Msc1 facilitates glucose starvation-induced remodeling of the nucleus-vacuole junction

  1. Yusuke Mito
  2. Shintaro Fujimoto
  3. Saori Shinoda
  4. Yasushi Tamura

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

The nucleus-vacuole junction (NVJ) is a membrane contact site between the nuclear envelope and the vacuole in yeast that undergoes dynamic remodeling in response to nutrient starvation. Here, we report that Msc1 is a glucose starvation (GS)-responsive NVJ factor. GS strongly induced Msc1 expression and promoted its accumulation at the NVJ. Although Msc1 is not essential for NVJ formation itself, loss of Msc1 impaired GS-dependent functional maturation of the NVJ, including stabilization and recruitment of multiple NVJ-associated proteins. Notably, GS-induced transcriptional activation of NVJ1 was markedly attenuated in msc1Δ cells, suggesting that proper NVJ remodeling contributes to the execution of stress-responsive transcriptional programs. Together, these findings establish Msc1 as an upstream regulator linking GS to functional remodeling of the NVJ and associated transcriptional responses.

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

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

    Reply to the Reviewers

    1. General Statements

    We thank the reviewers for their careful evaluation of our manuscript and for the many constructive suggestions. Overall, the reviewers found the identification of Msc1 as a glucose starvation-responsive NVJ-associated factor to be novel and potentially important, while also raising several important concerns regarding the mechanistic interpretation of our findings and the topology/localization of Msc1. We particularly appreciate the reviewers' comments regarding potential overinterpretation of several conclusions. In the revised manuscript we will substantially revise the wording throughout the text to more carefully distinguish correlation from causation and to avoid unsupported mechanistic conclusions. In addition, we plan to address the reviewers' concerns through a combination of additional experiments, revised data presentation, clarification of methodological details, and expanded discussion of alternative interpretations.

    1. Description of the planned revisions

    In the point-by-point response below, the reviewers' comments are presented in italics, and our responses are provided below each comment.

    Reviewer #1

    Comment:

    *This is a nice, smallish study of Msc1, a fungal protein of unknown function. The authors show it localises to the NVJ when that expands in late-log/stationary phase, at which stage its transcription is increased 80-fold - an induction one whole order of magnitude greater than shown by Nvj1 itself. This indicates that Msc1 may be a previously unappreciated master regulator of the NVJ. There are some interesting phenotypes of deleting Msc1, including some cell death and loss of Nvj1, mostly through destabilisation since the transcriptional effect is marginal. *

    *While no mechanism for Msc1 is discovered, that might be too much to ask for in this first paper. However, there are ways to begin to address this that the authors should look into. *

    *My major issue with the paper is that it makes no link to the previously studied homologues of Msc1 in S pombe (Ish1/Les1 - see Asakawa et all 2022). Admittedly, S. pombe has no Nvj1 homolog, but there is a physical relationship between nucleus and vacuole (Chadwick et al (2020) 10.1088/1478-3975/aba510). Also, the paper on Ish1/Les1 developed a phenotype to test Ish1 (toxicity of over expression) that might be useful for studies of Msc1. *

    *The current MS should link to work on Ish1/ Les1 in S. pombe, relating to several features: *

    *Topology. Given the high similarity between Msc1 and Ish1/Les1, they are (a priori) likely to share considerable form and function. If Msc1 is a soluble protein in the ER lumen, then the previous report that Ish1/Les1 have TMDs is wrong. The report here should make that link and carefully explain how the Pombe paper is wrong. Also explain how is it possible for Msc1 (and Ish1/Les1) to stay restricted to the nuclear envelope? (in many images it is diffuse throughout the NE). The only mechanism I can think of is binding an integral protein that sorts to the inner-NE by known mechanisms (or possibly binding to an outer-NE protein that binds to an inner-NE one, like SUN/KASH). I cannot think of any other example of a soluble proteins restricted to the NE - so this is quite a claim. An alternative view that could be investigated and should definitely be discussed is that Msc1 (and by implication Ish1 and Les1) has a TMD even though it is extracted by carbonate. Something similar has been reported for some single TMD proteins in mitochondria (Kim et al (2015) 10.1002/pro.2817). Investigations would include proteomics showing whether the protein is normally full length (as coded by the open reading frame) or clipped (indicating the signal sequence is removed for a soluble protein). Such data may already be available in published mass spec datasets. *

    We agree that the relationship between Msc1 and the previously characterized S. pombe homologs Ish1/Les1 should be discussed more carefully, particularly with respect to membrane topology. In the revised manuscript, we will cite and discuss the Ish1/Les1 studies and further investigate the topology and localization of Msc1 through several additional experiments. First, as suggested by the reviewer, we will examine whether the N-terminal region of Msc1, which is predicted to function as a signal sequence and as a weak transmembrane domain, undergoes proteolytic processing. To address this, we plan to perform mass spectrometry-based analyses and examine whether the N-terminus is retained in the mature protein. As an alternative approach in case the N-terminal peptide cannot be reliably detected by mass spectrometry, we will generate an Msc1 mutant lacking the predicted N-terminal 22-amino-acid signal sequence and compare its migration on SDS-PAGE with that of the full-length protein. This analysis should provide an additional assessment of whether the predicted signal sequence is removed during Msc1 maturation.

    In addition, following comments from multiple reviewers, we will repeat the alkaline carbonate extraction experiments using additional ER membrane protein controls to more carefully evaluate the membrane association properties of Msc1.

    Furthermore, we plan to perform additional split-GFP localization analyses to test whether Msc1 localizes within the perinuclear ER lumen. Specifically, we will express GFP1-10 either within the ER lumen or within the nucleoplasm and examine in which compartment co-expression of Msc1-GFP11 results in GFP fluorescence.

    Finally, as suggested by the reviewer, we agree that interactions with integral membrane proteins may explain the restricted localization of Msc1 within the nuclear envelope/NVJ region. In our preliminary experiments, we obtained results suggesting a physical interaction between Msc1 and Nvj2. Therefore, we plan to further investigate the interaction of Msc1 with Nvj2, as well as with other known NVJ-associated proteins, to better understand the mechanism underlying its localization and enrichment at the NVJ.

    *Minor Issues *

    *The Abstract switches from response to lack of glucose to terminology about 'stress-response'. This could appear to be an effort to appear more interesting. If the idea is to remain, it needs some support with the introduction of the idea that yeast experiences stress (as opposed to "normal" transcription driven programmatic changes in relation to changing levels of glucose in normal cultures. *

    To avoid overstating our findings, we will revise the Abstract and related text to use more precise terminology and to more clearly describe the observed responses.

    Introduction para 1 seems to be dedicated to the idea that a set of intracellular structures (here MCS) are 'dynamically and coordinately remodeled in response to metabolic and stress conditions'. This conclusion applies widely and may not be noteworthy. The paragraph needs a bit of rethinking.

    While nutrient-dependent changes in the NVJ itself have long been recognized, we believe that dynamic remodeling of multiple MCSs in response to environmental and metabolic conditions has only more recently become appreciated more broadly in the field. We therefore think that discussing the emerging concept that diverse MCSs undergo dynamic reorganization under different physiological conditions provides important context for the present study. Nevertheless, we will revise the Introduction to explain this point more clearly and concisely.

    Figure 2D: I could not find Nsg1 result described in the text.

    We will repeat the experiment independently and quantify the immunoblot results shown in Figure 2D. The resulting quantitative data will be added to Figure 2D, and the Results section will be revised accordingly to describe these findings, including the Nsg1 phenotype.

    P6: "Strikingly, GS-dependent transcriptional activation of NVJ1 was significantly suppressed in msc1∆ cells (Fig. 4B)." This overstates the strength of the result. Instead state that the induction diminishes from 6-fold to 4-fold, and give the p value.

    We will revise the text to provide a more quantitative description of the result, including the corresponding p value.

    *Language: Avoid use of rhetorical wording (e.g. dramatic): just state the results (e.g. 80-fold induction) and let the results be dramatic/striking etc. all by themselves.

    We will also revise the text to avoid rhetorical wording and instead describe the results in a more direct and quantitative manner.

