Strains, Plasmids, Media, and Genetic Manipulations
The S. cerevisiae
strains and plasmids used in this study are listed in Supplementary Tables 1 and 2, respectively. All strains from our laboratory are isogenic with TB50. Yeast manipulations, including cell cultures, sporulation, tetrad dissections, and genetic techniques, were carried out essentially as described by Guthrie and Fink (1991)
. The media were YPD (1% yeast extract, 1% peptone, 2% dextrose, plus 2% agar for solid media) and minimal synthetic medium (SD; yeast nitrogen base at 6.7 g/l, 2% dextrose, relevant amino acids and 2% agar for plates). YP medium was used for the glucose depletion experiment. SDS in YPD was 0.01%. Cells were treated with rapamycin at 200 ng/ml final concentration (added from a 1 mg/ml stock solution in 90% ethanol–10% Tween20) and/or 8-Bromo-cAMP at 5 mM final concentration (from 250 mM stock solution in water). Before 8-Bromo-cAMP treatment, cells were centrifuged and resuspended in 5 ml of the required medium. In most experiments, yeast strains carrying a plasmid were precultured in SD medium lacking the corresponding amino acids for plasmid maintenance and subsequently diluted into YPD medium. Cells were then grown for 4–5 h (to OD600
about 0.8) before treatment. For SILAC labeling, yeast cells were grown in SD medium containing 13
-arginine and 13
-lysine (Cambridge Isotope Laboratories, Andover, MA). Transformations of S. cerevisiae
cells were according to the lithium acetate method with single-strand carrier DNA and dimethyl sulfoxide (DMSO; Hill et al., 1991
). All deletion and genomically tagged strains were constructed by PCR-based gene targeting (Wach et al., 1994
). One-step site directed mutagenesis was performed as described previously (Zheng et al., 2004
Fluorescence microscopy and indirect immunofluorescence on whole fixed cells was performed as described previously (Schmelzle et al., 2004
Protein Extraction, Immunoprecipitation, Western Blotting, and Phosphatase Treatment
Protein extracts were prepared from exponentially growing cells. At least 25 OD cells were broken in a bead beater in lysis buffer (PBS containing 0.5% Tween 20 and 10% glycerol) containing 1× Roche protease inhibitor cocktail, 1 mM PMSF, and phosphatase inhibitors (10 mM NaF, 10 mM NaN3, 10 mM p-nitrophenyl phosphate, 10 mM sodium pyrophosphate, and 10 mM β-glycerophosphate). hemagglutinin (HA)-, MYC-, or tandem affinity purification (TAP)-tagged proteins were immunoprecipitated from cell extracts with protein A-Sepharose beads coupled to anti-HA antibody (12C2A), protein G-Sepharose beads coupled to anti-MYC antibody (1-9E10.2), or human IgG-Sepharose beads (GE Healthcare, Waukesha, WI), respectively. For the phosphatase experiment, beads were further washed with phosphatase buffer (50 mM Tris-Cl, pH 7.5, 1 mM MgCl2), resuspended in 100 μl phosphatase buffer, and incubated for 10 min at 30°C. Twenty units of calf intestine alkaline phosphatase was added, and the reaction was incubated for 15 min 30°C. For the gel-shift assay, the acrylamide-bisacrylamide ratio for SDS-PAGE was changed to 172:1. The primary antibodies for Western blotting were as follows: anti-HA (mouse monoclonal, 12C2A), anti-MYC (mouse monoclonal, 1-9E10.2), anti-protein A (rabbit polyclonal, Sigma-Aldrich, St. Louis, MO), anti-BCY1 (goat polyclonal, Bcy1 [yN19] sc-6765, Santa Cruz Biotechnology, Santa Cruz, CA), anti-RxxS/T (rabbit polyclonal phospho-[Ser/Thr] PKA substrate antibody no. 9621, Cell Signaling, Beverly, MA), anti-RRxS/T (phospho-PKA Substrate [RRXS/T; 100G7E], rabbit mAb no. 9624, Cell Signaling), anti-MPK1 (goat polyclonal, Mpk1 [yC-20] sc-6803, Santa Cruz Biotechnology), anti-phospho MPK1 (rabbit polyclonal, phospho-p44/42 MAPK [Thr202/Tyr204] antibody no. 9101, Cell Signaling). (The phosphorylated residues are underlined.)
