Supplementary Figure 1. Processing and quality control for recombinant proteins.

Transcription

1 Supplementary Figure 1 Processing and quality control for recombinant proteins. (a) Schematic representation of processing of a recombinant protein library. Recombinant proteins were generated individually. Portions of each protein preparation were subjected to purification followed by SDS- PAGE in order to check protein quality and to determine protein concentration. The remaining portions of the preparations were mixed and purified. The obtained protein mixture was digested and used for peptide identification or as an internal standard for MRM assays. (b) Recombinant proteins were synthesized in the presence of [ 13 C 6, 15 N 2 ]lysine and [ 13 C 6, 15 N 4 ]arginine, digested, and subjected to determination of protein concentration with an MRM assay for a peptide (LGPLVPR) corresponding to the MAFG tag. The log 2 ratio of protein concentration determined by SDS-PAGE to that determined by MRM was calculated.

2 Supplementary Figure 2 DDA of the in vitro proteome achieves unbiased peptide identification. The percentage of proteins in the in vitro proteome identified by DDA was determined for each of the indicated pathways (a) and molecular functions (b). Proteins were assigned to pathways and molecular functions with the Kyoto Encyclopedia of Genes and Genomes (KEGG) database, and the coverage was then compared with that for data in PeptideAtlas.

3 Supplementary Figure 3 Labeling efficiency of mtraq. (a) A recombinant protein digest was labeled (or not) with mtraq 4, and the peptides were then subjected to MRM analysis together with a digest of the same protein containing [ 13 C 6, 15 N 2 ]lysine and [ 13 C 6, 15 N 4 ]arginine. The MRM chromatogram of the peptide AGAHLQGGAK is shown as an example. The red and blue lines correspond to the non mtraq-labeled peptide and the internal standard containing a [ 13 C 6, 15 N 2 ]lysine residue, respectively. (b) A whole HeLa cell digest (100 g) was incubated for 2 h at room temperature with the indicated amounts of mtraq reagent, after which the amount of non mtraq-labeled peptides was quantified by MRM analysis together with a digest of proteins containing [ 13 C 6, 15 N 2 ]lysine and [ 13 C 6, 15 N 4 ]arginine. The labeling efficiency was calculated from the percentage of non mtraq-labeled peptides remaining after the reaction. Efficiency values are means ± s.d. for an average of 102 peptides in three independent measurements.

4 Supplementary Figure 4 Properties of nonlabeled and mtraq-labeled peptides in MS. (a) Nonlabeled and mtraq 4-labeled digests of a mixture of recombinant proteins (108 proteins) were subjected to DDA, and identified unique peptide sequences were compared. (b) The numbers of peptides identified from nonlabeled and mtraq-labeled digests were compared for each charge state. (c) Patterns of MS/MS spectra for nonlabeled and mtraq-lableled peptides were compared with Spearman s correlation coefficient on the basis of the intensities of all observed fragment ions (left panel) or only y-ions (right panel). (d) MRM assays developed with nonlabeled and mtraq-labeled peptides were compared on the basis of signal intensity. Data for the protein ACADVL are presented as a typical example.

5 Supplementary Figure 5 Validation of MRM-based absolute quantification. (a) Reproducible detection rate in repeated MRM analysis of 198 peptides. (b) Coefficient of variation for repeated MRM measurement. (c) Correlation between the absolute abundance of 10 proteins estimated by MRM analysis and the corresponding values obtained by immunoblot analysis. (d) Correlation between the absolute abundance of core metabolic enzymes determined by mtraq-based or AQUA protein based strategies.

6 Supplementary Figure 6 impaqt platform. The impaqt platform includes impaqt-knowledge DB and impaqt-quant software. impaqt-knowledge DB contains all MS/MS spectra assigned to peptides, MRM chromatograms for verification, and a search interface based on functional annotation (such as pathway, biological process, molecular function, and cellular localization) or on ID/accession number. Verified MRM transitions for each protein of interest can be downloaded in csv format. Each assay is specified as impaqt probe ID. Acquired data are processed by impaqt-quant software, which detects co-eluted peaks and calculates cosine similarity by comparison with corresponding MS/MS spectra. impaqt-knowledge DB also contains the concentration of each recombinant protein to assist estimation of the absolute abundance of each protein automatically.

7 Supplementary Figure 7 Absolute quantification of metabolic enzymes in human diploid fibroblasts. (a) The absolute values for the abundance of metabolic enzymes in HDFs as determined by MRM in a single experiment were compared between culture lots (A and B) and process replicates (1 and 2). (b) RNA abundance as determined by RNA sequencing for proteins not detected in HDFs by MRM analysis. FPKM, fragments per kilobase of exon model per million mapped fragments.

8 Supplementary Figure 8 Relationship between protein and RNA amounts for metabolic enzymes in mammalian cells and budding yeast. (a) Comparison of protein copy number per RNA molecule between human (this study) and mouse (Schwanhäusser et al. 20 ). (b) Protein/RNA ratio for orthologs quantified in human (this study), mouse (Schwanhäusser et al. 20 ), and budding yeast (Lawless et al. 26 ).

9 Supplementary Figure 9 Establishment of oncogene-induced transformation in HDFs. (a) Immunoblot analysis of expressed oncogenes and Hsp90 (loading control) in parental HDFs (TIG-3 cells), immortalized cells expressing htert and SV40 ER proteins (TS cells), and transformed cells expressing htert, SV40 ER proteins, and either c-myc or RasG12V (TSM and TSR cells, respectively). (b) Phasecontrast microscopy of TIG-3 cell lines. Scale bars, 200 μm. (c) Proliferation curves for parental, TS, TSM, and

10 TSR cells. Data are means ± s.d. for three independent determinations. (d) Colony formation by cells cultured in soft agar for 12 days. Scale bars, 200 μm. (e) Number of colonies formed by parental (P) HDFs and derived cell lines in experiments similar to that in d. Data are means ± s.d. for three independent determinations. ND, none detected. (f, g) Lactate production (f) and glucose consumption (g) in parental HDFs and derived cell lines maintained under normoxic or hypoxic conditions. Data are means ± s.d. for three independent experiments. (h, i) Rates of nucleic acid (h) and fatty acid (i) synthesis as determined by measurement of the incorporation of 14 C from 14 C-labeled glucose into DNA or RNA or into fatty acids, respectively. Data are means ± s.d. for three independent experiments.

11 Supplementary Figure 10 Significance of transformation-related changes in the abundance of metabolic enzymes. Volcano plots of fold change in protein abundance and accompanying q-value are shown for TSM cells (a) or TSR cells (b) relative to parental cells. Proteins are color-coded according to metabolic pathway. Red: glycolysis, TCA cycle, and oxidative phosphorylation. Purple: pentose phosphate pathway. Orange: nucleic acid metabolism. Pink: fatty acid metabolism. Blue: serine, glycine, and one-carbon metabolism.