    __Reviewer #2 __* The study describes the finding of the nuclear envelope protein Msc1 as a new component of the membrane contact site nucleus vacuole junction (NVJ) under the conditions of glucose starvation. Msc1 has previously only been known as a nuclear envelope protein, presumably localizing to the nuclear lumen, and its role in DNA damage repair. The main finding of this study is the glucose starvation-induced upregulation and NVJ-localization of Msc1 (Figure 1). The second main finding is that the loss of Msc1 results in an impaired induction of the expression of Nvj1 (the main component of the NVJ, responsible for the formation of NVJ via direct interaction with Vac8) upon glucose starvation (Fig. 3 A). The effect of Msc1-loss on the Nvj1 expression levels is transcriptional (Fig. 4 B). The glucose starvation-mediated expression induction of some other previously identified NVJ components, Nsg1 and Nsg2 is also impaired in the msc1D mutant, while the expression of Ypf1 is affected to a lesser degree. The data supporting these two main findings are solid (Figure 1; Figure 3 A; Figure 4 A, B).

    The study further shows that the loss of Msc1 results in a loss of NVJ-localization of NVJ components Tsc13, Ypf1 and to a lesser degree Hmg2. The microscopy data looks solid, however the interpretation of this finding is not clear. In my view, the most likely explanation is that the effect of Msc1 loss on the localization of NVJ components to the NVJ is due to the impaired glucose starvation-induced Nvj1 expression in the msc1D mutant.

    MAJOR COMMENTS:

    Here are suggested experiments that would strengthen the study:

    • It is difficult to imagine how a NE protein could affect expression levels of other NVj proteins - this key finding would be supported by a complementation experiment where MSC1 is expressed from a vector - to test whether this rescues the phenotype (to make sure that the observed phenotype is not due to an off-target effect of msc1D deletion) *

    As suggested by the reviewer, we plan to perform complementation experiments by expressing Msc1 from a plasmid in msc1∆ cells to confirm that the observed phenotypes are specifically caused by loss of MSC1.

    *- If technically feasible under the glucose starvation conditions, this hypothesis could be tested by overexpressing Nvj1 from an inducible or some other promoter. *

    We agree that this is an important point. As suggested by the reviewer, we plan to overexpress Nvj1 using a constitutive promoter and examine whether this suppresses the phenotypes observed in msc1∆ cells.

    *- The effect of msc1D deletion on Tsc13 proteins levels (preferentially using the same Tsc13-GFP strain as used in microscopy - anti Tsc13 or anti-GFP antibodies could be used) *

    We will examine Tsc13 protein levels in msc1∆ cells using the same Tsc13-GFP strain used for microscopy.

    *- The results concerning the localization of Msc1-GFP in elo3D mutant have been interpreted as "accelerated localization", "expansion of the the size of Msc1-NVJ domain" etc. However, the levels of Msc1-GFP in the elo3D mutant are higher compared to WT (Figure 2 D). Considering this, it is very likely that the larger surface area measured in the elo3D mutant is a consequence of this. This could be potentially checked by comparing images set of WT and elo3D that are set to a similar fluorescence intensity. In any case, this possibility should be definitely addressed in the interpretation of the result. *

    We agree that the increased Msc1-GFP signal in elo3∆ cells could contribute to the apparent increase in NVJ area. However, in our previous study (Fujimoto and Tamura, 2026, J. Cell Biol.), we observed accelerated NVJ expansion under glucose starvation and in elo3∆ cells using Ypf1, whose expression levels are largely unchanged under these conditions. We therefore think that the observed phenotype is unlikely to be explained solely by increased Msc1 expression. Nevertheless, because Msc1 protein levels are clearly elevated in elo3∆ cells, we will revise the text to describe these results more carefully and fairly, while citing our previous findings.

    *- There is an impression that the data has been overinterpreted, and the conclusions should be written much more carefully. Examples: o "Here, we show that Msc1 is a GS-responsive NVJ factor that plays an important role in functional NVJ remodeling." - based on data shown, the effect of Msc1 could be indirect. The statement above should be re-written or argumented much better. o "we find that GS-dependent induction of NVJ1 transcription is attenuated in msc1Δ cells, suggesting that proper NVJ remodeling contributes to the execution of stress-responsive transcriptional programs" - this is unclear; which data support this? o "Together, these findings position Msc1 as an upstream regulator linking GS signaling to functional maturation of the NVJ and associated cellular adaptation responses." - same comment as above o "...suggesting that Msc1 functions as a GS-responsive regulator of NVJ functions." o "...these findings suggest that Msc1 acts upstream of Ypf1 in orchestrating GS-induced NVJ functional maturation." o "Collectively, these results indicate that Snf1 acts upstream of Msc1 to drive GS-induced NVJ remodeling, whereas reduced Elo3 activity further accelerates this process and promotes Msc1 accumulation." - not sure if the available data support this. o "These results indicate that although Msc1 ...... it is required for efficient GS-dependent functional maturation of the NVJ domain." o "These observations suggest that loss of Msc1 does not cause a general defect in transcriptional activation but rather impairs the proper execution and dynamic range of GS-dependent transcriptional responses." - this is unclear o "Within this context, the robust induction of NVJ1 appears to be particularly sensitive to Msc1 deficiency." - this sentence would benefit from being re-written. o "Together, these results indicate that Msc1 contributes to transcriptional reprogramming associated with NVJ remodeling during GS." - this sounds overstated. o "the observation that loss of Msc1 attenuates GS-dependent induction of NVJ1 raises the possibility that NVJ remodeling influences stress-responsive gene expression programs." *

    We appreciate the reviewer's concern that several interpretations in the current manuscript may extend beyond what is directly supported by the available data. We will therefore revise these statements throughout the manuscript to provide more balanced interpretations and avoid overstating our conclusions. In addition, several planned experiments, including complementation analyses, Nvj1 overexpression experiments, additional localization analyses, quantitative protein analyses, and identification of NVJ-associated proteins that interact with Msc1, may further clarify the relationship between Msc1, NVJ remodeling, and glucose starvation responses. We will revise the text accordingly based on the results obtained from these additional experiments.

    *OTHER COMMENTS FIGURE BY FIGURE - SOME ARE MAJOR (overlapping to the above comments), SOME ARE MINOR: *

    *Figure 1: *

    *Figure 1 A and B shows that Msc1-GFP expression is upregulated in cells starved for glucose for 24h, but not in nitrogen-starved cells. *

    *o Size of the markers (protein ladder) would be helpful. * We will reprocess the immunoblot images from the original data and revise the figure layout to include molecular weight markers.

    *Figure 2:

    • Comment: It is not clear if these are the same strains as analyzed by microscopy (GFP-tagged Msc1). This should be specified in the Figure legend 2 D. *

    *- Comment: o Since the levels of Msc1-GFP in the elo3D mutant are higher compared to WT (Figure 2 D), the larger surface area measured in C may be a consequence of this. *

    *o It is not clear if Figure A and D analyze the same strains (western blot and microscopy - do both show GFP-tagged Msc1? - using anti-GFP?). This should be specified in the Figure legend 2 D. Since the increased area measured in Figure 2 C could be due to increased Msc1-GFP levels in this mutant strain, the WB should check the levels of Msc1-GFP in the same strain and under same conditions as analyzed in Figure 2 C.

    o Does Tim23 serve as a loading control in Figure 2 D? *

    We added "Tim23 was used as a loading control." In the legend of Figure 2D.* o Would be good to have protein ladder sized marked in Western blots o Since the increase in Msc1 levels in the elo3D mutant could be significant for the interpretation of the results, it would be helpful to have quantification of the protein levels in WB (normalized to a loading control). *

    We will clarify in the Figure 2 legend that Figure 2A shows GFP-tagged Msc1 expressed cells analyzed by fluorescence microscopy, whereas Figure 2D shows untagged strains analyzed by immunoblotting using an anti-Msc1 antibody. We will also clarify that Tim23 was used as a loading control and add molecular weight markers to the Western blots. We agree that the increased Msc1-GFP levels in elo3∆ cells could influence the apparent increase in NVJ area measured in Figure 2C. As noted above, our previous findings using Ypf1 suggest that accelerated NVJ expansion in elo3∆ cells is unlikely to be explained solely by increased Msc1 expression (Fujimoto and Tamura, J. Cell Biol., 2026). Nevertheless, we acknowledge that elevated Msc1-GFP levels could influence the apparent NVJ area measured in Figure 2C. We will therefore revise the text to more carefully describe these results and discuss them in the context of our previous findings.

