(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



SO2 and copper tolerance exhibit an evolutionary trade-off in Saccharomyces cerevisiae [1]

['Cristobal A. Onetto', 'The Australian Wine Research Institute', 'Glen Osmond', 'South Australia', 'Dariusz R. Kutyna', 'Radka Kolouchova', 'Jane Mccarthy', 'Anthony R. Borneman', 'School Of Agriculture', 'Food']

Date: 2023-04

Copper tolerance and SO 2 tolerance are two well-studied phenotypic traits of Saccharomyces cerevisiae. The genetic bases of these traits are the allelic expansion at the CUP1 locus and reciprocal translocation at the SSU1 locus, respectively. Previous work identified a negative association between SO 2 and copper tolerance in S. cerevisiae wine yeasts. Here we probe the relationship between SO 2 and copper tolerance and show that an increase in CUP1 copy number does not always impart copper tolerance in S. cerevisiae wine yeast. Bulk-segregant QTL analysis was used to identify variance at SSU1 as a causative factor in copper sensitivity, which was verified by reciprocal hemizygosity analysis in a strain carrying 20 copies of CUP1. Transcriptional and proteomic analysis demonstrated that SSU1 over-expression did not suppress CUP1 transcription or constrain protein production and provided evidence that SSU1 over-expression induced sulfur limitation during exposure to copper. Finally, an SSU1 over-expressing strain exhibited increased sensitivity to moderately elevated copper concentrations in sulfur-limited medium, demonstrating that SSU1 over-expression burdens the sulfate assimilation pathway. Over-expression of MET 3/14/16, genes upstream of H 2 S production in the sulfate assimilation pathway increased the production of SO 2 and H 2 S but did not improve copper sensitivity in an SSU1 over-expressing background. We conclude that copper and SO 2 tolerance are conditional traits in S. cerevisiae and provide evidence of the metabolic basis for their mutual exclusivity. These findings suggest an evolutionary driver for the extreme amplification of CUP1 observed in some yeasts.

Completing a commercial wine fermentation is a tough job for a yeast. Grape juice is a highly variable environment and to cope with that variability, a large number of different yeast strains have been generated exhibiting different features. Two of the most distinguishing physical characteristics of wine yeast are copper and SO 2 tolerance, which appear to be mutually exclusive. The genetic underpinnings of these two traits are individually well-characterised, but there doesn’t appear to be an obvious reason why copper tolerance and SO 2 tolerance could not co-exist. We performed a genetic analysis that showed how over-expression of the SO 2 transporters responsible for SO 2 tolerance induced copper sensitivity. Our analysis of RNA and protein levels in SO 2 -tolerant yeast showed that they could still produce the molecules that would usually protect them when exposed to copper stress. However, the constant activation of the transporter that provides SO 2 tolerance also induced a sulfur limitation that could not be overcome when combined with copper stress.

Competing interests: The authors certify that they have no affiliations with or involvement in any organisation or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.

Funding: This work was supported by a grant from Wine Australia (AWRI 1701-3.2.2 to SA), with levies from Australia’s grapegrowers and winemakers and matching funds from the Australian Government. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability: All relevant data are within the manuscript or have been deposited with the following specialist or general data repositories. The genome sequence of strains AWRI 3811, AWRI 3471, AWRI 3807, AWRI 3001 and AWRI 3470 used in the analysis of QTL following bulk segregant sequence analysis and the raw RNA-seq reads are available in NCBI under Bioproject PRJNA877711. Quantitative proteomic data have been made available via the Proteomics Identifications Database (PRIDE) at http://www.ebi.ac.uk/pride/archive/projects/PXD037997 . Raw data is available at Dryad at https://doi.org/10.5061/dryad.wdbrv15s5 .

Copyright: © 2023 Onetto 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.

To determine the contribution of CUP1-2 copy number variation to copper tolerance in wine yeast, we determined the copy number status of 94 wine yeast strains and compared copy number variation to competitive fitness scores in a copper-containing medium. The comparison identified many strains that exhibited copper sensitivity despite carrying a high number of CUP1-2 copies. The genetic contributions to copper sensitivity in selected strains were determined by mating strains with equivalent or differential CUP1-2 copy numbers and/or fitness in high copper media, followed by bulk segregant analysis of progeny. Potential genetic contributions to copper sensitivity were confirmed through reciprocal deletions and over-expression analysis in parental lines. Transcriptomic and proteomic analysis identified a potential metabolic limitation induced by growth in high copper in SO 2 tolerant wine yeast. The degree to which SO 2 tolerance contributed to the metabolic limitation was evaluated using a series of fermentation trials.

