Molecular mechanism of ethanol fermentation inhibition via protein tyrosine nitration of pyruvate decarboxylase by reactive nitrogen species in yeast

Protein tyrosine nitration (PTN), in which tyrosine (Tyr) residues on proteins are converted into 3-nitrotyrosine (NT), is one of the post-translational modifications mediated by reactive nitrogen species (RNS). Many recent studies have reported that PTN contributed to signaling systems by altering the structures and/or functions of proteins. This study aimed to investigate connections between PTN and the inhibitory effect of nitrite-derived RNS on fermentation ability using the yeast Saccharomyces cerevisiae. The results indicated that RNS inhibited the ethanol production of yeast cells with increased intracellular pyruvate content. We also found that RNS decreased the activities of pyruvate decarboxylase (PDC) as a critical enzyme involved in ethanol production. Our proteomic analysis revealed that the main PDC isozyme Pdc1 underwent the PTN modification at Tyr38, Tyr157, and Tyr344. The biochemical analysis using the recombinant purified Pdc1 enzyme indicated that PTN at Tyr157 or Tyr344 significantly reduced the Pdc1 activity. Interestingly, the substitution of Tyr157 or Tyr344 to phenylalanine, which is no longer converted into NT, recovered the ethanol production under the RNS treatment conditions. These findings suggest that nitrite impairs the fermentation ability of yeast by inhibiting the Pdc1 activity via its PTN modification at Tyr157 and Tyr344 of Pdc1.

Nitric oxide (NO·) is a highly permeable free radical produced in a wide range of cells and tissues. NO• is now known to regulate various biological functions in cells when present at low or appropriate concentrations, including neurotransmission, gene expression, protein translation, apoptosis, and signal transduction [1][2][3][4] . In terms of industrial applications, the enhancement of NO• production has previously been reported to improve stress tolerance and fermentation ability in baker's yeast 5 . Exogenous NO· also increases an oxidative stress response involving a negative feedback system to regulate the intracellular ROS levels in the fission yeast Schizosaccharomyces pombe 6 . However, at high concentrations, NO· and various NO-derived molecules known as reactive nitrogen species (RNS) can cause cellular dysfunctions and/or cell death via a process known as nitrosative stress 7 . For example, Almeida and co-workers reported that NO-induced cell death was mediated by S-nitrosation of GAPDH as an apoptotic trigger 8 .
After translation, proteins can be edited enzymatically or non-enzymatically by post-translational modifications (PTMs), which dynamically alter the structures, stabilities, and functions of proteins 9 . Protein tyrosine nitration (PTN) is one of the RNS-dependent PTMs. PTN is a chemical process that selectively introduces a nitro group (NO 2 ) to an ortho carbon to a phenolic hydroxyl group on the aromatic ring of a tyrosine (Tyr) residue, leading to the conversion of Tyr into 3-nitrotyrosine (NT) 10 . PTN is caused by several combinatorial reactions of RNS with reactive oxygen species (ROS) 11 , including the production of peroxynitrite (ONOO ) 12 . The protein structure, nitration mechanism, and cellular environment of the protein all can influence the biological specificity of PTN 13 . PTN contributes to cellular signaling mechanisms because of its specificity, reversibility, and controlled rate 14,15 . It affects protein structures and functions by the steric hindrance of its nitro group, the deprotonated phenolic hydroxyl group caused by its reduced pKa value, or the crosstalk with Tyr phosphorylation signaling, leading to an immune response modulation 16 . It also plays a role in the pathogenesis of inflammatory responses, www.nature.com/scientificreports/ cytoskeletal dysfunction, platelet activation, age-related macular degeneration (AMD), and neurodegenerative disorders 11,[17][18][19][20][21] . There have been few reports about the demodification of PTN. However, a previous study showed that an NT-containing peptide was reduced to a peptide containing 3-aminotyrosine (AT), which is a reduced form of NT, in cells, probably through enzymatic activities 22 . Numerous studies on PTN in mammals and plants have been conducted. However, the NO· signaling mechanisms mediated by PTN in the lower eukaryotic yeasts have remained unclear. A previous study identified yeast mitochondrial proteins including proteins in the TCA cycle and respiratory system, such as aconitase and isocitrate dehydrogenase, as a target of PTN 23 . Furthermore, 14 nitrated proteins have been identified by proteomic analysis using the yeast Saccharomyces cerevisiae treated with a mating pheromone; these include proteins related to gene expression, actin cytoskeleton organization, cell wall organization, and spermidine transport 24 . However, the effects of PTN on these yeast proteins have not been investigated. S. cereivisae has