Regulatory T Cells Secrete IL10 to Suppress Neuroinflammation in Early Stage after Subarachnoid Hemorrhage

Round 1

Reviewer 1 Report

The authors investigate the mechanisms underlying the beneficial effects of Tregs in reducing neuroinflammation after SAH. Using transgenic mice, they found that TREG exerted its neuroprotective effect by infiltrating the brain early after SAH and secreting IL-10, suppressing neuroinflammation and reducing apoptosis of neurons.

The study is well done and presented well. I have some concerns.

1.     How much volume of DT did you administrate to the FoxP3-DTR mice? You should also state clearly where it was obtained from.

2.     You should note the reference paper of these two genetically modified mice so that everyone can verify reproducibility.

3.     The inflammatory phenotype of Il10 KO mice is strongly influenced by the genetic background and the husbandry conditions of the vivarium in which mice are maintained. How did you deal with it?

4.     Why did you use male mice? Is there any gender difference? 

5.     You should indicate the sequence of the primers used in qPCR in the method section.

6.     For better understanding, you need to add HE staining of mouse brain sections in Figure 5.

 

7.     All abbreviations must be spelled out on the first appearance.

Author Response

Response to Reviewer#1

 

The authors investigate the mechanisms underlying the beneficial effects of Tregs in reducing neuroinflammation after SAH. Using transgenic mice, they found that TREG exerted its neuroprotective effect by infiltrating the brain early after SAH and secreting IL-10, suppressing neuroinflammation and reducing apoptosis of neurons.

The study is well done and presented well. I have some concerns.

Response: Thank you so much for your comments and criticisms on our manuscript. We tried our best to respond to these point-to-point as follows:

1. How much volume of DT did you administrate to the FoxP3-DTR mice? You should also state clearly where it was obtained from.

Response: The following part about DT experimental protocol has been added in the Methods part.

As for Tregs depletion, Diphtheria toxin (DT, intraperitoneal, 0.05 mg/g body weight) was injected 3 days prior to SAH to deplete Tregs, and repeated every 3 days to maintain Tregs depletion until mice sacrifice. And DT(D0564) was purchased from Sigma-Aldrich (USA).

2. You should note the reference paper of these two genetically modified mice so that everyone can verify reproducibility.

Response: the references of two transgenic mice in this study were added in the references list: reference (9) for Foxp3-DTR-mice and reference (10) for Il10 gene KO mice.

Concrete references:

Foxp3-DTR mice: Kim JM, Rasmussen JP, Rudensky AY. Regulatory T cells prevent catastrophic autoimmunity throughout the lifespan of mice. Nat Immunol. 2007 Feb;8(2):191-7.

Il10 gene KO mice: Kühn R, Löhler J, Rennick D, Rajewsky K, Müller W. Interleukin-10-deficient mice develop chronic enterocolitis. Cell. 1993 Oct 22;75(2):263-74.

3. The inflammatory phenotype of Il10 KO mice is strongly influenced by the genetic background and the husbandry conditions of the vivarium in which mice are maintained. How did you deal with it?

Response: Il10 gene KO mice were susceptible to immune-related diseases which might influence the accuracy of the results due to their genetic background. To avoid this deviation, we housed Il10 gene KO mice at high health status conditions (high SPF) according to the manuscription of Jackson Laboratory and reference (12), which has been added in the methods experiment animals part.

4. Why did you use male mice? Is there any gender difference?

Response: Using male mice is one of the limitations of this study, We did take the gender differences into consideration and related experiments are carrying out now. Besides, there are some controversial opinions about the gender differences of Tregs. For instance, Lucia et al. reported that female Tregs were more accessible into brain and displayed stronger immunosuppressive effect in neonatal hypoxic-ischemic brain injury. However, George et al. found that Tregs from men owned a significantly higher sup-pressive capacity than those from women in a clinical cross-sectional study. Thus, the gender differences of Tregs and whether those differences would influence the protective effect of Tregs in SAH will be further elucidated in our future studies.

5. You should indicate the sequence of the primers used in qPCR in the method section.

Response: The following part about qPCR protocol has been added in Methods-real time fluorescence quantitative PCR (qPCR) part and the sequence of primers are listed in table1.

