The Androgen Hormone-Induced Increase in Androgen Receptor Protein Expression Is Caused by the Autoinduction of the Androgen Receptor Translational Activity

Round 1

Reviewer 1 Report

The authors tested 3 prostate cancer cells lines for responses to androgen exposure and confirmed biphasic response previously described in LNCaP cells but found this effect was not seen in the other two cell lines. The conducted further experiments to clarify AR responses to androgen exposure.

Generally the manuscript flows well and the methods are well described.

The wording of 3rd paragraph in the abstract (lines 30-33) is very convoluted and needs to be clarified/simplified.

Phrase "even when not significant" used several times in results (eg Line 232) - what does this mean? When not statistically significant?

The final paragraph of the discussion is very dense and needs to be broken into smaller paragraphs for better readability.

Author Response

Reviewer 1

On behalf of all authors, we would like to take this opportunity to express our sincere gratitude to the reviewers who identified areas of our manuscript that needed correction or modification. Their insightful comments have led to an improvement in our manuscript. Below you find the detailed response to the reviewers' comments:

The authors tested 3 prostate cancer cells lines for responses to androgen exposure and confirmed biphasic response previously described in LNCaP cells but found this effect was not seen in the other two cell lines. The conducted further experiments to clarify AR responses to androgen exposure.

Generally the manuscript flows well and the methods are well described.

The wording of 3rd paragraph in the abstract (lines 30-33) is very convoluted and needs to be clarified/simplified.

The 3rd paragraph was partially rephrased to increase clarity.

Phrase "even when not significant" used several times in results (eg Line 232) - what does this mean? When not statistically significant?

The phrase was demanded in several previous manuscripts by reviewers. However, as the phrase led to confusion, we deleted the paragraph, as it does not impact the manuscript's content.

The final paragraph of the discussion is very dense and needs to be broken into smaller paragraphs for better readability.

We broke up the conclusion into smaller paragraphs to increase readability.

Reviewer 2 Report

This is a study where the effects of induction with the synthetic androgen R1881 on AR expression at both the RNA and protein level as well as the transcriptional level of some downstream genes known to be regulated by AR are studied on three different cell lines. Translocation of the AR receptor into the nucleus is also studied.

I find the Introduction and Materials and Methods sections well written, but have some problems with the Result section and description on what is actually shown in the different figures.

I also think the authors use the word “peak” at several occasions in a somewhat misleading context. In my mind, a peak is the top concentration/level of something, where the levels of whatever you’re measuring are lower before and after. As I interpret the figures, you can instead see what more resembles a plateau reaching over several concentrations.

Specific comments:

1. I would appreciate if all the supplementary figures could be clearly marked so it is easier for the reader to understand what the figures belong to -the first marking comes on page 3 of the supplementary files so it is very hard for the reader to understand what the two first pages of supplementary figures represent. It would also greatly improve the understanding for the reader if figure legends describing what you see in the different supplementary figures were added.
2. Line 78: Should “three” be replaced by “two”?
3. Line 160 Section “eosFP-AR transfection”: it would be nice if the authors could give some more information about the plasmid used and how it works.
4. Fig 1 lanes 184 and 191: The authors claim they have a peak in the cell viability at 1 nM for LNCaP – I think it would be more correct to describe it as highest cell viability is found between 10^-1 to 10⁰ nM, as it looks like it is slightly higher, although not statistically significant, at 10⁰
5. Line 185: Please rephrase the sentence “However, even when not significant, this effect was…” as it is a bit difficult to understand.
6. Line 186 and 277: Authors state that viability in C4-2 is not affected by increased levels of R1881 in contrast to LNCaP. Looking at Figure 1, does it not look like there is a tendency, although not significant, that C4-2 is even more sensitive to high R1881 concentrations as you get a reduction in cell viability already at 1 nM?
7. Figure legend 2 lanes 207-208: Please correct so it fits with what is shown in the figure. In the legend it states that relative changes are shown after 16 h at 1 nM, but at the x-axis several different concentrations of R1881 is displayed.
8. Heading of Figure legend 3 is misleading when stating that R1881 increase nuclear localization in a time-dependent manner as only results from time-point 2 h is shown in the figure.
9. It would have been nice to complement the concentration dependent results in Figure 3 with a time dependent experiment to show that the AR in LNCaP behaves in the same way as the fusion protein in the supplementary data of PC3 cells.
10. Line 233: Should Fig 3B be 3C?
11. Line 234: I miss the unit for the time in the induction experiment.
12. Lines 238-239: Should Figure 3D be 4D and Figure 3E be 4E?
13. Lines 237-239: I do not agree with the authors interpretation that AR expression is peaking at 1 nM. In Fig 4B, it looks like more AR is expressed at 10 nM than 1 nM, and in 4C already at 0.1 nM. I think this is in line with what is shown in Fig 4D as well. And I do not see that >1 nM R1881 results in lower expression of AR, which is somehow insinuated when using the expression that you have a peak at 1 nM. But when it comes to the induction of PSA, I agree that this is peaking at 1 nM for LNCaP, but not for C4-2 where you seem to have a more even increase in expression related to the R1881 concentration.
14. Fig 4A: What is LAPC4 shown in the small figure legend?
15. Fig 4D y-axis: Are the AR levels really adjusted to PSA mRNA levels? If so, why?
16. Lines 287 and 296: Also here, I think it is wrong to state you have a peak at 1 nM, when you do not get a reduction at higher concentrations. It would be more correct to describe this as you reach the plateau at 1 nM and no more expression could be induced using higher concentrations.
17. Line 297: Could the authors please add some information of where they get the information for the conclusion “As this regulation is independent of changes in AR transcription and starts after 2 h”?
18. It would be interesting to get more information about the differences between the LNCaP and C4-2 cells. Proteomic and expression array differences between the two cell lines could give valuable information about the differences in cell viability and induction of downstream proteins in relation to different R1881 concentrations.