been used in the fermentation processes to produce various goods, including alcoholic beverages, breads, and biofuels. When yeast cells promote fermentation to synthesize ethanol, sugars are converted into pyruvate via the glycolytic pathway and then further metabolized to ethanol by the two additional enzymatic reactions, decarboxylation of pyruvate to acetaldehyde by pyruvate decarboxylase (PDC) and reduction of acetaldehyde to ethanol by alcohol dehydrogenase (ADH) 25 . The ethanol production is crucial for oxidation of NADH to maintain the appropriate NAD + /NADH redox balance for glycolysis. S. cerevisiae cells prefer the fermentative pathway when the glucose concentration is above 150 mg/L to the respiration via the TCA cycle even under aerobic conditions, which is known as the Crabtree effect 26,27 . Only a small amount of acetaldehyde is converted into acetate by aldehyde dehydrogenase 28 . PDC is a cytosolic enzyme that catalyzes the conversion of pyruvate into acetaldehyde and carbon dioxide with thiamin diphosphate (TPP) and Mg 2+ as cofactors 29 . PDC plays an essential role in the diversion of pyruvate toward fermentation and lipid synthesis in some bacteria, yeasts, fungi, and plants. The genome of S. cerevisiae contains three genes encoding PDC, PDC1, PDC5, and PDC6. Pdc1, a major isozyme consisting of 563 amino acids, is highly expressed under most physiological conditions 30 . The catalytically active enzyme in yeast is composed of four identical subunits of Pdc1; however, the dimer is required as the smallest catalytically active unit 31 . Pdc1 in S. cerevisiae represents the allosteric regulation of catalytic activity, in which its substrate pyruvate binds to the regulatory site to induce the protein conformational changes 32 .
Molasses is a viscous by-product of the processing of sugar cane or sugar beets into sugar. It has been widely used as one of the most cost-effective carbon sources to produce ethanol and other fermentative products in the industry. In some types of molasses, such as beet molasses, the toxic amounts of organic or inorganic compounds inhibit the rate of ethanol production. Also, molasses contains various amounts of nitrate and nitrite. A previous study showed that the concentration of nitrite, which has been identified as a yeast fermentation inhibitor, in undiluted beet molasses was around 9 mM 33 . Nitrite dramatically induces dose-dependent growth inhibition, the decreased ATP levels, and the inhibition of some enzyme activities, such as glyceraldehyde-3-phosphate dehydrogenase and glutamate dehydrogenase, when present at concentrations above 0.65 mM with pH below 5.0 [34][35][36][37] . The combination of an acidic condition and nitrite, called acidified nitrite, results in RNS production. Nitrous acid, a protonated form of nitrite under an acidic condition, is converted into dinitrogen trioxide, followed by heterolysis to NO· and nitrogen dioxide radicals 38 . In contrast, the pH of the culture medium during fermentation with yeast using glucose can fall below pH 2 37 . These results suggest that the inhibitory effect of nitrite during ethanol production is caused by nitrosative stress derived from acidified nitrite.
In this study, we analyzed PTN and its effect on ethanol production in S. cerevisiae cells under nitrosative stress conditions induced by acidified nitrite. Our findings show that S. cerevisiae cells produced less ethanol, which was caused by the decreased Pdc1 activity via PTN of Pdc1 at specific Tyr residues, when exposed to nitrosative stress. These findings provide new insights into the molecular mechanism underlying the negative effect of nitrite in medium on ethanol synthesis via the RNS-mediated PTN.

Results
RNS produced from acidified nitrite decreases ethanol production and PDC activity. We first examined the effect of nitrosative stress on ethanol production in S. cerevisiae. When yeast cells were treated with 1 mM NaNO 2 at pH 4.0 for 1 h, the ethanol concentration in culture medium was significantly lower than that of untreated cells (Fig. 1A). This result indicates that nitrosative stress inhibits ethanol production, consistent with the previous study 33 . Subsequently, the intracellular concentration of pyruvate, the end product of glycolysis, was measured. Interestingly, the pyruvate content dramatically increased when yeast cells were treated with acidified nitrite (Fig. 1B). To investigate the mechanism by which RNS generated from acidified nitrite decreases the ethanol production ability, we measured the activity of the enzymes involved in the ethanol synthetic pathway from glycolysis, PDC and ADH. Importantly, the PDC activities in the total protein extracts from the cells treated with acidified nitrite were approximately 50% lower than those from untreated cells (Fig. 1C), although ADH was fully active regardless of stress treatment (Fig. 1D). These results suggest that PDC activity was specifically decreased by RNS, resulting in the lower ethanol content.