In other work, Fay et al [ 39 ] observed upregulation of sulfate assimilation pathway genes in a subset of strains during a study of copper stress but could not directly associate this as a response to copper, but rather with the capacity of those strains to become discolored through the formation of CuS. de Freitas et al [ 40 ] showed that addition of copper could rescue the ability of Δssu1 yeast to grow on non-fermentable carbon sources. This was attributed to depletion of copper or iron by accumulation of intracellular sulfur compounds. Taken together these works provide a picture of the strong interdependence of copper and sulfur metabolism in yeast.

Hodgins-Davis et al. [ 35 ] show that both copper starvation and toxicity elicit responses from genes associated with sulfur metabolism. CUP1 poor strains were shown to respond to increased copper concentrations with upregulation of genes related to mitochondrial activity or oxidative stress. These cellular functions are also critical contributors to survival in multiple forms of starvation [ 37 ] and are part of a larger response to starvation which ultimately results in cell cycle arrest [ 38 ].

By constitutively exporting SO 2, over-expression of SSU1 directly intervenes in the sulfate assimilation pathway, acting to remove the immediate precursor to hydrogen sulfide [reviewed in 34]. A dependency on sulfur metabolism genes as a response to extremes of copper stress has previously been noted [ 35 ] as has a relationship between SO 2 tolerance and copper sensitivity in wine yeast [ 36 ].

The sulfate assimilation pathway has also been implicated in resistance to SO 2 in a coordinated response mediated by COM2 [ 33 ]. It is suggested that components of this pathway contribute to SO 2 resistance by reducing SO 2 to H 2 S which can then either exit the cell or further metabolised by condensation with o-acetyl homoserine [for review see 34 ].

SO 2 induces its deleterious effects on yeast by compromising energy metabolism. Inhibition of glycolysis decreases ATP production, while membrane leakage that results from membrane damage increases its consumption [reviewed in 24 ]. SO 2 tolerance in S. cerevisiae is mediated by the efflux pump Ssu1p [ 25 ]. Unlike the mechanism that gives rise to copper tolerance, variation in SO 2 tolerance in wine strains of S. cerevisiae is predominantly the result of reciprocal translocations between chromosomes VIII and XVI [ 26 , 27 ] although translocations between chromosomes XV and XVI [ 28 ] and an inversion within chromosome XVI [ 29 ] have also been identified in a limited number of isolates. SSU1 expression in wild type S. cerevisiae is regulated by Fzf1p [ 30 ]. As a result of its fusion with the promoter of ECM34, control of SSU1 expression is freed from Fzf1p regulation in strains carrying the VIII::XVI translocation [ 31 ]. Substantial heterogeneity exists in the precise structure of the ECM34 promoter with variable expression of SSU1 and subsequent variance in SO 2 tolerance as a result [ 31 , 32 ].

The activities of copper in yeast are diverse and complex, as is its regulation [reviewed in 10 ]. Copper resistance is mediated predominantly by the metallothionein Cup1p [ 11 ] and, to a lesser extent, Crs5p [ 12 ], Sod1p [ 13 ] and glutathione [ 14 ]. Copy number variation of the CUP1 gene is commonly observed in S. cerevisiae and has been estimated at between 0–70 copies per cell [ 6 , 11 , 15 – 17 ], with higher copy numbers being associated with higher levels of resistance to otherwise inhibitory concentrations of copper [ 18 , 19 ]. Despite the primary association of CUP1 with copper tolerance, copy number expansion explains only 44.5% of phenotypic variation in copper tolerance [ 20 ]. Other mechanisms by which S. cerevisiae responds to copper excess include Mac1p dependent control of copper import [ 21 ], manipulation of the copper oxidation state via Fet3p [ 22 ] and SLF1-mediated mineralisation of copper into copper sulfide [ 23 ].

