NFAT2 drives both Orai3 transcription and protein degradation by harnessing the differences in epigenetic landscape

  1. Sharon Raju
  2. Akshay Sharma
  3. Gyan Ranjan
  4. Rajender K Motiani

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

The functional significance of any protein in physiological processes and pathological conditions is largely dependent on its expression profile. Therefore, nature has evolved several autonomous mechanisms to regulate protein expression such as transcription, translation, post-translational modifications and epigenetic changes. These processes are typically controlled by distinct molecular players with no overlapping roles. Here, we reveal that same transcription factor, NFAT2 regulates both transcription and lysosomal degradation of Orai3 oncochannel in a context dependent manner. We demonstrate that NFAT2 drives Orai3 transcription and thereby increases Orai3 levels in non-metastatic cancerous cells. While in invasive and metastatic cancerous cells, NFAT2 induces Orai3 lysosomal degradation by transcriptionally enhancing the levels of MARCH8 E3 ubiquitin ligase. Our biochemical and super-resolution microscopy data show that MARCH8 physically interacts with Orai3 eventually resulting in its degradation. Mechanistically, the dichotomy in regulation of Orai3 expression emerges from the differences in the epigenetic landscape of MARCH8. We uncover that the MARCH8 promoter is highly methylated in non-metastatic cancerous cells and hence NFAT2 does not induce MARCH8 mediated Orai3 degradation in these cells. Importantly, we demonstrate that MARCH8 restricts pancreatic cancer metastasis by targeting Orai3 degradation thereby highlighting pathophysiological importance of this signaling module. Taken together, we report a unique and clinically relevant scenario wherein nature has commissioned the same transcription factor to both enhance and curtail the expression of a target protein.

Highlights

  • ➢ NFAT2 transcriptionally upregulates Orai3 Ca 2+ channel in non-metastatic cells

  • ➢ NFAT2 induces Orai3 lysosomal degradation via MARCH8 E3 ubiquitin ligase in metastatic cells

  • ➢ The dichotomy in NFAT2’s function is an outcome of differences in the methylation status of MARCH8 promoter

  • ➢ MARCH8 inhibits pancreatic cancer metastasis by driving Orai3 degradation

    • Graphical Abstract

    Article activity feed

    1. Review Commons

      Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

      Learn more at Review Commons

      Reply to the reviewers

      1. General Statements [optional]

      We* thank all three Reviewers for appreciating our work and for sharing constructive feedback to further enhance the quality of our work. It is really gratifying to read that the Reviewers believe that this work will be of interest to broad audience and will be suitable for a high profile journal. Further, the experiments suggested by the reviewers will add value to the work and will substantiate our findings. It is important to highlight that we have already performed most of the suggested experiments except a couple of experiments that we have plan to carry out during full revision. Please find below the details of experiments performed and planned to address the reviewers comments. *

      2. Description of the planned revisions

      Reviewer #1

      Comment 6. In Figure 6A, B, does the Orai3 western blot show any of the heavier bands seen in the ubiquitination IP if you show the whole blot? It should.

      Reviewer #2

      Comment 5. Fig. 6A and 6B. Show the full Orai3 and Ubiquitin WBs. As presented the figure current just shows that there are ubiquitin proteins in Orai3 pull down, not that Orai3 is ubiquitinated.

      Reviewer #3

      Comment 3. In the scheme in Fig. 10, the authors highlight that Orai3 is ubiquitinated. Do they have any idea where the site of action of ubiquitination in Orai3 is located?

      Response: We thank the Reviewer 1, 2 and 3 regarding their query on the co-immunoprecipitation assays performed for studying Orai3 ubiquitination. The reviewers are asking for ubiquitination status of Orai3 and the potential sites for Orai3 ubiquitination. To address these comments, we are planning to perform co-immunoprecipitation assays with mutated Orai3 with mutations of potential ubiquitination sites. We have already performed bioinformatic analysis and it revealed presence of three potential ubiquitination sites on Orai3: K2 (present on N-terminal region), K274 and K279 (present on C-terminal region). We would mutate these lysine residues on Orai3 protein via site-directed mutagenesis and check the Orai3 ubiquitination status. These experiments will answer the question raised by Reviewers and strengthen the Orai3 ubiquitination data.