Pdc1 is the target of PTN at positions Tyr38, Tyr 157, and Tyr344. Pdc1, the main isozyme of three PDCs in S. cerevisiae 39 , was further analyzed as a protein responsible for the decreased PDC activity under nitrosative stress conditions. Western blotting analysis revealed that nitrite treatment did not change the protein level of Pdc1 (Fig. S1). This result raised the possibility that PTMs of Pdc1 are involved in the decrease of its activity. In addition, we subjected the total protein extract from cells treated with acidified nitrite to a proteomic analysis using the stable isotope labeling with amino acids in cell culture (SILAC) method 40 . As a result, we identified Pdc1 with 76.02% of the protein sequence coverage, including 14 of the 17 Tyr residues of Pdc1 (Fig. S2 www.nature.com/scientificreports/ were detected, suggesting that the intracellular reduction of NT occurred as previously reported 22 . Therefore, we concluded that these AT-containing peptides were derived from Pdc1 with PTN. From these results, we detected three AT-containing peptides in the sample treated with nitrite, which respectively contained PTN at Tyr38, Tyr157, or Tyr344, although no quantitative information was available because the signals of AT-containing peptides were detected in only either the control or nitrite-treated samples (Tables 1 and S1-S3, Fig. S3). The raw data of proteomic analysis were deposited to a public repository Japan ProteOme STandard Repository (jPOS-Trepo, https:// repos itory. jpost db. org/) and available with the accession number of JPST001473 41 . Subsequently, to confirm the PTN modification of Pdc1, yeast cells expressing Pdc1 fused with the myc7His epitope tag at its C-terminus were subjected to a pull-down assay, followed by western blot with anti-NT antibody. The results revealed that an acidified nitrite treatment increased the PTN level of Pdc1, consistent with the above results of proteomic analysis, but the increase in the PTN level was abolished by dithionite reduction (Figs. 2 and S4). These results indicated that Pdc1 was nitrated in response to NO· treatment in yeast cells.

RNS induce PTN and the inactivation of Pdc1. A wild-type (WT) enzyme of the recombinant Pdc1
was prepared using a heterologous expression system in Escherichia coli. The purified Pdc1 was treated with ONOOand then its activity was analyzed. The result indicated that the incubation with ONOOinhibited the enzymatic activity of Pdc1 in a dose-dependent manner (Fig. 3A). Samples prepared by the same method were analyzed by western blot using anti-NT antibody. The results showed that the ONOOtreatment induced dosedependent increases of the PTN level on Pdc1 without any decrease in the protein amount of Pdc1 (Figs. 3B and S5). These results suggest that the RNS-dependent PTN inactivates Pdc1.
PTN at Tyr157 or Tyr344 inhibits Pdc1 activity. In order to examine the effect of PTN on the Pdc1 activity at the Tyr residues that were identified as AT in proteomic analysis, the Pdc1 variants with the site-specific incorporation of NT at Tyr38, Tyr157, or Tyr344, which were named Tyr38NT-, Tyr157NT-, or Tyr344NT-Pdc1, respectively, were expressed and purified using the heterologous expression system in E. coli. Our immunoblotting analysis with anti-NT antibody showed a clear signal from each Pdc1 variant, indicating that these variants successfully contained NT (Figs. 3C and S6). Next, we measured the enzymatic activities of these Pdc1 variants (Fig. 3D). We found that the activity of Tyr38NT-Pdc1 was almost the same as that of the WT enzyme. ADH activity in the cell-free extract from yeast exposed to acidified nitrite was measured. The values represent the averages and standard deviations from three independent experiments. **ρ < 0.01 by Student's t-test. Table 1. Identified peptides of Pdc1 containing trace of PTN in the proteomic analysis. Y* = Tyr residues detected as AT, # = SILAC labeling of Lys (K) and Arg (R), ǂ = N-terminal acetylation.

Prevention of PTN at Tyr157 or Tyr344 recovers ethanol production under RNS treated conditions.
