Proteomic analysis demonstrated transcription factor YRR1 gene deletion in Saccharomyces cerevisiae enhances its resistance to vanillin through the upregulation of transcriptional activator Haa1, coactivator Mbf1, and proteasome assembly chaperone Tma17 expression

Background: Vanillin is one of the major phenolic inhibitors in Saccharomyces cerevisiae for cellulosic ethanol production. Deleting transcription factor gene YRR1 improves vanillin resistance by promoting some translation-related processes that were confirmed at the transcription level in our previous studies. However, the known genes regulated by Yrr1 are not related to translation process. Therefore, in this work, we investigated the effects of proteomic changes on vanillin stress and YRR1 deletion to provide different perspectives from transcriptome analysis for comprehending the mechanisms of YRR1 deletion in yeast protective response to vanillin. Results: In wild-type cells, vanillin reduced the numbers of ribosomal proteins quantities and thereby inhibited cells’ translation. YRR1 deletion changed the quantities of 121 proteins which have no overlaps with transcriptomic changes. Of 112 proteins were up-regulated; 48 of 112 up-regulated proteins are involved in stress response, translational and basal transcriptional regulation. Fermentation data showed that the overexpression of HAA1 , MBF1 , and TMA17 , which encode transcriptional activator, coactivator, and proteasome assembly chaperone, respectively, enhanced resistance to vanillin in S. cerevisiae. Conclusions: These results showed how YRR1 deletion increase vanillin resistance at protein level. This may advance our understanding of molecular mechanisms for YRR1 deletion to protect yeast from vanillin stress and offer novel targets of genetic engineering for designing inhibitor-resistant ethanologenic yeast strains.


Background
Saccharomyces cerevisiae is a traditionally competitive cell factory for bioethanol and the production of other desired chemicals due to its superior robustness and its ease of genetic manipulation [1,2].
Lignocellulosic materials have been viewed as a major resource for bioethanol production worldwide because they are massive, renewable, and do not compete with food resources [3]. However, the pretreatment of lignocellulose is inevitable to generate many inhibitors that hamper microorganism growth and fermentation, such as organic acid, furans, and phenolic compounds [4]. Vanillin, one of the major phenolic aldehyde compounds, has been considered a more effective repressor of ethanol fermentation than other lignocellulose-derived toxic compounds [5]. Thus, it is essential to understand the physiological mechanism of how vanillin inhibits S. cerevisiae growth and fermentation during the assembly of vanillin-resistant S. cerevisiae cell factories for bioethanol production.
In our previous work, we reported that the deletion of the multi-drug resistance-related transcription factor gene YRR1 remarkably improved S. cerevisiae vanillin resistance. The transcriptome analysis was initiated between the YRR1-deletion strain and the wild-type one either with or without vanillin stress. The outcome showed that vanillin repressed ribosome biogenesis and rRNA processing in the wild-type strain [6]. However, the deletion of YRR1 could prevent this repression. The genes related to ribosome biogenesis and rRNA processing are more up-regulated in the YRR1-deletion strain compared to the wild-type under vanillin stress. Ribosomes and rRNA are vital for translation and protein synthesis [6]. Thus, YRR1 was considered to impact translation and protein synthesis. In addition, several studies have revealed that many inhibitors, including vanillin, furfural, 5hydroxymethyl furfural (HMF), and acidic stress, repress translational initiation by blocking ribosome assembly, which leads to the accumulation of cytoplasmic processing bodies (P-bodies) and stress granules (SGs) [7][8][9].
Therefore, global mRNA profiles alone are not sufficient to present the changes in the biological system because translational, post-translational regulation mechanisms, protein degradation, and other events have not been taken into account. It is a common phenomenon that transcripts do mismatch with protein levels [10,11].
Taken together, to gain a complete insight into the mechanism behind how YRR1 deletion protects yeast from vanillin stress, we conducted a quantitative proteomic analysis of yeast treated with specific vanillin stress and deleting YRR1, respectively. These proteomic data were analyzed in combination with previous transcriptome data obtained under the same conditions. This study demonstrated that when facing vanillin stress, S. cerevisiae cut down translation by reducing ribosomal proteins to save energy and promote energy production pathways for survival. The YRR1 deletion protective mechanism of vanillin resistance up-regulates proteins related to the stress response as well as basal transcriptional and translational regulation. The overexpression of three proteins (transcriptional activator Haa1, proteasome assembly chaperone Tma17 and transcriptional coactivator Mbf1) remarkably improved the resistance of yeast to vanillin. These results enriched the mechanism of the YRR1 deletion protective response to vanillin stress and offered novel strategies for genetic engineering manipulation to obtain high inhibitor-resistant microbes for second-generation fuel ethanol technology.

