Sample Collection Method Bias Effects in Quantitative

Jun 4, 2015 - Current advances in selective enrichment, fractionation, and MS detection of phosphorylated peptides allowed identification and quantita...
3 downloads 9 Views 912KB Size
Subscriber access provided by NEW YORK UNIV

Technical Note

Sample collection method bias effects in quantitative phosphoproteomics Evgeny Kanshin, Michael Tyers, and Pierre Thibault J. Proteome Res., Just Accepted Manuscript • Publication Date (Web): 04 Jun 2015 Downloaded from http://pubs.acs.org on June 7, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Technical Note

TITLE Sample collection method bias effects in quantitative phosphoproteomics

AUTHORS & AFFILIATIONS Evgeny Kanshin1, Michael Tyers1,2,* and Pierre Thibault1,3,* 1

Institute for Research in Immunology and Cancer, Université de Montréal, C.P. 6128, Succursale centre-ville, Montréal, Québec, H3C 3J7, Canada.

2

Department of Medicine, Université de Montréal, C.P. 6128, Succursale centre-ville, Montréal, Québec, H3C 3J7, Canada.

3

Department of Chemistry, Université de Montréal, C.P. 6128, Succursale centre-ville, Montréal, Québec, H3C 3J7, Canada.

CONTACT *Correspondence should be addressed to M.T. ([email protected], Phone: 514 343-6668, Fax: 514 343-6843) or P.T. ([email protected], Phone: 514 343-6910, Fax: 514 3436843)

Running title: Sample collection bias in phosphoproteomics

Keywords: Phosphoproteomics - Sample collection – Quantitative proteomics - Saccharomyces cerevisiae - Osmotic shock

1 ACS Paragon Plus Environment

Journal of Proteome Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 23

Technical Note

ABSTRACT Current advances in selective enrichment, fractionation and MS detection of phosphorylated peptides allowed identification and quantitation of tens of thousands phosphosites from minute amounts of biological material. One of the major challenges in the field is preserving invivo phosphorylation state of the proteins throughout sample preparation workflow. This is typically achieved by using phosphatase inhibitors and denaturing conditions during cell lysis. Here we determine if the upstream cell collection techniques could introduce changes in protein phosphorylation. To evaluate the effect of sample collection protocols on global phosphorylation status of the cell, we compared different sample workflows by metabolic labeling and quantitative mass spectrometry on Saccharomyces cerevisiae cell cultures. We identified highly similar phosphopeptides for cells harvested in ice cold isotonic phosphate buffer, cold ethanol, trichloroacetic acid, and liquid nitrogen. However, quantitative analyses revealed that the commonly used phosphate buffer unexpectedly activated signalling events. Such effects may introduce systematic bias in phosphoproteomics measurements and biochemical analysis.

2 ACS Paragon Plus Environment

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Technical Note

INTRODUCTION Recent developments in mass spectrometry (MS) and affinity purification of phosphorylated peptides enabled the identification and quantitation of tens of thousands of phosphosites from mg amounts of material1. These large phosphoproteomics data sets have delineated cellular signaling responses to various external stimuli. Typical experimental workflows are complex and consist of numerous steps including sample collection, cell lysis, protein extraction, enzymatic digestion, phosphopeptide purification, fractionation techniques, and MS analysis2. Preserving the in vivo status of phosphoproteins throughout sample processing is critical to monitor stimulus-specific changes in protein phosphorylation profiles. Sample collection is typically performed at cold temperatures to preserve sample integrity but includes centrifugation and cell washing steps that may introduce changes in protein phosphorylation due to inadvertent cell stimulation in response to cold shock, changes in osmolarity and nutrient depletion3. Treatment of cell cultures with different protein phosphatase inhibitors during cell lysis can also produce distinct phosphopeptide populations, indicating that significant enzymatic activities still remain under different inhibitory conditions4. In a recent study, prolonged exposure of MCF7 breast cancer cells to ambient conditions was reported to trigger an environmental stress reaction that result in changes of the phosphoproteome5. Similar observations were also reported when analyzing changes in the phosphoproteome of human ovarian tumor and breast cancer xenograft tissue without vascular interruption after defined ischemic intervals6. These analyses suggest that caution should be exerted when interpreting changes in protein phosphorylation as cell activation could be associated to sample collection protocols. Previous reports indicated that rapid heat 3 ACS Paragon Plus Environment

Journal of Proteome Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 23

Technical Note

stabilization could prevent changes in levels of protein phosphorylation, variation in the patterns of protein sumoylation, and differences associated to levels of proteolytic cleavage in tissue samples7,5,

8.