To clarify the physiological significance of PTN of Pdc1 at each Tyr residue, we further measured the ethanol content in the culture medium of yeast strains. When Tyr157 or Tyr344 was substituted to phenylalanine (Phe) to prevent the PTN formation on these residues, the inhibitory effect of RNS on ethanol production was significantly abolished, although the ethanol content without acidified nitrite treatment was not affected by these www.nature.com/scientificreports/ substitutions (Fig. 4A). Additionally, it was found that the protein level of Pdc1 was not altered by either acidified nitrite treatment or the amino acid substitution to Phe on each PTN site (Figs. 4B and S7). These results suggest that PTN on Tyr157 and Tyr344 is critical for the inhibition of ethanol production. Therefore, we concluded that the inhibition of Pdc1 activity via its PTN modification at Tyr157 and Tyr344 induced by RNS led to the decreased ethanol productivity in S. cerevisiae cells.

Discussion
Here, we revealed the molecular mechanism by which nitrosative stress affected the ethanol production in yeast cells, which involved PTN of Pdc1 at Tyr157 or Tyr344 inhibiting the enzymatic activity of Pdc1. To the best of our knowledge, this is the first biochemical characterization of the functions of PTN modification on yeast proteins. This study also provides new insights for future research on PTN in yeast.
The previous studies reported that nitrite affected ethanol production and that the culture medium was acidified by prolonged cultivation. Because nitrite is converted into RNS under acidic conditions, the acidified nitrite treatment used in this study could be a model experimental condition to mimic yeast cultivation using molasses, which contains a large amount of nitrite. Our findings clarified the mechanisms by which nitrite impaired the ethanol fermentation of yeast, at least in part. In terms of industrial applications, the yeast strains expressing the Pdc1 variants with Tyr157Phe and Tyr344Phe substitutions are a promising tool to boost ethanol production, because these strains are resistant to the RNS-induced inhibition of ethanol production. Since it is considered that molasses fermentation proceeds under nitrosative stress conditions, such yeast strains may increase the ethanol productivity.
We showed that the intracellular pyruvate content was elevated under nitrosative stress conditions (Fig. 1B). There could be three explanations for pyruvate accumulation under nitrosative stress conditions. First, a decrease in Pdc1 activity causes pyruvate accumulation, as observed in this study. Second, the inhibition of the mitochondrial respiratory chain by the nitrosative stress could induce ATP depletion. The resulting higher AMP/ ATP ratio could potentially stimulate glycolysis via the AMP-activated protein kinase to produce ATP, leading to the acceleration of pyruvate synthesis 42 . Third, the pyruvate accumulation could be caused by the impairments of the TCA cycle that consumes pyruvate in mitochondria. The metabolites in glycolysis and the TCA cycle were not measured in this study, and thus it is unsure whether these metabolic pathways were affected by RNS. However, a previous study reported that the activities of some TCA cycle enzymes, including α-ketoglutarate dehydrogenase and pyruvate dehydrogenase, were inhibited by nitric oxide in Salmonella enterica 43 . Moreover, it has been reported that the yeast aconitase and NAD + -dependent isocitrate dehydrogenase were nitrated in vivo, although the biological significance of these findings has yet to be elucidated 23 . These previous findings imply that pyruvate was accumulated by the inhibition of the TCA cycle enzymes via their PTN modifications, in addition to the inhibition of Pdc1.
Pdc1 is the main isozyme of PDC, which is highly expressed under most conditions 39 and is also induced by the presence of glucose 44 . Aside from the transcriptional control of PDC1, PTMs such as phosphorylation are another means of modulating the function of Pdc1. Dephosphorylation of Pdc1 mediated by the serine/ threonine-protein phosphatase increases the ethanol fermentation and the Pdc1 activity by altering the apparent affinity of Pdc1 for TPP and pyruvate in S. cerevisiae cells 45 . We did not analyze the phosphorylation status www.nature.com/scientificreports/ of Pdc1, but it is possible that not only PTN but also the protein phosphorylation are responsible for the control of the Pdc1 enzymatic activity. Our proteomic analysis identified the peptides containing AT, which is a reduced form of NT, but not those containing NT itself, even though the PTN of Pdc1 was confirmed by immunoblotting analysis (Tables 1 and S1-S9, Figs. 2 and S4). We also did not detect the immonium ion with m/z of 181.06, which is a useful footprint of NT-containing peptides, from the peptides assigned to Pdc1 46 . A previous research reported that the PTMs with high electron affinity, such as PTN, prevent the cleavage of the peptide backbone, limiting the sequence information from the MS/MS analysis with electron capture dissociation or electron transfer dissociation 47,48 . On the other hand, it has been reported that proteins with PTN modification were reduced in cells to the corresponding proteins with AT 22 . From this information, it is possible that Pdc1 was nitrated in vivo and then partly reduced to its AT-containing form, which alone was detected in our proteomic analysis.