The commercial and environmental prevalence of copper and SO 2 has provided an environment where resistance to these two compounds has manifested in wine yeast and environmental isolates, including medical specimens [ 6 ]. Freshly prepared grape juice typically contains between 0.5 and 1.5 mg/L of copper, although concentrations above 7 mg/L [ 7 , 8 ] can be observed, depending on copper usage in the vineyard [ 9 ].

Copper and SO 2 are used nearly ubiquitously in the wine industry. Their usefulness is, at least in part, due to their varied activities. In the vineyard, copper- and sulfur-based sprays are applied to control both downy [ 1 ] and powdery mildews [ 2 ]. In the form of SO 2 , sulfur is used during grape processing to help protect harvested grapes, juice and must against unwanted microbial activity [ 3 ] and oxidation [ 4 ]. Likewise, it is used after fermentation to stabilise the finished wine. Copper is used in finished wine to moderate aromas derived from low molecular weight sulfur compounds [ 5 ]. As a result of this multitude of applications, copper and sulfur have been part of the grape grower and winemaker tool kit for generations.

Results and discussion

CUP1 copy number alone is a poor predictor of copper tolerance in wine yeast The CUP1 copy number in 94 wine yeast strains, normalised against a single insert molecular barcode, was determined using qPCR. CUP1 copy number varied dramatically between 2 (SD 0.1) and 55 (SD 2.8) absolute copies per cell [41, file T04]. This is consistent with the 0–26 haplotype copies (CUP1-1 + CUP1-2) for strains in the wine yeast clade (Fig B in S1 Text) estimated from whole-genome sequence data [15]. With the previously observed diversity in wine yeast copper tolerance [36] and the high diversity in CUP1 copy number among these strains we expected a strong relationship between copy number and fitness. However, no correlation between CUP1 copy number and tolerance to copper was observed (Fig 1). Many strains exhibited both a significant positive fitness attribute and high CUP1 copy number (between 6 and 18 copies). However, an equally large number of strains with a high CUP1 copy number exhibited poor fitness in 10 mg/L of copper. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. Relationship between yeast strain fitness in high copper medium and CUP1 copy number. The fitness value shown on the x-axis is the mean of two independent experiments. Significant observations are recorded as 0, 1 or 2 if there was evidence (P < 0.05) that the log2 fold change differed relative to the control condition in independent fitness experiments (n = 3 for each of 2 independent experiments). The y-axis shows the mean absolute CUP1 copy number of each strain. Error bars show standard deviation (n = 3). Strains pictured with a larger point size (3019, 3032 and 3029) were the parental strains used in subsequent bulk segregant QTL experiments. https://doi.org/10.1371/journal.pgen.1010692.g001

