Linkage mapping of yeast cross protection connects gene expression variation to a higher-order organismal trait

Natural variation in yeast reveals multiple paths for acquiring higher stress resistance

BACKGROUND

Organisms frequently experience environmental stresses that occur in predictable patterns and combinations. For wild Saccharomyces cerevisiae yeast growing in natural environments, cells may experience high osmotic stress when they first enter broken fruit, followed by high ethanol levels during fermentation, and then finally high levels of oxidative stress resulting from respiration of ethanol. Yeast have adapted to these patterns by evolving sophisticated "cross protection" mechanisms, where mild 'primary' doses of one stress can enhance tolerance to severe doses of a different 'secondary' stress. For example, in many yeast strains, mild osmotic or mild ethanol stresses cross protect against severe oxidative stress, which likely reflects an anticipatory response important for high fitness in nature.

RESULTS

During the course of genetic mapping studies aimed at understanding the mechanisms underlying natural variation in ethanol-induced cross protection against H2O2, we found that a key H2O2 scavenging enzyme, cytosolic catalase T (Ctt1p), was absolutely essential for cross protection in a wild oak strain. This suggested the absence of other compensatory mechanisms for acquiring H2O2 resistance in that strain background under those conditions. In this study, we found surprising heterogeneity across diverse yeast strains in whether CTT1 function was fully necessary for acquired H2O2 resistance. Some strains exhibited partial dispensability of CTT1 when ethanol and/or salt were used as mild stressors, suggesting that compensatory peroxidases may play a role in acquired stress resistance in certain genetic backgrounds. We leveraged global transcriptional responses to ethanol and salt stresses in strains with different levels of CTT1 dispensability, allowing us to identify possible regulators of these alternative peroxidases and acquired stress resistance in general.

CONCLUSIONS

Ultimately, this study highlights how superficially similar traits can have different underlying molecular foundations and provides a framework for understanding the diversity and regulation of stress defense mechanisms.

Pan-transcriptome reveals a large accessory genome contribution to gene expression variation in yeast

Gene expression is an essential step in the translation of genotypes into phenotypes. However, little is known about the transcriptome architecture and the underlying genetic effects at the species level. Here we generated and analyzed the pan-transcriptome of ~1,000 yeast natural isolates across 4,977 core and 1,468 accessory genes. We found that the accessory genome is an underappreciated driver of transcriptome divergence. Global gene expression patterns combined with population structure showed that variation in heritable expression mainly lies within subpopulation-specific signatures, for which accessory genes are overrepresented. Genome-wide association analyses consistently highlighted that accessory genes are associated with proportionally more variants with larger effect sizes, illustrating the critical role of the accessory genome on the transcriptional landscape within and between populations.

Species-wide quantitative transcriptomes and proteomes reveal distinct genetic control of gene expression variation in yeast

Gene expression varies between individuals and corresponds to a key step linking genotypes to phenotypes. However, our knowledge regarding the species-wide genetic control of protein abundance, including its dependency on transcript levels, is very limited. Here, we have determined quantitative proteomes of a large population of 942 diverse natural Saccharomyces cerevisiae yeast isolates. We found that mRNA and protein abundances are weakly correlated at the population gene level. While the protein co-expression network recapitulates major biological functions, differential expression patterns reveal proteomic signatures related to specific populations. Comprehensive genetic association analyses highlight that genetic variants associated with variation in protein (pQTL) and transcript (eQTL) levels poorly overlap (3.6%). Our results demonstrate that transcriptome and proteome are governed by distinct genetic bases, likely explained by protein turnover. It also highlights the importance of integrating these different levels of gene expression to better understand the genotype-phenotype relationship.

Highlights

At the level of individual genes, the abundance of transcripts and proteins is weakly correlated within a species ( ρ = 0.165). While the proteome is not imprinted by population structure, co-expression patterns recapitulate the cellular functional landscapeWild populations exhibit a higher abundance of respiration-related proteins compared to domesticated populationsLoci that influence protein abundance differ from those that impact transcript levels, likely because of protein turnover.

