An integrated strategy for highly sensitive phosphoproteome analysis from low micrograms of protein samples

Wendong Chen ab, Lan Chen a and Ruijun Tian *ac
aDepartment of Chemistry, Southern University of Science and Technology, Shenzhen 518055, China. E-mail: tianrj@sustc.edu.cn
bSUSTech Academy for Advanced Interdisciplinary Studies, Southern University of Science and Technology, Shenzhen 518055, China
cGuangdong Provincial Key Laboratory of Cell Microenvironment and Disease Research, Southern University of Science and Technology, Shenzhen 518055, China

Received 28th April 2018 , Accepted 9th June 2018

First published on 11th June 2018


Phosphoproteomics has become a popular proteomic technology for exploring cellular signaling networks. However, current approaches often require milligrams of protein samples which hamper their applications for translational studies with limited starting materials. In this study, we aimed to challenge the lowest starting material limit for phosphoproteome profiling. By carefully optimizing the well-established high-pH reversed-phase (RP) fractionation plus Ti4+-IMAC enrichment strategy, we achieved the identification of 15[thin space (1/6-em)]260 and 8936 unique phosphopeptides from only 500 μg and 250 μg predigested peptides, respectively. To further improve the sensitivity of phosphoproteome analysis for low micrograms of protein samples, we developed an integrated strategy, termed Phospho-SISPROT. This technology integrates three tips in tandem for protein digestion by the simple and integrated spintip-based proteomics technology (SISPROT), phosphopeptide enrichment by the Ti4+-IMAC tip, and desalting by the StageTip, respectively, which could dramatically reduce the phosphoproteome analysis time from a couple of days to only 6 hours and improve the system sensitivity. The flow through of Phospho-SISPROT could be reused for the global protein identification, which is very helpful for accurate phosphoproteome analysis with limited starting materials. More than 5500 and 600 unique phosphopeptides were respectively identified from 20 μg and 1 μg pervanadate treated HEK 293T cell lysates processed by the Phospho-SISPROT. To the best of our knowledge, this performance is the highest reported to date by using the standard LC-MS/MS setup. We expect that the Phospho-SISPROT and the optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy will be well suited for highly sensitive phosphoproteome analysis of rare biological samples.


Introduction

Protein phosphorylation is one of the most important post-translational modifications (PTMs) and plays essential roles in many cellular processes, such as signal transduction, cell division, apoptosis, and so on.1–3 Aberrant phosphorylation arising from dysfunctional kinase signaling has been found to be involved in many human diseases.4,5 Hence, the global profiling of protein phosphorylation is necessary to fundamentally understand various signaling networks and biological processes. Mass spectrometry (MS)-based proteomics has become the major technology for the global identification and quantification of protein phosphorylation.3,6 Current MS-based proteomics technologies are able to routinely identify more than 10[thin space (1/6-em)]000 phosphosites from milligrams of samples by combining with various chromatography fractionation approaches.6–8 However, the amount of starting material is generally limited in many cases of in vivo and translational studies. Therefore, it is desirable to develop high-sensitive technologies for the phosphoproteome study of rare biological samples.

Diverse off-line chromatography fractionation strategies have been applied for the large-scale phosphoproteome analysis.6,9,10 The strong cation exchange (SCX) fractionation followed by the titanium dioxide (TiO2) enrichment method was capable of identifying 15[thin space (1/6-em)]861 phosphosites.11 The phosphopeptide separation based on high-pH reversed-phase (RP) chromatography is a powerful fractionation approach for global phosphoproteomics without tedious desalting steps.12–14 As an example, more than 20[thin space (1/6-em)]000 phosphosites were identified from several milligrams of samples which were fractionated into 12 fractions.7 Although these fractionation strategies could easily identify >10[thin space (1/6-em)]000 phosphosites, they often require relatively large amounts of samples. A three-dimensional RP-strong anion exchange (SAX)-RP fractionation strategy was developed for the high efficiency phosphopeptide analysis, by which 2521 phosphosites were identified from only 10 μg of the peptide sample. However, the complicated system setup hinders its applications for routine phosphoproteome analysis.15