      Please refer to below diagrammatic illustration showing potential ubiquitination sites on Orai3:

      Reviewer #2

      Comment 7. Also, all the imaging and pull down do not prove conclusively direct interaction between MARCH8 and Orai3, they rather show that the proteins are in the same complex. Although it is unlikely best for the text to be moderated accordingly.

      Response: We understand the concern raised by Reviewer 2 regarding direct or indirect interaction of MARCH8 and Orai3. Hence, we are planning to perform co-immunoprecipitation assays in which we delete the MARCH8 interacting domain in Orai3 protein and check the for direct interaction of these proteins. Bioinformatic analysis and literature survey have highlighted two possible MARCH8 interacting domains in Orai3. The first domain is present in 2nd loop region, present between the 2nd and 3rd transmembrane domains at the LMVXXXL (AA113-120) motif and the second domain is present at the GXXXG (AA235-239) motif, present in the 3rd loop region of Orai3. We will remove these domains from Orai3 protein individually and check its effect on MARCH8 interaction. These experiments will provide conclusive evidence of direct interaction between Orai3 and MARCH8.

      Please refer to below diagrammatic illustration displaying potential MARCH8 binding sites on Orai3:

      3. Description of the revisions that have already been incorporated in the transferred manuscript

      Reviewer #1

      Comment 1. The observation that both transcriptional regulation and protein degradation of Orai3 is regulated downstream of one transcription factor is not, in and of itself, entirely surprising. All proteolytic components are transcriptionally regulated and this phenomenon is likely relatively common. However, what I do think is both impressive and important is that the authors have characterized both components of the pathway within a disease context. While I am not going to search the literature for how often transcription and proteolysis are co-regulated for other proteins, it is the case for many short-lived proteins and perhaps many others. As such, discussion throughout the abstract and introduction that co-regulation of these processes is unprecedented should be removed.

      Response: We thank the Reviewer for thinking that our work is both impressive and important. Further, we understand the Reviewer’s point that transcription and proteolysis may be co-regulated for other proteins. However, our extensive literature search did not resulted in such scenarios. Therefore, to best of our knowledge, we are revealing for the first time that same transcription factor regulates both transcription and protein degradation of the same target in a context dependent manner in a single study. In case, Reviewer would still recommend to modify the text in abstract and introduction, we would do it.

      Comment 2. In discussing figure 1, the authors switch from claiming to be studying NFATc binding to studying NFAT expression. This use of 2 different naming conventions is certain to confuse readers; the authors should use the approved current naming system in referring to NFAT isoforms. In which case NFAT2 is NFATc1.

      Response: We would like to thank the Reviewer for highlighting this point. We have effectively addressed this comment by changing the nomenclature of NFAT2 to NFATc1 throughout the manuscript text and figures.

      Comment 3. The ChIP analyses in figures 1H and 7D are important findings, however, there is missing information. Typically, ChIP is used to validate putative binding sites; as such, one would expect 3 separate qPCR reactions for Orai3, not one. It is also important to note that qPCR products should be uniform in size and under 100 bp; here, the product size is not stated. Finally, demonstrating that an antibody targeting ANY other NFAT isoform fails to pull down whatever product this is would increase confidence considerably.

      Also, the gold standard for validating ChIP is to mutate the sites and eliminate binding. The "silver" standard would be to mutate them in your luciferase vector and demonstrate that NFATc1 no longer stimulates luciferase expression. Since neither of these was done, the ChIP data provided should not be considered formally validated.

      Response: We thank the Reviewer for raising this highly relevant concern. In this revised manuscript, we have addressed this comment by performing several additional experiments. The new data provided in the revised manuscript corroborates our earlier results. Indeed, this data has notably strengthen our work.

      In the revised manuscript, we performed ChIP assay where we increased the number of sonication cycles to 35 so as to make sheared chromatin of around 100 bp. Next, we designed primers to amplify individual NFATc1 binding sites on Orai3 promoter, but due to close proximity of the NFATc1 binding sites, we could design two primer sets. The primer first set to amplify the -1017 bp binding site and the second set to amplify the -990 and -920 bp. Further, as suggested by the Reviewer, we performed immunoprecipitation with the four isoforms of NFAT. Our results show that only NFATc1 pulldown shows significant enrichment of Orai3 promoter with both the primer sets as compared to the IP mock samples and other NFAT isoforms (Figure 1J). Hence, our data reveals that only NFATc1 binds to these predicted sites on the Orai3 promoter and it doesn’t show a preference among these binding sites.

