Expression of G2019S LRRK2 in Rat Primary Astrocytes Mediates Neurotoxicity and Alters the Dopamine Synthesis Pathway in N27 Cells via Astrocytic Proinflammatory Cytokines and Neurotrophic Factors

Round 1

Reviewer 1 Report

The authors of this manuscript study the effect of overexpression version mutant G2019S to LRRK2 kinase, in rat astrocytes, to evaluate its neurotoxicity. This LRRK2 mutant induces astrocyte alteration related to inflammatory parameters and changes its phenotype from normal to reactive phenotype or astrogliosis. Moreover, when collected from reactive astrocyte media and after incubated in N27 rat dopaminergic neuronal, it altered its dopaminergic phenotype. Is an interesting work, but could have been more developed, because the authors were limited to releasing simple experiments to demonstrate your hypothesis. Even though, the current version needs some revisions. A few comments are listed to improve the current version:

2. Methods and Materials section

The authors should correct the worst point in title 2.5. Lactose dehydrogenase activity assay by 2.5. Lactate dehydrogenase activity assay

Results:

Fig.1B. If the authors had evaluated the cytotoxicity of the rASTRO transfected with LRRK2 WT and mutant, it should have obtained an increase in dead cell value in GS-hLRRK2.  However, it seems to be that the LRRK2 mutant is cytoprotective because the histogram is lower than the vector. The authors should correct to representation the histogram representation. Could it complete these results to release one more experiment using MTT and determine cell viability? In this form, the fig.1 would be more complete.

Fig.2. To evaluate in-depth this point, and analyze neurotrophic factors in rASTRO, the authors should complete this figure, measuring the GFs excreted to media cellular. For instance, an ELISA kit or other assay.

Fig.3 Again, as before figure, the authors are going to be able to measure the pro-inflammatory marker excreted to cellular media with other assays (ELISA assay). Moreover, there are some antibodies very great for measuring this pro-inflammatory marker in western blot, such as pre-IL-1B (RD Systems, AF-401-NA), INOS (NOS2 Santa Cruz Biotechnology sc-650), COX2 (Santa Cruz Biotechnology sc-1747), p65, etc. Could the authors release any Western blots and complete these figures?

Fig.4.

4. A. In this figure, there are some problems, focused on the blots images.  LRRK2 blot is not great, because the blot lines belonging to cell transfection with GS-hLRRK2 are not clear or rASTRO is not well transfected. I invite the authors to redo these western blots and get one good image. Consequently, the rest blot images are not concluyent, because GFAP, Vimentin, and LC3B images would have that it repeat.

On another hand, I want to recommend that the authors replace the Coomassie blue with other housekeeping proteins, for instance, GAPDH (35 kDa), tubulin (55 kDa), or Vinculin (120 kDa). All before antibodies are very good in WB, and there are many companies that can sell them.

For the evaluation of autophagy flux, it´s not enough to analyze LC3B protein levels. The main component and the protein that can start the autophagosome complex for correct development to the autophagy flux is p62. Could the authors analyze in WB the p62 protein? Moreover, the increases the both proteins, LC3B and p62 can mean two mechanisms, increase the protein expression or autophagy block. Could the authors release a simple experiment using NHCl4 / leupeptin, Bafilomicn B, Cloroquine, rapamycin, etc…and it can conclude whether the overexpression GS-LRRK2 blocks the autophagy flux or increase the expression of the proteins at least, LC3B and p62 in qRT-PCR?

Fig.5 This figure could be included in joint to Fig.1B

Fig.6. I cannot evaluate Figure 6 because the protein bands in the blot images are very faint. The authors should be able to increase the contrast and brightness the all panels. On another hand, the authors could do one complementary analysis of this figure, and add some mRNA analysis of the NURR1 target gene, for instance, TH, DAT, AADC, RET, VMAT, etc. Thus, it could increase the value of this experiment.