    In addition, we will quantify the Western blot signals in Figure 2D normalized to the loading control and include these data in the revised manuscript.

    *Figure 3 ** Together these data show that localization of other NVJ-proteins to the NVJ depends on the presence of Msc1. Comment:

    • From the available data it is possible that Msc1 recruits these components by direct interaction, or by modifying the structure of NVJ, or functions in an indirect manner - this should be discussed in the Discussion. Comment:
    • The signal of Tsc1-GFP in log-growing cells is very weak, therefore the quantification may be unreliable. I would remove this condition (log-grown cells) form the quantification in C) due to the low signal, since it is not crucial to the interpretation of the data. If the authors prefer to leave it, that is fine.
    • The title of the Figure 3 is "Msc1 supports stability and recruitment of NVJ-associated proteins" - I am not sure what "stability" is; the data don't address stability or recruitment in a direct manner - I suggest to change the figure title into a statement describing what is shown in the Figure, for example: "The loss of Msc1 results in decreased Nvj1 levels and a decreased localization of NVJ proteins to the NVJ). And have a comment that this data suggests that Msc1 supports recruitment of NVJ-associated proteins, likely in an indirect manner, based on the finding that the loss of Msc1 leads to a lower expression of Nvj1, in the main text (e.g. in the Discussion).
    • Is it possible that the loss of Msc1 on the loss of NVJ-localized Tsc13 is due to the downregulation of Tsc13 expression? Considering the effect of msc1D deletion on the expression of some NVJ proteins (Figure 3 A), Tsc13 expression levels would be good to be checked, considering the effect of msc1D on Tsc13-GFP localization. It would be optimal to do the WB with the same Tsc13-GFP-expressing strain and under the same growth conditions as was used in the microscopy in the Figure 3 B.
    • Expression levels of Ypf1 are lower in the msc1D strain, than in the WT (Fig. 3 A) - could this affect lower NVJ-area in his mutant? (Fig. 3 B)*

    We agree that the current data do not distinguish whether Msc1 affects localization of NVJ-associated proteins directly, indirectly through changes in NVJ structure, or through other indirect mechanisms. We also agree that the term "stability" used in the current Figure 3 title is not sufficiently supported by the available data, as our experiments do not directly address protein stability. To address this issue, we plan to overexpress Nvj1 in msc1∆ cells and examine the expression and localization of NVJ-associated proteins including Nsg1 and Nsg2. Based on the results obtained from these additional experiments, we will revise the Figure 3 title and discuss these possibilities more carefully in the revised manuscript.

    Regarding the quantification of Tsc13-GFP localization in log-growing cells, although the NVJ signal is relatively small and weak under these conditions, we confirmed the signal carefully during quantification. In addition, we consider this dataset important because it suggests that the effect of Msc1 is relatively limited during logarithmic growth. Therefore, we currently prefer to retain these data in the revised manuscript.

    As suggested by the reviewer, we will revise the Figure 3 title to more directly describe the observed phenotypes.

    We will also examine Tsc13 protein levels in msc1∆ cells using the same Tsc13-GFP strain and growth conditions used for the microscopy analyses. In addition, we will quantitatively analyze expression levels of Ypf1 and other NVJ-associated proteins in msc1∆ cells and discuss how these changes may contribute to the observed localization phenotypes.

    *Figure 4. Figure 4 A shows mRNA levels in glucose starved cells compared to log-.growing cells for MSC1, NVJ1 and YPF1.

    • Comment: I would move Figure 4 A to Figure 1. Figure 4 B shows mRNA levels of proteins expressed in WT and msc1D mutant strain, in log-growing cells in under glucose starvation. The data show that the loss of Msc1 leads to a decrease in NVJ1 mRNA under the conditions of glucose starvation. Th expression of other NVJ proteins analyzed are not affected.
    • Comment: Would this Figure 4 A-B better fit together with the data showing Nvj1 levels in the msc1D mutant from a previous figure (3 A)? *

    *Figure 4 C shows PI staining of cells after 5 days of glucose starvation. The loss of Msc1 leads to a double increase in PI-positive cells (in contrast to the nvj1D mutant, which is similar to WT), indicating that the viability of cells after 5 days of glucose starvation is decreased in the absence of Msc1.

    • Comment: Since there is no phenotype of nvj1D, this is likely not due to the non-functional NVJ, but another function of Msc1 - the question is which. This could be discussed in the Discussion.
    • Comment: This is informative, however it is not sure why this data is placed together with the mRNA data within the Figure 4. *

    We appreciate these suggestions and agree that the figure organization could be improved. Following the reviewer's recommendations, and taking into account the results of the additional experiments described above, we will reorganize the figure layout to better align related datasets and improve the overall flow of the manuscript. We also agree that the increased PI staining observed in msc1∆ cells is unlikely to be explained solely by loss of NVJ function, since nvj1∆ cells do not show a comparable phenotype. We will therefore discuss this point more carefully in the revised Discussion and consider additional functions of Msc1 that may contribute to cell survival during glucose starvation.

    *Figure S1.

    • Comment - as in Figure 2 - Msc1-GFP has a much stronger signal in elo3D mutant, than in WT, which could influence (or likely influences) the measured area. Perhaps one way to test this is to image WT cells with higher % of laser "a "longer exposition"), to get a stronger signal similar to that seen in the elo3D mutant, a*nd then repeat the quantification.

    • Taken the result as it is presently, I suggest taking the Figure S1 out.

    As discussed above for Figure 2C, increased Msc1-GFP levels in *elo3∆ *cells could influence the apparent increase in NVJ area. We agree that this analysis is not central to the main conclusions of the current manuscript. Therefore, together with the additional experiments described above, we will re-evaluate the organization of the supplementary figures and revise the figure layout accordingly. Based on the revised dataset, we will determine whether Figure S1 should be removed, relocated, or incorporated into a more appropriate context in the revised manuscript.

    *Figure S3 . Validation of anti-Msc1 antibody

    • Could be moved as S1. *

    We will move the current Figure S3 to Figure S1 in the revised manuscript.

    Reviewer #3

    *Summary: In this study, the authors identify Msc1 as a factor associated with nucleus-vacuole junctions (NVJs) during glucose starvation. Using Saccharomyces cerevisiae as a model system, and combining immunoblotting and microscopy approaches, they report a functional connection between Msc1 and the NVJ component Nvj1.

    Major comments:

    • Are the key conclusions convincing? Overall, the main conclusions are largely convincing. However, several interpretations are overstated and should be phrased more cautiously (see specific comments below).

    • Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether? Yes. In several instances, the data support correlation rather than causation, and the authors should clearly indicate when conclusions are speculative.

    • Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation. For some conclusions, either:

    • the interpretation should be weakened, or

    • additional experiments are needed to fully support the claims

    • Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments. If Western blot membranes are available, additional controls could likely be addressed by reprobing, which would require minimal effort and a short timeframe. Suggested microscopy experiments would require strain construction and are therefore expected to take approximately 2-3 weeks.

    • Are the data and the methods presented in such a way that they can be reproduced? Some methodological details are insufficiently described and should be clarified to ensure reproducibility.

    • Are the experiments adequately replicated and statistical analysis adequate? The authors do not specify which tests for normality were performed. It is therefore difficult to assess whether the use of Student's t-test is appropriate. In at least one case (comparison of three groups), a t-test is not appropriate and should be replaced with a suitable multiple-comparison test.

    Minor comments:

    • Specific experimental issues that are easily addressable. See below

    • Are prior studies referenced appropriately? Yes, mostly/ The authors should provide a reference supporting NVJ expansion during nitrogen starvation.

    • Are the text and figures clear and accurate? Yes

    • Do you have suggestions that would help the authors improve the presentation of their data and conclusions? see below *

    We appreciate the reviewer's overall positive evaluation of our study and the recognition that the main conclusions are largely convincing. We also appreciate the reviewer's careful and constructive suggestions regarding interpretation, experimental support, and presentation of the data. As also pointed out by other reviewers, several interpretations in the current manuscript may extend beyond what is directly supported by the available data. We will therefore revise the manuscript throughout to more clearly distinguish between observations directly supported by the data and more speculative interpretations, and to avoid overstating our conclusions. In addition, we are currently performing several additional experiments, including complementation analyses, Nvj1 overexpression experiments, quantitative protein analyses, and additional localization studies, which may further strengthen some of the interpretations. We will also revise the Methods section to provide more detailed information regarding experimental procedures, statistical analyses, and reproducibility. In addition, we will reanalyze the data using appropriate statistical methods where necessary. In addition, we will revise and reorganize several figures and figure legends, and methodological details, and improve the overall presentation and clarity of the manuscript. We will also add references regarding NVJ expansion during nitrogen starvation as suggested by the reviewer.