      Further, as suggested by the Reviewer, we mutated the Orai3 promoter in luciferase vector with deletions of the individual NFATc1 binding sites and also cloned a truncated Orai3 promoter with no NFATc1 binding sites into the luciferase vector. The luciferase assays with these mutant and truncated promoters show that upon co-expression of NFATc1, the luciferase activity of the mutant Orai3 promoter with deletion of individual NFATc1 binding site is significantly reduced in comparison to wild type Orai3 promoter. Furthermore, the maximum decrease in luciferase activity was seen with the truncated Orai3 promoter with no NFATc1 binding sites (Figure 1I). These results show that NFATc1 binds to the predicted binding sites on Orai3 promoter. Taken together, the additional ChIP assays with the four isoforms of NFAT and luciferase assays with mutated & truncated Orai3 promoters validates the transcriptional regulation of Orai3 by NFATc1.

      Comment 4. In figures 2 and 3, only one cell line is used to represent each of 3 conditions of pancreatic cancer. That is insufficient to make generalized conclusions; some aspects of this figure (expression and stability, not function) should be extended to 2 to 3 cell lines/condition. TCGA data validating this point would also be helpful.

      Response: We really appreciate the feedback given by Reviewer 1. To strengthen our manuscript, we have addressed this comment by performing experiments in 2 cell lines/condition of pancreatic cancer. This new data in the revised manuscript provides substantial evidence for the dichotomous regulation of Orai3 by NFATc1.

      In the revised manuscript, we carried out NFATc1 overexpression and NFAT inhibition via VIVIT studies in three additional cell lines: BXPC-3 (non-metastatic), ASPC-1 (invasive) and SW1990 (metastatic). The results in these cell-lines support our earlier findings as both overexpression of NFATc1 and VIVIT mediated NFAT inhibition leads to transcriptional upregulation of Orai3 in BXPC-3 (non-metastatic) (Figure S3A, D), ASPC-1 (invasive) (Figure S3G, J) and SW1990 (metastatic) (Figure S3M, P). These results are similar to our earlier data from MiaPaCa-2 (non-metastatic), PANC-1 (invasive) and CFPAC-1 (metastatic) cells. Further, NFATc1 overexpression leads to an increase in Orai3 protein levels in BXPC-3 (non-metastatic) (Figure S3B, C) and a decrease in Orai3 protein levels in ASPC-1 (invasive) (Figure S3H, I) and SW1990 (metastatic) (Figure S3N, O). Moreover, VIVIT transfection leads to a decrease in Orai3 protein levels in BXPC-3 (non-metastatic) (Figure S3E, F) and an increase in Orai3 protein levels in ASPC-1 (invasive) (Figure S3K, L) and SW1990 (metastatic) (Figure S3Q, R). The findings in these cell lines recapitulates the data obtained earlier from MiaPaCa-2 (non-metastatic), PANC-1 (invasive) and CFPAC-1 (metastatic) cell lines. Therefore, this new data supports our conclusion regarding the dichotomous regulation of Orai3 by NFATc1 across the three conditions of pancreatic cancer.

      Comment 5. Upon finding that NFAT inhibition stimulates Orai3 transcription (same as O/E), the authors essentially conclude that this confirms regulation of Orai3 by NFAT and that there must be compensation. This is not supported by any data; the use of siRNA validates that Orai3 has some dependence on NFATc1 for transcription, but the nature of this relationship is not adequately explained.

