Methionine Metabolism Alters Oxidative Stress Resistance via the Pentose Phosphate Pathway

1The ability of cells to counteract oxidative stress is fundamental for survival in an ever changing environment and is implicated in growth and aging. Metabolism is of importance for cellular tolerance to oxidants, providing the reducing power for the antioxidant machinery, including the glutathione and thioredoxin systems, and acting as a main source of intracellular oxidants itself. A central player in stress resistance is the amino acid methionine. Methionine is a direct target of reactive oxygen species (ROS), as its sulfur can be oxidized to sulfoxide; furthermore, via repair of methionine sulfoxide through methionine sulfoxide reductases, methionine can act as scavenger and protect cells from oxidative stress. Recent studies analyzing proteomic methionine usage show this amino acid to be enriched in protein at sites of major free radical production, such as the mitochondrial respiratory chain. Methionine has therefore been exploited by cells, acting as an ROS scavenger to protect proteins from oxidation, and has been implicated in caloric restriction phenotypes and aging.

2Genetic experiments have connected methionine biosynthesis to the pentose phosphate pathway (PPP), a pathway with essential roles in the antioxidative metabolism. This connection appears to be predominantly caused by the redox cofactor nicotinamide adenine dinucleotide phosphate (NADP). De novo methionine biosynthesis, via the assimilation of inorganic sulfate, requires three molecules of NADPH per molecule of methionine, and the PPP is key to regenerating NADPH and maintaining redox balance. Two PPP enzymes responsible for NADPH production are glucose 6-phosphate dehydrogenase (G6PDH) and 6-phosphogluconate dehydrogenase (6PGDH). When budding yeast is deleted for G6PDH (ZWF1), the rate limiting enzyme of this NADPH producing oxidative PPP, cells are no longer able to synthesize methionine and become methionine auxotrophs. Growth of zwf1Δ cells therefore depends on supplementation with organic sources such as cysteine and methionine.

3Methionine plays an essential role in oxidant resistances as it can be oxidized and has been implicated in caloric restriction and aging. We observed altered activity in the oxidative branch of the pentose phosphate pathway (PPP) after increasing the supplementation of methionine, detecting elevated levels of PPP metabolites, and increased abundance of the nicotinamide adenine dinucleotide phosphate (NADPH) producing enzyme 6-phosphogluconate dehydrogenase (6PGDH). Methionine preincubation also increased cellular tolerance to the thiol oxidizing agent diamide in dependency of the oxidative pentose phosphate. At least some oxidative stress resistance phenotypes caused by methionine hence involve the PPP.

4We hence questioned whether the effect of methionine on the PPP was related to the stress response. Direct NADPH quantification in yeast provides limited information about the PPP as (i) a major fraction of NADP(H) is localized to the vacuole or protein bound, (ii) different NADPH sources exist, and (iii) NADPH is readily oxidized upon cell lysis. We therefore addressed this problem by combining mass spectrometry-based metabolite quantification, protein quantification, and genetics. First, PPP metabolite abundance was determined using a targeted LC-MS/MS method, which gives absolute quantities for the majority of PPP metabolites (Table 1). Wild-type cells supplemented with methionine showed a distinct increase in PPP metabolites downstream of the NADPH producing enzyme 6PGDH (Fig. 1A, left panel). Auxotrophic yeast, unable to synthesize methionine by themselves, possessed similar concentrations confirming that the effect is mediated by methionine supplementation (Fig. 1A, right panel). Next, protein expression was analyzed by SWATH-MS, a target proteomics method based on LC/MS-MS. We detected a clear upregulation of 6PGDH in methionine-supplemented cells (Fig. 1B). Methionine treatment thus increases both metabolite concentration and enzyme expression in the oxidative PPP.

5Finally, we tested for a causal relationship between methionine and the PPP by testing oxidant resistance upon methionine supplementation. The thiol oxidizing agent diamide was considered ideal for testing PPP-dependent oxidant resistance, as cells have increased resistance to this oxidant when PPP activity is increased; however, in contrast to hydrogen peroxide or other peroxides, diamide does not block glycolysis or activate the PPP directly. When methionine supplementation was increased from 0 to 200 mg/L, resistance to diamide was increased (Fig. 1C). The upregulated 6PGDH cannot be fully deleted in yeast as it is an essential gene; therefore, to prevent flux in the oxidative PPP, ZWF1 was deleted instead. In zwf1Δ cells, methionine no longer increased diamide tolerance (Fig. 1C).

6Stress resistance phenotypes that follow methionine supplementation hence involve the oxidative PPP. Effects of methionine on oxidative metabolism, stress signaling, and aging have thus to be seen in the context of an altered activity of this NADPH providing pathway.

7All yeast strains and plasmids used are listed in Table 2. For metabolomic and proteomic studies, YSBN5, a prototrophic haploid variant of Saccharomyces cerevisiae S288c, was used alongside its met15Δ derivative. Deletion of MET15 in this strain was performed by homologous recombination using a kanMX marker (Table 3). To determine the effect of methionine supplementation on oxidative tolerance, BY4741 complemented with the centromeric vector (minichromosome) pHLUM (Addgene ID #40276) was used alongside its zwf1Δ derivative. Yeast was cultivated if not otherwise indicated at 30°C, in minimal supplemented synthetic media (SM; YNB yeast nitrogen base [6.8g/L; Sigma]), with 2% glucose (Sigma) as the carbon source.