    *Figures and data presentation • Figure 1A: The image is difficult to interpret. The authors should improve visibility, for example by: o using grayscale instead of magenta/green for single channels, or o applying an intensity LUT. This is particularly important as the Nvj1 signal is barely visible.

    We will revise Figure 1A to improve visibility of the fluorescence signals, including the Nvj1 signal, by adjusting the image presentation methods as suggested by the reviewer.

      • Figure 1B: The use of Tim23 as a loading control is not appropriate. The authors should justify why a mitochondrial protein was used as a reference.* Although Tim23 is a mitochondrial protein, we previously confirmed that its abundance is not substantially affected by glucose starvation conditions and therefore serves as a suitable loading control in this experimental setting (Fujimoto and Tamura, J. Cell Biol., 2026). In the revised manuscript, we will clarify the rationale for using Tim23 as a loading control. We will also normalize immunoblot signals to Tim23 and explicitly state this in the text.
    • Figure 1C: The experimental design and interpretation are problematic: o Using an ER protein together with mitochondrial markers in the proteinase K protection assay is not appropriate for the stated conclusions. * Because ER and mitochondrial membranes are both present in the membrane fraction used for the proteinase K protection assay, we believe that mitochondrial marker proteins can still serve as controls for proteinase K accessibility. However, we agree that the integrity of the ER membrane itself was not directly assessed in the current experiment. We therefore plan to repeat the experiment using appropriate ER membrane protein controls.

    *o The claim that Msc1 is not an integral membrane protein is not sufficiently supported, particularly if a polyclonal antibody was used. *

    Similar concerns regarding the topology and membrane association of Msc1 were also raised by other reviewers. To address these issues, we are currently performing additional experiments, including detailed analyses of the N-terminal region of Msc1 and further localization studies (see also our response to the first comment from Reviewer #1). We also plan to examine the fission yeast Msc1 homolog Les1, whose localization has been analyzed in greater detail previously (Asakawa et al. Genes Cells. 2022, 27(11):643-656. doi: 10.1111/gtc.12981). In addition, we are currently investigating NVJ-associated proteins that interact with Msc1, which may provide further mechanistic insight into the localization and function of Msc1 at the NVJ.

    *o The authors should provide additional evidence for localization (or use alternative approaches). *

    As also mentioned in our response to Reviewer #1, we plan to perform additional localization analyses using a split-GFP approach. Specifically, we will express GFP1-10 either within the ER lumen or within the nucleoplasm and examine in which compartment co-expression of Msc1-GFP11 results in GFP fluorescence.

      • Figure 1D: o The authors conclude that deletion of NVJ1 and VAC8 reduces Msc1 colocalization. However, an alternative explanation is that NVJs are not formed under these conditions.* *o This conclusion should therefore be phrased more cautiously. Alternatively, a known NVJ marker should be included to demonstrate NVJ formation. *

    We agree that reduced Msc1 localization in nvj1∆ and vac8∆ cells could simply reflect impaired NVJ formation itself. To address this possibility, we plan to examine NVJ formation in these mutants using split-GFP-based NVJ probes that we previously developed (Tashiro et al., *Front Cell Dev Biol. *2020, doi: 10.3389/fcell.2020.571388). If NVJ formation is indeed disrupted under these conditions, we will revise the interpretation more cautiously.

    *o The argument involving Ypf1 is weak, as the observed effect could be indirect and mediated via another factor. *

    The relationship between Msc1 and Ypf1 will be described more cautiously in the revised manuscript.

      • Figure 2B: The statistical analysis (Student's t-test) is not appropriate for the dataset presented.* The statistical analysis for Figure 2B will be revised using a more appropriate method.
      • Additional point: The authors again use a mitochondrial protein as a loading control in Figure 1D, which requires justification.* As mentioned above, Tim23 used as a loading control was selected based on our previous study showing that its abundance remains unchanged during glucose starvation (Fujimoto and Tamura,* J. Cell Biol.* 2026). This explanation will be added to the revised manuscript.

    *Conceptual interpretation • The link between transcriptional reprogramming and NVJ remodeling is not convincingly demonstrated. The data suggest a temporal correlation but do not establish causality. • The PI staining experiments show increased cell death in the absence of Msc1. However, a causal relationship to NVJ function is not demonstrated. An alternative explanation (e.g., an additional role of Msc1 in processes such as DNA repair) should be considered or discussed. *

    These points will be appropriately discussed in the revised manuscript, taking into account the results of additional experiments, including those examining the effects of Nvj1 expression in msc1∆ cells.

    • The claim that Msc1 localizes to the perinuclear space is not sufficiently supported: o Appropriate ER/nuclear envelope controls are missing. As noted above, we will perform additional split-GFP-based analyses to further investigate the localization of Msc1.

    we will perform additional experiments to further examine the membrane topology of Msc1, including controls using antibodies against ER proteins and alkaline extraction analysis of Les1, a fission yeast homolog of Msc1 with a characterized membrane topology. In addition, we will test whether Les1 can complement the msc1∆ mutant.

    *o As an alternative, structural predictions (e.g., transmembrane helix prediction) could strengthen this claim. *

    The N-terminal region of Msc1 is predicted to function as a weak transmembrane segment and a signal sequence. We will incorporate these predictions into the revised manuscript and perform additional experiments to examine the topology and potential processing of this region as mentioned above.

    *Literature and references • The authors should provide a reference supporting NVJ expansion during nitrogen starvation.

    The appropriate reference will be cited in the revised manuscript.

    *Methods • The antibody section is incomplete; all antibodies used need to be specified. *

    The antibody information will be completed in the revised Methods section.

      • Cultivation conditions require more detail: o duration of growth o timing and conditions of glucose starvation shift

    The cultivation conditions will be described in greater detail in the revised Methods section.

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

    Reviewer#1

    Previous reports of Msc1 in patches (page 3): the citation of Breker et al (LOQATE) seems wrong because that database shows Msc1 at the ER not at NVJ; Medina-Suarez et al is also not great: it shows NE w some patches - not high penetrance + some cER. So I suggest the authors simply rely on their own BioRxiv paper.* *

    We agree that the LOQATE database only weakly shows punctate localization of Msc1 and will therefore remove this citation. However, we believe that Medina-Suarez et al. still provides relevant support because Msc1 exhibits a localization pattern resembling the NVJ in a subset of cells, and therefore we plan to retain this reference.

    Table S1: needs Msc1-GFP adding to some lines.

    We revised Table S1 accordingly.

    *Avoid unnecessary abbreviations: GS creates a novel word that has no obvious meaning and makes the manuscript hard to read rapidly. It would be better to use "glucose starvation" in all cases, especially the abstract. *

    P7: "These results indicate that loss of Msc1 impairs NVJ function more severely than loss of Nvj1 alone." Here NVJ function might not be the target of Msc1 deletion, since nvj1-deletion does not show increased cell death. Also, in general very little is known about NVJ function as very few phenotypes can be pinned down to loss of the NVJ. Better here to say "cell function" (that may involve some aspect of Msc1's interactions at NVJs) instead.

    We revised the wording accordingly.

    Reviewer#2

    *- It is not certain what the term "stability of multiple NVJ proteins" means. Could another term be used, or this explained? *

    We agree that the term "stability" could imply a specific mechanism that is not directly demonstrated in our study. Therefore, we have revised the text to more accurately reflect our findings by referring to the abundance of NVJ proteins rather than their stability: "Together with these observations, our results suggest that Msc1 plays a central role in maintaining the abundance of multiple NVJ proteins, including Nvj1, Ypf1, Nsg1, and Nsg2, during glucose starvation."

    *o The title of the Figure 2 is: "Snf1 signaling and VLCFA metabolism modulate NVJ partitioning of Msc1" - what is "NVJ partitioning" - for me it would be clearer to write "Snf1 signaling and VLCFA metabolism modulate the localization of Msc1 to NVJ" *

    As suggested by the reviewer, we will revise the Figure 2 title from "Snf1 signaling and VLCFA metabolism modulate NVJ partitioning of Msc1" to "Snf1 signaling and VLCFA metabolism modulate the localization of Msc1 to the NVJ."