Transcriptional noise adjusted for expression levels reveals genes with high transcriptional noise that are highly expressed, functionally related, and co-regulated in yeast

Understanding the relationship between variability in single-cell and non-single-cell gene expression studies will aid in understanding the role of and mechanisms that lead to variability in biological systems. Studies on the variation of gene expression levels in yeast normally focus on single cells and use the coefficient of variance (CV) as a measure of noise. The CV is typically negatively correlated with gene expression levels, so most of the studies using yeast find that genes with high transcriptional noise are lowly expressed. We find adjusting noise for expression levels using linear/natural log polynomial, and local fits and analyzing many non-single-cell RNA-seq sets identifies genes with high median transcriptional noise that are different than those that have high median CVs. Interestingly, these genes are heavily regulated by transcription factors that are related to variability and stochastic processes based on single-cell studies, including Msn2p, Msn4p, Hsf1p, and Crz1p but are not associated with genes with high median CVs based on non-single-cell gene expression data. In addition, adjusting noise for expression levels in a single-cell RNA-seq data set adds value by finding genes that have noisy gene expression levels and their associated transcription factors that are not found to be associated with genes with high CVs in the single-cell expression data or a comparable non-single-cell gene expression data. Lastly, S. cerevisiae genes with noisy expression tend to have orthologs with noisy gene expression in C. albicans, indicating transcriptional noise is evolutionarily conserved.

Saccharomyces cerevisiae Requires CFF1 To Produce 4-Hydroxy-5-Methylfuran-3(2H)-One, a Mimic of the Bacterial Quorum-Sensing Autoinducer AI-2

Quorum sensing is a process of cell-to-cell communication that bacteria use to orchestrate collective behaviors. Quorum sensing depends on the production, release, and detection of extracellular signal molecules called autoinducers (AIs) that accumulate with increasing cell density. While most AIs are species specific, the AI called AI-2 is produced and detected by diverse bacterial species, and it mediates interspecies communication. We recently reported that mammalian cells produce an AI-2 mimic that can be detected by bacteria through the AI-2 receptor LuxP, potentially expanding the role of the AI-2 system to interdomain communication. Here, we describe a second molecule capable of interdomain signaling through LuxP, 4-hydroxy-5-methylfuran-3(2H)-one (MHF), that is produced by the yeast Saccharomyces cerevisiae Screening the S. cerevisiae deletion collection revealed Cff1p, a protein with no known role, to be required for MHF production. Cff1p is proposed to be an enzyme, with structural similarity to sugar isomerases and epimerases, and substitution at the putative catalytic residue eliminated MHF production in S. cerevisiae Sequence analysis uncovered Cff1p homologs in many species, primarily bacterial and fungal, but also viral, archaeal, and higher eukaryotic. Cff1p homologs from organisms from all domains can complement a cff1Δ S. cerevisiae mutant and restore MHF production. In all cases tested, the identified catalytic residue is conserved and required for MHF to be produced. These findings increase the scope of possibilities for interdomain interactions via AI-2 and AI-2 mimics, highlighting the breadth of molecules and organisms that could participate in quorum sensing.IMPORTANCE Quorum sensing is a cell-to-cell communication process that bacteria use to monitor local population density. Quorum sensing relies on extracellular signal molecules called autoinducers (AIs). One AI called AI-2 is broadly made by bacteria and used for interspecies communication. Here, we describe a eukaryotic AI-2 mimic, 4-hydroxy-5-methylfuran-3(2H)-one, (MHF), that is made by the yeast Saccharomyces cerevisiae, and we identify the Cff1p protein as essential for MHF production. Hundreds of viral, archaeal, bacterial, and eukaryotic organisms possess Cff1p homologs. This finding, combined with our results showing that homologs from all domains can replace S. cerevisiae Cff1p, suggests that like AI-2, MHF is widely produced. Our results expand the breadth of organisms that may participate in quorum-sensing-mediated interactions.

