Advertisement

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Active compensation for changes in TDH3 expression mediated by direct regulators of TDH3 in Saccharomyces cerevisiae

  • Pétra Vande Zande,

    Roles Conceptualization, Data curation, Formal analysis, Software, Validation, Writing – original draft, Writing – review & editing

    Current address: Department of Microbiology and Immunology, University of Minnesota, Minneapolis, Minnesota, United States of America

    Affiliation Department of Molecular, Cellular, and Developmental Biology, University of Michigan, Ann Arbor, Michigan, United States of America

    ⨯
  • Mohammad A. Siddiq,

    Roles Formal analysis, Investigation, Validation, Visualization, Writing – review & editing

    Affiliations Department of Molecular, Cellular, and Developmental Biology, University of Michigan, Ann Arbor, Michigan, United States of America, Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, Michigan, United States of America

    ⨯
  • Andrea Hodgins-Davis,

    Roles Investigation, Methodology, Resources, Writing – review & editing

    Affiliation Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, Michigan, United States of America

    ⨯
  • Lisa Kim,

    Roles Resources, Writing – review & editing

    Affiliation Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, Michigan, United States of America

    ⨯
  • Patricia J. Wittkopp

    Roles Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing

    wittkopp@umich.edu

    Affiliations Department of Molecular, Cellular, and Developmental Biology, University of Michigan, Ann Arbor, Michigan, United States of America, Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, Michigan, United States of America

    ⨯

Abstract

Genetic networks are surprisingly robust to perturbations caused by new mutations. This robustness is conferred in part by compensation for loss of a gene’s activity by genes with overlapping functions, such as paralogs. Compensation occurs passively when the normal activity of one paralog can compensate for the loss of the other, or actively when a change in one paralog’s expression, localization, or activity is required to compensate for loss of the other. The mechanisms of active compensation remain poorly understood in most cases. Here we investigate active compensation for the loss or reduction in expression of the Saccharomyces cerevisiae gene TDH3 by its paralog TDH2. TDH2 is upregulated in a dose-dependent manner in response to reductions in TDH3 by a mechanism requiring the shared transcriptional regulators Gcr1p and Rap1p. TDH1, a second and more distantly related paralog of TDH3, has diverged in its regulation and is upregulated by another mechanism. Other glycolytic genes regulated by Rap1p and Gcr1p show changes in expression similar to TDH2, suggesting that the active compensation by TDH3 paralogs is part of a broader homeostatic response mediated by shared transcriptional regulators.

Author summary

Living things have a remarkable ability to modify their molecular and cellular processes to reduce the negative impacts of genetic and environmental perturbations. Here, we examine how cells of baker’s yeast respond to mutations changing the expression of a key metabolic gene, which affects cell division rates (i.e., fitness). We find that the expression of other metabolic genes, including a gene with ancestrally similar sequence, function, and regulation (a paralog), changes its expression in ways complementary to the mutated gene, reducing the impact of the mutation on fitness. This type of compensatory expression change has also been reported for other genes, but the molecular mechanisms behind such changes are generally unknown. In this case, we find that upregulation of the paralog is mediated by transcription factors that also directly regulate the mutated gene. This observation suggests that ancestrally shared regulation of paralogous genes can play a key role in homeostasis and might help maintain genes in the genome following gene duplication.

Citation: Vande Zande P, Siddiq MA, Hodgins-Davis A, Kim L, Wittkopp PJ (2023) Active compensation for changes in TDH3 expression mediated by direct regulators of TDH3 in Saccharomyces cerevisiae. PLoS Genet 19(12): e1011078. https://doi.org/10.1371/journal.pgen.1011078

Editor: Kelly A. Dyer, University of Georgia, UNITED STATES

Received: November 10, 2023; Accepted: November 22, 2023; Published: December 13, 2023

Copyright: © 2023 Vande Zande et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: RNA-sequencing data is available at GEO accession GSE175398. Code used for analysis and to generate figures in this manuscript is available at Github and in a permanent zenodo release (URL: https://github.com/pvz22/Compensation_TDH3, Zenodo DOI: 10.5281/zenodo.10223579). All flow cytometry data is deposited at flowrepository.org and publicly accessible at http://flowrepository.org/id/FR-FCM-Z72G.

Funding: This work was supported by the National Institutes of Health (T32GM07544 to P.V.Z; F32CA261115 to M.A.S; and R35GM118073 and R01GM108826 to P.J.W.) and the National Science Foundation (MCB-1929737 to P.J.W.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Biological systems are often robust to genetic and environmental perturbations [1,2]. This robustness is conferred in part by the presence of multiple genes in the genome with overlapping functions [2–4]. Such genes often arise through duplication events that give rise to two or more paralogous genes [5,6]. As described in Diss et al. [7], paralogous genes can contribute to phenotypic robustness through either passive or active mechanisms. In passive paralogous compensation, the normal activity of one of the paralogs is sufficient to minimize the phenotypic impact of losing the activity of the other paralog. By contrast, active paralogous compensation occurs when the activity of one paralog changes in response to loss of activity of the other paralog, reducing the phenotypic impact of this loss. For example, a gene may respond to loss of a paralogous gene’s function by increasing its expression level, producing more protein capable of performing the function of the mutated gene.

Multiple examples of active compensation by upregulation of a paralog have been identified [8–13], but the molecular mechanisms responsible for such transcriptional compensation remain largely unknown. One notable exception is loss of the CLV1 receptor kinase in Arabidopsis thaliana, which is compensated for by the upregulation of paralogous receptor kinases BAM1, BAM2, and BAM3. Under normal circumstances the BAM genes are negatively regulated by CLV1, and loss of CLV1 removes this transcriptional repression, resulting in upregulation of the BAM genes that compensates for the loss of CLV1 [14]. This exact mechanism of active compensation for loss of CLV1 is not conserved in tomato or maize, but other steps in the CLV signaling pathway show evidence of active or passive compensation by paralogs within these species [11]. For example, in tomato, upregulation of SlCLE9, the closest paralog to SlCLV3, in response to loss of SlCLV3 reduces the phenotypic impact of the SlCLV3 mutation, although the mechanism causing this upregulation is unclear [11].

Large-scale synthetic genetic interaction studies in the baker’s yeast Saccharomyces cerevisiae have also shown that paralogs with overlapping function are frequently able to compensate for each other [15,16]. Up-regulation of paralogous genes with overlapping functions when one paralog is deleted has been reported in S. cerevisiae, and paralogs with partially overlapping regulatory motifs are more likely to be dispensable than those without overlap suggesting compensation for their loss [17]. A proposed model posits that paralogous enzymes that catalyze the same metabolic step and are regulated by the same transcription factors may act via active compensation. In this framework, accumulation of a metabolic substrate due to reduction in activity of one paralog would trigger feedback mechanisms that increase the activity of shared transcriptional regulators, which in turn cause upregulation of the other paralog, and thus active compensation [17]. There are many examples of feedback circuits from yeast to mammals with the potential to function this way, making the model potentially of wide relevance to many biological systems [18]. To the best of our knowledge, however, the proposed dependency on a shared regulator for active compensation by upregulation of paralogous genes has yet to be demonstrated empirically.