QTL analysis of segregants identifies SSU1 as a contributing factor in copper sensitivity The observation here and elsewhere [20] that CUP1 copy number is a poor predictor of copper tolerance in yeast raises the obvious question; why are strains with a high CUP1 copy number so poorly tolerant of copper in the medium? The question of copper sensitivity among CUP1 containing yeast was addressed using bulk segregant analyses of haploid strains derived from crosses between a single copper tolerant parent with high CUP1 copy number (AWRI 796, 25.6 [SD 5.3] copies) and two copper sensitive parents, with either low (AWRI 1537, 2.5 [SD 0.9]) or high (AWRI 1487, 31.7 [SD 2.7]) copies of CUP1, respectively (highlighted with large spots in Fig 1). Stable haploid progeny were obtained for each of the parental genotypes, through the prior inclusion of a molecular barcode that disrupted HO [described in 36]. The parental copper-resistance phenotype was present in all spores isolated from AWRI 3019, AWRI 3029 and AWRI 3032 (Fig C in S1 Text) corresponding to barcoded versions of AWRI 796, AWRI 1537 and AWRI 1487, respectively. It should be noted that the copper tolerance phenotype was only assessed in a subset of the isolated spores (i.e. those containing a barcode and containing complementary mating type genes). A copper-tolerant haploid derived from AWRI 3019 (Cutol: CUP1high) was mated with copper sensitive haploids derived from AWRI 3029 (Cusen: CUP1low) and AWRI 3032 (Cusen: CUP1high) to yield two diploid strains, AWRI 3001 and AWRI 3811. In each case, the diploid derivatives were copper sensitive, indicating that this is the dominant phenotype. To map F1 phenotypes, both AWRI 3001 and AWRI 3811 were sporulated. Tetrad dissection of spores routinely recovered 3 to 4 viable spores. In total, 80 and 146 spores were isolated and phenotyped for copper sensitivity for the two crosses, which were subsequently divided into either copper-tolerant (n = 41 and 67) or copper-sensitive (n = 32 and 51) pools (excluding a small number of spores with an ill-defined phenotype) (Fig A in S1 Text). An analysis of the SNP frequency in F1 progeny from AWRI 3001 (Cutol:CUP1high x Cusen:CUP1low) was undertaken. One hundred and one positions across the genome were identified that exhibited a SNP ratio greater than 0.85. Sixty-five of those were on Chr VIII with a peak between 210,000 and 218,000 bp. This position corresponds to the location of CUP1-1 and CUP1-2 (Fig 2A). The major association on Chr VIII was consistent with the mean difference in CUP1 copy number between the two strains (Δ copy number = 23.1 copies [95CI, 19.9, 26.3]). This data explains the copper sensitivity of the AWRI 3029 strain and supports previous observations [11,12,17,18] that CUP1 is a key determinant of copper tolerance in yeast. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Single nucleotide variant (SNV) frequency in spores generated from diploids following crosses between A) AWRI 3471 and AWRI 3470, and B) AWRI 3471 and AWRI 3807. SNV frequency is shown for each parent, the diploid generated from each cross (AWRI 3001 and AWRI 3811) and spore pools where each spore in the pool was classified as either copper tolerant (Cu-tol) or copper sensitive (Cu-sens). https://doi.org/10.1371/journal.pgen.1010692.g002 Of the remaining genomic locations with SNP ratios greater than 0.85, eleven of them were on Chr IV between positions 1,163,450 and 1,163,515 bp. However, the limited breadth of the change in SNP frequency around this location and that this, and other locations with high SNP ratios, were not mirrored in Cu-tol and Cu-sens pools suggest that there is no association with the phenotype. QTL analysis in the progeny of the Cutol: CUP1high x Cusen: CUP1high cross (AWRI 3811) showed a divergence in SNP frequencies approaching 100% (for the sensitive and resistant parental genotypes in the copper sensitive and resistant pools) on the extreme left arm of Chr VIII and between 350,000 bp and 400,000 bp on Chr XVI (Fig 2B). There were few other genomic locations where either parental SNP frequency exceeded 0.75 in this data set. The two positions of the QTLs on Chr VIII and XVI are consistent with the position of the previously described translocation between the genes SSU1 and ECM34 [26]. It is noteworthy that the two parents of this cross have either wild-type (AWRI 3019) or translocated (AWRI 3032) chromosomes at the SSU1 locus [41, file T05]. Translocations at this position have previously been associated with increased SO 2 tolerance in yeast [26] but associations with copper sensitivity have not been reported. A divergence from the expected 0.5 SNP ratios for the entirety of Chr I was observed in diploids derived from both crosses (AWRI 3001 and AWRI 3811) and copper tolerant and copper sensitive pools prepared from the respective F1 progeny (Fig 2A and 2B). The divergence from expected SNP ratios for Chr I can be explained by a whole chromosome duplication that has been reported previously [42] in the progenitor strain used for this work (AWRI 796).