With the recent development of modern MS instrumentation and high-performance LC separation columns, deep phosphoproteome analysis by single-shot LC-MS/MS has become possible. The EasyPhos technology facilitated high-throughput phosphoproteome analysis in diverse cells and tissues at a depth of >10[thin space (1/6-em)]000 sites by a single 4 hour gradient LC-MS/MS analysis.3 A robust protocol implementing the enzymatic digestion of DNA and RNA using benzonase prior to immobilized metal ion affinity chromatography (IMAC) column loading enabled the identification of around 12[thin space (1/6-em)]500 phosphosites in 1.4 mg human cell lysates.16,17 Nevertheless, these technologies still needed ≥1 mg of protein as the starting material, which limited their application to only a few biological samples. An integrated multistep enzyme digestion, enrichment, and database search strategy was used to process 500 μg cell lysates; about 8000 phosphopeptides were identified by mass injection of 5% of the enrichment sample (a similar amount enriched from 25 μg pre-digested proteins).18 Two highly sensitive technologies, the titanium(IV) monolithic tip19 and the Ti4+-IMAC spin tip,20 were developed for the phosphoproteome analysis of minute amounts of samples. About 1000 phosphosites and 936 unique phosphopeptides were identified from 5 μg HeLa cell protein digest, respectively. However, these two technologies were only used to enrich the phosphopeptides from predigested protein samples. Fortunately, the phosphoproteomic reactor was able to solve this issue by combining efficient protein processing and phosphopeptide enrichment with SCX bead-packed and Ti4+-IMAC bead-packed capillary columns, respectively.21 However, the operation of this pressure-driven system could be complicated for routine phosphoproteome analysis.

In this study, we aimed to develop an easy-to-use strategy for the highly sensitive phosphoproteome profiling of low micrograms of protein samples. Firstly, we optimized the high-pH RP fractionation plus Ti4+-IMAC enrichment strategy to improve the depth of phosphoproteome analysis from sub-milligrams of samples (e.g. 500 μg). To challenge the lowest starting material limit for phosphoproteome profiling, we developed Phospho-SISPROT which integrated three tips in tandem for protein digestion by the simple and integrated spintip-based proteomics technology (SISPROT),22 phosphopeptide enrichment by the Ti4+-IMAC tip,20 and desalting by the StageTip,23 respectively. This integration feature of Phospho-SISPROT would greatly help to enhance the overall sensitivity, reproducibility, and throughput and reduce the phosphoproteome analysis time from a couple of days to only 6 hours. The optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy was firstly exemplified for the global phosphoproteome profiling of 500 μg HEK 293T peptides and 250 μg Jurkat T peptides, respectively. Then, the performance of Phospho-SISPROT was evaluated by directly processing 100 μg cell lysates. Finally, the sensitivity and reproducibility of Phospho-SISPROT were systematically evaluated with less than 20 μg cell lysates.

Experimental

Cell culture, harvest, and lysis

Human embryonic kidney (HEK) 293T cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS) and incubated at 37 °C in 5% CO2. Jurkat T cells were cultured in RPMI 1640 medium supplemented with 10% FBS and incubated at 37 °C in 5% CO2. Before harvest, HEK 293T cells and Jurkat T cells were treated by serum starvation for 3 hours and then stimulated with 1 mM pervanadate for 10 minutes. Finally, the cells were lysed with the lysis buffer containing 1% (w/v) n-dodecyl β-D-maltoside (DDM), 600 mM guanidine HCl, 25 mM HEPES, pH 7.4, 150 mM NaCl, 1 mM pervanadate, and protease inhibitor mixture (Sigma-Aldrich, USA). The cell lysates were briefly sonicated and then incubated on ice for 30 min. After centrifugation at 17[thin space (1/6-em)]000g for 15 min at 4 °C, the supernatant was transferred to a new tube and the protein concentration was measured by the BCA assay (Thermo Fisher Scientific, USA).

Design and operation of the Phospho-SISPROT

As shown in Fig. 1, Phospho-SISPROT integrated three tips packed with different materials. The SISPROT tip was composed of POROS SCX beads (Applied Biosystems, USA) and C18 disks (3 M Empore, USA) packed in tandem in the same 200 μL pipette tip; the Ti4+-IMAC tip was made of Ti4+-IMAC beads and one plug of C8 disk (3 M Empore, USA) packed in the same 200 μL pipette tip; the C18 StageTip was made of C18 disks packed in a 200 μL pipette tip. The number of C18 disks and the amount of SCX beads and Ti4+-IMAC beads were used according to the protein or peptide amount processed. Typically, one plug of C18 disk is used for 5 μg peptides and 0.4–1.5 mg SCX beads is used for 1–20 μg proteins. The Ti4+-IMAC beads were provided by the Key Laboratory of Separation Science for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of Sciences and used with a bead-to-peptide ratio of 20/1 (w/w).20
image file: c8an00792f-f1.tif
Fig. 1 Integrated strategy for the highly sensitive phosphoproteome analysis. Left: Optimized high-pH RP fractionation plus Ti4+-IMAC enrichment workflow for sub-milligrams of samples; Right: Phospho-SISPROT workflow for low micrograms of samples.