As shown in this study, PTN of Pdc1 at Tyr157 and Tyr344 was responsible for the decrease in the enzymatic activity of Pdc1 (Fig. 3D). The findings in a previous study 49 suggest that PTN of Pdc1 is likely to contribute to the alteration of the Pdc1 structure and thus its activity. The crystal structure of Pdc1 has been determined at 1.71 Å of resolution in the presence of its substrate pyruvate and cofactor TPP (PDB ID code: 2VK1). In the Pdc1 structure, neither Tyr157 nor Tyr344 directly interacts with the binding site for pyruvate or TPP. On the other hand, because the aromatic ring of Tyr157 is located near Gly66 and Met187, the addition of a nitro group to the side chain of Tyr157 is likely to induce a steric hindrance with these residues (Fig. 5A). Tyr344 is buried inside the protein molecule and interacts with several residues, including Asn213, Pro214, and Val347 or Phe240, through hydrogen bonds or π-π stacking, respectively (Fig. 5B). These intramolecular interactions of Tyr157 or www.nature.com/scientificreports/ Ty344 imply that the PTN modifications at these sites disturb the correct direct interaction of these Tyr residues, leading to an indirect alteration of the whole structure and enzymatic activity of Pdc1. Our enzyme assay using the site-specifically NT-incorporating Pdc1 exhibited that PTN at either Tyr157 or Tyr344 decreased its activity (Fig. 3D), indicating that both PTN modifications are capable of inhibiting the Pdc1 activity. On the other hand, the ethanol production was recovered by either Tyr157Phe or Tyr344Phe substitution, suggesting that PTN modification at both Tyr residues is required to reduce the ethanol synthesis in yeast cells (Figs. 4B and S7). An additive inhibition of Pdc1 activity by the combined PTN modifications at these two sites might be necessary to inhibit the ethanol production. The Tyr157 and Tyr344 residues are completely conserved among yeast species (Fig. 5C), suggesting that the inhibition of ethanol production by PTN of Pdc1 is a common phenomenon in yeasts. In contrast, Tyr344 is also conserved in plants, although Tyr157 is not. Plants might possess some functions mediated by the PTN modification at Tyr344. In contrast to our findings, a previous study using 0.5 mM NaNO 2 for acidified nitrite treatment found that yeast cells exposed to nitrosative stress produced more ethanol in a rich medium by increasing the alcohol dehydrogenase activity under stress conditions, by using 0.5 mM NaNO 2 for acidified nitrite treatment 50 . Lower concentrations of acidified nitrite could produce lower concentrations of RNS. A milder nitrosative stress might be insufficient to cause Pdc1 nitration but adequate to inhibit the mitochondrial electron transport chain, and then activate glycolysis to produce ATP 51 . Therefore, a lower concentration of nitrite under acidic conditions might increase ethanol production through the enhancement of glycolytic metabolic flow, providing pyruvate a substrate for ethanol synthesis.
The alcohol dehydrogenase involved in ethanol production is an essential enzyme for NAD + regeneration in yeast cells 52 . Our findings here suggest that the nitrosative stress impaired ethanol synthesis by the decreased supply of acetaldehyde through the inhibition of Pdc1. This process could lead to an increase in the NADH/ NAD + ratio by inhibiting the oxidation of NADH to NAD + by ADH. NO dioxygenase (NOD) encoded by the YHB1 gene, which utilizes not only NADPH but also NADH as its reducing force, is the main enzyme in NO· detoxification to nitrate 53 . It is possible that the increased concentration of NADH contributes to the nitrosative stress tolerance via the upregulation of NOD activity.