Deletion of SSU1 restores copper tolerance in copper sensitive hybrid The contribution of SSU1 to copper sensitivity was evaluated by reciprocal deletion of SSU1 in the copper-sensitive hybrid AWRI 3811 (Cutol: CUP1high: SSU1WT x Cusen: CUP1high: SSU1trans) to generate AWRI 3901 (SSU1WT/Δ) and AWRI 3902 (SSU1Δ/trans). The growth of each of these strains, in addition to the haploid parents of AWRI 3811, was evaluated in a defined medium with copper concentrations of 0.25 mg/L or 10 mg/L (Fig 3). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Heritability of SSU1 dependent copper tolerance and sensitivity assessed in defined medium containing either 0.25 or 10 mg/L copper. (A) Growth of yeast AWRI 3471 (●) and AWRI 3807 (■), haploid derivatives of AWRI 796 and AWRI 1487 respectively. (B) Growth of AWRI 3811, a diploid derived from a cross between 3471 x 3807. (C) Growth of AWRI 3901 and AWRI 3902, two derivatives of the AWRI 3811 diploid each containing a deletion of SSU1 at chromosome XVI and VIII::XVI respectively. Filled lines; standard defined medium, dashed lines; defined medium containing 10 mg/L copper. Points show mean of three replicates with error bars indicating standard deviation. https://doi.org/10.1371/journal.pgen.1010692.g003 While deleting the wild type copy of SSU1 from Chr VIII had no effect on the sensitivity of AWRI 3811, deleting SSU1 from the translocated VIII::XVI chromosome restored copper tolerance to the hybrid (Fig 3). This demonstrates that SSU1 on the translocated chromosome is the causative factor of copper sensitivity in the AWRI 3811 hybrid. The negative effect of SSU1 on copper tolerance is entirely due to its level of expression. This is demonstrated in Fig 4A which compares the growth of the copper-tolerant haploid AWRI 3471 with AWRI 4052 (ssu1(pr)Δ::ECM34(pr)), a strain in which the wild-type promoter of SSU1 in the AWRI 3471 background was exchanged for the ECM34 promoter from AWRI 1487. These near isogenic strains were compared in a defined medium containing 0.25 mg/L and 10 mg/L of copper. As in Fig 3, AWRI 3471 does not show any growth inhibition in the presence of elevated copper concentrations. Except for the promoter of SSU1, AWRI 4052 is genetically identical to AWRI 3471. However, AWRI 4052 is copper sensitive, exhibiting a mean biomass decrease of 2.1 g/L DCW [95CI, 1.8, 2.4]. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. Effect of copper on fermentation, gene expression and protein production in AWRI 3471 and AWRI 4052. (A) Growth as indicated by absorbance at 600 nm and (B) Sugar consumption of SSU1 wild type (AWRI 3471, blue) and ssu1(pr)Δ::ECM34(pr) (AWRI 4052, red) yeast strains during growth in defined medium with and without copper at 10 mg/L. Filled lines; standard defined medium, dashed lines; defined medium containing 10 mg/L copper. Arrow in (A) and (B) show sampling points used for RNAseq analysis. Points show mean of three replicates with error bars indicating standard deviation. (C) Relative transcript abundance grouped by Gene Ontology summary category for the contrast AWRI 4052 HCu–AWRI 3471 HCu, expressed as Log 2 Fold Change (Log 2 FC) with red showing increased and blue decreased expression. Colour intensity highlights the P value score obtained from differential expression analysis undertaken with DEseq2. SSU1 was omitted from over-representation analysis and therefore does not appear in enriched ontology summaries. (D) Relative protein abundance shown as volcano plots with colours indicating increased (red) and decreased (blue) relative fold change for four contrasts i) AWRI 4052 –AWRI 3471, ii) AWRI 4052 HCu–AWRI 3471 HCu. Colours indicating increased (orange) and decreased (light blue) are used in plots iii) AWRI 4052 HCu–AWRI 4052, iv) AWRI 3471 HCu–AWRI 3471. Vertical dotted lines indicate Log 2 FC = 1 and horizontal dotted lines indicate an adjusted P value = 0.005. n = 3 in all cases. SSU1 is shown despite having P value > 0.005 in panel ii). https://doi.org/10.1371/journal.pgen.1010692.g004 Fig 4B compares the fermentation performance of AWRI 3471 and AWRI 4052. In addition to the effects on growth, elevated copper concentrations in grape juice have been shown to impede sugar utilisation [43,44], an effect that is particularly relevant to commercial winemaking. Elevated copper concentrations effected fermentation progress in both AWRI 3471 and AWRI 4052, however, a severe delay in fermentation onset and significantly higher residual sugar concentrations were observed in fermentations by strain AWRI 4052 in copper excess relative to strain AWRI 3471 (mean increase of 78 g/L [95CI, 67, 86]) at day 17.