All the sample preparation steps of Phospho-SISPROT were carried out using a centrifuge. For centrifugation, the tips were placed into a 1.5 mL Eppendorf tube by using an adapter. The operations of three tips are as follows.

The SISPROT tip was used for the protein digestion. The SISPROT tip was firstly washed with methanol and equilibrated with 10 mM potassium citrate buffer (pH 3). The cell lysates were acidified to pH 2–3 with 2% trifluoroacetic acid (TFA) and loaded onto the tip. After washing with 20% acetonitrile (ACN) in 8 mM potassium citrate buffer (pH 3), the proteins were reduced by infusing 10 mM tris(2-carboxyethyl) phosphine hydrochloride (TCEP) in 9 mM potassium citrate buffer (pH 3) into the tip with a syringe and incubating for 15 min at room temperature. The tip was then washed with water. The protein digestion was activated by infusing 2 μg μL−1 trypsin (Promega, Madison, WI) in 10 mM iodoacetamide, 100 mM Tris-HCl, pH 8 into the tip and incubating for 1 hour at room temperature (in darkness). The resultant peptides were transferred from the SCX beads to the C18 disk with 200 mM ammonium bicarbonate. After a desalting wash with 0.1% TFA, the peptides were directly eluted with the loading buffer (80% ACN, 6% TFA) of the Ti4+-IMAC tip.

The Ti4+-IMAC tip was used to purify the phosphopeptides. Before enrichment, the Ti4+-IMAC tip was equilibrated with loading buffer (80% ACN, 6% TFA). Next, the digested peptides from the SISPROT tip were loaded onto the tip. Afterwards, the tip was washed with washing buffer 1 (50% ACN, 6% TFA, 200 mM NaCl) to remove non-specific binding peptides and washing buffer 2 (30% ACN and 0.1% TFA) to remove the salt. Finally, the bound phosphopeptides were successively eluted with 10% NH3·H2O and 50% ACN. The phosphopeptides were lyophilized to dryness prior to the C18 StageTip desalting.

The C18 StageTip was firstly activated with methanol and 50% ACN, 0.5% acetic acid, respectively. After being equilibrated with 1% formic acid (FA), the phosphopeptide sample reconstituted in 1% FA was loaded onto the tip. Then, the phosphopeptides bound onto the tip were washed with 1% FA and subsequently eluted with 50% ACN, 0.5% acetic acid. Finally, the phosphopeptides were lyophilized to dryness and reconstituted in 5% ACN, 4% FA for LC-MS/MS analysis.

Global protein identification with the flow through of Phospho-SISPROT

To investigate whether the flow through of Phospho-SISPROT could be used for the global protein identification, 10 μg HEK 293T cell lysates were processed by the Phospho-SISPROT and the flow through of sample loading during the phosphopeptide enrichment was collected and lyophilized to dryness. Then, the flow through was reconstituted in 1% FA, desalted with the C18 StageTip, and detected by LC-MS/MS. For comparison, 10 μg HEK 293T cell lysates processed by the SISPROT22 and peptides prepared by the in-solution digestion method24 were also analyzed by LC-MS/MS.

Optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy

The optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy includes multiple steps. The cell lysates were firstly digested by the in-solution digestion method.24 After desalting, the digested peptides were fractionated by the high-pH RP fractionation approach.14 We used an Agilent 1260 pump equipped with a Waters XBridge BEH130 C18 column (5 μm particles, 2.1 mm i.d. and 15 cm long) and a photodiode array detector (set at 214 nm wavelength) to fractionate the peptides. The peptides were separated by using a binary buffer system of 2% ACN in 10 mM ammonium bicarbonate, pH 8 and 90% ACN in 10 mM ammonium bicarbonate, pH 8 at a flow rate of 0.4 mL min−1. Peptides were subjected to a 60 min linear gradient from 5 to 35% ACN and fractionated into a total of 60 fractions, which were finally consolidated to 6 fractions by combining every 6th fraction (1, 7, 13, 19, 25, 31, 37, 43, 49, 55; 2, 8, 14, 20, 26, 32, 38, 44, 50, 56; …). These fractions were subsequently acidified with 10% FA and lyophilized to dryness. Each fraction was enriched with the same Ti4+-IMAC tip as Phospho-SISPROT and desalted with the C18 StageTip for LC-MS/MS analysis.