      Response: We thank the Reviewer for asking this question. In our manuscript, we performed NFATc1 inhibition studies using VIVIT and siRNA-mediated NFATc1 knockdown. Both of these assays show increase in Orai3 mRNA levels in all non-metastatic, invasive and metastatic pancreatic cancer cell lines. To understand if the increase in Orai3 mRNA levels is via transcriptional regulation, we performed luciferase assay which showed that VIVIT mediated NFAT inhibition leads to increase in luciferase activity suggesting the binding of other transcription factors on the Orai3 promoter. To corroborate this hypothesis, in our revised manuscript, we performed luciferase assay in wild type Orai3 promoter and truncated Orai3 promoter with no NFATc1 binding sites. NFAT inhibition via VIVIT transfection led to an increase in luciferase activity in both wild type and truncated Orai3 promoter (Figure S2A). Hence, removal of NFATc1 binding sites had no significant effect on luciferase activity suggesting that apart from NFATc1, other endogenous transcription factors are involved in regulating Orai3 transcription. We have not identified all the transcription factors that can modulate Orai3 upon NFAT inhibition as it is beyond the scope of this study. We sincerely hope the Reviewer 1 would be satisfied with this additional data.

      Reviewer #2

      Comment 1. Figure 1 all overexpression no evidence of endogenous NFAT2 regulating Orai3. I realize there may be limitations on available NFAT isoform specific antibodies so it is not essential to directly show this but a comment to that effect in the paper would be useful.

      Response: We apologize to the Reviewer for not highlighting the NFAT2 (NFATc1) loss of function data effectively. Actually, in the __Figure 3 __and __Supplementary Figure 2 __of the original manuscript, we showed VIVIT mediated NFAT inhibition and siRNA induced NFATc1 silencing data to provide the evidence that endogenous NFATc1 regulates Orai3.

      Comment 2. Figure 1F. Show RNA levels of Orai3 following overexpression of the other NFAT isoforms.

      Response: As suggested by the Reviewer, in the revised manuscript, we overexpressed the four NFAT isoforms: NFATc2, NFATc1, NFATc4 & NFATc3 and checked Orai3 mRNA levels. qRT-PCR analysis shows that overexpression of NFATc1 results in the highest and significant increase in Orai3 mRNA levels compared to the empty vector and other NFAT isoforms (Figure 1F). This data corroborates the western blot data of NFAT isoforms overexpression highlighting the transcriptional regulation of Orai3 by NFATc1.

      Comment 3. Fig. S3D, E. For both MARCH3 and 8 higher expression levels correlate with better survival whereas in the text it is stated that this is the case only for MARCH8. Please correct.

      Response: The survival analysis of pancreatic cancer patients with low MARCH3 and MARCH8 levels shows that around 30% of patients with low MARCH3 levels survived for 5.5 years, whereas in case of MARCH8 30% of patients with high MARCH8 levels survived for >7.5 years. Hence high MARCH8 expression in pancreatic cancer patients provided significant survival advantage compared to high MARCH3 levels. Therefore, in the text, we meant that compared to MARCH3, higher MARCH8 levels correlate with better survival. As suggested by the Reviewer, we have modified the text to make this point clearer.

      Comment 4. For the 2APB stimulation experiments there is a large variation in the level of the response between experiments even for the same cell type. For example, compare the level of the 2APB-stimulated Orai3 influx between Fig. 4H and 5C on the MiaPaCa-2 cells. Also there doesn't seem to be a correlation between the levels of Orai3 protein from WB and the 2APB stimulated entry among the different cell lines. This needs to be addressed and differences explained.

      Response: We understand the concern raised by Reviewer 2 regarding calcium imaging experiments in MiaPaCa-2 cell line. Therefore, in the revised manuscript, we repeated calcium imaging experiments in MiaPaCa-2 and updated the representative traces as well as quantitative analysis (Figure 2D, E, 3D, E, 4H, I, S2L, M). Further, we have discussed this point in the text of the manuscript.

      Comment 6. Fig. 6C and 6D. Show the line in 6C from which the intensity profile in 6D was generated. Also give the details of the imaging setup in methods: size of the pinhole, imaging mode, etc. The colocalization is not very convincing.

      Response: As recommended by the Reviewer, in the revised manuscript, we have indicated the region used for intensity profile generation by drawing a line in the representative image (Figure 6D). Further, we have updated the methodology of colocalization microscopy with details of the size of the pinhole and imaging mode.

      Comment 8. May be worth showing that overexpression of MARCH8 in the metastatic cell lines decreases their migration and metastasis as the argument is that these cells need high Orai3 but not too high. So, it would be predicted that overexpression of MARCH8 should lower Orai3 levels enough to prevent their metastasis.