Minor point:

Miss reference in section 3.3. Astrogliosis and abnormal autophagic flux in G2019S LRRK2-expressing rASTRO. Line 213

Comments for author File: Comments.pdf

Author Response

Review 1

The authors of this manuscript study the effect of overexpression version mutant G2019S to LRRK2 kinase, in rat astrocytes, to evaluate its neurotoxicity. This LRRK2 mutant induces astrocyte alteration related to inflammatory parameters and changes its phenotype from normal to reactive phenotype or astrogliosis. Moreover, when collected from reactive astrocyte media and after incubated in N27 rat dopaminergic neuronal, it altered its dopaminergic phenotype. Is an interesting work, but could have been more developed, because the authors were limited to releasing simple experiments to demonstrate your hypothesis. Even though, the current version needs some revisions. A few comments are listed to improve the current version:

à We really appreciated your suggestion and pointing out. All issues were changed or replaced via the re-evaluation or re-do experiments.      

----------------------------------------------------------------------------------------------------------------

2. Methods and Materials section

The authors should correct the worst point in title 2.5. Lactose dehydrogenase activity assay by 2.5. Lactate dehydrogenase activity assay

à We deleted LDH data and methods section.

----------------------------------------------------------------------------------------------------------------

Results:

Fig.1B. If the authors had evaluated the cytotoxicity of the rASTRO transfected with LRRK2 WT and mutant, it should have obtained an increase in dead cell value in GS-hLRRK2.  However, it seems to be that the LRRK2 mutant is cytoprotective because the histogram is lower than the vector. The authors should correct to representation the histogram representation. Could it complete these results to release one more experiment using MTT and determine cell viability? In this form, the fig.1 would be more complete.

à As your suggestion, we tested the cell viability. And we found that there were some issues. Cytotoxicity was affected by the cell viability at day 5, because the decrease cell number for 5 days incubation brought out the decrease od LDH level in culture media, too. So, we deleted our previous results and replace to the cell viability results for 4 days incubation followed the change of culture media a day after transfection.    

----------------------------------------------------------------------------------------------------------------

Fig.2. To evaluate in-depth this point, and analyze neurotrophic factors in rASTRO, the authors should complete this figure, measuring the GFs excreted to media cellular. For instance, an ELISA kit or other assay.

à To address the issues you pointed out, we performed ELISA for neurotrophic factors including GDNF, BDNF, and NGF. VEGF was excluded from the overall analysis results because it is an unchanging neurotrophic factor similar to other GDNF and BDNF.

----------------------------------------------------------------------------------------------------------------

Fig.3 Again, as before figure, the authors are going to be able to measure the pro-inflammatory marker excreted to cellular media with other assays (ELISA assay). Moreover, there are some antibodies very great for measuring this pro-inflammatory marker in western blot, such as pre-IL-1B (RD Systems, AF-401-NA), INOS (NOS2 Santa Cruz Biotechnology sc-650), COX2 (Santa Cruz Biotechnology sc-1747), p65, etc. Could the authors release any Western blots and complete these figures?

à To solve the problem you pointed out, we performed ELISA of proinflammatory cytokines, including TNFα and IL-1β. Our intention was to show that TGFβ, an anti-inflammatory cytokine, was also unchanged, but TGFβ was excluded because the description of anti-inflammatory cytokines was not necessary for the content of this study.

----------------------------------------------------------------------------------------------------------------

Fig.4.

4. A.In this figure, there are some problems, focused on the blots images.  LRRK2 blot is not great, because the blot lines belonging to cell transfection with GS-hLRRK2 are not clear or rASTRO is not well transfected. I invite the authors to redo these western blots and get one good image. Consequently, the rest blot images are not concluyent, because GFAP, Vimentin, and LC3B images would have that it repeat.

On another hand, I want to recommend that the authors replace the Coomassie blue with other housekeeping proteins, for instance, GAPDH (35 kDa), tubulin (55 kDa), or Vinculin (120 kDa). All before antibodies are very good in WB, and there are many companies that can sell them.

For the evaluation of autophagy flux, it´s not enough to analyze LC3B protein levels. The main component and the protein that can start the autophagosome complex for correct development to the autophagy flux is p62. Could the authors analyze in WB the p62 protein? Moreover, the increases both proteins, LC3B and p62 can mean two mechanisms, increase the protein expression or autophagy block. Could the authors release a simple experiment using NHCl4 / leupeptin, Bafilomicn B, Cloroquine, rapamycin, etc…and it can conclude whether the overexpression GS-LRRK2 blocks the autophagy flux or increase the expression of the proteins at least, LC3B and p62 in qRT-PCR?