LC-MS/MS analysis

An Easy-nLC 1000 system coupled with an Orbitrap Fusion mass spectrometer (Thermo Fisher Scientific, USA) was used to analyze the samples. The peptide sample was directly loaded onto the analytical column (100 μm i.d. × 20 cm) with an integrated spray tip packed with 1.9 μm and 120 Å ReproSil-Pur C18 resins (Dr Maisch GmbH, Ammerbuch, Germany). The peptides were separated with a binary buffer system of 0.1% FA in water (buffer A) and 0.1% FA in ACN (buffer B) at a flow rate of 250 nL min−1. The gradient for the phosphopeptide sample was set as follows: from 3 to 7% buffer B in 2 min, from 7 to 25% buffer B in 55 min, from 25 to 35% buffer B in 5 min, from 35 to 90% buffer B in 2 min, holding at 90% buffer B for 6 min, declining to 3% buffer B in 2 min, and holding at 3% buffer B for 8 min. For the global protein identification sample, the separation gradient time was set to 50 min from 7 to 22% buffer B and 10 min from 22 to 35% buffer B, respectively.

The peptides were detected using an Orbitrap Fusion mass spectrometer. Full MS scans were performed in the Orbitrap mass analyzer over the m/z range of 350–1550 at a mass resolution of 120[thin space (1/6-em)]000, while the MS/MS scans were implemented in the ion trap mass analyzer with an isolation window of 1.6 Da by using the quadrupole mass filter. The data-dependent acquisition method was used in the top speed mode with a cycle time of 3 s. The normalized collision energy of HCD fragmentation was set to 30 and the dynamic exclusion time was set to 30 s.

MS data analysis

The raw files were compared against the Human fasta database (70[thin space (1/6-em)]698 entries, downloaded on June 01, 2017) by MaxQuant (version 1.5.5.1). The oxidation (M), deamidation (NQ), and Phospho (STY) were selected as the variable modifications for the phosphopeptide identification, while the oxidation (M) and deamidation (NQ) were selected for the global protein identification. The carbamidomethyl was set as the fixed modification. The maximum missed cleavage for trypsin digestion was set to 2. Label-free quantification (LFQ) and match between runs were set for the duplicate or triplicate analysis data. FDR based on posterior error probability (PEP) was determined by searching a reverse database and was set to 0.01 for proteins and peptides. Other parameters were set as default. Confidently assigned phosphorylation sites with Localization Probability ≥0.75 and Score Difference ≥5 were defined as Class I sites.

Gene Ontology (GO) terms for the identified phosphoproteins of HEK 293T cells were determined using DAVID (version 6.8) and categorized based on their cellular localization. The corresponding Uniprot identifiers to phosphoproteins were uploaded as a gene list. A cut-off of p < 0.05 was utilized for all GO categories.

Results and discussion

Design of the integrated strategy for highly sensitive phosphoproteome analysis

In this study, we attempted to develop an easy-to-use strategy for the highly sensitive phosphoproteome profiling of low micrograms of protein samples. As shown in Fig. 1, this strategy has two different ways to process the cell lysates depending on their amounts. For sub-milligrams of samples (e.g. 500 μg), we optimized the high-pH RP fractionation plus Ti4+-IMAC enrichment strategy to achieve the identification of >10[thin space (1/6-em)]000 unique phosphopeptides. To challenge the lowest starting material limit for phosphoproteome analysis, we developed an integrated technology, termed Phospho-SISPROT.

For the high-pH RP fractionation plus Ti4+-IMAC enrichment strategy, cell lysates were digested by the in-solution digestion method.24 After desalting, the digested peptides were fractionated into 6 fractions by the high-pH RP fractionation approach with modifications.14 A smaller column was used to fractionate sub-milligrams of samples to reduce sample loss. The pH of the mobile phase was adjusted from 10 to 8, at which phosphopeptides were much more stable. Then, the phosphopeptides in each fraction were purified with the Ti4+-IMAC tip and finally desalted with the C18 StageTip to enhance the overall sensitivity. As demonstrated below, the optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy is a powerful method for the global phosphoproteomics.