The Saccharomyces cerevisiae TDH1, TDH2, and TDH3 genes are paralogs with overlapping protein function and partially overlapping regulation that might make them likely to show active compensation. All three of these proteins act as glyceraldehyde-3-phosphate dehydrogenases (GAPDHs) [19,20], catalyzing a central step in both glycolysis and gluconeogenesis. The TDH2 and TDH3 proteins are most similar to each other, retaining 94% amino acid sequence identity [21,22], whereas the TDH1 and TDH3 proteins have 89% amino acid sequence identity [22,23]. TDH2 and TDH3 are expressed during exponential growth, with TDH3 expressed at a much higher level, while TDH1 is expressed primarily during stationary phase [24,25]. The divergence in expression patterns and levels, as well as differences in the sensitivity of TDH1 to trans-regulatory mutations [26], indicates divergence in underlying regulatory control of the paralogs, particularly for TDH1. Deletion of TDH3 reduces fitness to ~93–98% of wild type [27,28] whereas deletion of TDH1 or TDH2 alone has little to no effect [28]. Deletion of TDH1 and TDH3 together does not exacerbate the fitness defect of deletion of TDH3 alone; however, deletion of TDH2 and TDH3 together shows a strong negative interaction, with growth at only 20% of wild type levels [28]. This nonadditive impact on fitness suggests that the functional overlap of TDH2 and TDH3 allows TDH2 to help compensate for loss of TDH3.

Here, we investigate the molecular mechanisms responsible for this compensation. We find that expression of TDH2 is upregulated when TDH3 expression is reduced, and downregulated when TDH3 expression is increased, suggesting that TDH2 provides active compensation for changes in TDH3 expression. We show that this active compensation requires functional transcription factor proteins Rap1p and Gcr1p, which directly regulate TDH3, and requires binding sites for Gcr1p. We also observe upregulation of TDH1 when TDH3 expression is reduced, but this upregulation seems to be independent of Gcr1p, suggesting that there are differences in the molecular mechanisms causing upregulation of the two paralogs. This involvement of Rap1p and/or Gcr1p in the upregulation of TDH2 provides empirical support for the model proposed by Kafri et al. [17] in which active compensation by paralogous genes is facilitated by one or more shared regulators and feedback loops. But this upregulation is not limited to paralogous genes; we also see upregulation of other genes regulated by Gcr1p and Rap1p that encode proteins that function in the same metabolic pathway. These results suggest that active compensation for changes in TDH3 expression via upregulation of paralogs is not a specific regulatory program, but rather part of a general activation of the glycolytic regulon. Consequently, this study shows how shared regulators controlling expression of paralogs with overlapping function can provide mutational robustness through active compensation that contributes to homeostasis.

Results

Active compensation for loss of TDH3 by TDH2

To determine whether the compensation for loss of TDH3 activity by TDH2 might be mediated by changes in their expression, we examined TDH2 expression in a TDH3 deletion strain of S. cerevisiae (tdh3Δ) previously analyzed using RNA-seq [29]. We found that TDH2 showed significantly higher expression in the tdh3Δ strain than in the unmutated wild-type strain (Fig 1A, Wald test P-value for TDH2 = 0.04). To determine whether this upregulation also affects protein abundance, we engineered strains in which a cyan fluorescent protein (CFP) was fused to the native TDH2 protein and assayed expression level of this fusion protein using flow cytometry. We found that fluorescence increased upon deletion of TDH3, indicating an increase in Tdh2p (Fig 1B, one sided t-test p-value = 2.23x10-7). To determine whether the degree of upregulation correlates with the extent to which TDH3 expression is altered, we used additional RNA-seq data from Vande Zande et al [29] to examine TDH2 expression in strains of S. cerevisiae carrying changes in the TDH3 promoter that cause more moderate alterations in TDH3 expression. Three of these strains carry a single point mutation in the TDH3 promoter that drives either 20%, 50%, or 85% of wild-type TDH3 expression [29]. A fourth strain carries a duplication of the TDH3 gene with each copy carrying a single promoter mutation reducing expression levels from each promoter, resulting in a strain expressing TDH3 at 135% of wild-type levels. We found that TDH2 expression was negatively correlated with TDH3 expression among these strains, with TDH2 showing both increased expression when TDH3 expression was decreased and decreased expression when TDH3 expression was increased (Fig 1C). We also examined the expression levels of the TDH3 paralog TDH1 and found that TDH1 is also upregulated in the tdh3Δ deletion mutant (Fig 1A, Wald test P-value for TDH1 = 2x10-5). Unlike TDH2, however, TDH1 showed more of a threshold-like relationship with TDH3 expression: TDH1 expression was strongly increased in the TDH3 null strain, but only mildly (and similarly) increased in the mutant strains expressing TDH3 at 20%, 50%, and 85% of wild-type levels (Fig 1D). Like TDH2, TDH1 expression decreased in the strain overexpressing TDH3 (Fig 1D). Differences in the expression changes observed for TDH1 and TDH2 in these TDH3 mutants are consistent with divergence in the regulation of TDH1 and TDH2 (S1 Fig). Taken together, these data provide evidence of active compensation when TDH3 expression is altered, with expression of its paralog TDH2 changing in ways expected to minimize the impacts of these TDH3 mutations on fitness.

thumbnail
Download:
Fig 1. TDH1 and TDH2 are upregulated in response to reductions in TDH3 expression.

(A) Changes in expression of TDH1, TDH2, and TDH3 in response to the deletion of TDH3 are shown, measured as fold change in expression relative to a wild type. Error bars represent one standard error of the mean. Statistical significance of expression changes was assessed using Wald tests in DESeq2, with the P-value for TDH1 = 2x10-5, TDH2 = 0.04, and TDH3 = 7x10-107. (B) Fluorescence normalized to cell size (arbitrary units) is shown for a strain bearing a PTDH2:CFP-TDH2 fusion protein with TDH3 intact (TDH3) or deleted (tdh3Δ). Each violin plot represents data from four biological replicates, each containing 15,000 singlet cells measured by flow cytometry. White points indicate medians from each of the four replicates. Expression in the tdh3Δ strain is significantly higher than the TDH3 strain (one sided t-test p-value = 2.23x10-7) (C) Changes in expression of TDH3 and TDH2 are shown for strains with cis-acting mutations causing 0%, 20%, 50%, 85%, and 135% of wild type TDH3 expression. Error bars show one standard error of the mean. (D) Changes in expression of TDH3 and TDH1 are shown for strains with cis-acting mutations causing 0%, 20%, 50%, 85%, and 135% of wild type TDH3 expression. Error bars show one standard error of the mean. RNA-sequencing data in panels A, C, and D from [29]. As described previously, data from each strain is composed of 2 (TDH3 deletion) or 4 (others) biological replicates.

https://doi.org/10.1371/journal.pgen.1011078.g001

Active compensation might be caused by direct regulators of TDH3

For historical reasons [27], the control strain and TDH3 mutant strains profiled for expression using RNA-seq in Vande Zande et al [29] all carried a reporter gene composed of the wild-type TDH3 promoter allele driving expression of a yellow fluorescent protein (PTDH3-YFP). Surprisingly, we found that expression of this reporter gene was increased when native TDH3 expression was decreased by mutations in its promoter and not by the duplication of TDH3 with promoter mutations causing over-expression of TDH3 (Fig 2A). This negative correlation between expression of the native TDH3 gene harboring cis-acting mutations and expression driven by a wild-type allele of the TDH3 promoter suggests that factors regulating expression of TDH3 itself might be involved in the mechanism of active compensation.

thumbnail
Download:
Fig 2. Feedback regulating TDH3 expression is mediated by Gcr1p TFBSs.