Results And Discussion
Quantitative proteomics reveals remarkable alterations in the protein abundance response to vanillin To investigate the changes in protein abundance induced by vanillin stress perturbation, the proteomic differences were examined in the BY4741 laboratory strain in samples treated with and without vanillin. Based upon our stringent data mining criteria, a threshold of 1.3-fold was applied to filter out proteins that are differentially expressed. Among them, 135 proteins were up-regulated, and 83 proteins were down-regulated in response to vanillin (P-value < 0.05). Functional annotation of all the identified proteins was conducted based on the Gene Ontology (GO) database [12] and the Kyoto Encyclopedia of Genes and Genomes (KEGG) database [13]. The KEGG pathway enrichment of the different expressed proteins (DEPs) suggests that activated protein genes are mainly involved in the functional categories of energy production, amino acid metabolism and transport, the biosynthesis of secondary metabolites and others (Fig. 1a). In contrast, repressed protein genes are largely involved in the functional categories of ribosome, amino acid metabolism and biosynthesis of secondary metabolites (Fig. 1b).

Proteomes exhibited visible differences from transcriptomes in the presence of vanillin
Considering of the repression of the translation initiated by vanillin, mRNA most likely cannot represent a protein situation. Thus, the proteomic dataset was compared to data from a previous transcriptomic outcome (GEO accession number: GSE89854) to investigate the correlation between mRNA and protein levels [6]. Overlaps (i.e., homo-directional) between DEPs and differentially expressed genes (DEGs) are shown in the Venn diagrams (Fig. 2a). Only 23.8% of these DEPs exhibited a homo-directional change at the level of their transcript counterparts. A scatter plot was used to compare high-quality protein with the transcript expression ratios for all 218 DEPs (Fig. 2b).
The positive Spearman rank correlation coefficient (Sr) of 0.47 that was calculated for these 218 DEPs revealed a weak correlation between transcriptome and proteome. This phenomenon is likely due to vanillin's ability to repress translation [7].
The DEPs and DEGs enriched (p < 0.05) in KEGG pathways are listed in Fig. 2c. Glycolysis-related genes/proteins and ABC transporters were up-regulated both at the mRNA and protein levels, which indicates that vanillin stimulates these two categories at the transcriptional level. Both the "pentose phosphate pathway (PPP)" and the "ribosome biogenesis pathway" were down-regulated only at the mRNA level. Ribosomal proteins and RNA polymerases were significantly decreased only at the protein level, which indicates that vanillin might repress the translation of the corresponding proteins or influence their post-transcriptional regulation.
Ribosomal proteins and rRNA processing-related proteins significantly decreased under vanillin stress Although the transcriptome showed that 13 genes related to ribosome biogenesis were notably downregulated (Table 1), the corresponding proteins exhibited no changes in protein abundance [6].
However, 20 ribosomal proteins were sharply suppressed by vanillin, including 5 small subunits of ribosomes (40S) and 15 large ribosomal subunits (60S) ( Table 1). Moreover, one ribosome biogenesisrelated protein (Rlp24) [14], three proteins (Esf1, Nsa2, Pop8) involved in rRNA processing, two RNA polymerase I subunits (Rpa12, Rpa190), one RNA polymerase III subunit (Rpc11), tRNA 4demethylwyosine synthase Tyw1, and ATPase for tRNA processing Nbp35 also displayed a visible decrease in protein abundance. These down-regulated proteins were constituents of the cellular translational machine, which indicates that vanillin blocks translation efficiency and protein synthesis by reducing the quantity of ribosomal proteins available. This phenomenon coincided with the facts that vanillin inhibits translation initiation, so the cells do not need many more ribosomes. Therefore, the reduction of ribosome biogenesis is likely representative of the cell's effort to conserve energy for survival since ribosome biogenesis demands a great supply of energy [15,16]. a DEPs ratio (BY4741with/without vanillin stress) which is less than 1 means down-regulation at protein level. b log2 ratio (BY4741with/without vanillin stress) which is less than 0 means down-regulation at mRNA level.
