Heterochromatin epimutations impose mitochondrial dysfunction to confer antifungal resistance

  1. Andreas Fellas
  2. Alison L. Pidoux
  3. Pin Tong
  4. Harriet H. Hewes
  5. Emma C. Wallace
  6. Robin C. Allshire

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Global health and food supply are endangered by an increasing frequency of antifungal resistance in pathogenic fungi. Wild-type fission yeast, Schizosaccharomyces pombe , can gain resistance to insults such as caffeine and antifungal compounds through reversible epimutations. Resistant epimutants exhibit histone H3K9 methylation-dependent heterochromatin islands at various chromosomal locations, reducing expression of underlying genes. Two genes whose heterochromatin island-induced repression causes resistance encode mitochondrial proteins: the LYR domain protein Cup1 and the Cox1 translation regulator Ppr4. Genetic mutations, cup1-tt and ppr4Δ , that phenocopy their respective epimutants, cause mitochondrial dysfunction, including respiratory deficiency, poor growth on non-glucose carbon sources, and elevated reactive oxygen species. RNA-Seq analyses indicate that cup1-tt and ppr4Δ cells activate the mitonuclear retrograde pathway and the Pap1 transcription factor-dependent oxidative stress response pathways. Both mutants show increased nuclear localisation of Pap1 and its recruitment to promoters of genes encoding oxidoreductases and membrane transporters, causing increased efflux activity. cup1 and ppr4 epimutants also show mitochondrial dysfunction phenotypes and increased efflux, explaining how heterochromatin-island epimutations cause drug resistance. Thus, wild-type cells harness epimutations that impose mitochondrial dysfunction to bypass external insults. As mitochondrial dysfunction has been linked to antifungal resistance in several fungi, similar epimutations likely contribute to development of resistance in fungal pathogens.

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

    We prefer not to post our response to reviewers on BioRXiv as it is optional

  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

    This study applies cellular and molecular assays, together with transcriptome analysis, to dissect how certain heterochromatin-based epimutations can confer resistance to caffeine and other drugs in fission yeast cells. The findings indicate that compromising the function of two mitochondrial proteins, Cup1 and Ppr4, leads to increased oxidants and the activation of the mito-nuclear retrograde response, which in turn causes the activation of the Pap1-mediated oxidative stress response, including the induction of transmembrane transporters to increase the efflux of drugs. This provides mechanistic insights into how the chromatin-mediated silencing of mitochondrial factors can result in fungal drug resistance. The authors also show that these phenotypes are variable within a cell population, allowing phenotypic plasticity to changing environments. This is a straightforward and clearly presented study, and the conclusions are generally justified based on the experiments presented.

    Minor comments:

    1. Fig. 3A: The legend needs more information to understand what is shown here. Does this show the normalized read counts (cpm?) for each gene scaled per average counts in all samples? Another possibility would be to show relative data for the two mutants compared to wild-type. Also, the labels for the bottom two clusters seem the wrong way round, i.e. the last cluster should be cup1-tt only. How many genes are shown here which made the cutoff?
    2. To strengthen some of the conclusions, it would be meaningful to calculate the significance of overlaps between key gene lists, given the size of the lists involved and the background gene list (Fig. 3B; Fig. 4).
    3. The font size indicating the significance of differences is too tiny in some bar plots (Fig. 5C-E; Fig. 6D; Fig. 7C).

    Referees cross-commenting

    In response to issues raised by Reviewer 1:

    In my opinion, the growth, TPF and ROS assays applied are robust and diagnostic to show a mitochondrial dysfunction. Additional assays, like Seahorse, would provide more specific insights about particular aspects of mitochondrial dysfunction, but this is not really relevant to this study. The key point is that the epimutations compromise mitochondrial function by downregulating mitochondrial proteins, which, in turn, are exploited by the cell to trigger a stress response that protects against antifungal compounds. The exact nature of the mitochondrial dysfunction, any changes in morphology, or details of differentially expressed genes are not critical for this mechanism, as it relies on downstream processes like the retrograde response that is activated by diverse mitochondrial problems.

    The question of whether heterochromatin-mediated resistance phenotypes are prevalent in human fungal pathogens is interesting and an important avenue for future study. But it is not evident to me how this could be addressed bioinformatically.

