Galectin-3 Mediates Tumor Progression in Astrocytoma by Regulating Glycogen Synthase Kinase-3β Activity

Round 1

Reviewer 1 Report

Dear authors,

 

You have done a great and elegant work, nevertheless a have two suggestions and one question.

1. The term, glioblastoma multiforme is not more in use. The term glioblastoma would be the correct one.

2. In row 51, I suppose the authors refer that Galectin 3 is one of the Galectin family members…  

3. The final sentence of the discussion states, “Therefore, galectin-3 is upregulated by GSK3B in astrocytoma”, whereas in the conclusions is said, “galectin-3 mediated tumor progression by upregulating GSK3B protein expression”. I think there is a contradiction.  Which is the right one?

Author Response

You have done a great and elegant work, nevertheless a have two suggestions and one question.

Thank you for your affirmation and encouragement.

1. The term, glioblastoma multiforme is not more in use. The term glioblastoma would be the correct one.

Thanks for your remind. We modified it.

2. In row 51, I suppose the authors refer that Galectin 3 is one of the Galectin family members…  

Thanks for your remind. We modified it.

 

3. The final sentence of the discussion states, “Therefore, galectin-3 is upregulated by GSK3B in astrocytoma”, whereas in the conclusions is said, “galectin-3 mediated tumor progression by upregulating GSK3B protein expression”. I think there is a contradiction.  Which is the right one?

It is my mistake. “galectin-3 mediated tumor progression by upregulating GSK3B protein expression” is the right. Therefore, we modified the sentence “GSK3B is upregulated by galectin-3 in astrocytoma.”

Author Response File: Author Response.pdf

Reviewer 2 Report

Authors present a study of Galectin-3 expression in astrocytomas. As is well known, the activity of this marker positively regulates the carcinogenesis of many tumor species including certainly breast. 

I believe that selecting such a marker studied in other tumor species to gliomas needs a scientific rationale better described from the introduction. Indeed, it is not useful to choose a marker present in a tumor and apply it empirically to gliomas: is there a target therapy capable of crossing the blood-brain barrier? Are there any known mutations or signal cascades in glioma that might suggest direct Gal-3 involvement? Do gliomas after WHO 2021 have distinct molecular profiles delineated ab-initio across species, which ones might Gal-3 be involved in? Also it is mentioned that there is downregulation of ki67 which plays a very important role in the prognosis of GB patients (and less in low grade gliomas, see Armocida D, Frati A, Salvati M, Santoro A, Pesce A. Is Ki-67 index overexpression in IDH wild type glioblastoma a predictor of shorter Progression Free survival? A clinical and Molecular analytic investigation. Clin Neurol Neurosurg. 2020 Nov;198:106126. doi: 10.1016/j.clineuro.2020.106126. Epub 2020 Aug 3. PMID: 32861131.) would it be worth doing a study specifically on GBs?

 

The paper inherently is well written and the methodology is well described, we need to work better on describing the rationale for the study

Author Response

Authors present a study of Galectin-3 expression in astrocytomas. As is well known, the activity of this marker positively regulates the carcinogenesis of many tumor species including certainly breast. 

I believe that selecting such a marker studied in other tumor species to gliomas needs a scientific rationale better described from the introduction. Indeed, it is not useful to choose a marker present in a tumor and apply it empirically to gliomas: is there a target therapy capable of crossing the blood-brain barrier?

Thanks for your remind. In this study, we found that galectin-3 regulated tumor progression through Wnt pathway. In my next step, we want to used docking for galectin-3 to select the drug. About crossing the BBB, we design a nanoparticle for antibody-drug conjugate to target astrocytoma in our team. Therefore, we will combine this study and nanoparticle to make a new drug for inhibiting galectin-3 to attenuate tumor progression in astrocytoma.

Are there any known mutations or signal cascades in glioma that might suggest direct Gal-3 involvement?

Thanks for your remind. In our opinions, there was no report about the upstream of galectin-3. It is one of my future work.

Do gliomas after WHO 2021 have distinct molecular profiles delineated ab-initio across species, which ones might Gal-3 be involved in?

Thanks for your remind. In this study, we select the patients from 2008 to 2018 in our hospital. However, the patients were not identified by molecular pathology including IDH, ATRX, and 1p/19q. We will re-identify all classes by WHO 2021. For now, we don’t know which molecule is associated with galectin-3.

Also it is mentioned that there is downregulation of ki67 which plays a very important role in the prognosis of GB patients (and less in low grade gliomas, see Armocida D, Frati A, Salvati M, Santoro A, Pesce A. Is Ki-67 index overexpression in IDH wild type glioblastoma a predictor of shorter Progression Free survival? A clinical and Molecular analytic investigation. Clin Neurol Neurosurg. 2020 Nov;198:106126. doi: 10.1016/j.clineuro.2020.106126. Epub 2020 Aug 3. PMID: 32861131.) would it be worth doing a study specifically on GBs?

Thanks for your remind. In this study, ki-67 was one factor of tumor progression. Ki-67 is an important role in glioblastoma and is used for diagnosis in our hospital. I believed that ki-67 is worth for a study specifically on GBs.