(A) Changes in expression of TDH3 and a reporter gene with a wild type TDH3 promoter driving expression of YFP (PTDH3-YFP) are shown for strains with cis-acting mutations causing 0%, 20%, 50%, 85%, and 135% of wild type TDH3 expression. Error bars show one standard error of the mean. (B) Schematics and sequences of the TDH3 promoter in mutant strains bearing mutations in binding sites for Rap1p and Gcr1p at a distance of 510, 485, and 482 nucleotides upstream of the TDH3 ATG that result in TDH3 expression levels of 20%, 50%, and 85% relative to wild type are shown. No schematic is shown for the mutant strain expressing TDH3 expression at 135% of wild type levels; it contains two copies of the TDH3 gene separated by a copy of the URA3 gene, with both copies of TDH3 containing a mutation in the binding site for Rap1p at a distance of 505 nucleotides upstream of the ATG (GGTGTCTGaGT). (C) Changes in expression of RAP1, GCR1, and TDH3 are shown for strains with cis-acting mutations in the TDH3 promoter causing 0%, 20%, 50%, 85%, and 135% of wild type TDH3 expression, measured as fold change in expression relative to a wild type. Error bars represent one standard error of the mean. RNA-sequencing data in panels A and C are from [29]. (D) Fluorescence normalized to cell size (arbitrary units) is shown for a strains bearing a PTDH3:YFP construct with point mutations in transcription factor binding sites of either Gcr1p or Rap1p, each with TDH3 intact (TDH3) or deleted (tdh3Δ). These include two of the three mutant promoter alleles shown in (B), one mutant allele at the 505 position in the Rap1p TFBS resulting in ~70% expression, plus a mutant allele containing a mutation 444 nucleotides upstream of the TDH3 start codon, which affects the GCR1 binding site closer to the TATA box shown in (B). Each violin plot consists of four biological replicates each containing 15,000 singlet cells measured by flow cytometry. Points indicate medians of each of the four replicates. Asterisks indicate significantly higher expression in the tdh3dΔ strain as compared to the matched TDH3 strain (one-sided t-test for median fluorescence p-values from left to right are: 8.7x10-6, 0.78, 1.3x10-4, 6.7x10-4, 0.014).

https://doi.org/10.1371/journal.pgen.1011078.g002

The transcription factors Rap1p and Gcr1p regulate expression of TDH3 [30,31] as well as expression of other glycolytic genes, including TDH1 and TDH2 [32–35]. In fact, the mutations altering expression of TDH3 in the mutant strains expressing TDH3 at 20%, 50%, and 85% of wild-type expression levels all altered either Rap1p or Gcr1p binding sites in the TDH3 promoter (Fig 2B, [27,29]). We thus wondered whether transcription of RAP1 and/or GCR1 was changed in the strains with TDH3 promoter mutations. Using the same RNA-seq dataset described above, we found that GCR1 was upregulated linearly in response to reductions in TDH3 expression caused by mutations in the TDH3 promoter whereas expression of RAP1 was not (Fig 2C). If anything, expression of RAP1 was slightly and similarly reduced in all mutants with reduced TDH3 expression (Fig 2C).

We next tested whether the Rap1p and Gcr1p transcription factor binding sites (TFBS) were necessary for the upregulation of the reporter driven by the TDH3 promoter upon reduction in TDH3 expression. We engineered strains with mutations in either Rap1p or Gcr1p TFBSs in the TDH3 promoter driving YFP expression (S1 Table, S1 File). We then deleted the native TDH3 locus in these backgrounds and found that fluorescence did not significantly increase upon reduction in TDH3 in one of the Gcr1p TFBS mutants (Fig 2D, one-sided t-test for median fluorescence p-values from left to right are: 8.7x10-6, 0.78, 1.3x10-4, 6.7x10-4, 0.014), showing that the specific nucleotides in this Gcr1p binding site are necessary for compensatory upregulation via the TDH3 promoter.

Mutations in Rap1p and Gcr1p disrupt compensatory expression changes of TDH2

If Rap1p and/or Gcr1p are involved in the upregulation of TDH2 upon reduction of TDH3 expression, we expect that strains with mutations in Rap1p or Gcr1p causing a reduction in TDH3 expression would not show the same compensatory upregulation of TDH2 seen in strains with wildtype Rap1 and Gcr1 proteins. That is, if the upregulation of TDH3 paralogs requires Rap1p or Gcr1p, then mutations in these proteins that disrupt their ability to drive TDH3 expression at wild-type levels should also impair their ability to upregulate expression of other genes in response to reduced TDH3. To test this hypothesis, we examined RNA-seq data from 9 mutant strains of S. cerevisiae each carrying 1–6 mutations in the RAP1 (4 mutants) or GCR1 (5 mutants) gene previously shown to affect TDH3 expression [36]. These data were collected in parallel with the expression data for the TDH3 mutants [29]. One GCR1 mutant strain (GCR1.162) carried a single nucleotide deletion resulting in an early stop codon, suggesting it was likely to be a null mutation. This mutant expressed TDH3 at only 7% of wild-type expression levels (Fig 3A). The other GCR1 mutant alleles were more likely to be hypo- (GCR1.339, GCR1.281, GCR1.37) or hypermorphs (GCR1.241), causing TDH3 expression to range from ~22% to ~105% of wild type levels (Fig 3A). RAP1 null mutants are lethal [37], suggesting that all of the RAP1 mutants examined were either hypo- or hypermorphs. These RAP1 mutants showed TDH3 expression ranging from ~20% to ~115% of wild-type levels (Fig 3A).

thumbnail
Download:
Fig 3. Active compensation by TDH2 is mediated by RAP1 and/or GCR1.