    *Figure 1 A and B shows that Msc1-GFP expression is upregulated in cells starved for glucose for 24h, but not in nitrogen-starved cells.

    • Comments: o Is Tim23 used as a loading control? If yes, it should be stated in the figure legends and/ or main text. *

    We have revised the figure legends to indicate that Tim23 was used as a loading control.*

    *o Which antibody is used for Western in B? *

    We have revised the figure legend to specify that the immunoblots shown in Fig. B were probed with anti-Msc1, anti-Nvj1, anti-Ypf1, and anti-Tim23 antibodies.

    • Comment: It would be helpful to explain the abbreviation "PK" in Figure 1C Figure legend. *

    We have revised the Figure 1C legend to define PK as proteinase K.

    Figure 1 D: Msc1-GFP localization to the NVJ is dependent on Nvj1, Vac8, but not Nsg1 and 2 and Ypf1

    • Comment: a typo: "(D) Fluorescence microscopy images of the indicates strains..." should be "indicated".
    • Comment: "Single focal planes were shown." Would be better in present tense "are shown". *

    We have corrected "indicates" to "indicated" and revised "Single focal planes were shown" to "Single focal planes are shown" in the Figure 1D legend.

    *Figure S2.

    • The list of genes analyzed and the conditions analyzed are different in the figure and in the legend. Probably the figure is correct. *

    We revised the Figure S2 legend.

    3. Description of analyses that authors prefer not to carry out

    Reviewer#1

    Comment:

    Function/structural form: the manuscript is light on describing what Msc1 is: it shares the same repeat structure that has been described in Ish1/Les1. The S pombe work described the repeats wrongly as motifs, when AlphaFold2 confidently predicts them as structurally characteristic domains with 2 parallel helices separated by a loop. It would be interesting to speculate a bit on how these might function in the NVJ. One major mystery of the NVJ is the extreme uniformity, shown especially well by cryo-ET (MIllen et al (2008) 10.1111/j.1600-0854.2008.00789.x). This suggests some long-range oligomerisation: is it possible that Msc1 provides that? Possible experiments include expressing Pombe Ish1/Les1 either whole or chimeras with Msc1 to see if they function and are extractable. If that is not to be done here it should at least be discussed.

    Regarding the suggested experiments using S. pombe Ish1/Les1 or chimeric constructs, we agree that these would be interesting approaches. However, because we plan to prioritize additional analyses of Msc1 topology, including detailed characterization of the N-terminal region and repeated alkaline carbonate extraction experiments using ER membrane protein controls, we do not currently plan to pursue extensive chimera-based functional analyses within the scope of the present revision.

    With respect to the possibility that Msc1 contributes to long-range oligomerization underlying the structural uniformity of the NVJ, we currently consider this possibility less likely. In density-gradient centrifugation analyses performed under mild detergent conditions, the apparent molecular size of Msc1 was not particularly large, and we therefore did not obtain evidence supporting formation of a stable large oligomeric complex by Msc1.

    Reviewer#2

    - The authors refer to a previous study showing that nvj1D deletion does not affect protein levels of several NVJ proteins, however, it would be nice to have this data shown here - i.e. the localization of Tsc13, Ypf1 (and Hmg2) in the nvj1D mutant, especially since the study cited has not been peer-reviewed yet: "Notably, our previous work showed that loss of Nvj1 or Ypf1 does not affect the protein levels of each other or those of other NVJ-associated factors such as Nsg1 and Nsg2 (Fujimoto and Tamura, 2025)."

    We believe this point may reflect two partially distinct issues: (i) whether loss of Nvj1 affects the protein levels of NVJ-associated factors, and (ii) whether loss of Nvj1 affects their NVJ localization. In our previous study, we showed that loss of Nvj1 does not affect the protein levels of Ypf1, Nsg1, or Nsg2, whereas their NVJ localization does require Nvj1 (Fujimoto and Tamura, 2026; J Cell Biol. 225. doi:10.1083/jcb.202506071, now published). In addition, a previous study demonstrated that Tsc13 localizes to the NVJ in an Nvj1-dependent manner (Kvam et al., 2005). We also showed that loss of Ypf1 prevents efficient accumulation of Hmg2 at the NVJ (Fujimoto and Tamura, J. Cell Biol. 2026). We therefore believe that these localization dependencies have already been sufficiently established in previous studies.

  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

    Summary:

    In this study, the authors identify Msc1 as a factor associated with nucleus-vacuole junctions (NVJs) during glucose starvation. Using Saccharomyces cerevisiae as a model system, and combining immunoblotting and microscopy approaches, they report a functional connection between Msc1 and the NVJ component Nvj1.

    Major comments:

    • Are the key conclusions convincing?

    Overall, the main conclusions are largely convincing. However, several interpretations are overstated and should be phrased more cautiously (see specific comments below).

    • Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether?

    Yes. In several instances, the data support correlation rather than causation, and the authors should clearly indicate when conclusions are speculative.

    • Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation.

    For some conclusions, either:

    • the interpretation should be weakened, or
    • additional experiments are needed to fully support the claims
    • Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments.

    If Western blot membranes are available, additional controls could likely be addressed by reprobing, which would require minimal effort and a short timeframe. Suggested microscopy experiments would require strain construction and are therefore expected to take approximately 2-3 weeks.

    • Are the data and the methods presented in such a way that they can be reproduced?

    Some methodological details are insufficiently described and should be clarified to ensure reproducibility.

    • Are the experiments adequately replicated and statistical analysis adequate?

    The authors do not specify which tests for normality were performed. It is therefore difficult to assess whether the use of Student's t-test is appropriate. In at least one case (comparison of three groups), a t-test is not appropriate and should be replaced with a suitable multiple-comparison test.

    Minor comments:

    • Specific experimental issues that are easily addressable.

    See below

    • Are prior studies referenced appropriately?

    Yes, mostly/ The authors should provide a reference supporting NVJ expansion during nitrogen starvation.

    • Are the text and figures clear and accurate?

    Yes

    • Do you have suggestions that would help the authors improve the presentation of their data and conclusions? see below

    Figures and data presentation

    • Figure 1A: The image is difficult to interpret. The authors should improve visibility, for example by:

    • using grayscale instead of magenta/green for single channels, or

    • applying an intensity LUT. This is particularly important as the Nvj1 signal is barely visible.

    • Figure 1B: The use of Tim23 as a loading control is not appropriate. The authors should justify why a mitochondrial protein was used as a reference.

    • Figure 1C: The experimental design and interpretation are problematic:

    • Using an ER protein together with mitochondrial markers in the proteinase K protection assay is not appropriate for the stated conclusions.

    • The claim that Msc1 is an integral membrane protein is not sufficiently supported, particularly if a polyclonal antibody was used.

    • The authors should provide additional evidence for localization (or use alternative approaches).

    • Figure 1D:

    • The authors conclude that deletion of NVJ1 and VAC8 reduces Msc1 colocalization. However, an alternative explanation is that NVJs are not formed under these conditions.

    • This conclusion should therefore be phrased more cautiously. Alternatively, a known NVJ marker should be included to demonstrate NVJ formation.

    • The argument involving Ypf1 is weak, as the observed effect could be indirect and mediated via another factor.

    • Figure 2B: The statistical analysis (Student's t-test) is not appropriate for the dataset presented.

    • Additional point: The authors again use a mitochondrial protein as a loading control in Figure 1D, which requires justification.

    Conceptual interpretation

    • The link between transcriptional reprogramming and NVJ remodeling is not convincingly demonstrated. The data suggest a temporal correlation but do not establish causality. The PI staining experiments show increased cell death in the absence of Msc1. However, a causal relationship to NVJ function is not demonstrated. An alternative explanation (e.g., an additional role of Msc1 in processes such as DNA repair) should be considered or discussed. The claim that Msc1 localizes to the perinuclear space is not sufficiently supported: Appropriate ER/nuclear envelope controls are missing. As an alternative, structural predictions (e.g., transmembrane helix prediction) could strengthen this claim.

    Literature and references

    The authors should provide a reference supporting NVJ expansion during nitrogen starvation.