The paper inherently is well written and the methodology is well described, we need to work better on describing the rationale for the study

Thank you for your affirmation and encouragement.

Author Response File: Author Response.pdf

Reviewer 3 Report

Tsai and co-authors propose a study aiming to show a correlation between clinical parameters in astrocytoma and Galectin 3/GSK3B expression, and importantly between these two proteins with tumorigenic roles. It is shown that Galectin 3 and GSK3B are independent prognostic factors and that Galectin 3 controls GSK3B expression.

This work has potential interest for researchers in the field. Indeed, the relationship between Galectin 3 and GSK3B is poorly understood. However, the demonstration needs to be improved to determine the impact of Galectin 3 on GSK3B regulation and tumor cells behavior. Indeed, from a biological and pharmacological point of view, it is important to know if Galectin 3 controls the activation or inhibition of upregulated GSK3B, and if the inhibition of both proteins could induce death of astrocytoma cells. This knowledge is required before claiming that Galectin 3 and GSK3B represent anticancer targets for astrocytoma therapy.

Below are suggestions to improve the manuscript:

Main

-The number of experiments (n) in each biological assay must be indicated in the legend, as well as the statistical test used.

-It is crucial to determine if the MTT assay data indicate death or decreased proliferation of tumor cells. Experiments aiming to measure caspase activation should be performed.

-Serine 9 phosphorylation of GSK3B, as well as B catenin level, must be analyzed by WB to determine the activation status of GSK3B and Wnt pathway, respectively, after Galectin 3 siRNA.

-WB of Ki67 and VEGF are of very poor quality and not convincing. They must be improved.

Minor

-Correct references for figures: there is not figure 1D, figures 6-7-8-9

-It should be of interest to present the different cell lines used in this study and to comment for a possible relationship between their tumor potential (proliferation, invasion, resistance to therapies) and the expression levels of Galectin 3/GSK3B. 

-Affiliation 6 does not correspond to authors.

-Line 35: correct for might be considered

-Line 40: correct for were

-Line 55: correct for in addition

-Line 63: correct for reports

-Line 114: add a space

-Line 126: correct anti-B actin concentration

-Line 127: add a space

-Line 132: discard in vitro

-Line 142: replace was/by

-Figure 2 legend: Galectin 3 and GSK3B are inversed; there is no p<0.05

-Line 227: write migration of

-Line 232: write abilities of

-Figure 3B: improve the image contrast; discard siRNA in the title; precise the exact time of the wound healing assay

-Line 254: correct for binds

-Line 272: correct for previous

-Line 277: replace that regulated by to regulate

-Line 296: correct by GSK3B is upregulated

-Line 302: correct for of astrocytoma cells

Author Response

Tsai and co-authors propose a study aiming to show a correlation between clinical parameters in astrocytoma and Galectin 3/GSK3B expression, and importantly between these two proteins with tumorigenic roles. It is shown that Galectin 3 and GSK3B are independent prognostic factors and that Galectin 3 controls GSK3B expression.

This work has potential interest for researchers in the field. Indeed, the relationship between Galectin 3 and GSK3B is poorly understood. However, the demonstration needs to be improved to determine the impact of Galectin 3 on GSK3B regulation and tumor cells behavior. Indeed, from a biological and pharmacological point of view, it is important to know if Galectin 3 controls the activation or inhibition of upregulated GSK3B, and if the inhibition of both proteins could induce death of astrocytoma cells. This knowledge is required before claiming that Galectin 3 and GSK3B represent anticancer targets for astrocytoma therapy.

Thank you for your affirmation and encouragement.

Below are suggestions to improve the manuscript:

Main

-The number of experiments (n) in each biological assay must be indicated in the legend, as well as the statistical test used.

Thanks for your remind. We had filled in the number of experiments in all figure legend.

-It is crucial to determine if the MTT assay data indicate death or decreased proliferation of tumor cells. Experiments aiming to measure caspase activation should be performed.

Thanks for your remind. We added the results of western blot for PARP and cleaved caspase-3 in Figure 3.

-Serine 9 phosphorylation of GSK3B, as well as B catenin level, must be analyzed by WB to determine the activation status of GSK3B and Wnt pathway, respectively, after Galectin 3 siRNA.

Thanks for your remind. We added the results of western blot for p-GSK3B ser9 and β-catenin in Figure 4.

-WB of Ki67 and VEGF are of very poor quality and not convincing. They must be improved.

Thanks for your remind. We renewed the western blot of Ki67 and VEGF in Figure 4.

Minor

-Correct references for figures: there is not figure 1D, figures 6-7-8-9

Thanks for your remind. It is our mistake. We had modified it.

-It should be of interest to present the different cell lines used in this study and to comment for a possible relationship between their tumor potential (proliferation, invasion, resistance to therapies) and the expression levels of Galectin 3/GSK3B. 

Thanks for your remind. We added all data (including proliferation, invasion, migration and western blot) of U87-MG in Figure 2, 3 and 4.

 

-Affiliation 6 does not correspond to authors.

Thanks for your remind. We modified it.

-Line 35: correct for might be considered

Thanks for your remind. We modified it.