(A) Changes in expression of TDH3 in response to various mutations in either GCR1 (dark grey) or RAP1 (light grey), measured as log2 fold change in expression relative to a wild type. Specific mutation identities in each strain are described in S1 Table. Error bars represent one standard error of the mean. (B, C) Fold changes in expression of TDH3 and TDH2 (B) or a reporter gene with a wild type TDH3 promoter driving expression of YFP (PTDH3-YFP) (C), are shown for strains with mutations in either the RAP1 (squares) or GCR1 (empty triangles) coding sequences. Error bars show one standard error of the mean. RNA-sequencing data are from [29]. As described previously, RNA-sequencing data for each strain consists of 4 biological replicates. (D) Schematics of the TDH2 promoter driving YFP, indicating Gcr1p TFBSs that are either deleted or mutated. (E) Fluorescence normalized to cell size (arbitrary units) is shown for strains bearing a PTDH2-YFP construct that is wild type, with Gcr1p TFBSs mutated, or with Gcr1p TFBSs deleted (as shown in panel D), each with TDH3 intact (TDH3) or deleted (tdh3Δ). Each violin plot consists of four biological replicates each containing 15,000 singlet cells measured by flow cytometry. Points indicate medians of each of the four replicates. (One sided t-test for median fluorescence p-values from left to right are: 5.17x10-6, 0.18, 0.027) (F) As in panels B and C, fold changes in expression of TDH1 is shown for strains with mutations in either RAP1 or GCR1 coding sequences.

https://doi.org/10.1371/journal.pgen.1011078.g003

Consistent with Rap1p and Gcr1p mediating compensatory changes in paralog gene expression, we found that the TDH2 gene was not upregulated in either the Rap1p or Gcr1p mutants that decreased TDH3 expression (Fig 3B). TDH2 expression was also not reduced in mutants causing overexpression of TDH3 (Fig 3B). These observations suggest that both Rap1p and Gcr1p are required for the compensatory changes in TDH2 expression seen in strains carrying mutations in the TDH3 promoter. Changes in expression of the PTDH3-YFP reporter gene seen in the TDH3 mutants (Fig 2A) were also absent in the RAP1 and GCR1 mutants altering TDH3 expression (Fig 3C), again implying that Gcr1p and Rap1p were required for these changes.

We next tested whether regulation by Gcr1p was required for the compensatory upregulation of TDH2 by engineering strains with variations of the TDH2 promoter driving YFP expression, including an unmutated TDH2 promoter, a TDH2 promoter in which each putative Gcr1p TFBS was completely deleted, and a TDH2 promoter bearing single point mutations in each Gcr1p TFBS expected to disrupt Gcr1p binding (Fig 3D). The wild type TDH2 promoter drove higher expression upon deletion of the native TDH3 locus, as expected from our RNA-seq data (Fig 3E, one sided t-test for median fluorescence p-value = 5.17x10-6). Abolishing all Gcr1p TFBS greatly reduced pTDH2 activity, while point mutations in all Gcr1p TFBS reduced activity to a lesser extent (Fig 3E). In both cases, deletion of TDH3 failed to induce comparable increases in expression of the TDH2 promoter (Fig 3E, one sided t-test for median fluorescence p-values = 0.18, 0.027). Therefore, we conclude that the compensatory increase in TDH2 promoter activity upon deletion of TDH3 is dependent upon intact Gcr1p TFBS.

Expression of TDH1, on the other hand, showed increases in expression in GCR1 mutants with lowered TDH3 expression (Fig 2F), suggesting that Gcr1p is not required for the upregulation of TDH1 in response to reduced expression of TDH3. Rap1p might be required for this upregulation, however, because neither of the RAP1 mutants decreasing TDH3 expression showed an upregulation of TDH1 (Fig 3F). These data support a model in which Gcr1p is involved in the active compensation for changes in TDH3 expression via TDH2, but not TDH1, with Rap1p involved in the changes in expression of both genes.

Expression changes are also seen for other, non-paralogous, metabolic genes

Rap1p and Gcr1p are transcription factors that regulate expression of many metabolic genes [38,39], thus active compensation for altered TDH3 expression mediated by Rap1p and Gcr1p might affect more than just genes paralogous to TDH3. Indeed, the eight genes encoding enzymes that function in the glycolytic pathway at steps immediately surrounding the step controlled by the TDH proteins have all been annotated as targets of Gcr1p and Rap1p based on either gene expression and/or chromatin immunoprecipitation experiments [33–35]. We therefore examined the expression of these genes (Fig 4A) in the RNA-seq data from TDH3, RAP1, and GCR1 mutants described above. We found that each of these genes was upregulated in the tdh3Δ null mutant and their expression levels showed an inverse relationship with TDH3 expression in the other TDH3 promoter mutants (Fig 4B–4F, circles), although the magnitude of increased expression was variable between genes. The detected upregulation of other glycolytic genes was likely mediated by Rap1p and Gcr1p because none of the aforementioned genes were upregulated in strains bearing mutations in RAP1 or GCR1 in response to reduced levels of TDH3 (Fig 4B–4F, squares and triangles). The similarity of expression patterns between other glycolytic enzymes and TDH2 suggests that the compensatory upregulation of TDH2 is part of the larger homeostatic network regulating expression of genes in the glycolytic pathway and is not specific to compensation by paralogs. Activation of the entire pathway is consistent with active compensation for changes in TDH3 expression being mediated through homeostatic feedback mechanisms involving Gcr1p and Rap1p in place for the regulation of glycolysis.

thumbnail
Download:
Fig 4. Multiple enzymes in the glycolysis pathway are upregulated upon reduction in TDH3 expression in a RAP1/GCR1 dependent manner.

(A) A simple schematic of the glycolytic pathway surrounding the metabolic step catalyzed by TDH1, TDH2, and TDH3, showing other enzymes catalyzing adjacent reactions. Enzymes that are significantly upregulated upon reduction in TDH3 are in blue. Enzymes in this pathway that were not statistically significantly upregulated are shown in grey. Differences in the variance among replicates for PFK1 and PFK2 resulted in PFK2 but not PFK1 being statistically significantly upregulated even though the two genes showed similar magnitudes of upregulation. (B-F) Expression fold changes relative to wild type of the genes PFK1 and PFK2 (B), FBA1 and TPI1 (C), PGK1 (D), GPM1 (E), and ENO1 and ENO2 (F) in yeast strains with varying levels of TDH3 expression due to mutations in the native TDH3 promoter (circles connected by lines), and in the 9 yeast strains with varying levels of TDH3 expression due to mutations in the genes encoding RAP1 (solid boxes) or GCR1 (empty triangles) as estimated by RNA-sequencing data from Vande Zande et al [29]. Error bars are one standard error of the mean.

https://doi.org/10.1371/journal.pgen.1011078.g004

Discussion

Many genes with overlapping functions can compensate for each other’s loss, contributing to the genetic robustness of biological systems, but the mechanisms by which this compensation arises, operates, and is maintained over evolutionary time continues to be unclear [40–42]. In this study, we show that changes in TDH3 expression trigger feedback mechanisms that depend on the activity of transcription factors Rap1p and Gcr1p to offset the effects of these changes. Strains bearing cis-regulatory mutations in the TDH3 promoter that decrease its expression presumably fail to upregulate TDH3 because the transcription factor binding sites for Rap1p or Gcr1p are disrupted in these alleles (or because the locus is absent in the null mutant), yet expression of other genes regulated by Gcr1p and Rap1p is increased, including the TDH3 paralogs TDH2 and TDH1 and even a reporter gene driven by a wild-type TDH3 promoter. In other words, reduction in TDH3 expression results in active compensation by upregulation of its paralog TDH2 (Fig 5).

thumbnail
Download:
Fig 5. Model for active compensation by feedback and shared regulation.

