instead of neutrophil-specific conditional knockout mice to examine the role of neutrophilderived ROS in liver repair. We found that Nox2 -/-/Gcsf -/mixed bone marrow chimeric mice profoundly impaired liver regeneration and macrophage conversion during the resolution phase, as compared to WT/Gcsf -/chimeras (Fig. 4a-e).
Considering that there are partial depletion of Nox2 in bone marrow cells other than neutrophils in chimera models, as you mentioned in Question 13, we have performed an adoptive transfer experiment to confirm the role of neutrophil-derived ROS. We adoptively transferred WT or Nox2 -/neutrophils into neutrophil-depleted mice (Fig. 4f,   Supplementary Fig.4f, g). Adoptive transfer of WT neutrophils rescued the exacerbated damage and depressed hepatic regeneration in neutrophil-depleted mice during the resolution phase (Fig. 4g-i). In contrast, transfer of Nox2 -/neutrophils failed to do so. Moreover, transfer of WT but not Nox2-deficient neutrophils promoted macrophage conversion in neutrophil-depleted mice (Fig. 4j). These results may suggest that ROS-producing neutrophils contribute to macrophage skewing and liver repair. However, considering that we cannot use the ideal lineage-specific deletion method to obtain more unambiguous evidence about "neutrophil-derived ROS", we have tuned down our statements of relevant conclusions in the revised manuscript.

Question 2:
In general, the authors did not adequately address comments of Reviewer #1. For example, the reviewer correctly pointed out that although N-acetyl-L-cysteine (NAc) is an inhibitor of ROS, it also acts as an antidote for APAP overdose through a number of potential mechanisms other than inhibition of ROS. Importantly, it is now known that NAc is active and used clinically because it directly detoxifies acetaminophen (by binding the acetaminophen metabolite NAPQI). NAc cannot be used to simply detoxify ROS, the authors results with NAc are due to acetaminophen detoxication. Note also that given that NOX deficient mice (mice deficient in NADPH oxidase) inhibit the ability of neutrophils, as well as macrophages, from producing ROS, the authors cannot conclude that changes in toxicity of acetaminophen in these knockout mice are due to ROS production from neutrophils.

Response:
Thanks for your valuable comments. We acknowledge that NAC is not only an inhibitor of ROS, but also an antidote for APAP overdose. Thus, we agree that the use of NAC to assess the role of ROS in liver repair is not appropriate. However, it is worth noting that the increased severity of APAP-induced liver injury during the resolution phase after NAC treatment could not be simply explained by APAP detoxification. If NAC detoxified APAP, liver damage would be mitigated rather than aggravated during the resolution phase. Even so, using NAC could not clearly demonstrate the role of ROS in liver repair. Thus, we removed all the results of NAC related to liver repair.
On the other hand, we used a more appropriate strategy to ascertain the importance of ROS. In vivo, we took advantage of Nox2 -/mice, in which the gp91 subunit of NADPH oxidase is deleted ( Fig. 3d-g). In vitro, we used catalase to degrade hydrogen peroxide as well as Nox2 -/neutrophils for neutrophil-macrophage co-culture experiments (Fig. 5a, b).
Regarding the issue of neutrophil-derived ROS, we have performed adoptive transfer experiments in order to provide more convincing evidence, as mentioned in Response to Question 1.

Response:
We're sorry for the mistakes and we have corrected the error as below.

Response:
Thanks for your good suggestions. We have measured the catalase concentration in liver homogenates and found that there was no significant difference in hepatic catalase concentrations between normal and APAP-challenged mice (Supplementary Fig.4c).

Question 5:
Fig. 5, without appropriate controls (such as the addition of catalase) one cannot conclude that neutrophil derived ROS mediates the conversion of pro-inflammatory macrophage ex vivo; these are critical controls in experiment using conditioned medium that were not presented.

Response:
Thanks for your good suggestions. We have added catalase to degrade hydrogen peroxide in the neutrophil-macrophage co-culture system. Neutrophil-conditioned medium promoted pro-inflammatory Ly6C hi CX3CR1 lo monocytes/macrophages skewing toward a reparative phenotype, but the addition of catalase induced a decrease of pro-resolving marker and an increase of pro-inflammatory marker expression ( Fig.   5a).  Thanks for your good suggestions. We're sorry for the mistakes. Mean intensity of Ca 2+ influx after various concentrations of H2O2 stimulation has been presented in the revised manuscript (Fig. 6d).