      Response: We would like to thank the Reviewer for this highly relevant suggestion. In our revised manuscript, we carried out transwell migration assays with MARCH8 overexpression as well as MARCH8 knockdown in CFPAC-1 (metastatic) cells. Our data shows that stable lentiviral knockdown of MARCH8 increased the number of migrated CFPAC-1 cells compared to shNT CFPAC-1 cells while MARCH8 overexpression decreased the number of migrated CFPAC-1 cells compared to empty vector control cells (Figure 9F, G). Therefore, as pointed out by the Reviewer, MARCH8 overexpression lowers Orai3 levels in metastatic pancreatic cancer cells and hinders their metastatic potential.

      Comment 9. Fig. 10. Show higher levels of Orai3 protein in the metastatic side.

      Response: As suggested, we have updated the summary figure (Figure 10) showing higher Orai3 protein levels in the metastatic side.

      Comment 10. Please show all full WBs in the supplementary data.

      Response: As recommended by the Reviewer, we have provided all full western blots in a supplementary file (Supplementary File 1).

      Reviewer #3

      Comment 1. The authors show that MARCH8 physically associates with Orai3 using Co-IP and Co-localization studies. For the co-localization studies the authors should still provide a quantitative analysis. Furthermore, can the authors detect FRET between March and Orai3? Can you please state the labels used in the co-localization experiments also in the figure legend.

      Response: As suggested by Reviewer 3, in the revised manuscript, we have provided quantitative analysis of Orai3 and MARCH8 co-localization. Further, we have stated the labels used in the co-localization experiment in the figure legend of the revised manuscript. Unfortunately, we could not perform FRET assay between Orai3 and MARCH8 due to limited resources. Instead, as discussed in the planned revisions section, we are planning to perform co-immunoprecipitation assay with mutated Orai3 protein in which the MARCH8 interacting domains are deleted to investigate direct interaction of Orai3 and MARCH8. We believe that Reviewer 3 will be satisfied with this experiment.

      Comment 2. In the abstract it is only getting clear at the end that pancreatic cancer cells are used. It would be great if the authors could introduce this fact already more at the beginning of the abstract.

      Response: As recommended by the Reviewer, in the revised manuscript, we have introduced the use of pancreatic cancer cells at the beginning of the abstract.

      Comment 4. In other cancer types recent reports suggest a co-expression of Orai1 and Orai3 and even the formation of heteromers. Does only Orai3 or also Orai1 play a role in pancreatic cancer cells? Could there we difference in degradation when Orai3 forms homomers or heteromers with Orai1.

      Response: We thank the reviewer for asking this interesting question. There is only one report on Orai1’s role in pancreatic cancer. It was suggested that Orai1 can contribute to apoptotic resistance of pancreatic cancer cells (Kondratska et al. BBA-Molecular Cell Research, 2014). However, only one cell line i.e. PANC-1 was used in this study. While our earlier work and other studies have demonstrated that Orai3 drives pancreatic cancer metastasis (Arora et al. Cancers, 2021) and proliferation (Dubois et al. BBA-Molecular Cell Research, 2021) respectively. Therefore, emerging literature suggests that both Orai1 and Orai3 can contribute to different aspects of pancreatic cancer progression. But whether Orai1 and Orai3 form heteromers in pancreatic cancer cells remains unexplored. Further, we believe that the degradation machinery and the underlying molecular mechanisms would be analogous for both Orai3 homomers and heteromers. Nonetheless, the rate of degradation may differ for Orai3 homomers and heteromers as literature suggests that usually proteins are more stable in large heteromeric protein complexes.

    2. Review Commons

      Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

      Learn more at Review Commons

      Referee #3

      Evidence, reproducibility and clarity

      The study by Raju et al. demonstrated that NFAT2 drives both Orai3 transcription and protein degradation. They find a clearly distinct mechanism between non-metastatic cancerous and metastatic cells. While in non-metastatic cells NFAT2 drives Orai3 transcritpion and increases Orai3 expression, in invasive and metastatic cells degradation of Orai3 is driven. They find a physical interaction of MARCH8 with Orai3 resulting in degradation. This degradation is not happening in non-metastatic cells as MARCH8 promotor is highly methylated. This study is highly interesting for a broad readerships and provides a solid basis for the development of novel therapeutic strategies for cancer treatment. Before publication the authors should address a few minor comments.