Identification of Phosphorylation Sites in BCY1
Wild-type TB50a cells or cells expressing N-terminally H- tagged BCY1 (SA094) were grown in 500 ml YPD to OD600 nm of 1 and treated with rapamycin (200 ng/ml) or drug vehicle for 90 min. Cells were broken with glass beads with RIPA buffer (25 mM Tris-Cl, pH 7.5, 150 mM NaCl, 1% Nonidet NP-40, 1% Na-deoxycholate) supplemented with 1× Roche protease inhibitor cocktail and phosphatase inhibitors (10 mM NaF, 0.2 μM okadaic acid, and 20 nM calyculin A). HA-tagged BCY1 was immunoprecipitated from 40 mg of total protein extract with 1 μl of monoclonal anti-HA antibody (HA.11; Covance, Princeton, NJ) at 4°C for 18 h followed by incubation with 100 μl of Dynabeads protein G (Invitrogen, Carlsbad, CA) for 4 h at 4°C. The beads were washed and HA-BCY1 was eluted from the beads with 10 μl 3 M urea, and 20 mM DTT for 10 min at room temperature. HA-BCY1 was digested with trypsin for 18 h at 37°C. Proteolysis was stopped with 8.4 μl acetic acid, and the peptides were dried in a Speed Vac. Phosphopeptides were enriched with TiO2-magnetic beads (Perkin Elmer-Cetus, Waltham, MA) according to the manufacturer's instructions. The final peptides were analyzed by capillary liquid chromatography tandem MS (LC/MS/MS) using a 300SB C-18 column (0.3 × 50 mm; Agilent Technologies, Basel, Switzerland) connected on line to an LTQ-Orbitrap hybrid instrument (Thermo Finnigan, San Jose, CA; see below).
Staining of intracellular glycogen was performed with iodine vapor as previously described (Barbet et al., 1996
MPK1 Kinase Assay
Yeast cells expressing HA-tagged MPK1 were grown in 500 ml YPD to an OD600 nm
of 1 and then treated for 30 min with rapamycin or drug vehicle. Protein extraction, immunoprecipitation, and in vitro kinase assays were done as previously describe with slight modifications (Watanabe et al., 1997
). Modified RIPA buffer (25 mM Tris-Cl, pH 7.5, 150 mM NaCl, 1% Nonidet NP-40, 1% Na-deoxycholate) was used to break the cells, and 20 mg of total protein extract was used for the immunoprecipitation. The TB50a wild-type strain was used as mock control. A fraction of total lysate (20 μg) was kept to check MPK1-HA expression and phosphorylation on Western blots.
Zymolyase Sensitivity Assay
Phosphoproteome Analysis: SILAC Labeling and Protein Extraction
Two 200-ml cultures of YPJ2 cells (derived from TB50a) were grown at 30°C in SD medium supplemented with either 30 mg/l 12C6-arginine and 12C6-lysine (“light” culture) or 13C6 l-arginine and 13C6,15N2 l-lysine (“heavy” culture) to an optical density (600 nm) of ~0.7. The heavy culture was treated for 15 min with 200 nM rapamycin, whereas the light culture was treated mock-treated with the vehicle for 15 min. To produce reliable quantitative data, four biological replicates were performed, always starting from a single yeast clone. In one of the four experiments the labeling was swapped to check that the labeling influences neither protein expression nor phosphorylation. After rapamycin treatment, both cell cultures were centrifuged at 3500 × g for 10 min at 4°C, and the cell pellets were washed with ice-cold water.
The cell pellets were individually resuspended in 2 ml ice-cold lysis buffer, containing 100 mM Tris-HCl, pH 7.5, 2.5% SDS, 10% glycerol, 1× protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IN; dissolved in ddH2O), 1× phosphatase inhibitor cocktail 1 (Sigma-Aldrich, dissolved in 100% DMSO) and 1 mM PMSF (AppliChem, Darmstadt, Germany; dissolved in 100% DMSO). Total protein extraction from either light or heavy cultures was performed by bead-beating as described above. The lysates were cleared at 15,000 × g for 10 min at 4°C. Protein concentrations in the extracts were measured with the bicinchonic acid assay (BCA, Sigma-Aldrich). About 2.5 mg of light- or heavy-labeled protein extracts were mixed and after addition of 6× sample buffer were incubated at 95°C for 5 min and subjected to preparative electrophoresis.