The combined effect of SSU1 over-expression and high copper concentration on yeast gene expression during fermentation The effect of SSU1 over-expression on copper sensitivity was explored through an analysis of the transcriptional response of strains bearing SSU1 and ssu1(pr)Δ::ECM34(pr) grown in a medium containing 10 mg/L of copper [41, file T10]. Genes for which there was strong evidence of differential abundance (P < 0.005) and for which the magnitude of the change was greater than 2-fold (Log 2 FC > 1) were further subjected to over-representation analysis, using Gene Ontology (GO) Biological Process as a grouping category. Fig 4C shows the relative expression as Log 2 FC (AWRI 4052—AWRI 3471) of the filtered gene set, grouped according to the GO category with which they are associated. Notably, the two strains cannot be differentiated according to the Cup2p or Msn2p responsive genes. CUP1 transcript abundance, for example, is equivalent between the two genotypes. The most prominent distinguishing transcriptional features are related to sulfur compound transport (Fig D in S1 Text, enrichment log 10 (P) = -7.33). Specifically, increased expression of genes associated with sulfate uptake (SUL1, SOA1), sulfonate catabolism (JLP1) and sulfur-containing amino acid uptake (MUP1, MUP3, YCT1, OPT1, AGP3) in AWRI 4052 relative to AWRI 3471 indicate that over-expression of SSU1 increases the burden on sulfur metabolism beyond that imposed by copper alone. Furthermore, the genes AGP3, PDC6 and YRO2, which have previously been postulated to be markers of sulfur-limited growth [45], are all up-regulated in this contrast. The general down-regulation of cell wall structural components under copper stress is accentuated in cells over-expressing SSU1 with greater down-regulation of mannoproteins (TIR1, TIR2, TIR3, TIR4, DAN1), and seripauperins (PAU17, PAU5, PAU16). A feature of the expression profile of AWRI 4052 grown under high copper is the down-regulation of genes related to thiamine metabolism. Not only is the expression of THI4 decreased (Log 2 FC = -1.1, P = 0.006), but a down-regulation of the broader thiamine regulon is evident. Down-regulation of THI7 (Log 2 FC = -1.2), THI73 (Log 2 FC = -1.1), THI20 (Log 2 FC = -1.6) and PDC5 (Log 2 FC = -2.4) in the ssu1(pr)Δ::pECM34(pr) background suggest that these cells are sensing excess thiamine [46]. Thiamine accumulation may be a consequence of a slowing growth rate in yeast over-expressing SSU1 and experiencing copper stress. In summary, the copper sensitivity exhibited by the ssu1(pr)Δ::pECM34(pr) strain AWRI 4052 cannot be explained by mis-regulation of genes known to be critical in the maintenance of copper homeostasis, such as CUP1. The up-regulation of genes associated with either the direct import of sulfur (as sulfate or as sulfur-containing amino acids) or the scavenging of sulfur (from intra-cellular sulfonates) indicates that SSU1 over-expression exacerbates copper stress by inducing a sulfur limitation. A diagrammatic representation of the sulfate assimilation pathway showing the effect of copper on the expression of genes in the pathway is provided in Fig 5. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 5. Diagrammatic representation of the sulfate assimilation pathway. Gene names in red and blue indicate up-regulated and down-regulated genes, respectively, comparing SSU1 over-expressing (AWRI 4052) yeast with control yeast (AWRI 3471) growing in medium containing 10 mg/L copper. SSU1 over-expression is indicated by a red helix. Gene names in black indicate no change in expression. Green circles represent copper ions. The structure of Cup1p is adapted from the crystal structure determined by Calderone et al [48]. Cysteine residues in the structure are coloured yellow. APS; Adenosine-5’-phosphosulfate, PAPS; phosphoadenosine phosphosulfate. The image was created with BioRender.com. https://doi.org/10.1371/journal.pgen.1010692.g005 SSU1 encodes an SO 2 efflux pump (Ssu1p). SO 2 is an intermediate in the sulfate assimilation pathway that is required for the biosynthesis of cysteine and methionine [47]. It is possible that increased activity of the Ssu1p transporter may limit the flux through the sulfate assimilation pathway. If this were the case, sulfur limitation induced by SSU1 over-expression may contribute to copper sensitivity in a number of ways. Although CUP1 transcripts are not mis-regulated in SSU1 over-expressing cells, as demonstrated above, the 53 amino acid product of CUP1 contains 12 cysteines [48] and therefore sulfur limitation may place a constraint on Cup1p production. Glutathione, a metabolite critical in the response to copper stress [49,50] may be similarly constrained. Alternatively, a restriction of flux through the sulfate assimilation pathway may also constrain the production of hydrogen sulfide, another intermediate in the sulfate assimilation pathway. Hydrogen sulfide has been shown to moderate the toxicity of copper through SLF1-mediated CuS mineralisation [23]. In the following sections we will evaluate whether SSU1 mediated constraints on the sulfate assimilation pathway contribute to copper sensitivity in S. cerevisiae through the alternative possibilities discussed above, beginning with the production of Cup1p protein.