6. For better understanding, you need to add HE staining of mouse brain sections in Figure 5.

Response: The H&E staining of the corresponding brain sections have been carried out and added into Fig5 panel B “H&E staining”.

 

7. All abbreviations must be spelled out on the first appearance.

Response: we have revised our manuscript for abbreviations and polish our language as much as possible.

Reviewer 2 Report

The reviewer is  neurosurgeon with cerebrovascular specialization and from this point is this study interesting. 

But  from  mice  to human is still long distance.

 

Maybe will be interesting in future application this method  to another pathology especially  for intracerebral hypertonic hemorrhage, because this  filed of neurosurgery is still very unsatisfied.

 

 

Author Response

Response to Reviewer#2

The reviewer is neurosurgeon with cerebrovascular specialization and from this point is this study interesting.

But from mice to human is still long distance.

 

Maybe will be interesting in future application this method to another pathology especially for intracerebral hypertonic hemorrhage, because this filed of neurosurgery is still very unsatisfied.

 

Response: Thank you so much for your comments and criticisms on our manuscript. We tried our best to respond as follows:

Our study clarified the protective effect of Tregs in early stage in SAH in mice, which has been reported in ischemia stroke、Parkinson’s disease and other nervous system disease before. Despite the huge gap between human and mouse studies, mouse experiments provide the choice of therapeutic targets for clinical treatment. The rise of immunotherapy, especially Tregs cell therapy（engineered Tregs, CAR-T, etc.）will become the key research focus of SAH treatment in the future. Our study laid a certain foundation for the in clinical treatment of Tregs cell therapy in SAH. As for the effect of Tregs in intracerebral hypertonic hemorrhage, the associated research will be elucidated in our future study.

Reviewer 3 Report

Zhou et al present a very comprehensive investigation into the role of Tregs in neuroinflammation after SAH in mice. They employ many experimental techniques in an attempt to address their question from various angles, and their results and narrative are consistent across methodologies. Overall, I believe this manuscript discussed experiments that were well-conducted and comprehensive, but there are problems with the way the results are presented. As discussed below, the Results section is completely lacking in a discussion of the statistical analyses performed. Additional comments are below:

Methods - 

- How were the cells FACS sorted for Tregs? Were they permeabilized and stained for intracellular Foxp3 before sorting or was it just based on CD4 and CD25 expression?

Results - 

- Overall, throughout the results section, there are results mentioned using general statements (ex. "the neurological function of the mice in the Tregs depletion group were significantly poorer than those in the control group") but no statistical results are discussed including no p values, odds ratios, etc. On the Figures, asterisks are used to show statistical significance, but the meaning of these asterisks (* vs **, etc.) are not defined. 

- Did they look out past 5 days? If the Tregs were still increasing at that time it would be interesting to see the trajectory through the first 2 weeks, which is the peak vasospasm window in humans after SAH and a very clinically relevant period.

- Figure 1C needs to be explained more clearly and labeled more clearly in the figure caption. It should clearly state what down and up mean (I assume it means down/up in brain-infiltrating Tregs compared to peripheral blood Tregs?). There should also be a line on the graph showing the cutoff point for significance. 

- Section 3.3/3.4 - The methods described at the beginning of both paragraphs this section 3.3 and 3.4 should be discussed in the methods section, not in the results. It should also be clear whether SAH was induced before or after DT/PBS injection. 

- Section 3.4 - The modified Garcia's Score is not defined in the methods section or in the results section (it should be in methods)

- Figure 5 - It should be clearer in the description of this figure in the Results section and in the caption that these results are from different populations of mice (If I am understanding them correctly). It appears that 5A is from the initial experiments comparing native Tregs in the brain and peripheral blood, and then 5B and C (which are not even referenced in the text) are from the later experiments where mice are infused with WT vs IL10KO Tregs. 

- Figure 5 - It appears that IL-10 may not be the only mechanism by which infiltrating Tregs have a protective effect, as the SAH+IL10KO Tregs infused in panel C appear to have a neurologic function in between PBS treated mice and mice treated with Tregs that produce IL-10. The authors should comment on this.