While this technique requires special equipment and cannot be easily

applied to cultured cells, the effects of different sample collection methods on global phosphorylation profiles have not been quantitatively investigated. In the present study, we used metabolic labeling and quantitative phosphoproteomics to examine how common sample collection protocols applicable to cell cultures can alter profiles of protein phosphorylation.

EXPERIMENTAL SECTION. Cell culture To allow for SILAC quantitation we used S288 Saccharomyces cerevisiae strain in which genes encoding de novo arginine or lysine synthetic enzymes Argininosuccinate lyase (ARG4) and Saccharopine dehydrogenase (LYS1) are deleted (S288C LYS1Δ::kanMX; ARG4Δ::kanMX) - a generous gift of Ole Jensen, University of Southern Denmark. These deletions force cells to use arginine and lysine from the media and ensure complete incorporation of heavy isotopes. Cells were grown in Synthetic Dextrose (SD) medium (0.17% yeast nitrogen base without amino acids, 0.5% ammonium sulfate, 2% glucose, and appropriate amino acids) supplemented with either 12C, 14N (light), 2H (medium) or 13C, 15N (heavy) lysine (30 mg/L) and arginine (20 mg/L) (Cambridge Isotope Laboratories). In order to minimize arginine to proline metabolic conversion media were also supplemented with L-Proline (20 mg/l). Typically, after 7-9

4 ACS Paragon Plus Environment

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Technical Note

doublings, heavy culture incorporated almost 100% of heavy arginine and lysine. Cells were grown until late-exponential phase (OD600~1). Sample collection. TCA-based protocol. Cell cultures were mixed with 100% TCA to get final TCA concentration of 10% and centrifuged at 3,000 x g for 5 min at 4oC. Cell pellets were washed once with ice-cold 10% TCA. EtOH-based protocol. Cell cultures were mixed with 9-times excess of cold (-80oC) EtOH and centrifuged at 3,000 x g for 5 min @ 4oC. Cell pellets were washed twice with ice-cold EtOH and once with ice-cold PBS (brief vortex and 5 min centrifugation at 3,000 x g). PBS-based protocol. Cell cultures were centrifuged at 3,000 x g for 5 min at 4oC. Cell pellets were washed 2 times with ice-cold PBS (brief vortex and 5 min centrifugation at 3,000 x g). Collection in liquid nitrogen. In order to immediately stop all metabolic activity at specific times, cultures were frozen in excess of liquid nitrogen and obtained ice chunks were stored at 80oC. Cell stimulation All treatments were performed with yeast cultures at OD600=1 at 30oC. Cultures were treated with either 4 M NaCl dissolved in the culture medium (preheated to 30oC) to a final concentration of 0.4 M NaCl or with the corresponding volume of culture medium without NaCl (control).

5 ACS Paragon Plus Environment

Journal of Proteome Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 23

Technical Note

Cell lysis and protein extraction (Freezer Mill) Yeast cells were lysed by mechanical grinding under liquid nitrogen using a Freezer Mill apparatus (BioSpec). Efficiency of the lysis was accessed by microscopy and depended on cell concentration. In our experiment (diluted cell cultures) in order to break > 90% of cells, frozen cultures had to be ground for 32 cycles consisting of 2 minute grind at maximum intensity with 2 min cool down periods in between. In order to concentrate and purify proteins from other components of culture media, they were precipitated with trichloroacetic acid (TCA). Frozen culture powders were mixed with equal volume of 30 % TCA solution and incubated on ice for 2 hours. Then samples were centrifuged at 20,000 x g for 20 minutes at 4oC. Supernatants were discarded and protein pellets were washed in 10 ml of cold 10 % TCA and centrifuged one more time. Then pellets were washed in cold acetone and proteins were resolubilized in 8 M urea buffer (8 M urea, 100 mM Tris pH 8.0, supplemented with HALT phosphatase inhibitor cocktail (Pierce)). Samples were cleared by centrifugation at 40,000 x g for 10 min, the supernatants were transferred into clean tubes and the protein concentrations were measured by BCA assay (Thermo Fisher Scientific). Cell lysis and protein extraction (Bead beating) Yeast cells were lysed by bead beating for 10 min in lysis buffer (8 M urea, 50 mM Tris pH 8.0, supplemented with HALT phosphatase inhibitor cocktail, Pierce). Samples were centrifuged at 40,000 x g for 10 min, and the supernatants were transferred into clean tubes prior to determination of protein concentrations by BCA assay (Thermo Fisher Scientific).