(A) In a wild-type cell, the Gcr1p and Rap1p complex regulate expression levels of TDH2 and TDH3, and Rap1p likely regulates expression of TDH1. (B) When the native promoter of TDH3 is mutated, TDH3 levels decrease, leading to an upregulation of TDH2 and a PTDH3-YFP reporter gene via Gcr1p and Rap1p, and TDH1 via Rap1p. (C) When Gcr1p is mutated, levels of all its direct targets are reduced. Lower levels of TDH3 lead to an upregulation of TDH1, possibly via Rap1p. (D) When Rap1p is mutated, levels of all its direct targets are reduced. Despite lower levels of TDH3 expression, the paralogs are not upregulated due to lack of functional Rap1p.

https://doi.org/10.1371/journal.pgen.1011078.g005

The upregulation of TDH2 by Gcr1p/Rap1p might be achieved by increased expression of the GCR1 gene in response to reduced TDH3 expression. Transcriptional upregulation is not the only mechanism of activation of transcription factors [43], but GCR1 has been shown to be both transcriptionally and post-transcriptionally regulated by glucose availability [44] and we observed increased GCR1 expression in mutants with decreased TDH3 expression, demonstrating that this transcription factor is transcriptionally regulated under some circumstances. RAP1, on the other hand, performs roles in telomere maintenance and activation of ribosomal protein genes in addition to the activation of glycolytic genes [45,46], and is not known to be transcriptionally responsive to metabolic changes. Because Rap1p and Gcr1p act in a complex to activate target gene expression, with Gcr1p being the major activator of the complex [38], we propose that upregulation of GCR1 transcription upon reduction in TDH3 expression is primarily responsible for the upregulation of the Rap1p/Gcr1p complex’s target genes, while still being dependent on functional Rap1p for upregulation of its target genes.

Upregulation of TDH1 appears to occur via a different mechanism, as indicated by its more threshold-like response to reduction in TDH3 expression and its upregulation in strains bearing mutations in GCR1. These differences in how TDH1 and TDH2 respond to reduction in TDH3 expression may not be surprising given that the expression pattern of TDH1 has diverged from that of the other two paralogs (S1 Fig, [19]). TDH1 has been shown to be upregulated under various stress conditions causing slow growth [20], and might therefore be upregulated by a mechanism related to the slower growth of mutants with reduced TDH3 expression level rather than feedback specifically involving Gcr1p, although it does appear to be at least somewhat dependent on Rap1p function (Fig 3F).

The fact that the upregulation of TDH2 does not completely eliminate the fitness effect of deleting TDH3 suggests that either the upregulation of TDH2 does not produce as much GAPDH activity as the normal expression of the TDH3 gene (which is consistent with the data shown in Figs 1C and S1) or that there are pleiotropic effects of the compensation mechanism itself, such as the upregulation of other glycolytic enzymes, that causes a fitness cost [47]. Alternatively, the upregulation of other enzymes in the pathway could be beneficial, and any remaining fitness costs due to some divergence in the functions of the TDH2 and TDH3 proteins such that they cannot completely compensate for each other. Although TDH3 is best known for its roles in glycolysis and gluconeogenesis, it has also been implicated in transcriptional silencing [48], RNA-binding [49] and antimicrobial defense [50], functions which may not be able to be compensated for by TDH2 despite their high levels of protein conservation. More work assessing these ‘non-canonical’ or ‘moonlighting’ [51–53] functions of the GAPDHs in S. cerevisiae is needed to better understand their relative roles in the cell.

The redundancy of paralogous genes both imparts robustness to biological systems and simultaneously makes them evolutionarily unstable given that mutations in one gene are masked by the presence of the other gene. Yet, paralogous genes with overlapping functions are sometimes maintained over long evolutionary timescales [4,15,16,18,54–58]. Divergence in gene regulation and/or protein function might contribute to the maintenance of all three TDH paralogs over evolutionary time; however, in general, it remains to be seen how often the ability of paralogs to actively compensate for each other and contribute to genetic robustness is actively selected for or simply a side effect of their ancestrally shared regulators with sensitivity to feedback mechanisms. Decoding the molecular mechanisms responsible for active compensation among paralogous genes in other systems will help address this issue, revealing how living systems can thrive despite the inevitable changes in the environment and their genotype.

Materials and methods

Strains used in this study

The S. cerevisiae strains used in this study are haploid strains derived from S288C and include the 5 cis-regulatory mutants affecting expression of TDH3 containing changes in the S. cerevisiae TDH3 promoter and the 9 trans-regulatory mutants affecting expression of TDH3 that each carry 1–6 mutations in either the RAP1 or GCR1 gene described in Vande Zande et al. [29]. Construction of the cis-regulatory mutant strains, including the tdh3Δ strain, is described inx2 [27], and construction of the strains bearing mutations in the RAP1 or GCR1 genes is described in [36].

Reporter strains were constructed by fusing either the TDH1, TDH2, or TDH3 promoters to VenusYFP and engineering this construct into the HO locus of a haploid S288C strain, as described in [59]. Variants of the TDH3 and TDH2 promoters were constructed similarly: promoter alleles with desired mutations in the Gcr1p or Rap1p binding sites were cloned upstream of the YFP coding sequence in a plasmid with homology arms to the HO locus using Gibson assembly. The constructs were subsequently amplified using PCR and engineered into the HO locus using CRISPR-Cas9, following methods described in Laughery et al. [60]. Proper integration into the HO locus was verified using Sanger Sequencing. The CFP-TDH2 fusion strain was constructed using PCR SOE (Splicing by Overlap Extension) to generate a DNA fragment with CFP fused to the 5’ end of TDH2. CRISPR-Cas9 was used to introduce a double-stranded break at the 5’ end of the native TDH2 locus [60]. Insertion of the CFP-TDH2 DNA fragment containing homology arms to the TDH2 promoter and gene interrupted the gRNA PAM recognition site. Then, correct insertion was confirmed by Sanger sequencing of two amplicons spanning the edited locus. Sequences for each sgRNA and inserted construct are available in S1 File.

The collection numbers and specific mutations in each strain, as well as their impacts on TDH3 expression, are detailed in S1 Table.

Gene expression data

RNA-sequencing data presented in this paper is a subset of the data described Vande Zande et al. [29] and are available at GEO accession GSE175398. That dataset consists of RNA-sequencing data for cis-regulatory mutants and a larger set of trans-regulatory mutants affecting TDH3 expression. Details of data collection and processing are available in [29] and are summarized here. Briefly, yeast cells were grown to mid log phase in glucose media, pelleted, and frozen at -80C. polyA RNA was extracted from frozen cell pellets using oligodT magnetic beads. RNA libraries were prepared for sequencing using a ⅓ volume TruSeq RNA Sample Preparation v2 kit (Illumina) and sequenced on a HiSeq 4000 by the University of Michigan Sequencing Core. Each genotype (all mutants and non-mutated reference strains) was assayed in quadruplicate with each replicate consisting of a unique random array of genotypes and controls in a 96 well plate.

Measures of fluorescence changes over during population growth

Strains with different YFP reporter genes (driven by either PTDH1, PTDH2, or PTDH3) were patched from glycerol stocks onto YPG agar media (10g/L yeast extract, 20g/L peptone, 20g/L agar, 20% glycerol) and grown at 30C for 2–3 days. They were subsequently grown in YPD (10g/L yeast extract, 20g/L peptone, 20% dextrose) liquid culture for 48–72 hours, until all strains had reached stationary phase. The saturated strains were then diluted into 96 well plates, with 5ul of saturated culture added to 195ul of YPD. The plates were then incubated at 30C with shaking in a Synergy H1 plate reader. Culture growth and reporter gene expression were characterized by recording OD660 and YFP fluorescence readings, respectively, at 20-minute intervals for 48 hours. Three replicates were measured for each genotype using this method.