The goal of Phospho-SISPROT is to achieve highly sensitive phosphoproteome analysis with less than 100 μg cell lysates. Previously, we have achieved highly sensitive proteome profiling using the SISPROT by which all the protein sample preparation steps were integrated and completed within 2 hours.22 Since the phosphopeptides account for around 1% of the total protein digest, the phosphopeptide enrichment step is usually required for the phosphoproteome analysis. Here, the Phospho-SISPROT was developed by integrating three tips with different functionalities. The SISPROT tip was used for the protein digestion within 2 hours, the Ti4+-IMAC tip was used for the phosphopeptide enrichment within 3 hours, and the C18 StageTip was used for the phosphopeptide desalting within 1 hour. One significant advantage of Phospho-SISPROT is that it can shorten the phosphoproteome analysis time from a couple of days (traditional method) to only 6 hours. This feature will greatly help in controlling the sample quality, especially compared with the overnight in-solution digestion protocol. It was reported that biomolecular contaminants, such as nucleic acid molecules, could seriously interfere with IMAC based phosphopeptide enrichments.17 The Phospho-SISPROT was capable of solving this issue. At low pH, the proteins carry multiple positive charges and form strong ionic bonds with the SCX beads present in the SISPROT tip, which would efficiently concentrate the proteins and remove substantial interferents such as DNA, RNA, and other molecules. In addition, the digested peptides were transferred from the SCX beads to C18 membranes in the SISPROT tip, where the peptides could be desalted and directly eluted with the loading buffer (80% ACN, 6% TFA) of the Ti4+-IMAC tip. Hence, Phospho-SISPROT eliminated the lyophilization step between the protein digestion and phosphopeptide enrichment steps compared to the phosphoproteomic reactor,21 which would reduce the sample loss. Furthermore, Phospho-SISPROT integrated phosphopeptide desalting with C18 StageTips,23 which would eliminate the influence of residual ammonia salts on the MS detection efficiency. We expected that all these improvements would make Phospho-SISPROT sensitive for phosphoproteome profiling of rare biological samples.

Performance of the optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy

The optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy was firstly exemplified by processing a 500 μg HEK 293T peptide sample. As shown in Fig. 2(a), the distributions of unique phosphopeptides identified in each fraction are largely even, indicating an efficient fractionation strategy. About 4000 unique phosphopeptides were identified in each fraction. In total, 15[thin space (1/6-em)]260 unique phosphopeptides (Table S1) and 20[thin space (1/6-em)]669 phosphosites (Table S2) were identified, among which 13[thin space (1/6-em)]709 sites were Class I sites with Localization Probability ≥0.75 and Score Difference ≥5. These Class I sites were compared to the human phosphorylation sites in the PhosphoSitePlus database (http://www.phosphosite.org). 3880 phosphosites are unrecorded in the database and should be novel (Table S3). The phosphorylation site distributions of serine, threonine, and tyrosine are shown in Fig. 2(c), with proportions of 72.0%, 16.7%, and 11.3%, respectively. Particularly, 1987 Class I pTyr sites were identified. It can be concluded that our strategy is powerful for the in-depth analysis of tyrosine phosphoproteome compared to the SH2 domain derived pTyr superbinder based workflow, by which over 1800 Class I pTyr sites were identified from 2 mg pervanadate treated Jurkat T proteins.25 Then, we reduced the sample amount to 250 μg Jurkat T peptides. As illustrated in Fig. 2(b), about 2500 unique phosphopeptides were identified in each fraction. In total, 8936 unique phosphopeptides (Table S4) and 10[thin space (1/6-em)]454 phosphosites (Table S5) were identified, among which 6939 sites were Class I sites. The phosphorylation site distributions of serine, threonine, and tyrosine are similar to those of the HEK 293T peptide sample, with proportions of 77.3%, 11.5%, and 11.2%, respectively (Fig. 2(d)). Particularly, 1020 Class I pTyr sites were identified. It is obvious that 10[thin space (1/6-em)]000 phosphosites can be routinely identified from 250 μg samples by the optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy.
image file: c8an00792f-f2.tif
Fig. 2 Phosphopeptides identified by the optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy with (a) 500 μg HEK 293T cell digested peptides and (b) 250 μg Jurkat T cell digested peptides; F1–F6 represent fraction 1–fraction 6. Phosphorylation site distributions of serine, threonine, and tyrosine for the results of (c) 500 μg peptides and (d) 250 μg peptides.

GO analysis was performed for all the phosphoproteins and the tyrosine phosphorylated proteins identified from 500 μg HEK 293T peptides by the optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy. As shown in Fig. 3, it reveals that the phosphoproteins in different cellular components can be efficiently identified, especially for the phosphoproteins associated with the cytoplasm, nucleus, cytosol, nucleoplasm, and membrane (Fig. 3(a)). GO cellular component analysis of the tyrosine phosphorylated proteins demonstrates the enrichment for plasma membrane proteins, indicating that the optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy permits the coverage of membrane tyrosine phosphorylation without affinity enrichment (Fig. 3(b)).


image file: c8an00792f-f3.tif
Fig. 3 GO cellular component for (a) all the phosphoproteins and (b) the tyrosine phosphorylated proteins identified from 500 μg HEK 293T peptides by the optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy.