8Intracellular levels of sugar phosphates were quantified using a previously published LC-MS/MS method. 7.5 OD595 units of cells (YSBN5 MET15 ± methionine [20 mg/L] or YSBN5 met15Δ + methionine [20 mg/L]) were harvested in the exponential growth phase (OD595 = 1.5 ± 0.05) following cold methanol quenching. Cell pellets were resuspended in 200 μL of extraction buffer (75:25 [v/v] acetonitrile:methanol, 0.2% formic acid) and lysed during three FastPrep-24 (MP Biomedicals) cycles for 20 s at 6.5 m/s. After centrifugation for 5 min at 16,000 g, the pellet was extracted again with 200 μL ultra high pressure liquid chromatography-grade water. Combined supernatants from both extractions were evaporated in a Concentrator plus SpeedVac (Eppendorf), resuspended in 100 μL 7% acetonitrile, centrifuged, and submitted to LC-MS/MS analysis. One microliter of the metabolite extract was injected on a C8 column (ZORBAX SB-C8 Rapid Resolution HD, 2.1 × 100 mm, 1.8 μm (Agilent); column temperature: 20°C) for liquid chromatography separation (Agilent 1290). Water/acetonitrile mixtures (A: 10% [v/v] acetonitrile, B: 50% [v/v] acetonitrile) that were used as mobile phases contained 750 mg/L octylammoniumacetate as the ion pairing reagent. After 3.5 min of isocratic flow at 12% acetonitrile, a 2.5-min gradient to 38% acetonitrile was used to elute the sugar phosphate analytes, followed by a washing step to 42% acetonitrile for 0.5 min and re-equilibration at starting conditions. Quantification was achieved with an online coupled triple quadrupole mass spectrometer (Agilent 6460) operating in the selective reaction monitoring (SRM) mode. Identification of metabolites was ensured by comparison of retention times and fragmentation patterns with commercially available standards. For each individual compound, optimal analytical conditions (SRM transitions, ionization, and fragmentation energies) were determined separately. MassHunter Workstation software package (Agilent) was used for data analysis. Repeatedly recorded external calibration curves were used to calculate absolute metabolite concentrations. The SRM transitions used are shown in Table 4.

9Analysis of PPP enzyme abundances was performed using SWATH-MS as previously described. In short, cells equaling 10 OD595 of an exponentially growing yeast culture (YSBN5 MET15 ± methionine [20 mg/L] or YSBN5 met15Δ + methionine [20 mg/L]) were harvested and tryptic peptides prepared in accordance with the RapidACN method. LC-MS/MS analysis was performed on a Tandem Quadrupole Time-of-Flight mass spectrometer (AB/Sciex TripleTOF5600) equipped with a Turbo V ion source and 25 μm emitter needles (AB/Sciex), on line coupled to a HPLC system (Eksigent nanoLC 425) operating in the microflow mode. Chromatographic separation was achieved by direct injection of 10 μg yeast digest supplemented with 0.5× HRM-Kit (Biognosys) onto an analytical column (Eksigent ChromXP, 3 μm, C18CL, 300Å, 150 × 0.3 mm) followed by a linear gradient (0–28% acetonitrile/0.1% formic acid in 26 min) at a flow rate of 3 μL/min. Data acquisition was performed with 29 × 16 m/z SWATH windows, a range of 400–850 m/z, and 40 ms accumulation time resulting in 1.2 s cycle time. Extracted ion chromatograms were then extracted with the help of Spectronaut software (version 5.0.5900; Biognosys) by querying the data with an ion library, constructed by operating the mass spectrometer in the data-dependent acquisition mode and repeatedly injecting tryptic yeast digests prefractionated using high pH reverse phase chromatography. The chromatographic setup for the ion library was as outlined above. Data were validated by manually removing interfering fragment ions using Skyline software, and the three most abundant fragment ions were used for calculating peptide fold changes. To correct for amount of sample injected, data were normalized using the fold change of glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.12; TDH1) as a control. Protein fold changes were calculated by averaging three peptide fold changes per protein. Transitions of peptide DYFGAHTFR for 6PGDH (EC 1.1.1.44; GND1) taken from SWATH chromatographic spectra in Skyline were then illustrated with R package “ggplot2.” Fold change in protein abundance for YSBN5 MET15 and met15Δ strains + methionine was then based on protein abundance for YSBN5 MET15 strains untreated.

10For the oxidant tolerance test following methionine treatment, BY4741 ZWF1 and zwf1Δ strains, complemented with the centromeric vector pHLUM, were grown for two nights in SM ± methionine (ZWF1 with 0 and 200 mg/L, and zwf1Δ with 20 and 200 mg/L of L-methionine [Sigma], respectively, as the zwf1Δ strain is auxotrophic for organic sulfur (8)). Cell cultures were then normalized to 3.6e+06 cells in 200 μL SM and spotted in 1:5 serial dilutions on SM solid media with l-methionine (200 mg/L) ± diamide (Sigma). Growth was documented after 3 days incubation at 30°C.