Flow cytometry

All flow cytometry data is deposited at flowrepository.org and publicly accessible at http://flowrepository.org/id/FR-FCM-Z72G. Strains bearing fluorescent proteins were patched from glycerol stocks onto 5-FOA agar media (0.67% YNB, 0.2% SC-uracil dropout mix, 2% glucose, 50ug/mL uracil, 0.1% 5-FOA) and grown at 30C for 3 days then stored briefly at 4C. Strains were revived by inoculation in liquid YPD (10g/L yeast extract, 20g/L peptone, 20% dextrose) and grown to saturation: 1mL YPD cultures inoculated in 14mL culture tubes were incubated at 30C with 200 rpm shaking for 48h in 4 replicates. Saturated cultures were then back diluted to inoculate cultures for scoring by adding 50uL of saturated culture to 1mL fresh YPD in culture tubes and incubating at 30C with 200 rpm shaking. After 24h growth, strains were diluted for scoring by transferring 20uL saturated culture into 0.5mL phosphate-buffered saline (Thermo Scientific blood bank saline, pH 7.0–7.2 cat#8504). Fluorescence was quantified on a Cytek Aurora analyzer (U1359SP) located at the University of Michigan BRCF Flow Cytometry Core. Cultures were diluted and scored one replicate at a time using autosampling from a 40-tube rack (flow rate: medium, 6 sec data recording delay, 2s agitation every 2 wells). FCS files were exported and analyzed in R (version 4.1.3) using the flowCore [61] and flowClust [62] packages. Cells were filtered for size using hard gates set on FSC.A and SSC.A values and a clustering-based gate on FSC.A, and then a singlets gate applied. Raw fluorescence units were normalized to cell size by dividing log10 of fluorescence in the bandpass filter appropriate for each fluorochrome (CFP: V5 detector 405 excitation, 508/20 emission; YFP: B2 detector: 488 excitation, 525/17 emission) by the log10 FSC.A value. All samples were down-sampled to 15,000 singlet cells per sample to equalize the number of cells analyzed for each strain. Each violin plot is composed of all four biological replicates, each consisting of 15,000 single cell events for a total of 60,000 cells. A point at the median of each replicate (4 points total in each violin plot) is superimposed over the violin plots. Medians of each replicate were used for statistical tests for difference of means (Student’s t-test).

Statistical analysis

All statistical analysis was performed in R, (version 3.5.2). As described in Vande Zande et al. [29], RNA-seq reads were pseudomapped to the S.cerevisiae transcriptome (Ensemble, release 38, retrieved from http://ftp://ftp.ensemblgenomes.org/pub/release-38/fungi/fasta/saccharomyces_cerevisiae/cdna/), and DeSeq2 [63] was used to estimate log2 fold changes and significance values reported in the text. One-sided t-tests for flow cytometry data were performed in R using the base ‘t.test’ function (alternative = “greater”) to compare median fluorescence intensity between reporter strains with intact TDH3 and tdh3Δ. Each comparison consisted of four biological replicates. Code used in the analysis and to generate figures in this manuscript is available at Github and in a permanent zenodo release (URL: https://github.com/pvz22/Compensation_TDH3, Zenodo DOI: 10.5281/zenodo.10223579).

Supporting information

S1 Table. Strains used in this study.

Table including all mutant strains of S. cerevisiae used in this study, including those for which RNA-sequencing data was collected and strains bearing fluorescent protein reporters.

https://doi.org/10.1371/journal.pgen.1011078.s001

(XLSX)

S1 Fig. TDH1, TDH2, and TDH3 are expressed at different levels and under different growth conditions.

(A) Population growth curves (measured using the optical density (OD) at 660 nm) are shown for strains expressing a yellow fluorescent protein (YFP) driven by PTDH1 (light blue), PTDH2 (dark blue), or PTDH3 (black). All three strains showed similar growth dynamics. Data in (A) was used to demarcate lag phase (red), exponential growth (green), diauxic shift (blue), and respiratory growth (purple) phases for all panels. Three replicates of each strain are shown. (B-D) Fluorescence values normalized to OD660 to account for changes in cell density are plotted across the growth curve for strains expressing YFP driven by PTDH3 (B), PTDH2 (C), and PTDH1 (D). PTDH3 and PTDH2 dynamics are similar across the stages of the growth curve (lag phase in red, exponential growth in green, diauxic shift in blue, and respiratory growth in purple, see panel A), with expression increasing during early exponential growth and then declining, halting during the diauxic shift, and increasing at a slower rate throughout respiratory growth. PTDH1 shows different dynamics; as TDH2 and TDH3 begin to decline, TDH1 begins to increase and does so steadily throughout the diauxic shift and respiratory growth. (E) Fluorescence values normalized to OD660 to account for changes in cell density are plotted across the growth curve for strains expressing YFP driven by PTDH1 (light blue, same data as panel D), PTDH2 (dark blue, same data as panel C), or PTDH3 (black, same data as panel B) promoters. The TDH3 promoter drives expression at a much higher level (approx. 6x) than that of PTDH1 or PTDH2.

https://doi.org/10.1371/journal.pgen.1011078.s002

(PDF)

S1 File. Primers and guide RNA target sequences used to generate engineered strains.

Nucleotide sequences and brief descriptions of transformation protocols used to generate strains bearing fluorescent reporters and fluorescent fusion proteins under the control of various promoters.

https://doi.org/10.1371/journal.pgen.1011078.s003

(PDF)

Acknowledgments

We thank Abigail Lamb for constructive feedback on the manuscript and Holly Scheer for technical and intellectual support, and other members of the Wittkopp Lab for helpful discussions and feedback on drafts of this manuscript. Flow cytometry data was generated at the University of Michigan Flow Cytometry Core.