Performance of the Phospho-SISPROT

The performance of Phospho-SISPROT for phosphoproteomic analysis was evaluated by processing 100 μg HEK 293T cell lysates directly. For comparison, 100 μg predigested HEK 293T peptides (in-solution digestion) were purified with the Ti4+-IMAC tip and desalted with the C18 StageTip. Duplicate analysis was performed by these two methods. As illustrated in Fig. 4, about 5500 unique phosphopeptides were identified from 100 μg predigested peptides (Table S6), while approximately 5900 unique phosphopeptides (88.8% of total peptides, Table S7) were identified from 100 μg cell lysates directly processed by the Phospho-SISPROT. It should be noted that the Phospho-SISPROT started from the cell lysates and obtained the corresponding unique phosphopeptides to those of the peptides predigested by the in-solution digestion of milligrams of cell lysates. Therefore, the Phospho-SISPROT integrating protein digestion step is a very highly sensitive technology and suitable for the phosphoproteome profiling of minute amounts of samples.
image file: c8an00792f-f4.tif
Fig. 4 Phosphopeptides identified by the Ti4+-IMAC tip purifying 100 μg predigested HEK 293T peptides (in-solution digestion) and the Phospho-SISPROT processing 100 μg HEK 293T cell lysates directly. Duplicate analysis was performed; R1 and R2 represent replicate 1 and replicate 2.

Global protein identification with the flow through of Phospho-SISPROT

For the accurate quantitative analysis of phosphosites, it is necessary to quantitatively analyze the proteins to eliminate the changes in protein level. Normally, some sample is split for the global protein identification and quantification. The loading buffer for TiO2 enrichment contains lactic acid which is hard to remove. Whereas, the loading buffer for Ti4+-IMAC enrichment only includes 80% ACN and 6% TFA which is easy to remove. Therefore, it is expected that the flow through of Phospho-SISPROT could be reused for the global protein analysis.

The flow through of sample loading during the phosphopeptide enrichment was lyophilized to dryness, desalted, and detected by LC-MS/MS. For comparison, the cell lysates processed by the SISPROT22 and the in-solution digestion method24 were also analyzed. As summarized in Table 1, 3419 proteins and 20[thin space (1/6-em)]801 unique peptides were identified from the flow through of Phospho-SISPROT, which were comparable with those of the SISPROT sample and the in-solution digestion sample. It is concluded that the flow through of Phospho-SISPROT can be reused for the global protein identification and quantification, which is extremely important for accurate phosphoproteome analysis with limited starting materials.

Table 1 Comparison of the identified proteins and unique peptides among the flow through of Phospho-SISPROT, SISPROT sample, and in-solution digestion sample
Samples Identified proteins Identified unique peptides
Flow through of Phospho-SISPROT 3419 20[thin space (1/6-em)]801
SISPROT sample 3417 21[thin space (1/6-em)]019
In-solution digestion sample 3197 19[thin space (1/6-em)]115


Sensitivity of the Phospho-SISPROT

The sensitivity of Phospho-SISPROT was evaluated by processing 20 μg, 10 μg, 5 μg, and 1 μg HEK 293T cell lysates. As shown in Fig. 5, by taking advantage of the Orbitrap Fusion mass spectrometer with fast acquisition speed, more than 5500, 3000, 1900, and 600 unique phosphopeptides were identified from 20 μg, 10 μg, 5 μg, and 1 μg HEK 293T cell lysates, respectively (Table S8 to Table S11). To the best of our knowledge, this performance is the highest reported to date by using the standard LC-MS/MS setup.
image file: c8an00792f-f5.tif
Fig. 5 Sensitivity of the Phospho-SISPROT. Phosphopeptides identified from 20 μg, 10 μg, 5 μg, and 1 μg HEK 293T cell lysates processed by the Phospho-SISPROT. Duplicate analysis was performed; R1 and R2 represent replicate 1 and replicate 2.

It is indicated that the Phospho-SISPROT is more sensitive than the Ti(IV) monolithic tip19 and the Ti4+-IMAC spin tip,20 by which about 1000 phosphosites and 936 unique phosphopeptides were identified from 5 μg HeLa cell protein digest, respectively. More importantly, these two technologies were only used to enrich the phosphopeptides from the predigested protein samples, while the Phospho-SISPROT could be directly applied for the phosphoproteome analysis from cell lysates. In addition, >5000 unique phosphopeptides could also be identified when the cell lysates were reduced from 100 μg to as low as 20 μg. Particularly, 600 unique phosphopeptides were able to be identified from 1 μg cell lysates. As we know, 1 μg HeLa cell lysates approximately corresponds to 5000 HeLa cells,26 and the phosphopeptides could be adsorbed on the surface of the autosampler needle, switching valve, and even silica surface.27 Consequently, we expect that Phospho-SISPROT will be very useful for the phosphoproteome profiling of rare biological samples and even a single population of primary cells with an optimized LC-MS/MS system.