Vanillin stress increased the quantity of proteins involved in energy generation As shown in Table 2, a number of enzymes involved in energy generation pathways, such as glycolysis, oxidative phosphorylation, and the PPP, were dramatically up-regulated. Six glycolytic enzymes (Pyc1p, Tdh1, Tdh2, Hxk2, Eno1, and Gpm2) associated with glycolysis were remarkably increased. In particular, Hxk2, the predominant enzyme that catalyzes the first irreversible step of glycolysis [17], was up-regulated more than 1.5 times. Hxk2 is first of two rate-limiting enzymes in glycolysis [18]. Moreover, the proteins related to adenosine triphosphatase (ATP) synthesis, including the five ATP synthase subunits (Atp20, Atp7, Atp15, Atp14, and Tim11), three cytochrome c oxidase subunits (Cox4, Cox12, and Cox8), four ubiquinol cytochrome-c reductase subunits (Qcr9, Qcr2, Cor1, and Qcr6), and one ATPase (Pma2), also increased significantly. The up-regulated glycolysis and oxidative phosphorylation pathways tend to supply more ATP for cellular survival as the detoxification of vanillin consumes energy. For example, when facing vanillin stress, one strategy is to up-regulate some ATP binding cassette (ABC) transporters, such as Pdr5 and Snq2 (up-regulated 1.68 and 3.78 times, respectively, at the protein level), to extrude vanillin from the cells. These two transporters were also up-regulated at the mRNA level [6]. a DEPs ratio (BY4741with/without vanillin stress) which is less than 1 means down-regulation. Two enzymes (Gnd1 and Sol3) of the PPP increased over 1.5 times, especially the enzyme Gnd1, which generates nicotinamide adenine dinucleotide phosphate (NADPH) during its catalysis [19]. The PPP is the major source of NADPH, a vital cofactor for vanillin conversion to vanillyl alcohol, which is less toxic than vanillin [20,21]. In the face of vanillin stress, another resistance strategy of S. cerevisiae is utilization of NADPH-dependent alcohol dehydrogenase (ADH) enzymes, such as Adh6/7, for vanillin transformation [22,23].
Thus, the lower number of ribosomal proteins and ribosome biogenesis proteins to conserve energy as well as the higher level of glycolysis and oxidative phosphorylation pathway for producing more energy are thought to be an adaption of yeast for survival in response to vanillin stress.
Comparative proteomics analysis of the deletion of YRR1 Our previous study found that vanillin repressed ribosome biogenesis and pathways. In addition, the deletion of YRR1 could lower this repression and lead to improving resistance of S. cerevisiae to vanillin [6]. Thus, YRR1 was considered to be related to ribosome biogenesis or translation pathways.
However, the known YRR1 targets are mostly related to ABC transporters and permeases, not the two aforementioned pathways. Further, the deletion of YRR1 triggered only eight genes (YBR230W-A, CAR2, FMP45, YCL048W-A, SCS3, UTH1, PMP3, and YIL002W-A), which have no effect on vanillin resistance expressed differently at the mRNA level (data not shown). To investigate the protein quantity changes caused by deleting YRR1, a proteomic analysis was conducted between BY4741(yrr1Δ) and BY4741. The outcome revealed that there were 121 DEPs, including 112 upregulated and 9 down-regulated proteins that had no overlap with their corresponding mRNA level.
The up-regulated DEPs were divided into functional categories according to the MIPS Functional Catalog and mainly felt in transcription, translation, and energy generation categories ( Table 3). The functions of differentially expressed proteins are listed in Table 4. Significance was estimated with FunSpec (http://funspec.med.utoronto.ca/) based on the hypergeometric distribution of the MIPS functional categories of the differentially expressed proteins compared to the yeast proteome (P-value smaller than 0.01, bonferroni correction applied). The deletion of YRR1 stimulated quantity changes of translational proteins In the present proteomic data, a series of rRNA processing proteins were up-regulated after deleting YRR1. Among them, Nop6, Nhp2, Naf1, Lsm5, Nsr1, and Fyv7 participate in the synthesis of the 40S (small) ribosomal subunit. Cgr1 is involved in processing rRNA for the 60S ribosomal subunit [24]. The elements of rRNA synthesis also displayed increases, including the subunit of RNA polymerase I Rpa34, the RNA polymerase III assembly protein Rbs1, and two transcription factors (Hmo1 and Abf1) involved in activating rRNA and ribosomal protein transcription under the control of the TOR pathway [25][26][27]. Moreover, several regulators of translation, including three translation initiation factors (Tif11, Hcr1, and Gis2), were also up-regulated. Considering its ability to repress ribosome genesis and translation by vanillin stress, the deletion of YRR1 might strengthen translational elements to promote vanillin resistance in yeast. Intriguingly, three translational suppressors also increased, including two 4E-binding proteins, Caf20 and Eap1, which inhibit translational initiation by competing with translation initiation factor eIF4G to bind with eIF4E [28]. The other translational repressor was Stm1, which was previously demonstrated to repress translational elongation by limiting the