Phosphoproteome Analysis: Protein Fractionation and In-Gel Digestion
The mixed protein extracts were separated on a preparative 10% SDS slab gel. After electrophoresis, the gel was stained with SimplyBlue SafeStain (Invitrogen). The gel was then sliced horizontally into 16 regions, and the individual slices were further diced into 1-mm3 cubes. The gel pieces were destained overnight in 1 ml 50% acetonitrile/50 mM NH4HCO3, dehydrated with 500 μl 100% acetonitrile, and dried in a speed-vac. The proteins were in-gel reduced in 1 ml 50 mM NH4HCO3 containing 10 mM DTT at 55°C for 60 min. Alkylation was done in 1 ml 50 mM iodoacetamide (in 50 mM NH4HCO3) in the dark for 30 min. After the gel pieces were washed three times with 1 ml 50% acetonitrile/50 mM NH4HCO3, they were dehydrated with 500 μl 100% acetonitrile, dried in a speed-vac, and then rehydrated on ice for 1 h in 1 ml 50 mM NH4HCO3, pH 8.0, containing 15 ng/μl trypsin (Sigma). Digestion was carried out overnight at 37°C. Supernatants were collected in fresh tubes and the gel pieces were extracted three times with 50% acetonitrile/5% formic acid, followed by a final extraction with 100% acetonitrile. The volume of the individual digests was reduced in a speed-vac to ~10 μl to which 290 μl 1% acetic acid was added. A small drop was spotted onto pH paper, and if necessary the pH was adjusted to 2.0–2.5 with 10% acetic acid.
Phosphoproteome Analysis: Peptide Desalting and Phosphopeptide Enrichment
For phosphopeptide enrichment, the digests were desalted on C18 MacroSpin columns (500 μl packed resin, The Nest Group, Southborough, MA) according to the manufacturer's instructions. The peptides were eluted from the cartridge with 600 μl 60% acetonitrile/1% acetic acid. The eluates were speed-vac dried to ~10 μl to which 90 μl IMAC-buffer (30% acetonitrile/250 mM acetic acid) was added. One microliter of each digest was diluted 200-fold with 2% acetonitrile/0.1% acetic acid, and 10 μl was analyzed by LC/MS/MS for expression analysis.
For phosphopeptide enrichment, 40 μl IMAC slurry (PHOS-Select, Sigma-Aldrich) was washed five times with 1 ml IMAC-buffer and then loaded into a constricted GELoader tip (Thingholm et al., 2006
). The desalted digests were applied to the IMAC columns. The flow-throughs were collected and reloaded five times to ensure maximal binding. The resin was washed three times with 150 μL IMAC-buffer. Bound phosphopeptides were eluted with three successive 70-μl desorptions of 50 mM KH2
, pH 10.0, into Eppendorf tubes containing 30 μl 10% formic acid. The IMAC eluates were desalted on disposable C18 MicroSpin columns (MA100 μl packed volume, The Nest Group) according to the manufacturer's instructions. After elution with 200 μl 60% acetonitrile/1% acetic acid, the volume was reduced in a speed-vac to ~10 μl, and the phosphopeptides were diluted with 40 μl 2% acetonitrile/0.1% formic acid for LC-MS/MS analysis.
Phosphoproteome Analysis: LC-MS/MS Analysis
LC-MS/MS analysis was performed on an LTQ-Orbitrap hybrid instrument (Thermo Scientific). Ten microliters of IMAC eluate was injected with an autosampler (CTC Analytics, Agilent Technologies, Basel, Switzerland) onto a C18 trapping column (300SB C-18 0.3 × 50 mm, Agilent Technologies) that was connected to a separation column (0.1 mm × 10 cm) packed with Magic 300 Å C18 reverse-phase material (5-μm particle size, Michrom Bioresources, Auburn, CA). A linear 80-min gradient from 2 to 50% solvent B (80% acetonitrile/0.1% acetic acid) in solvent A (2% acetonitrile/0.1% acetic acid) was delivered with a Rheos 2200 pump (Flux Instruments, Basel, Switzerland) at a flow rate of 100 μl/min. A precolumn split was used to reduce the flow to ~300 nl/min. The eluting peptides were ionized at 1.7 kV.