    Methods

    • The antibody section is incomplete; all antibodies used need to be specified.
    • Cultivation conditions require more detail:
    • duration of growth
    • timing and conditions of glucose starvation shift

    Referee cross-commenting

    Rev#1:

    I generally agree with the other reviewers. I found an error (?typo) in one thing Reviewer 3 says about Fig 1C: "The claim that Msc1 is an integral membrane protein is not sufficiently supported, particularly if a polyclonal antibody was used." I think they mean: "The claim that Msc1 is NOT an integral membrane protein is not sufficiently supported, particularly if a polyclonal antibody was used." I see that my own review has lots of typos - I will write separately to the editor about those.

    Rev#2:

    I agree with the Reviewer 3 that the link between transcriptional reprogramming and NVJ remodeling is not convincingly demonstrated.

    I agree with the Reviewer 3 that the localization of Msc1 to the perinuclear space is not sufficiently supported. The authors may re-write the conclusion to include this uncertainty, or add experimental data.

    I am not sure if I agree with the Reviewer 1 in that the loss of Msc1 leads to the downregulation of Nvj1 "mostly through destabilisation since the transcriptional effect is marginal". Available data does not include the quantification of the Nvj1 protein levels in the msc1- mutant compared to WT, therefore, it is presently unclear how large the downregulation at the protein level is.

    I agree with the Reviewer 3 that the Methods section needs a more detailed description, especially of the growth conditions and glucose starvation protocol (at which OD600 were cells diluted to, were cells washed prior to media change, etc.).

    Rev#3:

    I find Reviewer 2's suggestion of a complementation experiment compelling; this assay would require minimal additional effort and would help exclude off-target effects of the msc1Δ phenotype.

    I agree with Reviewer 1 that the use of "GS" is unnecessary and hinders readability; "glucose starvation" should be used throughout.

    I agree with Reviewer 1 that a more thorough comparison with homologous proteins in S. pombe (Ish1/Les1), including topology and functional parallels, would substantially strengthen the manuscript.

    I thank Reviewer 1 for identifying the misleading phrasing regarding integral versus associated membrane proteins. However, I maintain that the assay in Figure 1C still requires stronger support.

    Significance

    Nature and significance of the advance

    The manuscript describes Msc1 as a novel factor associated with NVJs, contributing to the growing body of work characterizing these membrane contact sites. While this represents a potentially interesting addition, the mechanistic insight provided is currently limited.

    Context within the literature

    The study fits into a series of recent publications that systematically characterize NVJs. Compared to these, the present work adds a new component but does not substantially advance mechanistic understanding.

    Audience

    The primary audience will be researchers interested in:

    • membrane contact sites
    • NVJ biology
    • yeast cell biology The manuscript could reach a broader audience interested in cellular metabolism if the authors more strongly connect their findings to metabolic states and regulatory pathways.

    Reviewer expertise

    The reviewer has expertise in:

    • membrane contact sites (MCS)
    • nucleus-vacuole junctions (NVJs)
    • yeast as a model system
    • microscopy-based analysis
    • intracellular communication
  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

    The study describes the finding of the nuclear envelope protein Msc1 as a new component of the membrane contact site nucleus vacuole junction (NVJ) under the conditions of glucose starvation. Msc1 has previously only been known as a nuclear envelope protein, presumably localizing to the nuclear lumen, and its role in DNA damage repair. The main finding of this study is the glucose starvation-induced upregulation and NVJ-localization of Msc1 (Figure 1). The second main finding is that the loss of Msc1 results in an impaired induction of the expression of Nvj1 (the main component of the NVJ, responsible for the formation of NVJ via direct interaction with Vac8) upon glucose starvation (Fig. 3 A). The effect of Msc1-loss on the Nvj1 expression levels is transcriptional (Fig. 4 B). The glucose starvation-mediated expression induction of some other previously identified NVJ components, Nsg1 and Nsg2 is also impaired in the msc1D mutant, while the expression of Ypf1 is affected to a lesser degree. The data supporting these two main findings are solid (Figure 1; Figure 3 A; Figure 4 A, B).

    The study further shows that the loss of Msc1 results in a loss of NVJ-localization of NVJ components Tsc13, Ypf1 and to a lesser degree Hmg2. The microscopy data looks solid, however the interpretation of this finding is not clear. In my view, the most likely explanation is that the effect of Msc1 loss on the localization of NVJ components to the NVJ is due to the impaired glucose starvation-induced Nvj1 expression in the msc1D mutant.

    Major comments:

    Here are suggested experiments that would strengthen the study:

    • It is difficult to imagine how a NE protein could affect expression levels of other NVj proteins - this key finding would be supported by a complementation experiment where MSC1 is expressed from a vector - to test whether this rescues the phenotype (to make sure that the observed phenotype is not due to an off-target effect of msc1D deletion)
    • If technically feasible under the glucose starvation conditions, this hypothesis could be tested by overexpressing Nvj1 from an inducible or some other promoter.
    • The authors refer to a previous study showing that nvj1D deletion does not affect protein levels of several NVJ proteins, however, it would be nice to have this data shown here - i.e. the localization of Tsc13, Ypf1 (and Hmg2) in the nvj1D mutant, especially since the study cited has not been peer-reviewed yet: "Notably, our previous work showed that loss of Nvj1 or Ypf1 does not affect the protein levels of each other or those of other NVJ-associated factors such as Nsg1 and Nsg2 (Fujimoto and Tamura, 2025)."
    • The effect of msc1D deletion on Tsc13 proteins levels (preferentially using the same Tsc13-GFP strain as used in microscopy - anti Tsc13 or anti-GFP antibodies could be used)

    Other Major comments:

    • The results concerning the localization of Msc1-GFP in elo3D mutant have been interpreted as "accelerated localization", "expansion of the the size of Msc1-NVJ domain" etc. However, the levels of Msc1-GFP in the elo3D mutant are higher compared to WT (Figure 2 D). Considering this, it is very likely that the larger surface area measured in the elo3D mutant is a consequence of this. This could be potentially checked by comparing images set of WT and elo3D that are set to a similar fluorescence intensity. In any case, this possibility should be definitely addressed in the interpretation of the result.
    • There is an impression that the data has been overinterpreted, and the conclusions should be written much more carefully. Examples:
      • "Here, we show that Msc1 is a GS-responsive NVJ factor that plays an important role in functional NVJ remodeling." - based on data shown, the effect of Msc1 could be indirect. The statement above should be re-written or argumented much better.
      • "we find that GS-dependent induction of NVJ1 transcription is attenuated in msc1Δ cells, suggesting that proper NVJ remodeling contributes to the execution of stress-responsive transcriptional programs" - this is unclear; which data support this?
      • "Together, these findings position Msc1 as an upstream regulator linking GS signaling to functional maturation of the NVJ and associated cellular adaptation responses." - same comment as above
      • "...suggesting that Msc1 functions as a GS-responsive regulator of NVJ functions."
      • "...these findings suggest that Msc1 acts upstream of Ypf1 in orchestrating GS-induced NVJ functional maturation."
      • "Collectively, these results indicate that Snf1 acts upstream of Msc1 to drive GS-induced NVJ remodeling, whereas reduced Elo3 activity further accelerates this process and promotes Msc1 accumulation." - not sure if the available data support this.
      • "These results indicate that although Msc1 ...... it is required for efficient GS-dependent functional maturation of the NVJ domain."
      • "These observations suggest that loss of Msc1 does not cause a general defect in transcriptional activation but rather impairs the proper execution and dynamic range of GS-dependent transcriptional responses." - this is unclear
      • "Within this context, the robust induction of NVJ1 appears to be particularly sensitive to Msc1 deficiency." - this sentence would benefit from being re-written.
      • "Together, these results indicate that Msc1 contributes to transcriptional reprogramming associated with NVJ remodeling during GS." - this sounds overstated.
      • "the observation that loss of Msc1 attenuates GS-dependent induction of NVJ1 raises the possibility that NVJ remodeling influences stress-responsive gene expression programs."
    • It is not certain what the term "stability of multiple NVJ proteins" means. Could another term be used, or this explained?

    OTHER COMMENTS FIGURE BY FIGURE - SOME ARE MAJOR (overlapping to the above comments), SOME ARE MINOR:

    Figure 1: Figure 1 A and B shows that Msc1-GFP expression is upregulated in cells starved for glucose for 24h, but not in nitrogen-starved cells.