      1. The authors show that MARCH8 physically associates with Orai3 using Co-IP and Co-localization studies. For the co-localization studies the authors should still provide a quantitative analysis. Furthermore, can the authors detect FRET between March and Orai3? Can you please state the labels used in the co-localization experiments also in the figure legend.
      2. In the abstract it is only getting clear at the end that pancreatic cancer cells are used. It would be great if the authors could introduce this fact already more at the beginning of the abstract
      3. In the scheme in Fig. 10, the authors highlight that Orai3 is ubiquitinated. Do they have any idea where the site of action of ubiquitination in Orai3 is located?
      4. In other cancer types recent reports suggest a co-expression of Orai1 and Orai3 and even the formation of heteromers. Does only Orai3 or also Orai1 play a role in pancreatic cancer cells? Could there we difference in degradation when Orai3 forms homomers or heteromers with Orai1.

      Significance

      The authors highlight and decode a dual role of NFAT2 in controling Orai3 expression, which is highly interestingly to gain insight in different states of cancer cells (non-metastatic, metastatic). The findings form a great basis for a deeper understanding of potential therapeutic targets.

    3. Review Commons

      Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

      Learn more at Review Commons

      Referee #2

      Evidence, reproducibility and clarity

      Raju et al. presents a nice comprehensive study of the differential regulation of Orai3 at the transcriptional and stability levels in metastatic versus non-metastatic pancreatic cancer (PC) cells. They convincingly show that NFAT2 regulates Orai3 transcription in all PC cells but interestingly, in the metastatic PC cells NFAT2 also upregulates the expression of MARCH8 an E3 ubiquitin ligase that targets Orai3 for lysosomal degradation. The MARCH8 locus is hypermethylated in the non-metastatic cell line, thus preventing MARCH8 upregulation in those cells. The data is convincing and complementary. I only a few suggestions below.

      Specific Comments:

      1. Figure 1 all overexpression no evidence of endogenous NFAT2 regulating Orai3. I realize there may be limitations on available NFAT isoform specific antibodies so it is not essential to directly show this but a comment to that effect in the paper would be useful.
      2. Figure 1F. Show RNA levels of Orai3 following overexpression of the other NFAT isoforms.
      3. Fig. S3D,E. For both MARCH3 and 8 higher expression levels correlate with better survival whereas in the text it is stated that this is the case only for MARCH8. Please correct.
      4. For the 2APB stimulation experiments there is a large variation in the level of the response between experiments even for the same cell type. For example compare the level of the 2APB-stimulated Orai3 influx between Fig. 4H and 5C on the MiaPaCa-2 cells. Also there doesn't seem to be a correlation between the levels of Orai3 protein from WB and the 2APB stimulated entry among the different cells lines. This needs to be addressed and differences explained.
      5. Fig. 6A and 6B. Show the full Orai3 and Ubiquitin WBs. As presented the figure current just shows that there are ubiquitin proteins in Orai3 pull down, not that Orai3 is ubiquitinated.
      6. Fig. 6C and 6D. Show the line in 6C from which the intensity profile in 6D was generated. Also give the details of the imaging setup in methods: size of the pinhole, imaging mode, etc. The colocalization is not very convincing.
      7. Also all the imaging and pull down down do not prove conclusively direct interaction between MARCH8 and Orai3, they rather show that the proteins are in the same complex. Although it is unlikely best for the text to be moderated accordingly.
      8. May be worth showing that overexpression of MARCH8 in the metastatic cell lines decreases their migration and metastasis as the argument is that these cells need high Orai3 but not too high. So it would be predicted that overexpression of MARCH8 should lower Orai3 levels enough to prevent their metastasis.
      9. Fig. 10. Show higher levels of Orai3 protein in the metastatic side.
      10. Please show all full WBs in the supplementary data.

      Significance

      SIgnificant and relevant study that will be of great interest to the cancer and calcium signaling fields.

    4. Review Commons

      Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

      Learn more at Review Commons

      Referee #1

      Evidence, reproducibility and clarity

      The manuscript entitled, "NFAT2 drives both Orai3 transcription and protein degradation by harnessing the differences in epigenetic landscape of MARCH8 E3 ligase" offers an extensive study of how Orai3 levels are controlled during pancreatic cancer progression. The central hypothesis is that NFAT2 stimulates both Orai3 and MARCH8 transcription, resulting in both Orai3 transcription and degradation. They further establish that MARCH8 expression/Orai3 degradation is epigenetically regulated in PDAC, with a progressive loss of methylation during cancer progression leading to increased Orai3 transcription, stability and Ca2+ entry.