interaction of elongation factor eEF3 with ribosomes and stalling ribosomes [29]. Further, one recent study indicated that Stm1 also worked as a ribosome preservation factor in response to nutrient stress and facilitated ribosomal protein synthesis rates once nutrients are restored [30]. These three proteins assist with the formation of P-bodies [31,32]. P-bodies are viewed as reservoirs of mRNA after translation inhibition, and they are later released to reenter translation [33,34]. It is likely that the absence of YRR1 can protect yeast cells against vanillin stress by promoting the formation of P-bodies to store translation elements and saving energy for the denovo synthesis of translation machinery. The deletion of YRR1 promotes RNA polymerase II-directed transcription Table 4 Differentially expressed proteins induced by the deletion of YRR1  of TATA-containing genes [35][36][37][38]. TATA-containing genes tend to be stress-responsive [39,40].
Moreover, four transcription elongation factors (Gim5, Yke2, Spn1, and Npl3) also increased in conditions of protein abundance. Npl3, an mRNA-binding protein, is a positive transcription elongation factor shown to interact with the carboxy terminal domain (CTD) and have a direct stimulatory effect on the elongation activity of the polymerase [41]. Therefore, these results indicate that the deletion of YRR1 promotes TATA-containing gene transcription by the promotion of basal transcription factors to improve survival under vanillin stress.
The deletion of YRR1 improved S. cerevisiae's resistance to vanillin by increasing Tma17, Mbf1, and Haa1 at the protein level The results of above proteomic analysis led us to investigate possible novel mechanisms between the deletion of YRR1 and a vanillin-tolerant phenotype. First, two proteins attracted our attention since they exhibited reverse variation in two proteomes: Rtc3, which is involved in RNA metabolism, and plasma membrane protein Hsp12, which plays a role in maintaining membrane organization. They were both down-regulated by vanillin stress but up-regulated by YRR1 deletion. Their reverse variation is likely the reason why YRR1 deletion promotes the resistance of yeast to vanillin stress.
However, the overexpression of these proteins had no effect on vanillin resistance (data not shown).
We investigated several up-regulated proteins that were associated with transcription and the stress response in a YRR1-deletion strain as candidates. These candidates were six stress response-related proteins (Crz1, Dsk2, Hsp12, Haa1, Def1, and Tma17) and four transcription-related proteins (Taf10, Mbf1, Gis2, and Tif11). To test these proteins'/genes' function in a vanillin-resistant phenotype, these genes were overexpressed in BY4741. The overexpression of HAA1, TMA17, and MBF1 enhanced the strain growth under vanillin stress ( Fig. 3 and Fig. 4). The maximum specific growth rates of the transformants of HAA1, TMA17, and MBF1 were 18%, 33%, and 15% faster than the control in the presence of 6 mM vanillin, respectively ( Table 5). The specific consumption rates for vanillin of strains overexpressing HAA1, TMA17, and MBF1 were 0.041, 0.041, and 0.029 g g − 1 h − 1 , respectively.
Compared to the 0.034 g g − 1 h − 1 of the control (Table 5), only HAA1 and TMA17 significantly increased the specific vanillin consumption rates. MBF1 accelerated growth more than vanillin reduction. However, other genes related to translation did not significantly improve vanillin resistance (data not shown). Haa1 is a critical transcription factor involved in the acid stress adaption of yeast cells [42]. It also contributes to vanillin resistance, which indicates that vanillin and acid stress share some common features, such as stimulating reactive oxygen species accumulation and inhibiting translation [7,9,43,44]. Mbf1 is a transcriptional coactivator that mediates transcriptional activation by bridging the DNA-binding region of the activator and TBP (TATA binding protein) [45]. As mentioned above, TATAcontaining genes are mainly associated with stress responses. Therefore, it was expected that Mbf1 could improve S. cerevisiae vanillin resistance by facilitating stress-response gene transcription.
Under stress conditions, Tma17 (also named as Adc17) is the most important chaperone for yeast cellular proteasome biogenesis and assembly. The deletion of TMA17 aggravates proteasome defects and reduced the cells' fitness. The maintenance of adequately functional proteasomes is vital for cells and organisms to adapt to changes and survive in their environment [46]. Thus, the deletion of YRR1 stimulated greater Tma17 expression to allow cells to adapt to vanillin stress better. These results suggested that the deletion of YRR1 could up-regulate HAA1, TMA17, and MBF1 at the protein level to enhance vanillin resistance.