Discussion - 

- The use of the term "immune inflammation" is unconventional. Are the authors referring to a proinflammatory response, inflammation within the CNS, systemic inflammation affecting the CNS, inflammation in general? It seems like it may be sufficient just to say inflammation.

- The authors are likely missing important references regarding neuroinflammation after SAH, as they only reference 1 paper (11, a review paper) but there is substantial literature on this topic in both mice and humans. 

Overall the quality of the English was good. The only major grammatical edit I would suggest is there are some awkward tense changes in the Methods section (paragraphs start in past tense but then describe detailed methods in the present tense as if they are copied directly from experimental protocols).

Author Response

Response to Reviewer#3

Zhou et al present a very comprehensive investigation into the role of Tregs in neuroinflammation after SAH in mice. They employ many experimental techniques in an attempt to address their question from various angles, and their results and narrative are consistent across methodologies. Overall, I believe this manuscript discussed experiments that were well-conducted and comprehensive, but there are problems with the way the results are presented. As discussed below, the Results section is completely lacking in a discussion of the statistical analyses performed. Additional comments are below:

Response: Thank you so much for your comments and criticisms on our manuscript. We tried our best to respond to these points to points as follows:

 

Methods -

1.How were the cells FACS sorted for Tregs? Were they permeabilized and stained for intracellular Foxp3 before sorting or was it just based on CD4 and CD25 expression?

Response: As for FACS sorting of Tregs, we use Foxp3-DTR mice, in which Foxp3 of Tregs were labelled with GFP fluorescence. Thus, permeabilization was not necessary. And the concrete protocol of Tregs FACS sort has been added in methods part:

Single cell suspensions were isolated from brain or blood of Foxp3-DTR mice in SAH 3d or sham group as described for flow cytometry aforementioned. Single cells suspension was incubated with CD45-Pacific Blue antibody, CD3-PE-Cy7 antibody, CD4 APC-Cy7 antibody, CD25-BV605 antibody for 30 minutes on ice in the dark. As Tregs in Foxp3-DTR mice were GFP labeling, the CD45⁺CD3⁺CD4⁺CD25⁺GFP⁺cells in blood and CD45^highCD4⁺CD25⁺GFP⁺ cells in brain were sorted by FACSAriaIII (BD Bioscience).

 

Results -

2.Overall, throughout the results section, there are results mentioned using general statements (ex. "the neurological function of the mice in the Tregs depletion group were significantly poorer than those in the control group") but no statistical results are discussed including no p values, odds ratios, etc. On the Figures, asterisks are used to show statistical significance, but the meaning of these asterisks (* vs **, etc.) are not defined.

Response: The statistics definition of the symbol and the methods of each statistical analysis in each figure have been added in the figure legends.

3.Did they look out past 5 days? If the Tregs were still increasing at that time it would be interesting to see the trajectory through the first 2 weeks, which is the peak vasospasm window in humans after SAH and a very clinically relevant period.

Response: Actually, we observed the brain infiltration numbers of Tregs 1-7 days after SAH onset. The numbers of brain-infiltrated Tregs accumulated in 1-5 days, and decreased in 7 days (data not showed). Due to the significant pathology meaning of the early brain injury in SAH prognosis, our study focused on the early stage (1-5 days) of SAH. The effect of Tregs in the subacute phase needs to be further elucidated in future studies.

4.Figure 1C needs to be explained more clearly and labeled more clearly in the figure caption. It should clearly state what down and up mean (I assume it means down/up in brain-infiltrating Tregs compared to peripheral blood Tregs?). There should also be a line on the graph showing the cutoff point for significance. Response: We have revised Fig 1C now, clarifying that the mean of down/up is between SAH 3d brain-infiltrated Tregs and peripheral blood Tregs. Besides, a pair of cutoff lines have been added in Fig1 panel C. Thus, we renewed Fig1 in the revised manuscript.

5.Section 3.3/3.4 - The methods described at the beginning of both paragraphs this section 3.3 and 3.4 should be discussed in the methods section, not in the results. It should also be clear whether SAH was induced before or after DT/PBS injection.