Reproducibility and label-free quantification performance of the Phospho-SISPROT

Since the Phospho-SISPROT integrated the sample preparation steps into pipette tips, high-throughput analysis could be easily achieved by operating multiple tips in the same centrifuge. This design was expected to significantly increase the system reproducibility. To evaluate the reproducibility of the Phospho-SISPROT, triplicate analysis of 5 μg HEK 293T cell lysates was performed. As depicted in Fig. 6(a), 2382, 2361, and 2385 unique phosphopeptides were identified, with 2094 (80%) phosphopeptides identified in all three replicates (Table S12). This performance is comparable to those of SISPROT22 and a recent report using a spintip-based high-pH RP fractionation method28 for global protein identification, in which 73% and 80% of proteins were identified in three and two replicates respectively, indicating the good identification reproducibility of the Phospho-SISPROT. Fig. 6(b–d) show the correlation of the intensities of phosphopeptides between any two replicates. A good Pearson correlation coefficient of 0.94 is obtained, demonstrating the high label-free quantification precision of the Phospho-SISPROT.
image file: c8an00792f-f6.tif
Fig. 6 Reproducibility and label-free quantification performance of the Phospho-SISPROT in triplicate analysis of 5 μg HEK 293T cell lysates. (a) Common and unique phosphopeptides identified in each replicate. (b–d) Correlation of the intensities of phosphopeptides between any of two replicates. R1, R2, and R3 represent the intensities of phosphopeptides identified from replicate 1, 2, and 3, respectively.

Conclusions

An integrated strategy based on the Phospho-SISPROT and the high-pH RP fractionation plus Ti4+-IMAC enrichment was developed for the highly sensitive phosphoproteome analysis from low micrograms of protein samples. The optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy was powerful for the global phosphoproteomics with identification of >10[thin space (1/6-em)]000 unique phosphopeptides from <500 μg cell lysates. Phospho-SISPROT demonstrated the attractive features of high sensitivity, high reproducibility, high throughput, and a short sample preparation time (6 hours). With these advantages, the Phospho-SISPROT achieved the maximum identification of unique phosphopeptides from less than 20 μg cell lysates by using the standard LC-MS/MS setup. The flow through of Phospho-SISPROT could be reused for the global protein identification and quantification, which is very helpful for accurate phosphoproteome analysis with limited starting materials. In summary, the Phospho-SISPROT and the optimized high-pH RP fractionation plus Ti4+-IMAC enrichment strategy should be well suited for the global phosphoproteome analysis of rare biological samples and even a single population of primary cells with an optimized LC-MS/MS system.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Shenzhen Innovation of Science and Technology Commission (JSGG20160301103415523, JCYJ20150901153557178, and JCYJ20160229153100269), the National Natural Science Foundation of China (31700088 and 21575057), the Ministry of Science and Technology of China (2016YFA0501403 and 2016YFA0501404), and the funding from Guangdong Province (2016A030312016 and 2017B030301018).