References

  1. 1. Félix M-A, Barkoulas M. Pervasive robustness in biological systems. Nat Rev Genet. 2015;16: 483–496. pmid:26184598
  2. 2. Gibson G, Lacek KA. Canalization and Robustness in Human Genetics and Disease. Annu Rev Genet. 2020;54: 189–211. pmid:32867542
  3. 3. Ohya Y, Sese J, Yukawa M, Sano F, Nakatani Y, Saito TL, et al. High-dimensional and large-scale phenotyping of yeast mutants. Proc Natl Acad Sci U S A. 2005;102: 19015–19020. pmid:16365294
  4. 4. DeLuna A, Vetsigian K, Shoresh N, Hegreness M, Colón-González M, Chao S, et al. Exposing the fitness contribution of duplicated genes. Nat Genet. 2008;40: 676–681. pmid:18408719
  5. 5. Gu Z, Steinmetz LM, Gu X, Scharfe C, Davis RW, Li W-H. Role of duplicate genes in genetic robustness against null mutations. Nature. 2003;421: 63–66. pmid:12511954
  6. 6. Wagner A. Robustness against mutations in genetic networks of yeast. Nat Genet. 2000;24: 355–361. pmid:10742097
  7. 7. Diss G, Ascencio D, DeLuna A, Landry CR. Molecular mechanisms of paralogous compensation and the robustness of cellular networks. J Exp Zool B Mol Dev Evol. 2014;322: 488–499. pmid:24376223
  8. 8. Dohn TE, Cripps RM. Absence of the Drosophila jump muscle actin Act79B is compensated by up-regulation of Act88F. Dev Dyn. 2018;247: 642–649. pmid:29318731
  9. 9. DeLuna A, Springer M, Kirschner MW, Kishony R. Need-based up-regulation of protein levels in response to deletion of their duplicate genes. PLoS Biol. 2010;8: e1000347. pmid:20361019
  10. 10. Dong OX, Tong M, Bonardi V, El Kasmi F, Woloshen V, Wünsch LK, et al. TNL-mediated immunity in Arabidopsis requires complex regulation of the redundant ADR1 gene family. New Phytol. 2016;210: 960–973. pmid:27074399
  11. 11. Rodriguez-Leal D, Xu C, Kwon C-T, Soyars C, Demesa-Arevalo E, Man J, et al. Evolution of buffering in a genetic circuit controlling plant stem cell proliferation. Nat Genet. 2019;51: 786–792. pmid:30988512
  12. 12. Rudnicki MA, Braun T, Hinuma S, Jaenisch R. Inactivation of MyoD in mice leads to up-regulation of the myogenic HLH gene Myf-5 and results in apparently normal muscle development. Cell. 1992;71: 383–390. pmid:1330322
  13. 13. Denby CM, Im JH, Yu RC, Pesce CG, Brem RB. Negative feedback confers mutational robustness in yeast transcription factor regulation. Proc Natl Acad Sci U S A. 2012;109: 3874–3878. pmid:22355134
  14. 14. Nimchuk ZL, Zhou Y, Tarr PT, Peterson BA, Meyerowitz EM. Plant stem cell maintenance by transcriptional cross-regulation of related receptor kinases. Development. 2015;142: 1043–1049. pmid:25758219
  15. 15. Li J, Yuan Z, Zhang Z. The cellular robustness by genetic redundancy in budding yeast. PLoS Genet. 2010;6: e1001187. pmid:21079672
  16. 16. Kuzmin E, VanderSluis B, Nguyen Ba AN, Wang W, Koch EN, Usaj M, et al. Exploring whole-genome duplicate gene retention with complex genetic interaction analysis. Science. 2020;368. pmid:32586993
  17. 17. Kafri R, Bar-Even A, Pilpel Y. Transcription control reprogramming in genetic backup circuits. Nat Genet. 2005;37: 295–299. pmid:15723064
  18. 18. Kafri R, Levy M, Pilpel Y. The regulatory utilization of genetic redundancy through responsive backup circuits. Proc Natl Acad Sci U S A. 2006;103: 11653–11658. pmid:16861297
  19. 19. McAlister L, Holland MJ. Differential expression of the three yeast glyceraldehyde-3-phosphate dehydrogenase genes. J Biol Chem. 1985;260: 15019–15027. pmid:3905788
  20. 20. Linck A, Vu X-K, Essl C, Hiesl C, Boles E, Oreb M. On the role of GAPDH isoenzymes during pentose fermentation in engineered Saccharomyces cerevisiae. FEMS Yeast Res. 2014;14: 389–398. pmid:24456572
  21. 21. Holland JP, Holland MJ. Structural comparison of two nontandemly repeated yeast glyceraldehyde-3-phosphate dehydrogenase genes. J Biol Chem. 1980;255: 2596–2605. pmid:6244283
  22. 22. Engel SR, Dietrich FS, Fisk DG, Binkley G, Balakrishnan R, Costanzo MC, et al. The reference genome sequence of Saccharomyces cerevisiae: then and now. G3. 2014;4: 389–398. pmid:24374639
  23. 23. Holland JP, Labieniec L, Swimmer C, Holland MJ. Homologous nucleotide sequences at the 5’ termini of messenger RNAs synthesized from the yeast enolase and glyceraldehyde-3-phosphate dehydrogenase gene families. The primary structure of a third yeast glyceraldehyde-3-phosphate dehydrogenase gene. J Biol Chem. 1983;258: 5291–5299. pmid:6833300
  24. 24. Delgado ML O’Connor JE, Azorı N I, Renau-Piqueras J, Gil ML, Gozalbo D. The glyceraldehyde-3-phosphate dehydrogenase polypeptides encoded by the Saccharomyces cerevisiae TDH1, TDH2 and TDH3 genes are also cell wall proteins. Microbiology. 2001;147: 411–417.
  25. 25. Bradley PH, Gibney PA, Botstein D, Troyanskaya OG, Rabinowitz JD. Minor Isozymes Tailor Yeast Metabolism to Carbon Availability. mSystems. 2019;4.
  26. 26. Schubert OT, Bloom JS, Sadhu MJ, Kruglyak L. Genome-wide base editor screen identifies regulators of protein abundance in yeast. Elife. 2022;11. pmid:36326816
  27. 27. Duveau F, Toubiana W, Wittkopp PJ. Fitness Effects of Cis-Regulatory Variants in the Saccharomyces cerevisiae TDH3 Promoter. Mol Biol Evol. 2017;34: 2908–2912. pmid:28961929
  28. 28. Costanzo M, Baryshnikova A, Bellay J, Kim Y, Spear ED, Sevier CS, et al. The Genetic Landscape of a Cell. Science. 2010;327. pmid:20093466
  29. 29. Vande Zande P, Hill MS, Wittkopp PJ. Pleiotropic effects of trans-regulatory mutations on fitness and gene expression. Science. 2022;377: 105–109. pmid:35771906
  30. 30. Huie MA, Scott EW, Drazinic CM, Lopez MC, Hornstra IK, Yang TP, et al. Characterization of the DNA-binding activity of GCR1: in vivo evidence for two GCR1-binding sites in the upstream activating sequence of TPI of Saccharomyces cerevisiae. Mol Cell Biol. 1992;12: 2690–2700. pmid:1588965
  31. 31. Yagi S, Yagi K, Fukuoka J, Suzuki M. The UAS of the yeast GAPDH promoter consists of multiple general functional elements including RAP1 and GRF2 binding sites. J Vet Med Sci. 1994;56: 235–244. pmid:8075210
  32. 32. MacIsaac KD, Wang T, Gordon DB, Gifford DK, Stormo GD, Fraenkel E. An improved map of conserved regulatory sites for Saccharomyces cerevisiae. BMC Bioinformatics. 2006;7: 113. pmid:16522208