à As your suggestion, we re-do the entire experiments, and newly showed GFAP, vimentin blot. Additionally, we added the image of TNFα, the highest pro-inflammatory cytokine, and NGF, the significant low neurotrophic factor, and PARP1, the protein involved in apoptosis and TNFα-induced DNA damage. And all proteins were normalized to β-actin, alternative loading control of Coomassie Blue. However, we excluded the data of LC3B and p62. As you saw in our previous manuscript, the whole level of LC3B was decreased in GS-hLRRK2, and we found that p62 also significantly decreased by GS-hLRRK2. Interestingly, the LC3B II and p62 was dramatically in exosome, and we also found that the amount of exosome was increased by GS-hLRRK2 transfection. So, these results will be managed and published our next paper.       

----------------------------------------------------------------------------------------------------------------

Fig.5 This figure could be included in joint to Fig.1B

à Figure 5 results is about the cell viability of N27, which is treated by the conditioned media (CM) of transfected rat primary astrocyte, not about rat primary astrocyte itself. So, we added the ROS levels by the treatment with CM of transfected rat primary astrocyte to validate the effect of CM on cellular damage.   

----------------------------------------------------------------------------------------------------------------

Fig.6. I cannot evaluate Figure 6 because the protein bands in the blot images are very faint. The authors should be able to increase the contrast and brightness the all panels. On another hand, the authors could do one complementary analysis of this figure, and add some mRNA analysis of the NURR1 target gene, for instance, TH, DAT, AADC, RET, VMAT, etc. Thus, it could increase the value of this experiment.

à To address the issues you pointed out, we changed the brightness and contrast. And we added the results of mRNA quantification results and ELISA for dopamine measurement.

----------------------------------------------------------------------------------------------------------------

Minor point:

Miss reference in section 3.3. Astrogliosis and abnormal autophagic flux in G2019S LRRK2-expressing rASTRO. Line 213

à We deleted the sentence missed a reference.  

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors studied the effects of the mutant and wild type Lrrk2 on primary rat astrocytes, observing multiple growth factors and cytokines, as well as the impacts of N2A cells by conditioned medium. This is an interesting topic. However, there are serious defects in the current study.

1, transfection of GS Lrrk2 decrease NGF level, however, GS Lrrk2 significantly increase BDNF, GDNF and VEGF levels, compared with WT lrrk2 transfected and control. These growth factors also promote cell survival. The authors can not draw conclusion that GS Lrrk2 decrease NGF level, leading to toxicity, while overlook the increases of other growth factors.

2, the similar situation also observed in cytokines with some cytokines increaed and others decreased under GS lrrk2 transfection. Therefore it will be difficult to arrive at the conclusion that GS lrrk2 impact cell viability via modulation of growth factors and cytokines, as unconsistent results have been presented by the author.

3, in affects on astrogliosis by wt and GS lrrk2, it is obvious that wt lrrk2 have more significant influence. 

4, they desmonstrate that wt and GS lrrk2 have dustinct impacts, while GS lrrk2 has higher lrrk2 kinase activity than wt lrrk2 and overexpression of wt lrrk2 can also increase lrrk2 kinase activity in cells. So they only have different lrrk2 kinase increase and should not show opposite impacts on cells. The authors should exolain it.

5, based on evidence provided, it will be difficult to arrive at the conclusion the authors claimed.

Minor defects:

1,only Rt-PCR can not comfirm the expression changes of genes, western blot analysis is indispensible.

2, studies on only one kind of cell line will have limited significance. The authors should validate their findings in multiple cell lines and models.

Author Response

Reviewer 2

The authors studied the effects of the mutant and wild type Lrrk2 on primary rat astrocytes, observing multiple growth factors and cytokines, as well as the impacts of N2A cells by conditioned medium. This is an interesting topic. However, there are serious defects in the current study.

à Thank you for your comment. When we looked at the comments you pointed out below, we found that you either misread the indicated significance, or you simply judged it to be significant based on the shape of the bar in the graph. And because the schematic pointed out by Reviewer 1 and the schematic part you pointed out are consistent, we presented the experimental results based on a re-performed experiment rather than a refutation of the previous content.