    Significance

    This manuscript builds on a previous study by the same group, which showed that different heterochromatin-based epimutations can provide cellular resistance to caffeine (Torres-Garcia et al. Nature, 2000). Here they use the UR1 and UR2 epimutations to highlight an example of how such mutations can generate antifungal resistance and phenotypic plasticity by exploiting side effects of mitochondrial dysfunction. Epimutations are an interesting case of cellular adaptation that lasts longer than gene-expression responses but are more readily reversible and flexible than genetic mutations, allowing bet-hedging by generating variable phenotypes in a clonal cell population. This study provides fresh insights into the downstream effects of epimutations causing altered cellular traits, thus complementing previous studies focusing on the patterns and mechanisms of establishing heterochromatin-based genomic islands. The current study is of interest to researchers working on genome regulation, mitochondrial function, cellular adaption/evolution, and has possible applications to combat antifungal resistance.

    Field of expertise: genome regulation, gene function, fission yeast, stress response

  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

    This very interesting manuscript describes the impact of heterochromatin in triggering down-regulation of mitochondrial respiratory activity in S. pombe, and thereby causing increased efflux and consequently increased resistance to several compounds, including the azole class of antifungal drugs. The authors performed detailed mechanistic studies, focusing on two mitochondrial genes, cup1 and ppr4, which are under heterochromatin-dependent repression. Based on their findings, they conclude that reduced mitochondrial respiration causes increased levels of reactive oxygen species (ROS), which activates the transcription factor Pap1. Pap1 then upregulates several genes, including efflux pumps. The authors performed an excellent set of experiment to address heterogenous cell populations in the epimutants, which are described in the second part of the Results section and provide strong evidence of plastic drug resistance phenotypes.

    The manuscript is beautifully written and the data is presented well. Overall, the conclusions are supported by the data.

    I have a reservation with one particular conclusion that I discuss below under point 1. This can be addressed by modifications to the text. Under point 2, I suggest an easy to do experiment, which would strengthen the conclusion that ROS produced due to mitochondrial dysfunction are driving the drug resistance phenotype. This is an interesting mechanism, and the data in the manuscript supports it, but the authors' have not demonstrated it directly. They could do so by using antioxidants, as I suggest below.

    1. Most of the mechanistic analysis is centred around the transcription factor Pap1. The authors performed experiments to connect the production of ROS in mitochondrial mutants, with higher nuclear localisation of Pap1 and its activation of several genes, including the membrane transporter Cas5 and to a lesser extent Bfr1, which might be responsible for increased efflux. There is no question that efflux is elevated in mitochondrial mutants (a phenotype consistent with previous work in other yeast models). The authors also present data to show that inhibition of efflux reverses drug resistance. The data for Pap1's involvement is good in the cup1 mutant (one of the mitochondrial mutants that was studied) but not so much in the ppr4 mutant (the other mitochondrial mutant that was studied). There was little enrichment of Pap1 on the Cas5 promoter in the ppr4 mutant, and no effects of Pap1 on the expression of Cas5 in the ppr4 mutant (Fig 5C and 5D). While the pap1 mutation reduced resistance of the ppr4 mutant to drugs, the authors acknowledge that this could be due to increased sensitivity of the pap1 mutant to drugs. The enrichment of Pap1 on the bfr1 promoter was also modest in the mitochondrial mutants.

    I would therefore suggest that another transcription factor might be responsible for the upregulation of these efflux pumps and/or other efflux pumps are involved in Pap1' s contribution to drug resistance. The authors should consider modifying their conclusions on Pap1-dependent targets that are responsible for drug resistance in the mitochondrial mutants.

    1. The authors' conclusion that increased ROS levels upon dysfunctional respiration might be driving the drug resistance phenotype in S. pombe (via Pap1 but perhaps other mechanisms too), presents a novel mechanistic link between mitochondria and drug resistance. I would suggest solidifying this conclusion by asking if antioxidants can reduce ROS levels and thereby decrease drug resistance in S. pombe. N-acetyl-L-cysteine could be used for this purpose.

    Significance

    That mitochondrial dysfunction causes drug resistance has been known for over 20 years. This manuscript describes a new mechanism, which relies on the formation of semi-stable epimutants, whereby the expression of genes encoding key mitochondrial proteins is down-regulated. As the authors propose, the beauty of epimutations is that they cause a heterogenous phenotype and are reversible, which would create an opportunity for the organism to use a bet-hedging strategy in drug. The ability to reverse the phenotype would be particularly important with using mitochondrial dysfunction as a strategy to increase drug resistance, because mitochondrial dysfunction lowers metabolic flexibility and growth rates for the organism. Therefore, it is only beneficial in the presence of drugs. This is to my knowledge one of the first logical mechanistic explanations for how fungal cells (but likely applicable more broadly) might use mitochondrial dysfunction to their advantage when needed, and then this can be reversed back to respiratory competence to maintain metabolic flexibility when drug selection is no longer present.