6 ACS Paragon Plus Environment

Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Technical Note

Enzymatic digestion Disulfide bridges were reduced by adding dithiothreitol to a final concentration of 5 mM and incubating at 56 °C for 30 minutes. The samples were allowed to cool down to room temperature and the reduced cysteines were alkylated by adding iodoacetamide to 15 mM and incubating for 30 minutes in the dark at room temperature. Alkylation was quenched with 5 mM dithiothreitol (incubation for an additional 15 minutes). To reduce urea concentration samples were diluted 6 times with 20 mM TRIS, pH=8 containing 1 mM CaCl2 and trypsin were added to an enzyme to substrate mass ratio of 1:50. Proteins were digested overnight at 37oC. Because trypsin cleaves after arginine or lysine, heavy and light versions of every peptide (except the very C-terminal peptide of each protein) should be distinguishable based upon the mass difference between the heavy and light versions of at least one lysine or arginine. After digestion peptide mixtures were acidified by addition of formic acid (FA) to a final concentration of 1%, clarified by centrifugation (20,000 x g 10 min) and desalted on Oasis HLB cartridges (Waters) according to the manufacturer instructions. Peptide eluates were snapfrozen in liquid nitrogen, lyophilized in a speedvac and stored at -80oC. Phosphopeptide Isolation Due to the fact that only a small subset of peptides is phosphorylated, phosphopeptides must be enriched prior to MS analysis. The peptide samples were subjected to the TiO2 enrichment protocol as described in9. Briefly, sample loading, washing, and elution steps were performed in homemade spin columns assembled following the StageTip principle10 and comprised of 200 µl pipette tip with frit made of SDB-XC membrane (3M) and filled with TiO2 beads. SDB-XC 7 ACS Paragon Plus Environment

Journal of Proteome Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 23

Technical Note

material has similar hydrophobic properties to C18 and allows for combining phosphopeptide enrichment and desalting steps. Centrifugation speed was set to 2,000 x g. Before peptide loading, columns were equilibrated with 100 µl of loading buffer (250 mM lactic acid in 70% ACN 3% TFA). Peptides were solubilized in 100 µl of loading buffer and applied on a TiO2 column. Each column was washed with 100 μL of loading buffer followed by 2 x 100 μl of 125 mM asparagine and glutamine in 70% ACN 3% TFA and 100 µl of 70% ACN 3% TFA. Subsequent washing with 50 µl of 1% FA was used to equilibrate SDB-XC frit material. Phosphopeptides were eluted from TiO2 with 2 x 50 µl portions of 500 mM Na2HPO4, pH=7 and retained on SDBXC frit. Peptides were desalted in 50 µl of 1% FA and subsequently eluted from SDB-XC in 50 µl of 50% ACN 0.5% FA. Eluates were dried on a speedvac and stored at -80oC. Prior to MS analysis peptides were resuspended in 30 µl of 4% FA. NanoLC- MS/MS Phosphopeptides were analyzed by online reverse phase chromatography coupled with an electrospray ionization interface to acquire MS (measuring intensity and m/z ratio for peptides) and MS/MS (fragmentation spectra of peptides) scans. A nanoflow HPLC system (Eksigent, Thermo Fisher Scientific) was used for online reversed-phase chromatographic separation; peptides were loaded on 5 mm long trap column (inner diameter 300 µm) in buffer A (0.2% FA) and separated on 18 cm long fused silica capillary analytical column (inner diameter 150 µm), both packed with 3µm 200A Magic AQ C18 reverse-phase material (Michrom). Peptides were eluted by an increasing concentration of buffer B (0.2% FA in ACN) – from 5 to 40% in 100 min. Following the gradient elution, the column was washed with 80% buffer B and re-equilibrated

