The collection of yeast strains you've built is a great resource for exploring how more environmental factors influence FRET behavior. It will be interesting to compare future datasets to what you've found here and start to characterize IDRs as sensitive to specific stresses vs. generally sensitive to environmental perturbations.

Were there any interesting shared features of the IDRs that didn't express well, or was it seemingly random?

This is really beautiful work that is hypothesis-generating for both cell biology and climate change adaptation research, all in non-model organisms!

If possible, it would be nice to see L. pholis SYTO12 staining here as well. Whether it’s more like H. antarcticus or mammalian cells, either way is interesting and informative!

I'm curious about the species-specific differences in pH between these perinuclear bodies, as both nuclear proximity and temperature impact lysosomal pH in mammalian cells. If anything, I would expect H. antarcticus lysosomes and lysosome-related organelles to be more acidic. Lots of interesting follow-up directions here!

These images are so beautiful!

This is so interesting and really highlights the importance of assessing both RNA and protein levels, especially in murky contexts like this involving multiple RNA-binding proteins that in turn are regulating the expression of proteins involved in processes like splicing.

I'm curious how many RNAs were excluded with this step. How many RNAs showed variant-specific shifts, and are there any trends in these clusters that can inform our understanding of the consequences of specific mutations in TDP-43?

It's neat that you designed multiple variants predicted to be better or worse at phase separation rather than relying on one in each direction, it makes the datasets so much richer for informing future variant design!

I think this will definitely be an interesting future area of study, and would potentially benefit from examining the expression, localization, and activity of enzymes like superoxide dismutases, which are essential for copper, zinc, and manganese homeostasis.

This is such a neat technique and the images in figures 3 and 4 are so beautiful!

Has this been formally tested? If so, it would be nice to have a citation here.

The change in CDF-2::GFP intensity is also very striking. Do you think that excess manganese promotes increased transcription and/or translation of cdf-2 mRNA?

Is this trend pronounced in any tissues in particular, or is this random/across all tissue types?

In future work, it would be useful to collect brightfield/DIC microscopy data for these strains to examine the ratio of spherical to filamentous cells in these diverse environmental conditions.

This is an invaluable expansion of genetic information! I think it would be neat to highlight the 67 newly sequenced strains in Figure 1 to emphasize your contributions.

It's interesting to see that one of these grows relatively well in the presence of cycloheximide. It might be neat to dig into this strain more with additional phenotypic characterization and exploration of what's conferring this fitness.

I'm curious how many of your higher confidence dynamic proteins overlap with dynamic proteins identified in previous studies, and if there's any enrichment of proteins belonging to specific clusters in Figure 8.

This is a very valuable study and I'm excited to explore these data when they're publicly available!

Several studies investigate the transcriptomes of patients with not only endometriosis but also adenomyosis. I think it would be helpful to mention this here and provide some background on the co-occurrence rates of endometriosis and adenomyosis. It would be interesting if large, multi-study analyses like yours could parse out transcriptomic differences between individuals with one or both conditions.

Thank you for doing this work! Utilizing single-cell datasets for in silico drug prediction is a pragmatic way to help us better understand and treat endometriosis.

Would it be possible to include these gene lists in a supplemental table?

I'm curious if your analyses were able to provide any new information about specific EMT-related factors that are enriched in these cell types. What EMT genes were consistently identified? Were there enrichment signatures you found that bulk analyses may have missed?

This is very cool! Is this experiment with wild-type or fmp52∆ cells?

It looks like there's at least one putative UPRE upstream of FMP52 (425 bases upstream of the ATG, 5'-TACGTGT-3')! I'm curious if you tried looking at endogenous Fmp52-GFP with t-2-hex or other ER stress-inducing treatment, as it may get upregulated and be more visible than in unstressed conditions? This would be consistent with your pFMP52-luciferase results.

It will be interesting to see how organismal age and disease state impact hm5C saRNA stability/efficacy!

How does the number of saRNA-positive cells compare to the time points shown in Figure 2? It would be interesting to have micrographs from earlier time points at the same magnification of those in Supp. Fig. 1. I'm also curious if there's something measurably different about the cells still expressing saRNA after three months or if it's just a random decrease in cell cout over time.

These expression patterns between treatments and across time are so striking! How much does this vary between individuals? For example, the patterns between mRNA and saRNA on day 14 are very distinct from one another and I'm curious if these types of differences are consistent.

It could be useful to also assess proteomic data for these RBPs, as RNA abundance often doesn’t correlate with protein abundance, especially in the context of aging.

Since maximum lifespan is not a conserved phenotype, it would be interesting to see if other types of splicing events (e.g. alternative first exons) are enriched for specific biological functions, irrespective of their conservation.

Is it possible that under these conditions, neurons transfer damaged mitochondria and/or other cellular material to astrocytes? It could be interesting to do neuron-specific labeling prior to co-culturing to track mitochondria.

I’m curious if you looked at CHMP2B levels by western blot in addition to immunofluorescence. Is it possible that neurons express alternate mRNA isoforms and/or proteoforms (e.g. truncated CHMP2B) that disrupt canonical ESCRT recruitment?