Methods
Strains, plasmids, and medium. Table 2, 3, or S1 represents a list of the strains, plasmids, or primers used in this study, respectively. The yeast S. cerevisiae laboratory strain X2180-1A (MATa SUC2 mal mel gal2 CUP1) was a host strain to construct the yeast strains used in this study 54 . To generate the strain expressing Pdc1 fused with myc7His-tag at its C-terminus (Pdc1-myc7His) from the genome (the PDC1-myc7His strain), the DNA fragment amplified using the primers listed in Table S1 and plasmid pYM46 (Euroscarf) as a template was introduced into strain X2180-1A by the LiAc method 55,56 . For the construction of strain lacking both the PDC1 and URA3 genes (pdc1Δ ura3Δ), the DNA fragment amplified using primers listed in Table S1 and the plasmid pYM20 as a template was introduced to the strain lacking the URA3 gene constructed in the previous study 57 . In order to construct the strain lacking the ARG1 and LYS1 genes (arg1∆ lys1∆), which is capable to synthesize neither arginine (Arg) nor lysine (Lys), for the SILAC-based proteomic analysis, strain X2180-1A was transformed with the DNA fragment amplified from plasmid pFA6a-natNT2 (Euroscarf), followed by the further transformation with the fragment from plasmid pFA6a-hphNT1 (Euroscarf), using the primers shown in Table S1. Each positive clone was screened on the solid medium containing antibiotics G418 or hygromycin B. E. coli DH5α or BL21 (DE3) strain was used to construct plasmids or express Pdc1, respectively.
To construct the plasmid expressing Pdc1-myc7His under the control of its original promoter, the DNA fragment from 1000 bp upstream to downstream of PDC1 was amplified from the genomic DNA of the PDC1-myc7His strain using the primers listed in Table S1 and then introduced to pRS416 by the In-Fusion cloning system (Clontech) following the manufacturer's protocol. The resultant plasmid pRS416-PDC1-myc7His was introduced to pdc1Δ ura3Δ cells, generating the strain expressing Pdc1-myc7His (the WT-PDC1 strain). The expression plasmid pET55-PDC1 used to produce Pdc1 fused with N-terminal strep-tag and C-terminal Histag in E. coli was constructed as follows. The DNA fragment of PDC1 coding sequence was amplified from the genomic DNA of strain X2180-1A using the primers listed in Table S1 and then introduced to the entry plasmid pDONR221 by the BP reaction in Gateway technology (Invitrogen). The resultant plasmid pDONR221-PDC1 was subjected to the LR reaction with the plasmid pET55-DEST in Gateway technology, generating pET55-PDC1. Each reaction in Gateway technology was performed following the manufacturer's protocol.
Yeast cells were cultured at 30 °C in YPD (1% yeast extract, 2% peptone, and 2% glucose) or SD (0.17% yeast nitrogen base without amino acids and ammonium sulfate, 0.5% ammonium sulfate, and 2% glucose, pH 4.0) Construction of the plasmid expressing the PDC1 mutants. The site-directed mutagenesis to express each mutant of PDC1 was performed as previously described 58 , using the primers listed in Table S1 and pRS416-PDC1 to generate pRS416-PDC1 Tyr157Phe -myc7His or pRS416-PDC1 Tyr344Phe -myc7His, which expresses Pdc1-myc7His with Tyr157Phe or Tyr344Phe substitution, respectively. The plasmid pDONR221-PDC1 was used as a template with the primers listed in Table S1 to introduce the amber codon at the position corresponding to Tyr38, Tyr157, or Tyr344. The resultant pDONR221-PDC1 with mutation was subjected to the LR reaction in Gateway technology to generate the pET55-PDC1 plasmids with amber codon, pET55-PDC1 Tyr38NT , pET55-PDC1 Tyr157NT , and pET55-PDC1 Tyr344NT . Successful mutagenesis was confirmed by DNA sequencing.
Acidified nitrite treatment of yeast. Yeast cells cultured in SD medium (pH 4.0) until the exponential growth phase were treated with 1 mM NaNO 2 for 1 h at 30 °C. After treatment, cells were collected, frozen with liquid nitrogen, and stored at −80 °C for further analyses.
PDC and ADH activity. The PDC reaction was performed as previously described 59 . The assay mixture consisted of 40 mM imidazole-HCl buffer (pH 6.5), 0.2 mM thiamine pyrophosphate, 0.15 mM NADH, 88 U/ mL alcohol dehydrogenase, 5 mM MgCl 2 , and the cell-free extract or purified Pdc1. The reaction was started by the addition of 50 mM pyruvate. The mixture for ADH reaction consisted of 100 mM phosphate buffer (pH 7.6), 8 mM acetaldehyde, and the cell-free extract. Both enzymatic activities were analyzed spectrophotometrically by monitoring a decrease of absorbance at 340 nm derived from NADH over time at 30 °C, using a spectrophotometer DU-800 (Beckman). One unit of the specific activity of PDC or ADH was defined as the amount of enzyme to oxidize 1 μmol of NADH per min.