Acquired Resistance to Severe Ethanol Stress in Saccharomyces cerevisiae Protein Quality Control

Acute severe ethanol stress (10% [vol/vol]) damages proteins and causes the intracellular accumulation of insoluble proteins in Saccharomyces cerevisiae On the other hand, a pretreatment with mild stress increases tolerance to subsequent severe stress, which is called acquired stress resistance. It currently remains unclear whether the accumulation of insoluble proteins under severe ethanol stress may be mitigated by increasing protein quality control (PQC) activity in cells pretreated with mild stress. In the present study, we examined the induction of resistance to severe ethanol stress in PQC and confirmed that a pretreatment with 6% (vol/vol) ethanol or mild thermal stress at 37°C significantly reduced insoluble protein levels and the aggregation of Lsg1, which is prone to denaturation and aggregation by stress, in yeast cells under 10% (vol/vol) ethanol stress. The induction of this stress resistance required the new synthesis of proteins; the expression of proteins comprising the bichaperone system (Hsp104, Ssa3, and Fes1), Sis1, and Hsp42 was upregulated during the pretreatment and maintained under subsequent severe ethanol stress. Since the pretreated cells of deficient mutants in the bichaperone system (fes1Δ hsp104Δ and ssa2Δ ssa3Δ ssa4Δ) failed to sufficiently reduce insoluble protein levels and Lsg1 aggregation, the enhanced activity of the bichaperone system appears to be important for the induction of adequate stress resistance. In contrast, the importance of proteasomes and aggregases (Btn2 and Hsp42) in the induction of stress resistance has not been confirmed. These results provide further insights into the PQC activity of yeast cells under severe ethanol stress, including the brewing process.IMPORTANCE Although the budding yeast S. cerevisiae, which is used in the production of alcoholic beverages and bioethanol, is highly tolerant of ethanol, high concentrations of ethanol are also stressful to the yeast and cause various adverse effects, including protein denaturation. A pretreatment with mild stress improves the ethanol tolerance of yeast cells; however, it currently remains unclear whether it increases PQC activity and reduces the levels of denatured proteins. In the present study, we found that a pretreatment with mild ethanol upregulated the expression of proteins involved in PQC and mitigated the accumulation of insoluble proteins, even under severe ethanol stress. These results provide novel insights into ethanol tolerance and the adaptive capacity of yeast. They may also contribute to research on the physiology of yeast cells during the brewing process, in which the concentration of ethanol gradually increases.

Comparison of RNA isolation methods on RNA-Seq: implications for differential expression and meta-analyses

BACKGROUND

The increasing number of transcriptomic datasets has allowed for meta-analyses, which can be valuable due to their increased statistical power. However, meta-analyses can be confounded by so-called "batch effects," where technical variation across different batches of RNA-seq experiments can clearly produce spurious signals of differential expression and reduce our power to detect true differences. While batch effects can sometimes be accounted for, albeit with caveats, a better strategy is to understand their sources to better avoid them. In this study, we examined the effects of RNA isolation method as a possible source of batch effects in RNA-seq design.

RESULTS

Based on the different chemistries of "classic" hot phenol extraction of RNA compared to common commercial RNA isolation kits, we hypothesized that specific mRNAs may be preferentially extracted depending upon method, which could masquerade as differential expression in downstream RNA-seq analyses. We tested this hypothesis using the Saccharomyces cerevisiae heat shock response as a well-validated environmental response. Comparing technical replicates that only differed in RNA isolation method, we found over one thousand transcripts that appeared "differentially" expressed when comparing hot phenol extraction with the two kits. Strikingly, transcripts with higher abundance in the phenol-extracted samples were enriched for membrane proteins, suggesting that indeed the chemistry of hot phenol extraction better solubilizes those species of mRNA.

CONCLUSIONS

Within a self-contained experimental batch (e.g. control versus treatment), the method of RNA isolation had little effect on the ability to identify differentially expressed transcripts. However, we suggest that researchers performing meta-analyses across different experimental batches strongly consider the RNA isolation methods for each experiment.