      Overall, I'm certain that there is new information to be learned here. However, as detailed below, the manuscript makes a number of general claims about what happens during PDAC progression, but this is based on only one cell line per disease state. While they should not be expected to do a complete analysis in more cell lines, a demonstration that Orai3 and MARCH8 expression are correlated with disease progression in a panel of cell lines and/or on the TCGA database would increase enthusiasm considerably. In addition, although I found the work with MARCH8 to be highly convincing, the fact that NFAT2 knockdown increased rather than reduced Orai3 transcription does not support the central hypothesis. The explanation that this results from compensation is not very meaningful; that NFAT2 drives Orai3 transcription is in the title of the paper. These observations clearly demonstrate that this relationship is more complicated than suggested. Finally, there are a number of missing controls and unclear aspects to the authors' ChIP data that could help explain some of these discrepancies.

      Specific Comments:

      1. The observation that both transcriptional regulation and protein degradation of Orai3 is regulated downstream of one transcription factor is not, in and of itself, entirely surprising. All proteolytic components are transcriptionally regulated and this phenomenon is likely relatively common. However, what I do think is both impressive and important is that the authors have characterized both components of the pathway within a disease context. While I am not going to search the literature for how often transcription and proteolysis are co-regulated for other proteins, it is the case for many short-lived proteins and perhaps many others. As such, discussion throughout the abstract and introduction that co-regulation of these processes is unprecedented should be removed.
      2. In discussing figure 1, the authors switch from claiming to be studying NFATc binding to studying NFAT expression. This use of 2 different naming conventions is certain to confuse readers; the authors should use the approved current naming system in referring to NFAT isoforms. In which case NFAT2 is NFATc1.
      3. The ChIP analyses in figures 1H and 7D are important findings, however, there is missing information. Typically, ChIP is used to validate putative binding sites; as such, one would expect 3 separate qPCR reactions for Orai3, not one. It is also important to note that qPCR products should be uniform in size and under 100 bp; here, the product size is not stated. Finally, demonstrating that an antibody targeting ANY other NFAT isoform fails to pull down whatever product this is would increase confidence considerably.

      Also, the gold standard for validating ChIP is to mutate the sites and eliminate binding. The "silver" standard would be to mutate them in your luciferase vector and demonstrate that NFATc1 no longer stimulates luciferase expression. Since neither of these was done, the ChIP data provided should not be considered formally validated.

      1. In figures 2 and 3, only one cell line is used to represent each of 3 conditions of pancreatic cancer. That is insufficient to make generalized conclusions; some aspects of this figure (expression and stability, not function) should be extended to 2 to 3 cell lines/condition. TCGA data validating this point would also be helpful.
      2. Upon finding that NFAT inhibition stimulates Orai3 transcription (same as O/E), the authors essentially conclude that this confirms regulation of Orai3 by NFAT and that there must be compensation. This is not supported by any data; the use of siRNA validates that Orai3 has some dependence on NFATc1 for transcription, but the nature of this relationship is not adequately explained.
      3. In Figure 6A,B, does the Orai3 western blot show any of the heavier bands seen in the ubiquitinization IP if you show the whole blot? It should.

      Significance

      My expertise is in calcium signaling, particularly within the context of disease states. I currently have a PDAC study in its late stages, but I have worked more with melanoma.

      Issues about significance were raised in my comments above; generalization of these observations requires the appropriate use of a panel of cell lines and/or TCGA usage. In addition, some observations require additional investigation for confidence; necessary to achieve significance.

      The extent of the advance is quite reasonable for a high profile paper in this field, should the issues I and the other reviewers raise be formally and thoroughly addressed.

      Given that the study crosses lines between signaling, cancer, epigenetics, transcription and ubiquitination, I think that it is of potential interest to a general audience.

    5. Version published to 10.1101/2024.12.11.628081v2 on bioRxiv
    6. Version published to 10.1101/2024.12.11.628081v1 on bioRxiv