8 ACS Paragon Plus Environment

Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Technical Note

with 5% buffer B. Peptides were eluted into the mass spectrometer at a flow rate of 600 nl/min. The total run time was approximately 125 min, including sample loading and column conditioning. Peptides were analyzed using an automated data-dependent acquisition on a LTQ-Orbitrap Elite mass spectrometer. Each MS scan was acquired at a resolution of 120,000 fwhm (at 400 m/z) for mass range 300-2,000 Th and was followed by up to 12 MS/MS data dependent scans on the most intense ions using collision induced activation (CID). AGC target values for MS and MS/MS scans were set to 1e6 (max fill time 500 ms) and 1e5 (max fill time 50 ms) respectively. The precursor isolation window was set to 2 Th with CID normalized collision energy of 35; the dynamic exclusion window was set to 60 seconds. MS Data Processing and Analysis MS data were analyzed using MaxQuant11,12 software version 1.3.0.3 and searched against the SwissProt subset of the Uniprot database (Release Oct. 2012) containing 6,630 entries for S. cerevisiae. Database search was performed in Andromeda13 integrated in MaxQuant environment. A list of 248 common laboratory contaminants included in MaxQuant was also added to the database as well as reversed versions of all sequences. For searching, the enzyme specificity was set to trypsin with the maximum number of missed cleavages set to 2. The precursor mass tolerance was set to 20 ppm for the first search (used for non-linear mass recalibration14 and then to 6 ppm for the main search. Mass tolerance for fragment ions in CID spectra was set to 0.5 Da. Phosphorylation of serine, threonine and tyrosine residues was searched as variable modification; carbamidomethylation of cysteines was searched as a fixed modification. The maximal number of modifications per peptide was set to 5. The false

9 ACS Paragon Plus Environment

Journal of Proteome Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 23

Technical Note

discovery rate (FDR) for peptide, protein, and site identification was set to 1%. Additionally minimal Andromeda score for modified peptides was set on 40. The minimum peptide length was set to 6, cut-off for phosphosite localization confidence to 0.75 and the ‘peptide requantification’ function was enabled. To transfer identifications across different runs, the ‘match between runs’ option in MaxQuant was enabled with a retention time window of 0.7 minute and an alignment time window of 20 min. Subsequent data analysis were performed in either Perseus (http://www.perseus-framework.org/) or using R environment for statistical computing and graphics (http://www.r-project.org/).

RESULTS AND DISCUSSION. To determine changes in protein phosphorylation associated with sample collection, we used quantitative mass spectrometry to assess the effects of commonly used protocols on the global phosphoproteome in yeast. We assessed three sample collection workflows: two that used denaturating conditions and involved mixing of cell cultures with either an excess of ethanol (EtOH) at -80oC or trichloroacetic acid (TCA)15, and a third under non-denaturing conditions by suspension and washing cells in cold phosphate buffered saline (PBS)

4, 16

. We

used stable isotope labeling by amino acids in cell culture (SILAC)17 to compare changes in protein phosphorylation profiles to a reference condition in which cell cultures were directly frozen in liquid nitrogen to instantaneously halt all enzymatic activities18. Yeast cells were grown in media containing either heavy (for EtOH, TCA or PBS protocols) or light (liquid nitrogen reference protocol) isotopic forms of arginine and lysine amino acids (SILAC channels

10 ACS Paragon Plus Environment

Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Technical Note

were reversed for the second replicate) (Figure 1A). Cell culture harvested in EtOH, TCA or PBS were collected and mixed separately with an equal amount of reference culture harvested and maintained prior to lysis in liquid nitrogen. The resulting frozen samples were processed in a Freezer Mill to lyse cells by mechanical grinding in liquid nitrogen. The released proteins were precipitated by TCA, and solubilized in denaturing urea buffer prior to digestion with trypsin. Phosphopeptides were enriched on TiO2 resin, and analyzed by LC-MS/MS. (Experimental section). Although this workflow was labour intensive and required specialized equipment, the mixture and lysis of samples under liquid nitrogen minimized changes associated with downstream sample processing and enabled direct comparison between sample harvesting steps. Comprehensive phosphoproteomics analyses enabled the quantitation of 3227 phosphopeptides (1083 phosphoproteins) with an overlap of 80% of all phosphorylation sites across replicates and protocols (Figures 1B-C). We used interquartile ranges (IQR) to determine the variation of log2-transformed fold change (FC) distribution between individual protocols and the reference condition. IQR values measured for non phosphorylated peptides served as a proxy to determine technical variability, as no changes in protein abundance were expected within the time frame of sample collection (Figure 2A). These analyses revealed a progressive increase in protein phosphorylation from TCA to EtOH and PBS. For the TCA-based protocol, IQR values were superimposable to those associated with technical variability suggesting that minimal effect to protein phosphorylation is imparted by this sample collection protocol (Figure 2B and Table S1, Supporting information). In contrast, the PBS protocol yielded FC measurements with IQR values 25% higher than those observed for TCA. Comparison of results 11 ACS Paragon Plus Environment