    • Comments: o Is Tim23 used as a loading control? If yes, it should be stated in the figure legends and/ or main text. o Size of the markers (protein ladder) would be helpful. o Which antibody is used for Western in B?
    • Comment: It would be helpful to explain the abbreviation "PK" in Figure 1C Figure legend. Figure 1 D: Msc1-GFP localization to the NVJ is dependent on Nvj1, Vac8, but not Nsg1 and 2 and Ypf1
    • Comment: a typo: "(D) Fluorescence microscopy images of the indicates strains..." should be "indicated".
    • Comment: "Single focal planes were shown." Would be better in present tense "are shown".

    Figure 2:

    • Comment: It is not clear if these are the same strains as analyzed by microscopy (GFP-tagged Msc1). This should be specified in the Figure legend 2 D.
    • Comment: o Since the levels of Msc1-GFP in the elo3D mutant are higher compared to WT (Figure 2 D), the larger surface area measured in C may be a consequence of this. o It is not clear if Figure A and D analyze the same strains (western blot and microscopy - do both show GFP-tagged Msc1? - using anti-GFP?). This should be specified in the Figure legend 2 D. Since the increased area measured in Figure 2 C could be due to increased Msc1-GFP levels in this mutant strain, the WB should check the levels of Msc1-GFP in the same strain and under same conditions as analyzed in Figure 2 C. o The title of the Figure 2 is: "Snf1 signaling and VLCFA metabolism modulate NVJ partitioning of Msc1" - what is "NVJ partitioning" - for me it would be clearer to write "Snf1 signaling and VLCFA metabolism modulate the localization of Msc1 to NVJ" o Does Tim23 serve as a loading control in Figure 2 D? o Would be good to have protein ladder sized marked in Western blots o Since the increase in Msc1 levels in the elo3D mutant could be significant for the interpretation of the results, it would be helpful to have quantification of the protein levels in WB (normalized to a loading control).

    Figure 3 Together these data show that localization of other NVJ-proteins to the NVJ depends on the presence of Msc1. Comment:

    • From the available data it is possible that Msc1 recruits these components by direct interaction, or by modifying the structure of NVJ, or functions in an indirect manner - this should be discussed in the Discussion. Comment:
    • The signal of Tsc1-GFP in log-growing cells is very weak, therefore the quantification may be unreliable. I would remove this condition (log-grown cells) form the quantification in C) due to the low signal, since it is not crucial to the interpretation of the data. If the authors prefer to leave it, that is fine.
    • The title of the Figure 3 is "Msc1 supports stability and recruitment of NVJ-associated proteins" - I am not sure what "stability" is; the data don't address stability or recruitment in a direct manner - I suggest to change the figure title into a statement describing what is shown in the Figure, for example: "The loss of Msc1 results in decreased Nvj1 levels and a decreased localization of NVJ proteins to the NVJ). And have a comment that this data suggests that Msc1 supports recruitment of NVJ-associated proteins, likely in an indirect manner, based on the finding that the loss of Msc1 leads to a lower expression of Nvj1, in the main text (e.g. in the Discussion).
    • Is it possible that the loss of Msc1 on the loss of NVJ-localized Tsc13 is due to the downregulation of Tsc13 expression? Considering the effect of msc1D deletion on the expression of some NVJ proteins (Figure 3 A), Tsc13 expression levels would be good to be checked, considering the effect of msc1D on Tsc13-GFP localization. It would be optimal to do the WB with the same Tsc13-GFP-expressing strain and under the same growth conditions as was used in the microscopy in the Figure 3 B.
    • Expression levels of Ypf1 are lower in the msc1D strain, than in the WT (Fig. 3 A) - could this affect lower NVJ-area in his mutant? (Fig. 3 B)

    Figure 4. Figure 4 A shows mRNA levels in glucose starved cells compared to log-.growing cells for MSC1, NVJ1 and YPF1.

    • Comment: I would move Figure 4 A to Figure 1. Figure 4 B shows mRNA levels of proteins expressed in WT and msc1D mutant strain, in log-growing cells in under glucose starvation. The data show that the loss of Msc1 leads to a decrease in NVJ1 mRNA under the conditions of glucose starvation. Th expression of other NVJ proteins analyzed are not affected.
    • Comment: Would this Figure 4 A-B better fit together with the data showing Nvj1 levels in the msc1D mutant from a previous figure (3 A)? Figure 4 C shows PI staining of cells after 5 days of glucose starvation. The loss of Msc1 leads to a double increase in PI-positive cells (in contrast to the nvj1D mutant, which is similar to WT), indicating that the viability of cells after 5 days of glucose starvation is decreased in the absence of Msc1.
    • Comment: Since there is no phenotype of nvj1D, this is likely not due to the non-functional NVJ, but another function of Msc1 - the question is which. This could be discussed in the Discussion.
    • Comment: This is informative, however it is not sure why this data is placed together with the mRNA data within the Figure 4.

    Figure S1.

    • Comment - as in Figure 2 - Msc1-GFP has a much stronger signal in elo3D mutant, than in WT, which could influence (or likely influences) the measured area. Perhaps one way to test this is to image WT cells with higher % of laser "a "longer exposition"), to get a stronger signal similar to that seen in the elo3D mutant, and then repeat the quantification.
    • Taken the result as it is presently, I suggest taking the Figure S1 out. Figure S2.
    • The list of genes analyzed and the conditions analyzed are different in the figure and in the legend. Probably the figure is correct. Figure S3 . Validation of anti-Msc1 antibody
    • Could be moved as S1.

    *Referee cross-commenting

    Rev#1:

    I generally agree with the other reviewers. I found an error (?typo) in one thing Reviewer 3 says about Fig 1C: "The claim that Msc1 is an integral membrane protein is not sufficiently supported, particularly if a polyclonal antibody was used." I think they mean: "The claim that Msc1 is NOT an integral membrane protein is not sufficiently supported, particularly if a polyclonal antibody was used." I see that my own review has lots of typos - I will write separately to the editor about those.

    Rev#2:

    I agree with the Reviewer 3 that the link between transcriptional reprogramming and NVJ remodeling is not convincingly demonstrated.

    I agree with the Reviewer 3 that the localization of Msc1 to the perinuclear space is not sufficiently supported. The authors may re-write the conclusion to include this uncertainty, or add experimental data.

    I am not sure if I agree with the Reviewer 1 in that the loss of Msc1 leads to the downregulation of Nvj1 "mostly through destabilisation since the transcriptional effect is marginal". Available data does not include the quantification of the Nvj1 protein levels in the msc1- mutant compared to WT, therefore, it is presently unclear how large the downregulation at the protein level is.

    I agree with the Reviewer 3 that the Methods section needs a more detailed description, especially of the growth conditions and glucose starvation protocol (at which OD600 were cells diluted to, were cells washed prior to media change, etc.).

    Rev#3:

    I find Reviewer 2's suggestion of a complementation experiment compelling; this assay would require minimal additional effort and would help exclude off-target effects of the msc1Δ phenotype.

    I agree with Reviewer 1 that the use of "GS" is unnecessary and hinders readability; "glucose starvation" should be used throughout.

    I agree with Reviewer 1 that a more thorough comparison with homologous proteins in S. pombe (Ish1/Les1), including topology and functional parallels, would substantially strengthen the manuscript.

    I thank Reviewer 1 for identifying the misleading phrasing regarding integral versus associated membrane proteins. However, I maintain that the assay in Figure 1C still requires stronger support.

    Significance

    General assessment - strenghts and limitations:

    The identification of Msc1 as a new glucose starvation-induced protein that localizes to the NVJ is supported by strong data and represents a novel and a strong point of the paper. Furthermore, the finding that the loss of Msc1 results in the impaired expression of several other NVJ-localized proteins under glucose starvation is convincing, although the solidity of this latter data requires some more experimental controls (detailed above). The weak point is the interpretation of the Msc1 loss on the localization of other NVJ proteins - the present conclusions need to be modified, or supported by the additional experimental data.

    Advance - compare the study to existing knowledge - does it fill the gap?