  33. 33. Hu Z, Killion PJ, Iyer VR. Genetic reconstruction of a functional transcriptional regulatory network. Nat Genet. 2007;39: 683–687. pmid:17417638
  34. 34. Lickwar CR, Mueller F, Hanlon SE, McNally JG, Lieb JD. Genome-wide protein-DNA binding dynamics suggest a molecular clutch for transcription factor function. Nature. 2012;484: 251–255. pmid:22498630
  35. 35. Venters BJ, Wachi S, Mavrich TN, Andersen BE, Jena P, Sinnamon AJ, et al. A comprehensive genomic binding map of gene and chromatin regulatory proteins in Saccharomyces. Mol Cell. 2011;41: 480–492. pmid:21329885
  36. 36. Duveau F, Vande Zande P, Metzger BP, Diaz CJ, Walker EA, Tryban S, et al. Mutational sources of trans-regulatory variation affecting gene expression in Saccharomyces cerevisiae. Elife. 2021;10. pmid:34463616
  37. 37. Giaever G, Chu AM, Ni L, Connelly C, Riles L, Véronneau S, et al. Functional profiling of the Saccharomyces cerevisiae genome. Nature. 2002;418: 387–391. pmid:12140549
  38. 38. Piña B, Fernández-Larrea J, García-Reyero N, Idrissi F-Z. The different (sur)faces of Rap1p. Mol Genet Genomics. 2003;268: 791–798.
  39. 39. Uemura H, Koshio M, Inoue Y, Lopez MC, Baker HV. The role of Gcr1p in the transcriptional activation of glycolytic genes in yeast Saccharomyces cerevisiae. Genetics. 1997;147: 521–532. pmid:9335590
  40. 40. He X, Zhang J. Transcriptional reprogramming and backup between duplicate genes: is it a genomewide phenomenon? Genetics. 2006;172: 1363–1367. pmid:16322517
  41. 41. VanderSluis B, Bellay J, Musso G, Costanzo M, Papp B, Vizeacoumar FJ, et al. Genetic interactions reveal the evolutionary trajectories of duplicate genes. Mol Syst Biol. 2010;6: 429. pmid:21081923
  42. 42. Kuzmin E, Taylor JS, Boone C. Retention of duplicated genes in evolution. Trends Genet. 2022;38: 59–72. pmid:34294428
  43. 43. Hahn S, Young ET. Transcriptional regulation in Saccharomyces cerevisiae: transcription factor regulation and function, mechanisms of initiation, and roles of activators and coactivators. Genetics. 2011;189: 705–736. pmid:22084422
  44. 44. Hossain MA, Claggett JM, Edwards SR, Shi A, Pennebaker SL, Cheng MY, et al. Posttranscriptional Regulation of Gcr1 Expression and Activity Is Crucial for Metabolic Adjustment in Response to Glucose Availability. Mol Cell. 2016;62: 346–358. pmid:27153533
  45. 45. Sussel L, Shore D. Separation of transcriptional activation and silencing functions of the RAP1-encoded repressor/activator protein 1: isolation of viable mutants affecting both silencing and telomere length. Proc Natl Acad Sci U S A. 1991;88: 7749–7753. pmid:1881914
  46. 46. Shore D. RAP1: A protean regulator in yeast. Trends Genet. 1994;10: 408–412. pmid:7809947
  47. 47. Kovács K, Farkas Z, Bajić D, Kalapis D, Daraba A, Almási K, et al. Suboptimal Global Transcriptional Response Increases the Harmful Effects of Loss-of-Function Mutations. Mol Biol Evol. 2021;38: 1137–1150. pmid:33306797
  48. 48. Ringel AE, Ryznar R, Picariello H, Huang K-L, Lazarus AG, Holmes SG. Yeast Tdh3 (Glyceraldehyde 3-Phosphate Dehydrogenase) Is a Sir2-Interacting Factor That Regulates Transcriptional Silencing and rDNA Recombination. Pikaard CS, editor. PLoS Genet. 2013;9: e1003871. pmid:24146631
  49. 49. Shen X, De Jonge J, Forsberg SKG, Pettersson ME, Sheng Z, Hennig L, et al. Natural CMT2 variation is associated with genome-wide methylation changes and temperature seasonality. PLoS Genet. 2014;10: e1004842. pmid:25503602
  50. 50. Branco P, Francisco D, Chambon C, Hébraud M, Arneborg N, Almeida MG, et al. Identification of novel GAPDH-derived antimicrobial peptides secreted by Saccharomyces cerevisiae and involved in wine microbial interactions. Appl Microbiol Biotechnol. 2014;98: 843–853. pmid:24292082
  51. 51. Singh N, Bhalla N. Moonlighting Proteins. Annu Rev Genet. 2020;54: 265–285. pmid:32870732
  52. 52. Chauhan AS, Kumar M, Chaudhary S, Patidar A, Dhiman A, Sheokand N, et al. Moonlighting glycolytic protein glyceraldehyde-3-phosphate dehydrogenase (GAPDH): an evolutionarily conserved plasminogen receptor on mammalian cells. FASEB J. 2017;31: 2638–2648. pmid:28298336
  53. 53. Espinosa-Cantú A, Ascencio D, Barona-Gómez F, DeLuna A. Gene duplication and the evolution of moonlighting proteins. Front Genet. 2015;6: 227. pmid:26217376
  54. 54. Kafri R, Dahan O, Levy J, Pilpel Y. Preferential protection of protein interaction network hubs in yeast: evolved functionality of genetic redundancy. Proc Natl Acad Sci U S A. 2008;105: 1243–1248. pmid:18216251
  55. 55. Hanada K, Kuromori T, Myouga F, Toyoda T, Li W-H, Shinozaki K. Evolutionary persistence of functional compensation by duplicate genes in Arabidopsis. Genome Biol Evol. 2009;1: 409–414. pmid:20333209
  56. 56. Ihmels J, Collins SR, Schuldiner M, Krogan NJ, Weissman JS. Backup without redundancy: genetic interactions reveal the cost of duplicate gene loss. Mol Syst Biol. 2007;3: 86. pmid:17389874
  57. 57. Dean EJ, Davis JC, Davis RW, Petrov DA. Pervasive and persistent redundancy among duplicated genes in yeast. PLoS Genet. 2008;4: e1000113. pmid:18604285
  58. 58. Tischler J, Lehner B, Chen N, Fraser AG. Combinatorial RNA interference in Caenorhabditis elegans reveals that redundancy between gene duplicates can be maintained for more than 80 million years of evolution. Genome Biol. 2006;7: R69. pmid:16884526
  59. 59. Hodgins-Davis A, Duveau F, Walker EA, Wittkopp PJ. Empirical measures of mutational effects define neutral models of regulatory evolution in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A. 2019;116: 21085–21093. pmid:31570626
  60. 60. Laughery MF, Hunter T, Brown A, Hoopes J, Ostbye T, Shumaker T, et al. New vectors for simple and streamlined CRISPR-Cas9 genome editing in Saccharomyces cerevisiae. Yeast. 2015;32: 711–720. pmid:26305040
  61. 61. Hahne F, LeMeur N, Brinkman RR, Ellis B, Haaland P, Sarkar D, et al. flowCore: a Bioconductor package for high throughput flow cytometry. BMC Bioinformatics. 2009;10: 106. pmid:19358741
  62. 62. Lo K, Hahne F, Brinkman RR, Gottardo R. flowClust: a Bioconductor package for automated gating of flow cytometry data. BMC Bioinformatics. 2009;10: 145. pmid:19442304
  63. 63. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15: 550. pmid:25516281