----------------------------------------------------------------------------------------------------------------

1, transfection of GS Lrrk2 decrease NGF level, however, GS Lrrk2 significantly increase BDNF, GDNF and VEGF levels, compared with WT lrrk2 transfected and control. These growth factors also promote cell survival. The authors can not draw conclusion that GS Lrrk2 decrease NGF level, leading to toxicity, while overlook the increases of other growth factors.

à GS LRRK2 did not showed significant increase in BDNF, GDNF, and VEGF compared to WT LRRK2 ro Vector control. It is your misreading.

However, we did re-do of mRNA quantification and performed ELISA to address the issue with figure 2. VEGF was excluded from the overall analysis results because it is an unchanging neurotrophic factor similar to other GDNF and BDNF.

----------------------------------------------------------------------------------------------------------------

2, the similar situation also observed in cytokines with some cytokines increaed and others decreased under GS lrrk2 transfection. Therefore it will be difficult to arrive at the conclusion that GS lrrk2 impact cell viability via modulation of growth factors and cytokines, as unconsistent results have been presented by the author.

à It is your misreading. There was no difference in iNOS between the three experimental groups, and IL-1beta and TNFα were significantly increased in GS-hLRRK2. Even TGFβ, an anti-inflammatory cytokine but not a pro-inflammatory cytokine, was not significantly reduced in GS. Additionally, the release of IL-1beta and TNFα cytokines is significantly increased in GS-hLRRK2. This explains the differences due to GS-hLRRK2.

However, we performed ELISA to address the issue with figure 3. Our intention was to show that TGFβ, an anti-inflammatory cytokine, was also unchanged, but TGFβ was excluded because the description of anti-inflammatory cytokines was not necessary for the content of this study.

----------------------------------------------------------------------------------------------------------------

3, in affects on astrogliosis by wt and GS lrrk2, it is obvious that wt lrrk2 have more significant influence. 

à We did re-do of western blot to address the issue with figure 4.

----------------------------------------------------------------------------------------------------------------

4, they desmonstrate that wt and GS lrrk2 have dustinct impacts, while GS lrrk2 has higher lrrk2 kinase activity than wt lrrk2 and overexpression of wt lrrk2 can also increase lrrk2 kinase activity in cells. So they only have different lrrk2 kinase increase and should not show opposite impacts on cells. The authors should exolain it.

à You seem to be talking about experiments with LRRK2 knock down or kinase dead. This part is currently conducting a comparative study with GS LRRK2 on a human brian organoid, a kinase down-regulated mutant. This is research we plan to present at another opportunity.

To avoid misunderstandings regarding LRRK2 kinase, we added the results and descriptions of kinase activity between WT-hLRRK2 and GS-hLRRK2 in manuscript.

----------------------------------------------------------------------------------------------------------------

5, based on evidence provided, it will be difficult to arrive at the conclusion the authors claimed.

à Although this may be presumptuous, it is a misreading or misunderstanding on your part. You seem to be saying that it is inconsistent with the conclusion because you read the paper interpreting the significance shown in the results as a significant increase or decrease according to your standards rather than following the symbols I drew in the picture.

----------------------------------------------------------------------------------------------------------------

Minor defects:

1,only Rt-PCR can not comfirm the expression changes of genes, western blot analysis is indispensible.

à To solve the problem you pointed out, we performed neurotrophic factor ELISA and pro-inflammatory cytokine ELISA. And we added the image of TNFα, the highest pro-inflammatory cytokine, and NGF, the significant low neurotrophic factor.

2, studies on only one kind of cell line will have limited significance. The authors should validate their findings in multiple cell lines and models.

à First, even if there is a glioma cell line in the secondary cell line, there is no cell line with only astrocyte properties, so experiments must be conducted only with the primary cell line. However, to fully check the interactions between organic brain cells rather than experimenting only with cells, mutants with differences in kinase activity or GTPase activity, such as G2029S, R1441G, and G2385R LRRK2, are being studied in human organoids as described above. It is scheduled to be presented as a next paper.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors of this manuscript have addressed all of my significant concerns, and the current version can be accepted for publication in this journal. I don´t have more suggestions for them

No comments

Author Response

As you suggested, we have corrected the English text and will attach a certificate for it.