Conclusions
This work investigated the changes in general protein composition induced by vanillin stress via quantitative proteome. The results indicated that S. cerevisiae adapted to vanillin stress by repressing ribosomal protein abundance and enhancing energy production. YRR1 deletion up-regulated proteins related to stress response, transcriptional and translational regulation, which provide different perspectives from transcriptome for comprehending the mechanisms behind the YRR1 deletion protective response in yeast. The finding that three proteins (Haa1, Tma17, and Mbf1) could enhance vanillin resistance by their overexpression will provide novel strategies for designing inhibitorresistant ethanologenic yeast strains for bioethanol production.
Synthetic complete medium (SD) or Sc-URA medium (5 g L −1 ammonium sulfate, Sangon, China, 1.7 g L −1 yeast nitrogen base without amino acids, Sangon, China, CSM or CSM-URA, MP Biomedicals, Solon, OH, USA) was used for activation, batch fermentation, and protein sample preparation of recombinant strains by adding 20 g L −1 glucose. Vanillin was added to the medium as indicated. All of the cultures were incubated at 30℃.

Protein sample preparation for proteomic analysis
Yeast cells of BY4741 and BY4741(yrr1Δ) were harvested during the exponential growth phase (OD 600 =1.6-2.0) in SD medium with or without of 5 mM vanillin, respectively. All of the proteomic analyses were carried out in biological triplicate. The sample was ground by liquid nitrogen into cell powder and then transferred to a 5-mL centrifuge tube. Four volumes of lysis buffer (8 M urea, 1% protease inhibitor cocktail) were added to the cell powder, followed by sonication three times on ice using a high-intensity ultrasonic processor. The remaining debris was removed by centrifugation at 12,000 g for 10 min at 4°C. Finally, the supernatant was collected, and the protein concentration was determined using a BCA kit according to the manufacturer's instructions. For digestion, the protein solution was reduced with 5 mM dithiothreitol for 30 min at 56°C and alkylated with 11-mM iodoacetamide for 15 min at room temperature in darkness. The protein sample was then diluted by adding 100 mM TEAB to reduce the urea concentration to < 2 M. Finally, trypsin was added at a 1:50 trypsin-to-protein mass ratio for the first digestion overnight; the second 4-h digestion process had a 1:100 trypsin-to-protein mass ratio. A peptide of trypsin digestion was labeled with TMT, according to the manufacturer's protocol for a TMT kit.