    The study identifies protein Msc1, which was previously known as a nuclear envelope protein involved in DNA damage repair, as a new component of the membrane contact site nucleus vacuole junction (NVJ), whose expression and the localization to the NVJ is induced by glucose starvation. What kind of advance does it make - conceptual; incremental...? The finding that a nuclear lumen protein, which is required for DNA damage repair, under certain circumstances (glucose starvation) changes localization and potentially has new roles, has a potential of a conceptual advance, however, for that, more experimental data would be needed, specifically to determine the mechanistic role of Msc1 in glucose starvation, and compare it to it role in DNA damage response. The available data supports mainly an incremental advance in our understanding of the structure and regulation of the NVJ.

    Audience - broad; specialized; basic research...? The audience of this paper will be interested in the basic research. Especially interested may be scientists working with yeast.

    Describe your expertise: My expertise is in yeast genetics, in the field of degradation-mediated protein quality control.

  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

    This is a nice, smallish study of Msc1, a fungal protein of unknown function. The authors show it localises to the NVJ when that expands in late-log/stationary phase, at which stage its transcription is increased 80-fold - an induction one whole order of magnitude greater than shown by Nvj1 itself. This indicates that Msc1 may be a previously unappreciated master regulator of the NVJ. There are some interesting phenotypes of deleting Msc1, including some cell death and loss of Nvj1, mostly through destabilisation since the transcriptional effect is marginal.

    While no mechanism for Msc1 is discovered, that might be too much to ask for in this first paper. However, there are ways to begin to address this that the authors should look into.

    My major issue with the paper is that it makes no link to the previously studied homologues of Msc1 in S pombe (Ish1/Les1 - see Asakawa et all 2022). Admittedly, S. pombe has no Nvj1 homolog, but there is a physical relationship between nucleus and vacuole (Chadwick et al (2020) 10.1088/1478-3975/aba510). Also, the paper on Ish1/Les1 developed a phenotype to test Ish1 (toxicity of over expression) that might be useful for studies of Msc1. The current MS should link to work on Ish1/ Les1 in S. pombe, relating to several features:

    Topology.

    Given the high similarity between Msc1 and Ish1/Les1, they are (a priori) likely to share considerable form and function. If Msc1 is a soluble protein in the ER lumen, then the previous report that Ish1/Les1 have TMDs is wrong. The report here should make that link and carefully explain how the Pombe paper is wrong. Also explain how is it possible for Msc1 (and Ish1/Les1) to stay restricted to the nuclear envelope? (in many images it is diffuse throughout the NE). The only mechanism I can think of is binding an integral protein that sorts to the inner-NE by known mechanisms (or possibly binding to an outer-NE protein that binds to an inner-NE one, like SUN/KASH). I cannot think of any other example of a soluble proteins restricted to the NE - so this is quite a claim.

    An alternative view that could be investigated and should definitely be discussed is that Msc1 (and by implication Ish1 and Les1) has a TMD even though it is extracted by carbonate. Something similar has been reported for some single TMD proteins in mitochondria (Kim et al (2015) 10.1002/pro.2817). Investigations would include proteomics showing whether the protein is normally full length (as coded by the open reading frame) or clipped (indicating the signal sequence is removed for a soluble protein). Such data may already be available in published mass spec datasets.

    Function/structural form:

    the manuscript is light on describing what Msc1 is: it shares the same repeat structure that has been described in Ish1/Les1. The S pombe work described the repeats wrongly as motifs, when AlphaFold2 confidently predicts them as structurally characteristic domains with 2 parallel helices separated by a loop. It would be interesting to speculate a bit on how these might function in the NVJ. One major mystery of the NVJ is the extreme uniformity, shown especially well by cryo-ET (MIllen et al (2008) 10.1111/j.1600-0854.2008.00789.x). This suggests some long-range oligomerisation: is it possible that Msc1 provides that?

    Possible experiments include expressing Pombe Ish1/Les1 either whole or chimeras with Msc1 to see if they function and are extractable. If that is not to be done here it should at least be discussed.

    Minor Issues

    The Abstract switches from response to lack of glucose to terminology about 'stress-response'. This could appear to be an effort to appear more interesting. If the idea is to remain, it needs some support with the introduction of the idea that yeast experiences stress (as opposed to "normal" transcription driven programmatic changes in relation to changing levels of glucose in normal cultures.

    Introduction para 1 seems to be dedicated to the idea that a set of intracellular structures (here MCS) are 'dynamically and coordinately remodeled in response to metabolic and stress conditions'. This conclusion applies widely and may not be noteworthy. The paragraph needs a bit of rethinking.

    Previous reports of Msc1 in patches (page 3): the citation of Breker et al (LOQATE) seems wrong because that database shows Msc1 at the ER not at NVJ; Medina-Suarez et al is also not great: it shows NE w some patches - not high penetrance + some cER. So I suggest the authors simply rely on their own BioRxiv paper.

    Figure 2D: I could not find Nsg1 result described in the text.

    P6: "Strikingly, GS-dependent transcriptional activation of NVJ1 was significantly suppressed in msc1∆ cells (Fig. 4B)." This overstates the strength of the result. Instead state that the induction diminishes from 6-fold to 4-fold, and give the p value.

    P7: "These results indicate that loss of Msc1 impairs NVJ function more severely than loss of Nvj1 alone." Here NVJ function might not be the target of Msc1 deletion, since nvj1-deletion does not show increased cell death. Also, in general very little is known about NVJ function as very few phenotypes can be pinned down to loss of the NVJ. Better here to say "cell function" (that may involve some aspect of Msc1's interactions at NVJs) instead.

    Table S1: needs Msc1-GFP adding to some lines

    Language:

    Avoid unnecessary abbreviations: GS creates a novel word that has no obvious meaning and makes the manuscript hard to read rapidly. It would be better to use "glucose starvation" in all cases, especially the abstract. Avoid use of rhetorical wording (e.g. dramatic): just state the results (e.g. 80-fold induction) and let the results be dramatic/striking etc. all by themselves.

    Referee cross-commenting

    Rev#1:

    I generally agree with the other reviewers. I found an error (?typo) in one thing Reviewer 3 says about Fig 1C: "The claim that Msc1 is an integral membrane protein is not sufficiently supported, particularly if a polyclonal antibody was used." I think they mean: "The claim that Msc1 is NOT an integral membrane protein is not sufficiently supported, particularly if a polyclonal antibody was used." I see that my own review has lots of typos - I will write separately to the editor about those.

    Rev#2:

    I agree with the Reviewer 3 that the link between transcriptional reprogramming and NVJ remodeling is not convincingly demonstrated.

    I agree with the Reviewer 3 that the localization of Msc1 to the perinuclear space is not sufficiently supported. The authors may re-write the conclusion to include this uncertainty, or add experimental data.

    I am not sure if I agree with the Reviewer 1 in that the loss of Msc1 leads to the downregulation of Nvj1 "mostly through destabilisation since the transcriptional effect is marginal". Available data does not include the quantification of the Nvj1 protein levels in the msc1- mutant compared to WT, therefore, it is presently unclear how large the downregulation at the protein level is.

    I agree with the Reviewer 3 that the Methods section needs a more detailed description, especially of the growth conditions and glucose starvation protocol (at which OD600 were cells diluted to, were cells washed prior to media change, etc.).

    Rev#3:

    I find Reviewer 2's suggestion of a complementation experiment compelling; this assay would require minimal additional effort and would help exclude off-target effects of the msc1Δ phenotype.

    I agree with Reviewer 1 that the use of "GS" is unnecessary and hinders readability; "glucose starvation" should be used throughout.

    I agree with Reviewer 1 that a more thorough comparison with homologous proteins in S. pombe (Ish1/Les1), including topology and functional parallels, would substantially strengthen the manuscript.

    I thank Reviewer 1 for identifying the misleading phrasing regarding integral versus associated membrane proteins. However, I maintain that the assay in Figure 1C still requires stronger support.

    Significance

    This is a nice, smallish study of Msc1, a fungal protein of unknown function. The authors show it localises to the NVJ when that expands in late-log/stationary phase, at which stage its transcription is increased 80-fold - an induction one whole order of magnitude greater than shown by Nvj1 itself. This indicates that Msc1 may be a previously unappreciated master regulator of the NVJ. There are some interesting phenotypes of deleting Msc1, including some cell death and loss of Nvj1, mostly through destabilisation since the transcriptional effect is marginal.

  5. Version published to 10.64898/2026.03.13.711511 on bioRxiv