Label-free proteomic analysis demonstrates equivalent Cup1p production in S. cerevisiae containing either SSU1(pr) or ssu1(pr)Δ::ECM34(pr) To determine whether SSU1 over-expression inhibits Cup1p formation in high copper medium, a label free quantitative proteomic analysis was undertaken. A total of 3261 proteins were identified in each of the extracts of the same samples used in the analysis of gene expression, collected two days following inoculation (Fig 4D). The ssu1(pr)Δ::ECM34(pr) mutation in AWRI 4052 resulted in a 107-fold increase (P adj = 3.23e-13) in the abundance of Ssu1p relative to AWRI 3471 when grown in a medium containing standard copper concentrations [41, file T11]. The increase in Ssu1p abundance highlights the effectiveness of the ECM34 promoter in driving SSU1 expression. Differential abundance of an additional 77 proteins was observed relating to over-expression of SSU1 alone. The contrast in protein abundance between AWRI 4052 and AWRI 3471 in high copper medium identified the following pathways as being over-represented with differentially abundant proteins; primary alcohol biosynthesis, glycoprotein biosynthesis, apoptotic process, polysaccharide metabolic process and energy derived by oxidation of organic compounds (Fig E in S1 Text). There was limited overlap between transcriptomic and proteomic profiles. Six genes/proteins, excluding SSU1/Ssu1p, were common to both data sets (JLP1, SSA4, MSC1, AAC3, THI20, THI73). There was a 13-fold increase in Ssu1p abundance (P adj = 0.008), which is a decrease from that observed in a low copper medium [41, file T12]. There was strong evidence for an increase in the abundance of Jlp1p (log 2 FC = 7.8, P adj = 4.5e-6), supporting the idea that sulfur limitation is exacerbated in AWRI 4052 exposed to copper stress. However, there was no evidence for the differential abundance of other identifiers of sulfur-limited growth (Agp3p, Pdcp, Yro2p and Soa1p). A general decrease in the abundance of proteins involved in thiamine biosynthesis (Thi4p, Thi6p, Thi20p, and Thi73p) or thiamine precursor scavenging (Snz2p) was also observed, which is consistent with the gene expression profile of this strain under copper stress. As expected, the protein with the largest change in abundance in response to high-copper concentrations (high–low copper contrast) was Cup1p with a 10.4 (P adj = 2.4e-13) and 13.4 (P adj = 1.7e-13) Log 2 FC in AWRI 3471 and AWRI 4052, respectively. The 6.5-fold relative increase in Cup1p abundance (P adj = 0.02) in the ssu1(pr)Δ::pECM34(pr) background demonstrates that inability to produce sufficient metallothionein is not an explanation for copper sensitivity in this strain, but does suggest that increased copper stress is being perceived. There was no evidence (P adj > 0.5) for the differential abundance of proteins whose transcripts had previously been shown to be copper responsive (Oye3p, Fet3p, Ftr1p, Gto3p, Hsp12p, Fet4p and Sod1p). An interesting feature of the AWRI 4052 ‘high copper’- ‘low copper’ contrast was the apparent increase in abundance of MF(alpha)1 protein (Log 2 FC = 9.7, P = 0.004) despite strong evidence for a decrease in the abundance of its transcript (Log 2 FC = -2.8, P = 2.8e-268). MF(alpha)1 has previously been shown to be a copper-binding protein [51,52] but we cannot explain its increased abundance in SSU1 over-expressing cells growing in high copper concentrations. In summary, an examination of relative protein abundance data supports the idea that SSU1 over-expression exacerbates sulfur limitation in copper challenged yeast and rules out Cup1p limitation as a cause of copper sensitivity. It leaves open the question about the role of thiamine biosynthetic and uptake functions in copper sensitivity.