Question 7:
It is not helpful to compare the effects of ROS using LPS treated peritoneal macrophages or BMDM (Fig. 6) when the entire paper is on liver macrophages and acetaminophen, they are not the same macrophage populations and do not respond as liver macrophages.

Response:
Thanks for your comments. We agree that peritoneal macrophages and BMDMs cannot completely mimic liver macrophages. In the previous submission process, some reviewers suggested that we should explore the molecular mechanisms by which ROS regulate macrophage skewing in vitro. Because the number of hepatic recruited macrophages is too small for analyzing protein expression with various stimulating concentrations or at different time points, and cell viability after FACS sorting may be unstable in each experiment, it is very difficult to study molecular mechanisms using these macrophage populations isolated from livers. For these reasons, we used primary

Response:
Thanks for your good suggestions. We have measured protein level changes of various kinases in response to H2O2 at the indicated time points (Below Figures a, b). We found that at the time points observed, 20 μM H2O2 stimulation induced activation of AMPK within 15-120 min, but H2O2 did not significantly affect the phosphorylation of JAK1, JAK2, STAT3, STAT6, ERK1/2 and NF-κB. Additionally, 20 μM H2O2 stimulation may induce phosphorylation of STAT1 in a short time (within 15 min). Your concerns remind us that the conclusion is not accurate in the previous manuscript, so we changed the statement to "H2O2 treatment at indicated concentrations had no significant effects on the activation of JAK1, JAK2, STAT1, STAT3, STAT6, ERK1/2 and NF-κB at the time points observed".

Question 9:
Does acetaminophen/neutrophils/hydrogen peroxide activate AMPK in isolated hepatic macrophage populations (confocal images shown in Fig. 6, panel h, are not convincing and not comparable with isolated peritoneal and BMDM macrophages).

Response:
Thanks for your good suggestions. We performed western blots of p-AMPK in isolated hepatic macrophage populations. Consistent with in vivo labelling, increased phosphorylation of AMPK was observed in 72 hr Ly6C lo CX3CR1 hi macrophages compared to 24 hr Ly6C hi CX3CR1 lo monocytes/macrophages (Supplementary Fig.   5b).

Response:
Thanks for careful reading of the manuscript. We have deleted "p<0.01" in the legends of the Figure

Response:
Thanks for your comments. Your concerns remind us that the description of the results is not accurate, so we changed the statement to "we found that monocytes/macrophages and neutrophils were the major cell types among the infiltrating cells". Question 12: Fig. 1, neutrophils may contribute to tissue repair, but there is no evidence that "neutrophil depletion delays…liver repair after acute injury" (legend to Fig. 1).

Response:
Thanks for your good suggestions. We changed the statement to "neutrophils contribute to tissue repair".

Question 13:
In chimera studies using Gcsf and Nox2 ko bone marrow, there was no consideration of the fact that there is partial depletion of macrophages and this can markedly alter the conclusions from these studies.

Response:
Thanks for your valuable comments. We acknowledge that there are pitfalls in chimera models due to the partial depletion of Nox2 in macrophages. Given the limited specificity of existing neutrophil-specific conditional knockout mice, we performed an adoptive transfer experiment to confirm the role of neutrophil-derived ROS, as mentioned in Response to Question 1.
The authors addressed all of the comments of the reviewers and significantly improved the manuscript.

Response:
We recognize and deeply appreciate the effort and generosity that present in the constructive review. We wish to express our sincere thanks for your suggestions and comments, which are very valuable and helpful for revising and improving our manuscript.

Reviewer #3 (Neutrophil biology, liver inflammation)(Remarks to the Author):
This is an interesting manuscript with an interesting relationship between neutrophils and monocytes/macrophage in sterile injury and repair. I have a number of general comments that should be addressed.