Notes and references

  1. T. Yi, B. Zhai, Y. Yu, Y. Kiyotsugu, T. Raschle, M. Etzkorn, H.-C. Seo, M. Nagiec, R. E. Luna, E. L. Reinherz, J. Blenis, S. P. Gygi and G. Wagner, Proc. Natl. Acad. Sci. U. S. A., 2014, 111, E2182–E2190 CrossRef PubMed.
  2. J. G. Abelin, P. D. Trantham, S. A. Penny, A. M. Patterson, S. T. Ward, W. H. Hildebrand, M. Cobbold, D. L. Bai, J. Shabanowitz and D. F. Hunt, Nat. Protoc., 2015, 10, 1308–1318 CrossRef PubMed.
  3. S. J. Humphrey, S. B. Azimifar and M. Mann, Nat. Biotechnol., 2015, 33, 990–995 CrossRef PubMed.
  4. P. Blume-Jensen and T. Hunter, Nature, 2001, 411, 355–365 CrossRef PubMed.
  5. S. A. Johnson and T. Hunter, Nat. Methods, 2004, 2, 17–25 CrossRef PubMed.
  6. K. Sharma, R. C. J. D'Souza, S. Tyanova, C. Schaab, J. R. Wiśniewski, J. Cox and M. Mann, Cell Rep., 2014, 8, 1583–1594 CrossRef PubMed.
  7. P. Mertins, J. W. Qiao, J. Patel, N. D. Udeshi, K. R. Clauser, D. R. Mani, M. W. Burgess, M. A. Gillette, J. D. Jaffe and S. A. Carr, Nat. Methods, 2013, 10, 634–637 CrossRef PubMed.
  8. F. Sacco, A. Silvestri, D. Posca, S. Pirrò, P. F. Gherardini, L. Castagnoli, M. Mann and G. Cesareni, Cell Syst., 2016, 2, 159–171 CrossRef PubMed.
  9. J. R. Wiśniewski, N. Nagaraj, A. Zougman, F. Gnad and M. Mann, J. Proteome Res., 2010, 9, 3280–3289 CrossRef PubMed.
  10. B. Ruprecht, H. Koch, G. Medard, M. Mundt, B. Kuster and S. Lemeer, Mol. Cell. Proteomics, 2015, 14, 205–215 CrossRef PubMed.
  11. R. Tian, H. Wang, G. D. Gish, E. Petsalaki, A. Pasculescu, Y. Shi, M. Mollenauer, R. D. Bagshaw, N. Yosef, T. Hunter, A.-C. Gingras, A. Weiss and T. Pawson, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, E1594–E1603 CrossRef PubMed.
  12. C. Song, M. Ye, G. Han, X. Jiang, F. Wang, Z. Yu, R. Chen and H. Zou, Anal. Chem., 2010, 82, 53–56 CrossRef PubMed.
  13. C. Song, M. Ye, Z. Liu, H. Cheng, X. Jiang, G. Han, Z. Songyang, Y. Tan, H. Wang, J. Ren, Y. Xue and H. Zou, Mol. Cell. Proteomics, 2012, 11, 1070–1083 CrossRef PubMed.
  14. J. A. Paulo, A. Gaun and S. P. Gygi, J. Proteome Res., 2015, 14, 4246–4256 CrossRef PubMed.
  15. S. B. Ficarro, Y. Zhang, M. J. Carrasco-Alfonso, B. Garg, G. Adelmant, J. T. Webber, C. J. Luckey and J. A. Marto, Mol. Cell. Proteomics, 2011, 10, 1–19 CrossRef PubMed.
  16. C. M. Potel, M.-H. Lin, A. J. R. Heck and S. Lemeer, Nat. Methods, 2018, 15, 187–190 CrossRef PubMed.
  17. C. M. Potel, M.-H. Lin, A. J. R. Heck and S. Lemeer, Mol. Cell. Proteomics, 2018, 17, 1028–1034 CrossRef PubMed.
  18. H. Dohyun, J. Jonghwa, Y. Jiyoung, K. Kyunggon and K. Youngsoo, Proteomics, 2015, 15, 618–623 CrossRef PubMed.
  19. F. Liu, H. Wan, Z. Liu, H. Wang, J. Mao, M. Ye and H. Zou, Anal. Chem., 2016, 88, 5058–5064 CrossRef PubMed.
  20. Y. Yao, J. Dong, M. Dong, F. Liu, Y. Wang, J. Mao, M. Ye and H. Zou, J. Chromatogr., A, 2017, 1498, 22–28 CrossRef PubMed.
  21. H. Zhou, F. Elisma, N. J. Denis, T. G. Wright, R. Tian, H. Zhou, W. Hou, H. Zou and D. Figeys, J. Proteome Res., 2010, 9, 1279–1288 CrossRef PubMed.
  22. W. Chen, S. Wang, S. Adhikari, Z. Deng, L. Wang, L. Chen, M. Ke, P. Yang and R. Tian, Anal. Chem., 2016, 88, 4864–4871 CrossRef PubMed.
  23. J. Rappsilber, M. Mann and Y. Ishihama, Nat. Protoc., 2007, 2, 1896–1906 CrossRef PubMed.
  24. G. H. H. Borner and A. B. Fielding, Cold Spring Harb. Protoc., 2014, 2014, 1192–1195 Search PubMed.
  25. M. Dong, Y. Bian, Y. Wang, J. Dong, Y. Yao, Z. Deng, H. Qin, H. Zou and M. Ye, Anal. Chem., 2017, 89, 9307–9314 CrossRef PubMed.
  26. N. A. Kulak, G. Pichler, I. Paron, N. Nagaraj and M. Mann, Nat. Methods, 2014, 11, 319–324 CrossRef PubMed.
  27. T. Masuda, N. Sugiyama, M. Tomita and Y. Ishihama, Anal. Chem., 2011, 83, 7698–7703 CrossRef PubMed.
  28. R. T. Lawrence, E. M. Perez, D. Hernández, C. P. Miller, K. M. Haas, H. Y. Irie, S.-I. Lee, C. A. Blau and J. Villén, Cell Rep., 2015, 11, 630–644 CrossRef PubMed.

Footnote

Electronic supplementary information (ESI) available: Phosphopeptides and phosphosites identified in this study. All the raw mass spectrometry data have been deposited in the public proteomics repository MassIVE and are accessible at ftp://massive.ucsd.edu/MSV000082435. See DOI: 10.1039/c8an00792f

This journal is © The Royal Society of Chemistry 2018