SSU1 over-expression does not limit hydrogen sulfide production The contribution of H 2 S metabolism to copper sensitivity was assessed by plasmid-based over-expression of MET3, MET14 and MET16 ([MET+]) in the two strains AWRI 3471 and AWRI 4052. It was reasoned that if H 2 S was limited due to a decrease in the concentration of its precursor, then an increase in flux through the pathway should rectify this condition and restore copper tolerance. Growth in high and low copper medium was unaltered in either genetic background by increased expression of MET3, MET14 and MET16 with no alleviation of copper sensitivity evident in AWRI 4052 (Fig 6A and 6B). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 6. Effect of copper concentration and MET3-MET14-MET16 over-expression on growth and, SO 2 and H 2 S production in AWRI 3471 and AWRI 4052. (A) Growth of AWRI 3471 in medium containing 0.25 mg/L and 10 mg/L copper. (B) Growth of AWRI 4052 in medium containing 0.25 mg/L and 10 mg/L copper. (C) The concentration of SO 2 produced by AWRI 3471 and AWRI 4052 in medium containing 0.25 mg/L and 10 mg/L copper. (D) The concentration of H 2 S produced by AWRI 3471 and AWRI 4052 in medium containing 0.25 mg/L copper. Blue lines and bars; standard defined medium, red lines and bars; defined medium containing 10 mg/L copper, filled lines; plasmid carrying KANMX marker only, dashed lines; plasmid carrying KANMX, MET3, MET14 and MET16. Points and bars show mean of three replicates with error bars indicating standard deviation. https://doi.org/10.1371/journal.pgen.1010692.g006 Three-way analysis of variance indicated that yeast strain accounted for most of the variance in SO 2 production (P < 0.0001) (Fig 6C), an observation that is explained by SSU1 over-expression in AWRI 4052. Therefore, the effect of copper and MET 3/14/16 expression was analysed separately by strain using two-way ANOVA (Table C and Table D in S1 Text). There was strong evidence for a MET 3/14/16 dependent increase in SO 2 production (Fig 6C) in both AWRI 3471 (P < 0.0001) and AWRI 4052 (P = 0.046) with mean increases of 12.7 mg/L, [95CI, 9.7, 15.7] and 12.0 mg/L [95CI, 0.2, 23.8] respectively in low copper medium. Growth in high copper medium suppressed the MET 3/14/16 dependent changes in total SO 2 accumulation. In the absence of MET 3/14/16 over-expression, growth in high copper increased total SO 2 accumulation in AWRI 4052 only (mean increase = 13.3 mg/L [95CI, 1.9, 25.5], P = 0.024). The observed MET 3/14/16 dependent increase in total SO 2 production in low copper medium indicates that the modifications introduced into these strains successfully increase flux through the sulfate assimilation pathway. However, the data also suggests that copper may suppress either the activity of MET 3/14/16 or efflux of SO 2 via SSU1. Total H 2 S production could only be measured in low copper medium due to complexation between H 2 S and copper in the high copper condition. In standard defined medium AWRI 4052 produced more H 2 S than AWRI 3471 (mean diff = 5.7 mg/L, [95CI, 3.8, 7.5], P = 0.001). There was strong evidence (P = 0.0004) that over-expression of MET 3/14/16 increased total H 2 S production in AWRI 3471 (mean diff = 20.3 mg/L [95CI, 15.2, 25.5]). In AWRI 4052 there was a smaller increase in MET 3/14/16 dependent total H 2 S production (4.7 mg/L, [95CI, 1.5, 7.7], P = 0.01) (Fig 6D). This result suggests that SSU1 over-expression constricts flux of sulfur through to H 2 S. It should be noted that the parent of AWRI 3471 carries a mutation in MET2 (R301G) that decreases H 2 S production, presumably as a result of an improvement in the efficiency of H 2 S condensation with O-acetyl-homoserine [53]. This mutation is present in both AWRI 3471 and AWRI 4052 and explains the almost complete lack of H 2 S production in the AWRI 3471 [NatR+] empty vector strain. Overall, there is no evidence that H 2 S limitation is a causative factor of copper sensitivity in AWRI 4052.

[END]
---
[1] Url: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1010692

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.