Response:
First, we wish to express our sincere thanks for your suggestions and comments, which are very valuable and helpful for revising and improving our manuscript. Your questions and suggestions have been thoroughly discussed among all authors and we addressed your concerns in the following.

Question 1:
Not sure that a Ly6C high macrophage is an accepted term. Generally, these are referred to as monocytes. Throughout the manuscript the authors refer to Ly6C high and Ly6C low macrophage. However, immunologists would see this as markers of monocytes and macrophage respectively. While this could just be terminology it could also be a completely alternative interpretation ie., neutrophils help monocytes mature into macrophage. The authors must demonstrate at the beginning of their results that they truly are looking at macrophage and macrophage ie., is F4/80 and/or MerTK and/or other macrophage markers on the surface of both Ly6C high and Ly6C low cells.

Response:
Thanks for your valuable suggestions. We have assessed the expression of key macrophage markers F4/80, CD64, and the tyrosine kinase MerTK (Supplementary   Fig. 1b). Both 24 hr Ly6C hi cells and 72hr Ly6C lo cells expressed macrophage markers F4/80, CD64 and MerTK, but the expression of these macrophage markers was lower in 24 hr Ly6C hi cells than in 72 hr Ly6C lo cells. These results suggest that Ly6C hi cells may include infiltrating monocytes, monocytes in differentiation to macrophages and/or macrophages. Therefore, the identification of Ly6C hi cells as macrophages in the previous manuscript is not accurate. We changed the description "Ly6C hi macrophages" to "Ly6C hi monocytes/macrophages" in the revision.

Question 2:
In the abstract and throughout the manuscript the authors make the following claim verbatim: "We mechanistically demonstrate that neutrophil-derived reactive oxygen species (ROS) trigger the phenotypic switch between macrophages via AMP-activated protein kinase (AMPK)" This is not shown in this manuscript. The authors show that AMPK is oxidized/activated but whether it is responsible in the switch is not examined.
The authors only make one sentence statement saying that the knockout has fewer reapir genes. This needs to be better fleshed out. Was there less repair in this knockouts?

Response:
Thanks for your good suggestions. We acknowledge that we did not provide sufficient evidence about the role of AMPK in vivo. Because other cell types in the liver 1,2 , such as hepatocytes and hepatic stellate cells, are reported to express AMPK, AMPK systemic knockout mice may be not suitable for us to assess the role of AMPK in macrophages.
Since we could not obtain macrophage-specific conditional knockout mice in the short term, we examined the role of macrophagic AMPK by adoptive transfer of WT or AMPKα -/monocytes to Ccr2 -/mice. Adoptive transfer of WT monocytes rescued the exacerbated damage and depressed hepatic regeneration in Ccr2 -/mice during the resolution phase, whereas transfer of AMPKα -/monocytes failed to do so ( Fig. 6h-j).
These results may suggest that macrophagic AMPK is involved in liver repair. Given that our current results only suggest a possible mechanism by which ROS regulate macrophage transition and liver repair, we have tuned down our statements of relevant conclusions in the revised manuscript.

Question 3:
I am not sure the results of depletion of neutrophils in CCR2-/-mice provide a clear conclusion that neutrophil repair is related to monocytes.

Response:
Thanks for your valuable comments. In the previous manuscript, we showed that mice lacking neutrophils displayed exacerbated damage and depressed hepatic regeneration during the resolution phase, but this effect was lost upon monocyte/macrophage ablation in Ccr2 -/mice. Thus, we speculated that neutrophils mediate liver repair may be related to monocytes/macrophages. However, given that Ccr2 -/mice already resulted in increased liver damage, it may not be appropriate to see if neutrophil depletion would cause a worsen effect in these mice. Based on this, we performed an adoptive transfer experiment to investigate whether neutrophils could alleviate liver damage in WT or Ccr2 -/mice. We found that adoptive transfer of neutrophils alleviated liver damage in WT mice during the resolution phase (below Figures a-c), whereas transfer of neutrophils had no effect in Ccr2 -/mice (below Figures d-f). These results may indicate that the neutrophil function in liver repair requires the presence of monocytes/ macrophages. 1. Data does not strongly support the idea that neutrophil-derived ROS mediate effects on macrophages. The title should be changed to more accurately represent what the data actually shows...perhaps simply "Evidence that neutrophils regulate macrophage activity...." without reference to 'licensing?' or that the process is mediated by "ROS"; also eliminate references to ROS in the Introduction. There could be a separate section in the 'Results' asking the question of whether ROS mediate the effects of neutrophils, data on ROS could then be presented with a detailed discussion of the hypothesis and limitations of the data as well as the fact that neutrophils release many other mediators in addition to ROS. The Discussion should include a detailed review of the literature comparing our current understanding of the roles of neutrophils and macrophages in toxicity. 2. There is a disconnect between the timing of when neutrophils accumulate in the liver (and produce ROS) following exposure to a toxicant (first 24 hours) and when there are alterations in macrophage functioning (at much later time points). This should addressed in the Discussion. This reviewer notes that neutrophil-derived ROS could trigger a response after many hours and this should be addressed in the Discussion.