Response: We have deleted the beginning of 3.3 and 3.4 part in the revised manuscript. The following part about DT injection has been added in the methods：

As for Tregs depletion, Diphtheria toxin (DT, intraperitoneal, 0.05 mg/g body weight) was injected 3 days prior to SAH to deplete Tregs, and repeated every 3 days to maintain Tregs depletion until mice sacrifice. As for Tregs exogenous administration, Tregs were obtained from mice spleen by using a mouse Treg cell isolation kit according to the manufacturer’s instructions. 2×106 Tregs or equivalent volume of PBS were infused into mice through tail vein 2 hours after SAH onset.

6.Section 3.4 - The modified Garcia's Score is not defined in the methods section or in the results section (it should be in methods).

Response: The following part about the definition of Garcia’s score test has been added in the method:

The Garcia’s score test was performed as previously described. Six tests were per-formed: Spontaneous activity, Symmetry in four limb movement, Forepaw outstretching, Climbing, Body proprioception and Response to vibrissae touch. Each test was scored by a 3 points system according to the following criteria: 0, severe deficits; 1, moderate deficits; 2, mild deficits; 3, no deficits. The total score of the modified Garcia Score are 18 points, which ranges from 0 (serious injury) to 18 (no injury).

7.Figure 5 - It should be clearer in the description of this figure in the Results section and in the caption that these results are from different populations of mice (If I am understanding them correctly). It appears that 5A is from the initial experiments comparing native Tregs in the brain and peripheral blood, and then 5B and C (which are not even referenced in the text) are from the later experiments where mice are infused with WT vs IL10KO Tregs.

Response: The figure 5 panel A used the transcriptome data of SAH 3d brain-infiltrated Tregs vs. SAH 3d peripheral blood Tregs sorted from Foxp3-DTR mice. And the Fig 5 panel B and panel C used WT mice with PBS, Treg from WT mice or Treg from Il10 KO mice infusion after SAH. We have added these explanations in the results and figure caption, besides, Figure 5 has been refreshed in our revised manuscript.

 

- Figure 5 - It appears that IL-10 may not be the only mechanism by which infiltrating Tregs have a protective effect, as the SAH+IL10KO Tregs infused in panel C appear to have a neurologic function in between PBS treated mice and mice treated with Tregs that produce IL-10. The authors should comment on this.

Response: We have noted this result in our study, and the following part about the mechanism of protective effect of Tregs except from IL-10 has been added in the discussion part:

Tregs suppress neuroinflammation by various patterns and cytokines. Shi et al. reported Tregs secreted OPN to interact with microglia and promoted white matter repair in is-chemic stroke. Huang et al. indicated the CD45-galectin-1 interaction between Tregs and MN9D cells prevented MPP+ toxicity and reduce the neuron losses in MPTP induced Parkinson’s Disease mice with exogenous administration of Tregs. In this study, we performed Bulk RNA sequencing and in vivo experiment to confirm that Tregs sup-pressed neuroinflammation through IL-10 in SAH. However, while mice with exogenous administration of Il10 gene KO Tregs group showed poorer neurological function than mice with sham Treg administration, the neurological function of Il10 gene KO Tregs infusion group was better than SAH+PBS infusion group, indicating that Il10 is not the only mechanism of the protective effect of Treg. The other underlying mechanisms such as Areg, Vegfa, Tgfa, etc. which were highly expressed in brain-infiltrating Tregs in our transcriptome data require further studies.

 

Discussion -

8.The use of the term "immune inflammation" is unconventional. Are the authors referring to a proinflammatory response, inflammation within the CNS, systemic inflammation affecting the CNS, inflammation in general? It seems like it may be sufficient just to say inflammation.

Response: We tried to generalize the immune response and inflammation as “immune inflammation”, however, as the unconventional expression of “immune inflammation”, we replaced “immune inflammation” as “inflammation” in our revised manuscript.

9.The authors are likely missing important references regarding neuroinflammation after SAH, as they only reference 1 paper (11, a review paper) but there is substantial literature on this topic in both mice and humans.

Response: We did omit a large number of references closely related our study. After reviewing large amounts of literatures and our study, we added some references in our study. The renewed references list has been refreshed in revised manuscript.