Journal of Proteome Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 23

Technical Note

from biological replicates obtained using reverse SILAC labeling highlighted distinct sets of proteins that showed reproducible changes in phosphorylation (Figure 2C). In particular, PBS affected the phosphorylation of proteins associated with cold stress response (Nth1, Nup60) and availability of nutrients (Sip1, Gat1, Bcy1) (Figure 2D and Table S2, Supporting Information) 19

. These results indicated that sample collection protocols can inadvertently introduce specific

changes to the phosphoproteome that may confound measurement of genuine responses. In order to evaluate sample losses associated with each protocol we compared the distribution of SILAC ratios before and after normalization (i.e. median of log2(H/L)=0). This analysis showed no significant difference between protocols except for EtOH that gave 10-15% more protein compared to liquid nitrogen, a difference likely attributed to unavoidable mixing errors (Figure S1, Supporting Information).

To evaluate the impact of these variations on the identification of stimulus-specific changes in phosphorylation, we assessed the High Osmolarity Glycerol (HOG) response pathway in S. cerevisiae, a commonly used model to analyze systems level properties of signal transduction20. We profiled changes in protein phosphorylation of control cells (light) and cells stimulated with 0.4 M NaCl for 10 min (heavy) using metabolic labeling and quantitative proteomics (Figure 3A). All experiments were performed on 4 biological replicates to identify statistically relevant changes in phosphorylation for each protocol. In total, we have quantified 3683 unique phosphopeptides (Figure S2, Supporting information). The osmotic shock resulted in many expected changes of phosphorylation for members of the HOG response pathway,

12 ACS Paragon Plus Environment

Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Technical Note

though the extent of these changes varied between sample collection protocols (Table S3, Supporting Information). The pairwise comparison of log2FC between sample protocols showed similar distributions for EtOH and TCA, but differed significantly from those obtained with PBS (Figure 3B). To find regulated phosphorylation sites, we used a t-test with p-values adjusted for multiple hypothesis testing to obtain a FDR less than 5% (Figure 3C). A total of 27 phosphorylation sites were regulated upon osmotic shock using the TCA sample collection protocol compared to 12 sites for the EtOH protocol (Figure 3C). We also observed a good correlation of Log2 FC values for phosphopeptides regulated in TCA and EtOH protocols (R2=0.78) (Table S4, Supporting information). In marked contrast, the PBS method did not yield any regulated phosphorylation sites when using the same selection criteria. For example, we consistently observed an upregulation of Hog1 MAPK kinase phosphorylation on Y176 upon osmotic stress using the EtOH and TCA methods, whereas down regulation of this site was observed for the PBS protocol under the same conditions (Figure 3D). Interestingly, a previous study on osmotic shock in yeast by Soufi et al. used the PBS protocol with 5 and 20 min cell stimulation with 0.4 M NaCl, but did not detect any phosphorylation on HOG1 21. A comparison of phosphorylation sites identified in this study with that of Soufi et al. is presented in Figure S3, Supplementary Information. We also observed reduced phosphorylation of upstream kinases Pbs2 and Ssk2 for the PBS versus TCA and EtOH sample collection protocols. These specific differences are likely explained by the strikingly different composition of PBS compared to standard yeast culture media. Furthermore, we observed a greater variability of FC measurements across replicates for PBS samples compared to other protocols investigated (p-

13 ACS Paragon Plus Environment

Journal of Proteome Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 23

Technical Note

values: 0.11 for PBS vs. TCA and 0.04 for PBS vs. EtOH). This variability may arise from the variability in processing times across replicates, as cells remain viable in cold PBS.

CONCLUSIONS The above comparative analyses indicate that sample collection workflows can significantly affect quantitative phosphoproteomics measurements, and the conclusions drawn. Protocol differences affect both overall FC across the proteome and introduce bias in regulated phosphorylation sites. Our data suggest that either TCA or EtOH sample collection protocols introduce lower sample collection bias compared to the more commonly used ice cold PBS protocol. We note that yeast cultures are often harvested by the direct addition of ice to culture medium and or washing cell pellets in ice cold water, both of which are likely to further exacerbate perturbation of the phosphoproteome. Other dynamic PTMs such as acetylation, methylation or ubiquitylation may also be susceptible to sample protocol bias effects. It is noteworthy that we did not observed significant changes in protein abundances using the sample protocols outlined in this study due to the short time scale of the sample collection (