Author Response File: Author Response.pdf

Reviewer 2 Report

It is obvious that the revised manuscript has been significantly improved. However, there are some concerns need to be addressed.

1, the figure 4 is overlapped with old figure, so can not be read clearly

2, It is suggested that overexpression of mutant LRRK2 in rat primary astrocytes affect N27 cells via secretion of some cytokines, such as TNFa and IL-1b (Fig 3). However, to further establish the relationship between these cytokines and N27 cells, the authors should add purified TNFa and IL-1b to medium for N27 cells and see the effects subsequently. The cytokine only treatment should be similar as treatment from CM of mutant lrrk2 transfected astrocytes. 

Author Response

1, the figure 4 is overlapped with old figure, so can not be read clearly

I don't know why you said Figure 4 overlapped. If you pointed out overlap in the content part, then Western blot confirmed again that NGF, the neurotrophic factor that was most significantly decreased, and TNFα, the neuroinflammation-causing cytokine that was most significantly increased, were confirmed previously. In addition, the previous cell viability is also reconfirmed with PARP1, and this content has already been described in the results section. Furthermore, didn't you point out in the first decision that protein quantification was essential?  We proposed a result to address your previous point.

And if the overlap you are talking about means that the Western blot image in the previous manuscript before the first decision overlaps with the current image, then you are mistaken. This is because figure 4 in the revised manuscript was obtained by performing a completely new experiment with n=4.

Lastly, since it was confirmed that you had doubts about the demsitometer in figure 4 in the first decision, we also considered cases where you did not understand the quantitative part this time. So, we are going to capture and show the process of Western blot image quantification.

In case there are any parts that may be confusing due to our English language skills, we have corrected the English and a certificate of that is also attached.

If, despite our explanation, you say that Figure 4 does not read clearly and overlaps, there is nothing more we can do.

-----------------------------------------------------------------------------------------

2, It is suggested that overexpression of mutant LRRK2 in rat primary astrocytes affect N27 cells via secretion of some cytokines, such as TNFa and IL-1b (Fig 3). However, to further establish the relationship between these cytokines and N27 cells, the authors should add purified TNFa and IL-1b to medium for N27 cells and see the effects subsequently. The cytokine only treatment should be similar as treatment from CM of mutant lrrk2 transfected astrocytes. 

We already have been studying the effect of NGF reduction, TNFα increase on the dopaminergic neuronal healthy and its dopamine synthesis using human brain organoids composed of reprogrammed human dopaminergic neuronal progenitor and astrocyte as we told last rebuttal letter. The neutralizing antibody for TNFα or NGF were used in human brain organoid of LRRK2 G2019S patients or wild type. To put it briefly, we found that neutralization of beta-NGF in wild type organoid for a month significantly decreased the release of dopamine and TH levels compared to the treatment of beta-NGF. And the co-treatment of TNFα neutralizing antibody and beta-NGF protein in human brain organoid of LRRK2 G2019S patients exhibited significantly rescued the dopaminergic neuronal degeneration and dopamine release. We do not plan to describe the results of experiments on organoids in this study. Moreover, unfortunately, re-do the part you pointed out is no longer possible due to current financial and research resource issues in this institution. Since organoid research was not conducted at this institution from the beginning, it cannot be included in this study. I am already serving as the founder/CEO/CTO of another organization, and this research paper is my last one at this organization.

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

The revised manuscript has been significantly improved. However, the author has mentioned their results on brain organoid model under LRRK2 mutation to support their conclusions in the current study. Therefore the author should provide the brain organoid results or cite the papers on their brain organoid studies to support their conclusions. 

Author Response

The revised manuscript has been significantly improved. However, the author has mentioned their results on brain organoid model under LRRK2 mutation to support their conclusions in the current study. Therefore the author should provide the brain organoid results or cite the papers on their brain organoid studies to support their conclusions. 

-----------------------------------------------------------------------------------------

I understand your point. I'm really sorry, but that content cannot be included in this paper. And since the paper has not yet been published and is a paper that will be released in conjunction with our company's research on a new drug that increases NGF, it is impossible to respond to your request.