The LTQ-Orbitrap was operated in data-dependent mode. A survey scan between m/z 375 and 1600 was acquired in profile mode in the Orbitrap at 60,000 resolution, and the 10 most abundant ions were then selected for fragmentation in the LTQ part of the instrument at normalized collision energy of 35%. To facilitate fragmentation of phosphopeptides, a CID scan was followed by multistage activation. Singly charged ions were omitted from fragmentation, and previously selected ions were dynamically excluded for 25 s. Scan-to-scan calibration was allowed by setting the lock mass to m/z 445.120025 (Olsen et al., 2005
Phosphoproteome Analysis: Database Search, Statistics, and Quantitation
To obtain statistically relevant quantitative phosphoproteomes, four independent SILAC experiments from either untreated or rapamycin-treated cell cultures were collected. The MS data were processed with MaxQuant (version 188.8.131.52;Max Plank Institute of Biochemistry, Martinsreid, Germany) and searched with the Mascot search engine (version 2.2.04, Matrix Science, London, United Kingdom) against a databank containing 31,426 protein sequences (Perkins et al., 1999
; Cox and Mann, 2008
; Cox et al., 2009
). The databank contained forward and reverse S. cerevisiae
sequences as well as common contaminants (Swiss-Prot; from European Bioinformatics Institute; http://www.ebi.ac.uk/uniprot
). Precursor ion and fragment ion mass tolerances were set to 7 ppm and 0.6 Da, respectively. Two missed cleavages were allowed. Dynamic modifications were oxidation for methionine and phosphorylation of serine, threonine, and tyrosine, whereas cysteine carbamidomethylation was set as fixed modification. To increase identification stringency only peptides with a Mascot score above 20 were accepted. In addition, only MS/MS spectra with a posterior error probability (PEP) below 0.1 were accepted (Hilger et al., 2009
). The MaxQuant output was simplified by replacing oxidized with unmodified methionines. In addition, proteins matching the same (phospho)peptides were merged into one hit. Gene products of the Ty retrotransposons and the S. cerevisiae
virus L-A were manually removed. For expression analysis, all peptides belonging to the same protein were merged, whereas for phosphorylation analysis, all redundant phosphopeptides containing the same phosphorylation site(s) were grouped together. To evaluate the biological reproducibility of the four experiments, the ratios of the heavy labels to light labels obtained in one experiment were compared with all other experiments. A high correlation was obtained for all four experiments, indicating good reproducibility throughout the entire study (Supplementary Figure 1A). To find changes in expression or phosphorylation that are above the experimental error, an experiment was done without rapamycin treatment. The SILAC ratios were log2
transformed, and the experimental average and SDs were found to be 0.02 and 0.28 (SDcontrol
), respectively. Then, the SILAC ratios of the four cell cultures were log2
transformed, and for those proteins or phosphopeptides that occurred in more than one experiment, a 95% confidence interval was calculated along with the median of the log ratios to assess the influence of outliers. Up-regulation of expression or phosphorylation was considered significant when the lower confidence limits were more than or equal to 0.56 (twice the SDcontrol
). Likewise, down-regulation of protein expression or the extent of phosphorylation was considered significant when the upper confidence limit was less than or equal to −0.56. Those proteins or phosphopeptides that were found in only one experiment but also that showed a large change in abundance or phosphorylation were accepted only if the log2
-transformed ratios were more than 1.12 or less than −1.12 (four times SDcontrol
). The choice of these two thresholds (two and four times the SD of the control, respectively) was made in a conservative way, in order to minimize the number of false positive we obtained. Notably, by using these two selection criteria, among the regulated hits we found many proteins that were already known to be controlled by TORC1 (Supplementary Figure 1, C and D).