Pull-down assay. After  The unified concentrations of protein extracts were separated by SDS-PAGE under the non-reducing conditions using Laemmli's sample buffer without 2-mercaptoethanol and then transferred onto PVDF membranes. After the treatment with appropriate primary and secondary antibodies, the immunoblot was visualized using Amersham ECL prime reagents (GE Healthcare) and ImageQuant™ LAS4000 (GE Healthcare). For the negative control of PTN detection, the PVDF membrane was incubated with 10 mM sodium dithionite for 20 min before the treatment with anti-NT antibody. In this study, anti-NT antibody Nitro-Tyrosine Antibody (Cell Signaling Technology, 9691S), anti-myc antibody c-Myc (9E10) (Santa Cruz Biotechnology, sc-40), anti-Pgk1 antibody PGK1 Monoclonal Antibody (22C5D8) (Invitrogen, 459,250), or anti-mouse IgG antibody Anti-Mouse IgG (H + L), HRP Conjugate (Promega, W4021) was used as a primary or secondary antibody, respectively. Some membranes were cut out before the incubation with primary antibodies. www.nature.com/scientificreports/ Expression and purification of WT and NT incorporated Pdc1. In order to express the Pdc1 variants incorporating NT site-specifically, plasmid pET55-PDC1 Tyr38NT , pET55-PDC1 Tyr157NT , or pET55-PDC1 Tyr344NT , which expresses Tyr38NT-, Tyr157NT-, or Tyr344NT-Pdc1, respectively, constructed as described above was used. E. coli BL21 (DE3) strain harboring each of these plasmids and the plasmid pDule-3-nitroTyrosine (5B) (Addgene) 60 were cultured in M9CA medium at 37 °C. When OD 600 reached 0.6, 0.1 mM isopropyl-β-Dthiogalactopyranoside and 1 mM NT were added and E. coli cells were further cultured for 16 h at 16 °C. After harvesting cells by centrifugation, the C-terminally His-tagged Pdc1 was purified by Ni Sepharose™ 6 Fast Flow (Cytiva) following the manufacturer's protocol.
Sample preparation for mass spectrometry with SILAC method. Proteomic analysis with SILAC method was performed to identify and quantify the NT-containing peptides. The arg1∆lys1∆ strain was cultured in SD medium at pH 4.0 containing Arg and Lys or Arg (guanido-13 C) and Lys (4,4,5,5-D 4 ) for the control or nitrite treatment condition, respectively. Cells cultured until the exponential phase were treated with 1 mM NaNO 2 for 1 h and harvested. The same amount of extracted protein from each sample was mixed and subjected to SDS-PAGE under the non-reducing condition. Sliced gels were treated with 10 mM dithiothreitol at 56 °C for 45 min, 55 mM iodoacetamide at room temperature for 10 min, and then digested with Trypsin Gold, Mass Spectrometry Grade (Promega). The extracted peptides were filtered with a 0.45 μm syringe filter for the following LC-MS/MS analysis. Quantification of ethanol content. Yeast cells in the exponential growth phase were incubated in SD medium with or without nitrite for 1 h. The ethanol concentration of supernatant was determined using Ethanol Colorimetric/Fluorometric Assay Kit (Biovision), following the manufacturer's direction. Ethanol content in each sample was normalized by OD 600 of the culture medium.

LC-MS/MS analysis.
Quantification of pyruvate content. Yeast cells cultured until the logarithmic phase were treated with the acidified nitrite for 1 h, harvested, and suspended in the pyruvate assay buffer, provided in Pyruvate Assay Kit (Sigma-Aldrich). Cell-free lysate extracted using Multi-beads shocker was centrifuged at 15,000 g for 15 min at 4 °C and then the supernatant was subjected to ultrafiltration with Amicon Ultra (10 KDa) (Merck), to remove insoluble materials and proteins. Pyruvate concentration in the resultant solution was measured using Pyruvate Assay Kit. Pyruvate content in each sample was normalized by OD 600 of the culture medium.