Reviewer #3 (Remarks to the Author):
The manuscript is greatly improved and the authors have made a genuine effort to revise appropriately. I have two very small changes the authors may wish to consider. 1) As I mentioned previously, and the authors have taken under advisement they are watching monocytes convert to repair macrophage. The authors changed the word macrophage to monocyte/macrophage everywhere except the title. I am not going to force them to change their title but wonder whether the title reflects the manuscript.
2) I completely forgot that NAC was used to treat patients with APAP poisoning. As such calling it an anti-oxidant as the authors do in the resuts, undermines their very nice paper. They may wish to add a sentence saying that NAC may have other effects in the results section.

Reviewer comments:
Reviewer #1 (Remarks to the Author):

Question1:
Data does not strongly support the idea that neutrophil-derived ROS mediate effects on macrophages. The title should be changed to more accurately represent what the data actually shows...perhaps simply "Evidence that neutrophils regulate macrophage activity...." without reference to 'licensing?' or that the process is mediated by "ROS"; also eliminate references to ROS in the Introduction. There could be a separate section in the 'Results' asking the question of whether ROS mediate the effects of neutrophils, data on ROS could then be presented with a detailed discussion of the hypothesis and limitations of the data as well as the fact that neutrophils release many other mediators in addition to ROS. The Discussion should include a detailed review of the literature comparing our current understanding of the roles of neutrophils and macrophages in toxicity.

Response:
First, we wish to express our sincere thanks for your suggestions and comments, which are very valuable and helpful for revising and improving our manuscript. We recognize and deeply appreciate the effort and generosity that present in your constructive review.
Your comments and suggestions have been thoroughly discussed among all authors.
We realized that our previous statements about "neutrophil-derived ROS" may be misleading because we could not rule out that additional factors other than ROS might contribute to macrophage skewing and liver repair. To be more precise, we concluded that "our results suggested that neutrophils trigger macrophage skewing toward a reparative phenotype for optimal liver repair and the process is mediated by ROS".
In addition, we summarized several evidences on the role of ROS in macrophage conversion and liver repair, as shown below. We believe these findings could support the above conclusions and may be worth reconsidering.
(1) We found that ROS deficient Nox2 -/mice displayed delayed resolution of liver injury and impaired macrophage conversion.
(4) Using neutrophil adoptive transfer systems, we found that adoptive transfer of WT neutrophils rescued the exacerbated damage, depressed hepatic regeneration and impaired macrophage skewing in neutrophil-depleted mice during the resolution phase, whereas transfer of Nox2 -/neutrophils failed to do so.
(5) Using co-culture systems, we found that conditioned medium from WT neutrophils, but not Nox2 -/neutrophils, promoted pro-inflammatory monocytes/ macrophages skewing toward a reparative phenotype.
Although our results suggest that ROS play important roles in regulating macrophage skewing and liver repair, as you mentioned, we cannot exclude that other mediators released by neutrophils may have similar effects as ROS. Therefore, in order to present our results more objectively and accurately, we made the following changes in the writing of the manuscript according to your suggestions.
(1) Regarding the title and introduction, as you suggested, changing the title to "Neutrophils promote the development of reparative macrophages to orchestrate liver repair and the process is mediated by ROS" will more accurately represent our results.
Given the limitation of the title length in the manuscript preparation guidelines, we finally changed the title to "Neutrophils promote the development of reparative macrophages mediated by ROS to orchestrate liver repair". Additionally, for a more accurate summary in Introduction, we changed the related statements to "We show that neutrophils instruct inflammatory monocytes/macrophages to adopt a pro-regenerative phenotype for optimal liver repair and the process is mediated by ROS".
(2) Regarding the Results section about ROS, we changed the statement of our hypothesis to "Because activated neutrophils can potently produce ROS and phagocyte NADPH oxidase 2 (Nox2) complex-derived ROS are known to be crucial regulators of immune response, we asked whether ROS might mediate the regulation of the macrophage skewing". Importantly, we added a detailed discussion of the hypothesis and limitations of the data: "However, we cannot exclude that additional factors other than ROS might contribute to macrophage skewing and liver repair.
Since neutrophils are reported to generate many anti-inflammatory and pro-resolving mediators, whether these products have similar effects remains to be investigated. Increasing amounts of data indicate that neutrophils do not contribute to the liver injury during the early phase after APAP overdose 8 . Nevertheless, the role of neutrophils during the resolution phase remains largely unexplained. Here, we provide evidence that neutrophils are required for optimal liver regeneration and repair through promoting macrophage conversion." The following related literatures were cited:

Reviewer #3 (Remarks to the Author):
The manuscript is greatly improved and the authors have made a genuine effort to revise appropriately. I have two very small changes the authors may wish to consider.

Response:
First, we wish to express our sincere thanks for your suggestions and comments, which are very valuable and helpful for revising and improving our manuscript. We recognize and deeply appreciate the effort and generosity that present in your constructive review.

Question 1:
As I mentioned previously, and the authors have taken under advisement they are watching monocytes convert to repair macrophage. The authors changed the word macrophage to monocyte/macrophage everywhere except the title. I am not going to force them to change their title but wonder whether the title reflects the manuscript.

Response:
Thanks for your good suggestions. To make the title more accurately represent the manuscript, we have changed the title to "Neutrophils promote the development of reparative macrophages mediated by ROS to orchestrate liver repair".

Question 2:
I completely forgot that NAC was used to treat patients with APAP poisoning. As such calling it an anti-oxidant as the authors do in the resuts, undermines their very nice paper. They may wish to add a sentence saying that NAC may have other effects in the results section.

Response:
Thanks for your good suggestions. We're sorry that we did not explicitly explain the issue of NAC in the previous Point-by-point response, which make you feel confused.
NAC is not only an inhibitor of ROS, but also an antidote in acetaminophen intoxication.
However, the effect of NAC we observed during the resolution phase may be independent of its detoxification mechanism. There are two main reasons: (1) The detoxification mechanism of NAC is only effective during the early phase following APAP overdose. In the clinical setting, NAC is effective only for patients who present within eight hours of an acute overdose, and is less effective for late-presenting patients 1,2,3 . In mice, NAC only prevented toxicity when administered prior to or early following APAP 4 .
(2) Late or prolonged treatment with NAC is indeed toxic (detrimental) rather than detoxification (protective) in mice. Several studies in mice have reported that late or prolonged NAC treatment clearly decreased survival and impaired liver regeneration 5,6 . In our previous data, two doses of NAC were injected at 0 and 36 h to induce sustained ROS scavenging in APAP-challenged mice. We found that although NAC resulted in alleviated liver damage during the initial phase of inflammation, it exacerbated injury and impaired regeneration during the resolution phase of inflammation, which was consistent with previous results 6 . The increased severity of APAP-induced liver injury during the resolution phase could not be explained by APAP detoxification, but may be due to ROS scavenging.
We also have added a brief explanation about the effects of NAC in the Results section as below: "It is worth noting that although NAC also acts as an antidote in APAP intoxication, the effect of NAC we observed during the resolution phase may be independent of its detoxification mechanism. There are two main reasons: Firstly, the detoxification mechanism of NAC is only effective during the early phase following APAP overdose in humans and mice. Secondly, Late or prolonged treatment with NAC is indeed detrimental rather than protective in mice, which could not be explained by APAP detoxification."