Author Response

Reviewer 2:

On behalf of all authors, we would like to take this opportunity to express our sincere gratitude to the reviewers who identified areas of our manuscript that needed correction or modification. Their insightful comments have led to an improvement in our manuscript. Below you find the detailed response to the reviewers' comments:

Specific comments:

1. I would appreciate if all the supplementary figures could be clearly marked so it is easier for the reader to understand what the figures belong to -the first marking comes on page 3 of the supplementary files so it is very hard for the reader to understand what the two first pages of supplementary figures represent. It would also greatly improve the understanding for the reader if figure legends describing what you see in the different supplementary figures were added.

The missing labelling had been added to the supplementary figures. Moreover, a brief description has been added to each figure and the "Supplementary Materials" paragraph. We hope the added information will improve the understanding of the reader.

1. Line 78: Should "three" be replaced by "two"?

We apologise for this mistake. The issue has been corrected.

1. Line 160 Section "eosFP-AR transfection": it would be nice if the authors could give some more information about the plasmid used and how it works.

Additional information about the overexpression plasmid has been added to the eosFP-AR transfection paragraph.

1. Fig 1 lanes 184 and 191: The authors claim they have a peak in the cell viability at 1 nM for LNCaP – I think it would be more correct to describe it as highest cell viability is found between 10^-1 to 10⁰ nM, as it looks like it is slightly higher, although not statistically significant, at 10⁰

We agree with the author and corrected the issue.

1. Line 185: Please rephrase the sentence "However, even when not significant, this effect was…" as it is a bit difficult to understand.

The sentence has been rephrased.

1. Line 186 and 277: Authors state that viability in C4-2 is not affected by increased levels of R1881 in contrast to LNCaP. Looking at Figure 1, does it not look like there is a tendency, although not significant, that C4-2 is even more sensitive to high R1881 concentrations as you get a reduction in cell viability already at 1 nM?

We reevaluated the data and agree with the reviewer. We rephrased the paragraph.

1. Figure legend 2 lanes 207-208: Please correct so it fits with what is shown in the figure. In the legend it states that relative changes are shown after 16 h at 1 nM, but at the x-axis several different concentrations of R1881 is displayed.

We apologise for this mistake. The issue has been corrected.

1. Heading of Figure legend 3 is misleading when stating that R1881 increase nuclear localisation in a time-dependent manner as only results from time-point 2 h is shown in the figure.

We apologise for this mistake. The issue has been corrected.

1. It would have been nice to complement the concentration dependent results in Figure 3 with a time dependent experiment to show that the AR in LNCaP behaves in the same way as the fusion protein in the supplementary data of PC3 cells.