Independent Mechanisms for Acquired Salt Tolerance versus Growth Resumption Induced by Mild Ethanol Pretreatment in Saccharomyces cerevisiae

Microbes in nature frequently experience “boom or bust” cycles of environmental stress. Thus, microbes that can anticipate the onset of stress would have an advantage. One way that microbes anticipate future stress is through acquired stress resistance, where cells exposed to a mild dose of one stress gain the ability to survive an otherwise lethal dose of a subsequent stress. In the budding yeast Saccharomyces cerevisiae, certain stressors can cross protect against high salt concentrations, though the mechanisms governing this acquired stress resistance are not well understood. In this study, we took advantage of wild yeast strains to understand the mechanism underlying ethanol-induced cross protection against high salt concentrations. We found that mild ethanol stress allows cells to resume growth on high salt, which involves a novel role for a well-studied salt transporter. Overall, this discovery highlights how leveraging natural variation can provide new insights into well-studied stress defense mechanisms.

All living organisms must recognize and respond to various environmental stresses throughout their lifetime. In natural environments, cells frequently encounter fluctuating concentrations of different stressors that can occur in combination or sequentially. Thus, the ability to anticipate an impending stress is likely ecologically relevant. One possible mechanism for anticipating future stress is acquired stress resistance, where cells preexposed to a mild sublethal dose of stress gain the ability to survive an otherwise lethal dose of stress. We have been leveraging wild strains of Saccharomyces cerevisiae to investigate natural variation in the yeast ethanol stress response and its role in acquired stress resistance. Here, we report that a wild vineyard isolate possesses ethanol-induced cross protection against severe concentrations of salt. Because this phenotype correlates with ethanol-dependent induction of the ENA genes, which encode sodium efflux pumps already associated with salt resistance, we hypothesized that variation in ENA expression was responsible for differences in acquired salt tolerance across strains. Surprisingly, we found that the ENA genes were completely dispensable for ethanol-induced survival of high salt concentrations in the wild vineyard strain. Instead, the ENA genes were necessary for the ability to resume growth on high concentrations of salt following a mild ethanol pretreatment. Surprisingly, this growth acclimation phenotype was also shared by the lab yeast strain despite lack of ENA induction under this condition. This study underscores that cross protection can affect both viability and growth through distinct mechanisms, both of which likely confer fitness effects that are ecologically relevant.

IMPORTANCE Microbes in nature frequently experience “boom or bust” cycles of environmental stress. Thus, microbes that can anticipate the onset of stress would have an advantage. One way that microbes anticipate future stress is through acquired stress resistance, where cells exposed to a mild dose of one stress gain the ability to survive an otherwise lethal dose of a subsequent stress. In the budding yeast Saccharomyces cerevisiae, certain stressors can cross protect against high salt concentrations, though the mechanisms governing this acquired stress resistance are not well understood. In this study, we took advantage of wild yeast strains to understand the mechanism underlying ethanol-induced cross protection against high salt concentrations. We found that mild ethanol stress allows cells to resume growth on high salt, which involves a novel role for a well-studied salt transporter. Overall, this discovery highlights how leveraging natural variation can provide new insights into well-studied stress defense mechanisms.

Nutritional and meiotic induction of transiently heritable stress resistant states in budding yeast

Transient exposures to environmental stresses induce altered physiological states in exposed cells that persist after the stresses have been removed. These states, referred to as cellular memory, can even be passed on to daughter cells and may thus be thought of as embodying a form of epigenetic inheritance. We find that meiotically produced spores in the budding yeast S. cerevisiae possess a state of heightened stress resistance that, following their germination, persists for numerous mitotic generations. As yeast meiotic development is essentially a starvation response that a/alpha diploid cells engage, we sought to model this phenomenon by subjecting haploid cells to starvation conditions. We find also that haploid cells exposed to glucose withdrawal acquire a state of elevated stress resistance that persists after the reintroduction of these cells to glucose-replete media. Following release from lengthy durations of glucose starvation, we confirm that this physiological state of enhanced stress resistance is propagated in descendants of the exposed cells through two mitotic divisions before fading from the population. In both haploid starved cells and diploid produced meiotic spores we show that their cellular memories are not attributable to trehalose, a widely regarded stress protectant that accumulates in these cell types. Moreover, the transiently heritable stress resistant state induced by glucose starvation in haploid cells is independent of the Msn2/4 transcription factors, which are known to program cellular memory induced by exposure of cells to NaCl. Our findings identify new developmentally and nutritionally induced states of cellular memory that exhibit striking degrees of persistence and mitotic heritability.