Slice of Life: Porcine Kidney Slices for Testing Antifibrotic Drugs in a Transplant Setting

Round 1

Reviewer 1 Report

My congratulations to the authors. This is a beautiful example of the application of tissue slices. It shows how a decades-old pharmacological technique is viable and applicable today. The use case demonstrated in this manuscript is legitimate, and the pharmacological intervention with galunisertib to reduce pro-fibrotic tissue signals has, in my impression, clearly worked. I am convinced that the technique used in the species used has implications for application in humans.

The weak part of the manuscript is that the authors did not take the opportunity to either a) show application in human kidney biopsies/discarded kidneys or b) during pig kidney normothermic machine perfusion. Retransplantation of kidneys perfused with galunisertib would then show the real benefits of the treatment but may be beyond the scope of this manuscript.

I do have comments the authors need to address before acceptance of the manuscript is possible:

Major comments:

1.     Figure 2: The real control is missing from the figures. Please include data for the fresh tissue, either at the time of biopsy or at the time point before incubation, especially since this data is shown in figure 3. Thus, the reader can interpret the control (after incubation under standard conditions?) shown in figure 2. This should then additionally be discussed. Also, please normalize to control (scale bar).

2.     Data from the histological analysis is shown, but no exemplary images. This needs to be shown so the reader gets an accurate impression of tissue quality and staining intensity.

3.     The method for quantification of a-SMA is not very strong. I highly recommend performing western blots of tissue after incubation (normalized to protein content) for a more accurate assessment of relative intensities. The effects in a-SMA seen between baseline and control in figure 3, to me, seem to be an effect of tissue expansion rather than an actual decrease, which would be a highly unexpected finding.  

4.     Why do you show FN-2 in figure 2, but FN-1 in figure 5? Please explain. Also, two additional fibrosis markers are introduced in figure 5 that were not checked in figure 2. Why?

5.     a-SMA staining did not show improvement with galunisertib + TGF-B1. The discussion of this is legitimate, but please also include data for this (maybe in the supplement).

Minor comments:

1.     Please discuss/explain the application, useability, and previous use of kidney slices in the transplant setting more in-depth – how could the compounds be used during transplantation, and how do kidney slices help with that? The way this is introduced right now does not make it clear how the application here is any different from the in vivo situation. E.g. we can assume you do plan to go towards normothermic machine perfusion with this, but it is not clear.

2.     Page 5, line 195 – none of the treatments induced toxic effects is a very strong statement, please rephrase and be more specific.

3.     In graphs, please increase dot size and consider using dots, error bars, etc. in black. This would significantly increase the readability.

4.     Figure 5: Please also include statistical info for control vs. TGF-B1 + galunisertib

Author Response

Dear reviewer,

We are pleased to read the overall positive comments regarding our manuscript and would like to thank you for the opportunity to provide a revised version for consideration. We have addressed all issues raised and agree that with these changes, the quality of the manuscript is improved. Please find our point-by-point response and changes bellow as well as marked yellow in the manuscript.

My congratulations to the authors. This is a beautiful example of the application of tissue slices. It shows how a decades-old pharmacological technique is viable and applicable today. The use case demonstrated in this manuscript is legitimate, and the pharmacological intervention with galunisertib to reduce pro-fibrotic tissue signals has, in my impression, clearly worked. I am convinced that the technique used in the species used has implications for application in humans.

The weak part of the manuscript is that the authors did not take the opportunity to either a) show application in human kidney biopsies/discarded kidneys or b) during pig kidney normothermic machine perfusion. Retransplantation of kidneys perfused with galunisertib would then show the real benefits of the treatment but may be beyond the scope of this manuscript.

I do have comments the authors need to address before acceptance of the manuscript is possible:

We agree that normothermic machine perfusion would strengthen the manuscript and we are currently working on both porcine and discarded human kidney perfusion with galunisertib. We decided that it would be too many different experiments to add these experiments to this manuscript and just focus on our tissue slice model instead. Galunisertib has previously been tested in human PCKS as extensively described by Bigaeva et al. as mentioned this in the discussion (page 11, line 336-338).

Major comments:

1. Figure 2: The real control is missing from the figures. Please include data for the fresh tissue, either at the time of biopsy or at the time point before incubation, especially since this data is shown in figure 3. Thus, the reader can interpret the control (after incubation under standard conditions?) shown in figure 2. This should then additionally be discussed. Also, please normalize to control (scale bar).

Thank you for the pointed remark. We went back to our raw PCR data and made the following adjustments. First, we added the data for our baseline samples (before incubation/PCKS 0 hours). Second, we normalized our data to the control by equalizing the control to 1. These changes led to slight alterations in statistics. We agree that with these alterations, figures 2 and 5 are more complete (page 7, line 220-226 and page 9, line 259-260). We have adjusted our results and discussion based on the revised figures (page 6, line 211-220; page 9, line 252-261; page 10, line 281-286).