The authors fully agree that an immunofluorescence approach would be an excellent complement at each time point. Therefore, this approach was also discussed. However, real-time imaging with fusion proteins was previously performed in multiple studies with AR fusion proteins (e.g. PMID:17312014, PMID:16613485, PMID:21980429). This approach gives the advantage of obtaining images at a distinct time point without long staining protocols. However, we agree that due to the FP fusion protein, there is the possibility to change the biology of the protein of interest. Therefore, we decided to validate our finding with immunofluorescence at a time-point, showing 100 % nuclear localisation at almost all concentrations in our real-time imaging experiment. Our main reason to avoid these experiments was that we doubt that exact timing, especially at the critical early time points, is possible due to the necessary washing steps before fixation.

1. Line 233: Should Fig 3B be 3C?

Figure 3B shows the nuclear to cytoplasmic ratio evaluated with the cell profiler software. Figure 3C shows the increase in total AR evaluated with cell profiler.

1. Line 234: I miss the unit for the time in the induction experiment.

The time is displayed as the SI-Unit h for hour in line 234: (C) Relative change of total AR after 2 h of R1881 treatment.

1. Lines 238-239: Should Figure 3D be 4D and Figure 3E be 4E?

We apologise for these mistakes. The issues have been corrected.

1. Lines 237-239: I do not agree with the authors interpretation that AR expression is peaking at 1 nM. In Fig 4B, it looks like more AR is expressed at 10 nM than 1 nM, and in 4C already at 0.1 nM. I think this is in line with what is shown in Fig 4D as well. And I do not see that >1 nM R1881 results in lower expression of AR, which is somehow insinuated when using the expression that you have a peak at 1 nM. But when it comes to the induction of PSA, I agree that this is peaking at 1 nM for LNCaP, but not for C4-2 where you seem to have a more even increase in expression related to the R1881 concentration.

We want to thank the reviewer for his comment. We rephrased the paragraph according to his suggestions.

1. Fig 4A: What is LAPC4 shown in the small figure legend?

LAPC4 had been used for the mRNA experiment. However, as we did not use the cell line for the other experiments, we decided not to include it in the manuscript. Therefore, we deleted the leftover information about LAPC4.

1. Fig 4D y-axis: Are the AR levels really adjusted to PSA mRNA levels? If so, why?

We apologise for these mistakes in the axis labelling. The issues have been corrected.

1. Lines 287 and 296: Also here, I think it is wrong to state you have a peak at 1 nM, when you do not get a reduction at higher concentrations. It would be more correct to describe this as you reach the plateau at 1 nM and no more expression could be induced using higher concentrations.

We agree with the reviewer and changed the used wording.

1. Line 297: Could the authors please add some information of where they get the information for the conclusion "As this regulation is independent of changes in AR transcription and starts after 2 h"?

We added some data from the dataset of Massie et al. revealing no change in AR mRNA after R1881 after 2, 4, and 6 h (Figure S5A).

1. It would be interesting to get more information about the differences between the LNCaP and C4-2 cells. Proteomic and expression array differences between the two cell lines could give valuable information about the differences in cell viability and induction of downstream proteins in relation to different R1881 concentrations.

We agree that proteomic and expression arrays of these cell lines would be interesting to reveal changes leading to castration resistance of C4-2. However, array analysis already revealed that only transcription of a low number of genes is changed in C4-2 compared to LNCaP (https://cancerres.aacrjournals.org/content/64/7_Supplement/382.3). Interestingly, it has also been suggested that C4-2 arose from a subpopulation already present in LNCaP cells (PMID: 15162376). One hypothesis may also be the expression of the Androgen Receptor Variant 7, which is expressed in C4-2. The Variant has been related to changes in androgen responsiveness and a possible reason for castration resistance. We added this information into the discussion. However, it needs to be mentioned that our study did not aim to compare LNCaP with C4-2 cells and that a detailed transcriptome and proteome analysis would be beyond the scope of the manuscript.

Round 2

Reviewer 2 Report

I appreciate the changes the authors have made to the manuscript. 

I also think that the sentence "Expression analysis of Massie et al. dataset revealed no increase in AR mRNA levels after R1881 in the first 6 h after treatment (Figure S5A) [33]" added at line 277-278 contributes with valuable information, but I think it needs some rephrasing to be fully understandable. I also think it would be better to move this sentence from the Result section to the Discussion, where it could be inserted before the sentence in Line 329 "As this regulation is independent of changes in AR transcription and starts after 2 h, there is strong evidence that androgens influence AR stability or AR protein turnover.

Author Response

Again we would like to thank the reviewer for his time and help to improve our manuscript. We rephrased the sentence in the result section and hope it is now understandable. As the data had been added to Figure S5A, we think it is necessary to mention it in the result section. Therefore, we added the information also into the discussion.