-Line 40: correct for were

Thanks for your remind. We modified it.

-Line 55: correct for in addition

Thanks for your remind. We modified it.

-Line 63: correct for reports

Thanks for your remind. We modified it.

-Line 114: add a space

Thanks for your remind. We modified it.

-Line 126: correct anti-B actin concentration

Thanks for your remind. We modified it.

-Line 127: add a space

Thanks for your remind. We modified it.

-Line 132: discard in vitro

Thanks for your remind. We modified it.

-Line 142: replace was/by

Thanks for your remind. We modified it.

-Figure 2 legend: Galectin 3 and GSK3B are inversed; there is no p<0.05

Thanks for your remind. We modified it.

-Line 227: write migration of

Thanks for your remind. We modified it.

-Line 232: write abilities of

Thanks for your remind. We modified it.

-Figure 3B: improve the image contrast; discard siRNA in the title; precise the exact time of the wound healing assay

Thanks for your remind. We renewed the data in figure 3.

-Line 254: correct for binds

Thanks for your remind. We modified it.

-Line 272: correct for previous

Thanks for your remind. We modified it.

-Line 277: replace that regulated by to regulate

Thanks for your remind. We modified it.

-Line 296: correct by GSK3B is upregulated

Thanks for your remind. We modified it.

-Line 302: correct for of astrocytoma cells

Thanks for your remind. We modified it.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

It's ok add the reference suggested during revision: Armocida D, Frati A, Salvati M, Santoro A, Pesce A. Is Ki-67 index overexpression in IDH wild type glioblastoma a predictor of shorter Progression Free survival? A clinical and Molecular analytic investigation. Clin Neurol Neurosurg. 2020 Nov;198:106126. doi: 10.1016/j.clineuro.2020.106126. Epub 2020 Aug 3. PMID: 32861131.) would it be worth doing a study specifically on GBs?

Author Response

It's ok add the reference suggested during revision: Armocida D, Frati A, Salvati M, Santoro A, Pesce A. Is Ki-67 index overexpression in IDH wild type glioblastoma a predictor of shorter Progression Free survival? A clinical and Molecular analytic investigation. Clin Neurol Neurosurg. 2020 Nov;198:106126. doi: 10.1016/j.clineuro.2020.106126. Epub 2020 Aug 3. PMID: 32861131.) would it be worth doing a study specifically on GBs?

 

Thanks for your remind. I had added in ref [27].

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors have followed most of my suggestions to improve the manuscript. However, it remains some corrections/additions to do before publication:

-In the Abstract, it is important to mention that the downregulation of galectin and GSK3b decreases survival and b catenin in GB cells.

-In Mat et Met, antibodies used for the detection of PARP and Caspase 3 must be indicated. Also, correct for Abcam as supplier.

-Fig1: Again, Fig1D does not exist. So, discard it from the text please.

-Fig3 legend: In A, it is not a growth curve which is presented but the result of a MTT test. The part B with WB of PARP and caspase 3 is not described. In the title, there is one word si RNA to discard.

 

Author Response

The authors have followed most of my suggestions to improve the manuscript. However, it remains some corrections/additions to do before publication:

 

-In the Abstract, it is important to mention that the downregulation of galectin and GSK3b decreases survival and b catenin in GB cells.

Thanks for your remind. We modified it to “Galectin-3 or GSK3B downregulation induced apoptosis and decreased cell numbers, migration, and invasion. siRNA-mediated gene silencing of galectin-3 resulted in the downregulation of Ki-67, cyclin D1, VEGF, GSK3B, p-GSK3B Ser9 (p-GSK3B S9) and β-catenin. In contrast, GSK3B knock-down only decreased Ki-67, VEGF, p-GSK3B S9 and β-catenin protein expression but did not affect cyclin D1 and galectin-3 protein expression. The siRNA results indicated that GSK3B is downstream of the galectin-3 gene. These data support that galectin-3 mediated tumor progres-sion by upregulating GSK3B and β-catenin protein expression in glioblastoma. Therefore, galec-tin-3 and GSK3B are potential prognostic markers, and their genes may be considered to be anti-cancer targets for astrocytoma therapy.”

 

 

-In Mat et Met, antibodies used for the detection of PARP and Caspase 3 must be indicated. Also, correct for Abcam as supplier.

Thanks for your remind. We added “anti-PARP (1:500; Cell Signaling), anti-cleaved caspase-3 (1:200; Cell Signaling)” and modified “Abcan” to “Abcam”.

 

-Fig1: Again, Fig1D does not exist. So, discard it from the text please.

Thanks for your remind. We had finished it.

 

-Fig3 legend: In A, it is not a growth curve which is presented but the result of a MTT test. The part B with WB of PARP and caspase 3 is not described. In the title, there is one word si RNA to discard.

Thanks for your remind. In A, we modified “growth curve” to “cell numbers”. The part B, we added “(B) The protein expressions of PARP and cleaved caspase-3 between control, nonsense siRNA, GSK3B siRNA and galectin-3 siRNA for 3 days in following transfection with siRNA.” in figure legend. Finally, we detect one “siRNA”.

Thanks for your help.

Author Response File: Author Response.pdf