2. Data from the histological analysis is shown, but no exemplary images. This needs to be shown so the reader gets an accurate impression of tissue quality and staining intensity.

We agree that an exemplary image would strengthen figure 3. We have therefore added this to figure 3 (page 8 line 233-236).

3. The method for quantification of a-SMA is not very strong. I highly recommend performing western blots of tissue after incubation (normalized to protein content) for a more accurate assessment of relative intensities. The effects in a-SMA seen between baseline and control in figure 3, to me, seem to be an effect of tissue expansion rather than an actual decrease, which would be a highly unexpected finding.  

We agree that a western blot is a more accurate assessment of aSMA. However, we did not have sufficient samples left to perform a western blot. We will take your comment into consideration for our follow-up experiments. 

4. Why do you show FN-2 in figure 2, but FN-1 in figure 5? Please explain. Also, two additional fibrosis markers are introduced in figure 5 that were not checked in figure 2. Why?

Thank you for this comment. We choose the primers encoding for fibronectin based on the availability in our lab. As both primers encode for the fibronectin protein, we hypothesized that it would lead to similar results. We agree that it would have been more accurate to use one primer for both experiments. Since we did not look at fibrosis on a protein level in experiment two, we increased our panel to strengthen the results. We have added a paragraph to the discussion explaining why we chose the measured fibrosis markers (page 10, line 309-317).

5. a-SMA staining did not show improvement with galunisertib + TGF-B1. The discussion of this is legitimate, but please also include data for this (maybe in the supplement).

 We did not stain for a-SMA during the second part of the experiment as we did not see any differences during the first part of the experiment. We therefore chose a broader range of profibrotic mRNA markers.

Minor comments:

1. Please discuss/explain the application, useability, and previous use of kidney slices in the transplant setting more in-depth – how could the compounds be used during transplantation, and how do kidney slices help with that? The way this is introduced right now does not make it clear how the application here is any different from the in vivo situation. E.g. we can assume you do plan to go towards normothermic machine perfusion with this, but it is not clear.

Thank you for this insightful comment. We agree that this information is missing from the manuscript and have added an extra paragraph to the introduction explaining why we use PCKS (page 2, line 76-86)

2. Page 5, line 195 – none of the treatments induced toxic effects is a very strong statement, please rephrase and be more specific.

Thank you for this comment. Indeed, this is a very strong statement. We have removed it from the text and now only stated that viability was maintained (page 5, line 170).

3. In graphs, please increase dot size and consider using dots, error bars, etc. in black. This would significantly increase the readability.

We have increased the dot size and changed all dots and error bars to black. We agree that this improves the readability (see all figures).

4. Figure 5: Please also include statistical info for control vs. TGF-B1 + galunisertib

Thank you for this comment. We have performed a Dunnett’s multiple comparison analysis comparing the mean of each group to the mean of each other group. We have added this information to figure 5 (page 9 line 262-270).

Reviewer 2 Report

The authors propose a model for studying fibrosis after DCD-kidney transplantation,by using precision-cut kidney slices (PCKS) took from porcine kidneys after the exposure to 30min of warm ischemia followed by 3h of oxygenated hypothermic machine perfusion.

They incubated the PCKS for 48 hrs with different anti-fibrotic compounds (Galunisertib, taurine, febuxostat, and doxycycline) or a control solution. To assess the viability the authors measured the ATP production and found an increase, when compared to baseline, but no difference among antifibrotic agents or control. When analyzing the expression of fibrotic genes, only Galunisertib showed a reduction after incubation, while no compounds significantly reduced a-SMA expression.

The authors investigated then the anti-fibrotic effect of Galunisertib upon TGFB1 stimulation of PCKS. After 48 hrs of TGFB1 incubation profibrotic genes (TGF-β1, FN1, PAI-1, HSP47, and COL1A2) were signicantly increased, while in the presence of Galunisertib this process was limited. No difference  was found again in the a-SMA expression.

The paper is well written and it is of very interest. The authors declare that this model can have limitations regarding the sample size, the multicellular structure and the adjustment of the different compound concentrations (previously tested on animal models) that could have influence the significance in the other compounds.

My main concern is regarding the use of DMSO in the TGFB1+Galunisertib group and not in TGFB1 group that could have biased the profibrotic gene expression. Could the authors explain why did they use this strategy?

Author Response

Dear reviewer,

We are pleased to read your positive feedback regarding our manuscript and would like to thank you for the opportunity to provide a revised version for consideration. We understand your concern is regarding the use of DMSO and must admit that this was incorrectly stated in our manuscript. We did add DMSO to all of our groups to rule out any (negative) effects caused by DMSO. We have corrected table 2 (page 4, line 148-149 in yellow). Pardon me for the confusion.

Round 2

Reviewer 1 Report

The authors changed and, in my opinion, improved the manuscript according to the requests. I am happy to support the publication in present form and am looking forward to the future work of the authors. 

Reviewer 2 Report

The paper has been substantially improved after the revision.