    This study will be of high interest to researchers studying drug resistance and how phenotypic plasticity and bet-hedging mechanisms are used by cells to survive toxic compounds. This is applicable across fields. This study will further be very interesting to the fields of antifungal drug resistance and fungal pathogenesis, and will provide the foundation for studying similar mechanisms in relevant fungal pathogens of animals and plants.

    My expertise is in metabolism and mitochondrial roles in fungal pathogens. I really enjoyed reading the manuscript.

  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 under review appears to present findings on heterochromatin-mediated antifungal resistance, specifically focusing on the role of mitochondrial dysfunction in the model organism Schizosaccharomyces pombe. However, there are several significant concerns regarding the novelty and robustness of the conclusions drawn by the authors.

    The key conclusions of the paper lack sufficient convincing evidence. While the authors attribute resistance phenotypes to heterochromatin-mediated repression, the evidence presented does not strongly support these claims. Significant claims should be qualified as preliminary or speculative, particularly those that extend beyond the experimental results provided. For example, asserting a definitive link between heterochromatin status and antifungal resistance mechanisms requires more comprehensive empirical data.

    Additional experiments are crucial to bolster the claims made in the manuscript. The authors rely heavily on growth assays on nonfermentable carbon sources to supposedly elucidate respiratory function. However, this approach is outdated, and advancements in the field should be employed for a more robust assessment of mitochondrial integrity and function. Techniques such as the Seahorse assay could provide critical insights into the respiration capacity of the mutants. Furthermore, the use of electron transport chain (ETC) inhibitors like antimycin A would offer stronger evidence regarding mitochondrial dysfunction. The current use of generalized DCF staining to assess reactive oxygen species (ROS) lacks specificity. MitoSox and MitoTracker should be utilized to measure mitochondrial ROS levels and examine mitochondrial morphology effectively.

    The authors claim that mitochondrial dysfunction correlates with significant changes in the transcriptome related to aerobic respiration, yet this crucial aspect lacks adequate elaboration in their analysis. Given that mitochondrial function is a primary theme of the manuscript, in-depth discussion and interpretation of the differentially expressed aerobic respiratory genes in the transcriptome data are necessary to validate their conclusions.

    Additionally, as a pathogenic fungal microbiologist, I express interest in investigating whether heterochromatin-mediated resistance phenotypes are prevalent in human fungal pathogens, including Candida albicans and Cryptococcus neoformans. A bioinformatic analysis could help address this inquiry and potentially broaden the relevance of the findings.

    Lastly, in the section "Cup1 and Ppr4 deficiencies...retrograde gene repression," the conclusions are made primarily based on transcriptome analysis and lack empirical confirmation through molecular biology techniques. This section should be revised to include comprehensive molecular evidence supporting the claims.

    Significance

    General Assessment: Strengths and Limitations

    Strengths:

    The study introduces a potentially novel mechanism of antifungal resistance in Schizosaccharomyces pombe through heterochromatin-mediated epimutations. This is particularly relevant in the context of rising antifungal resistance globally.

    The focus on mitochondrial dysfunction as a contributor to drug resistance provides a deeper understanding of how fungi adapt to environmental stressors.

    The manuscript raises important questions about the epigenetic factors influencing fungal resistance, which could inspire subsequent investigations in the field.

    Limitations:

    The research relies heavily on traditional methods for assessing respiratory function, which may not fully characterize the complexities of mitochondrial integrity and function. This may weaken the overall conclusions regarding mitochondrial dysfunction.

    The evidence supporting key claims is not robust enough to confidently assert a direct link between heterochromatin changes and antifungal resistance. The lack of confirmatory experiments using more advanced techniques limits the study's impact.

    The analysis of transcriptome data is insufficiently detailed, leaving significant gaps in understanding the specific mechanisms at play.

    Advance: Comparison to Existing Published Knowledge

    This study contributes to the existing literature by exploring the role of epimutations in antifungal resistance, aligning with emerging interests in epigenetic mechanisms in microbial adaptation. While previous studies have focused on genetic mutations and efflux mechanisms, this research attempts to link heterochromatin dynamics to resistance pathways, thereby filling a conceptual gap in understanding how eukaryotic microorganisms may adapt to antifungal treatments.

    However, the advances made by the study appear to be incremental rather than groundbreaking. ​While it does shed light on the potential role of heterochromatin in drug resistance, further empirical evidence and a stronger methodological approach are required to substantiate these findings convincingly.

  5. Version published to 10.1101/2024.11.13.623381v1 on bioRxiv