High-performance liquid chromatography fractionation and mass spectrometric analysis
The labeling peptides were fractionated by high pH reverse-phase high-performance liquid chromatography (HPLC) using an Agilent 300 Extend C18 column (5-μm particles, 4.6 mm ID, 250 mm length). Briefly, the peptides were first separated into 60 fractions with a gradient ranging from 8-32% acetonitrile (pH 9.0) over 60 min. Then, the peptides were combined into 18 fractions and dried by vacuum centrifugation. The tryptic peptides were dissolved in solvent A (0.1% formic acid and 2% acetonitrile); the gradient was comprised of an increase from 6% to 20% solvent B (0.1% formic acid in 90% acetonitrile) over 20 min, from 20% to 33% in 11 min, and up to 80% in 4 min then holding at 80% for the last 3 min, all at a constant flow rate of 320 nL/min on an EASY-nLC 1000 UPLC system.
The peptides were subjected to an NSI source followed by tandem mass spectrometry (MS/MS) in a Q Exactive TM Plus (Thermo) coupled online to the UPLC. The electrospray voltage applied was 2.0 kV.
The m/z scan range was 350-1800 for a full scan, and intact peptides were detected in the Orbitrap at a resolution of 70,000. The secondary mass spectrometry scan range was set as a fixed first point of 100 m/z, and the resolution was 17,500. The data acquisition mode was a data-dependent scanning (DDA) program; the first 20 peptides with the highest signal intensity are selected to enter the HCD collision cell, and then 30% of the fragmentation energy is used for fragmentation after the first-stage scanning; the same step performed once more during the secondary mass spectrometry analysis. To improve the effective utilization of the mass spectrometer, the automatic gain control (AGC) is set as 5E4, the signal threshold is set as 1E4, and the maximum injection time is set as 200 ms with 30 seconds of dynamic exclusion time.

Database searching and data analysis
The resulting MS/MS data were processed using the Maxquant search engine (v.

Fermentation
A single colony was precultured in 3 mL SC-URA at 30°C for 24 h. Then the cultures were transferred into 10 mL fresh medium with an initial OD 600 at 0.2 and cultured overnight. Next, the cells were inoculated into 100-mL flasks containing 40 mL fermentation medium in the presence or absence of 6 mM vanillin with an initial OD 600 of 0.2. The fermentation was carried out at 30°C and 200 rpm.

Analyses of extracellular vanillin
The concentrations of vanillin and vanillyl alcohol were determined by HPLC Waters e2695 (Waters, USA) equipped with an Xbridge TM -C18 column (Waters, USA). The peaks were detected at room temperature using ultraviolet detection (PDA-2998) at 210 nm with a mobile phase containing 40% absolute methanol (Chromatographic grade, Fisher Chemical, USA) supplied at a flow rate of 0.6 mL min −1 [6].

Analyses of Physiological Parameters
The biomass concentrates were investigated to determine their correlation with OD 600 and the dry cell weight (DCW). Different strains exhibited different coefficients between OD 600 and the DCW (data not shown). The maximum growth rates were regarded as the linear regression coefficients of the ln (OD 600 ) versus time during the exponential growth phase. Specific consumption rates of vanillin were calculated as previously described [48].

Ethics approval and consent to participate
Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.