Site-Specific Quantification of Protein Ubiquitination on MS2 Fragment

Subscriber access provided by UNIV NEW ORLEANS

Article

Site-Specific Quantification of Protein Ubiquitination on MS2 Fragment Ion Level via Isobaric Peptide Labeling Ting Cao, Lei Zhang, Ying Zhang, Guoquan Yan, Caiyun Fang, Huimin Bao, and Haojie Lu Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b02654 • Publication Date (Web): 05 Oct 2017 Downloaded from http://pubs.acs.org on October 9, 2017

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.

Analytical Chemistry 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 16

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

Analytical Chemistry

SiteSite-Specific Quantification of Protein Ubiquitination on MS2 Fragment Ion Level via Isobaric Peptide Labeling Ting Cao1, Lei Zhang2, Ying Zhang2, Guoquan Yan1,2, Caiyun Fang1, Huimin Bao1, Haojie Lu1,2* 1 Shanghai

Cancer Center and Department of Chemistry, Fudan University, Shanghai 200032, P. R. China Institutes of Biomedical Sciences and Key Laboratory of Glycoconjugates Research Ministry of Public Health, Fudan University, Shanghai 200032, P. R. China 2

ABSTRACT: Proteome-wide quantitative analysis of protein ubiquitination is important to gain insight into its various cellular functions. However, it is still challenging to monitor how ubiquitination at each individual lysine residue is independently regulated especially whereabouts of peptides containing more than one ubiquitination site. In recent years, isobaric peptide termini labeling has been considered a promising strategy in quantitative proteomics, benefiting from its high accuracy by quantifying with series of b, y fragment ion pairs. Herein, we extended the concept of isobaric peptide termini labeling to large-scale quantitative analysis of protein ubiquitination. A novel MS2 fragment ion based quantitative approach was developed allowing quantifying ubiquitination at site level via isobaric K-ε-GG peptide labeling, which combined metabolic labeling, K-ε-GG immunoaffinity enrichment and site-selective N-terminus dimethylation. The feasibility of this proposed strategy was demonstrated through the ubiquitin proteome analysis of differently labeled MCF-7 cell digests. As a result, 2970 unique K-ε-GG peptides of 1383 proteins containing 2874 ubiquitinated sites were confidently quantified with high accuracy and sensitivity. In addition, we demonstrated that quantification on MS2 fragment ion level makes it possible to precisely quantify each individual ubiquitinated lysine residue in 39 K-ε-GG peptides bearing two ubiquitination sites by the use of specific ubiquitinated b, y ion pairs. It is expected that this proposed approach will serve as a powerful tool to quantify ubiquitination at site level, especially for those multi-ubiquitinated peptides. Ubiquitination is a process that refers to the covalent attachment of the 76-amino-acid protein ubiquitin to a substrate protein, which is catalyzed through the sequential action of three different enzymes: E1 ubiquitin activating enzymes, E2 ubiquitin conjugating enzymes, and E3 ubiquitin ligating enzymes.1,2 In addition to mono-ubiquitination, there are seven lysine residues in the ubiquitin sequence (K6, K11, K27, K29, K33, K48, and K63) and the α-amino group of its N-terminal methionine are known to participate in the formation of poly-ubiquitin chains.3,4 Protein ubiquitination regulates a wide variety of biological processes including protein degradation, immune response, signal transduction and DNA repair.5,6 Alterations of protein ubiquitination have been implicated in many diseases, such as cancer, inflammation, neurodegenerative diseases, and immune disorders.7,8 Advancements in mass spectrometry (MS)-based proteomics and the commercialization of antibodies recognizing lysine residues modified with a di-glycine remnant (K-ε-GG) following trypsin digestion have greatly improved the global analysis of protein ubiquitination.9-12 Recent reports utilizing these antibodies for enrichment of K-ε-GG peptides have identified thousands of in vivo ubiquitination sites.13-15 It’s worth noting that proteins modified by NEDD8 or ISG15 generate an identical di-glycine remnant on modified lysines after trypsin digestion. These modifications and ubiquitination are indistinguishable by K-ε-GG antibodies. However, expressions of NEDD8 and

ISG15-mediated modifications in cells seem to be restricted to a small fraction of proteins.16,17 A great majority of cellular peptides containing the di-glycine remnant are believed to stem from ubiquitinated proteins. And in this study, the di-glycine modified proteome and K-ε-GG peptides were referred to as the ubiquitinated proteome and ubiquitinated peptides, respectively. As signaling through the ubiquitin pathway is highly dynamic, quantitative analysis to determine the changes in endogenous protein ubiquitination at the proteome level would contribute to deep insight into various cellular processes of ubiquitination.18,19 Stable isotope labeling with amino acids in cell culture (SILAC) based quantification combined with K-ε-GG enrichment method has been mostly performed to detect changes at ubiquitination level.20,21 In addition, researches using isobaric labeling with iTRAQ (isobaric tag for relative and absolute quantitation) or TMT (tandem mass tag) reagents have been conducted to profile quantitative changes in the ubiquitinomes of cells and tissues and demonstrated that the di-glycine remnant IP variability was sufficiently low by carefully controlled sample handling.22,23 In the last few years, some innovative quantitative methods have been developed using b, y fragment ion pairs in the whole mass range of the MS2 spectra for quantification.24-29 Since isobaric peptide termini labeling (IPTL)24 was introduced, several variations of the strategy have been performed, such as pseudo-isobaric dimethyl 1

ACS Paragon Plus Environment


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

labeling (pIDL)30 and in vivo termini amino acid labelling (IVTAL)31. The basic principle of these strategies is to achieve isobaric labeling of peptides through crosswise labeling with different isotope reagents and produce identical molecular weight in the MS1 spectra but b, y fragment ion pairs in the MS2 spectra. Multiple quantification points per peptide with reciprocal ratios for N-terminal and C-terminal fragments in the MS2 spectra are obtained, hence, the robustness, reliability and accuracy of the quantification will be improved comparing to mass tags in the low mass range of the MS2 spectra or single precursor ion pairs in MS1 scan. Although IPTL has already been employed for the global proteome quantification analysis, there are no published studies to apply it for the investigation of protein posttranslational modifications. In theory, quantification on fragment ion level with b, y ion pairs has great advantages in posttranslational modifications. Because in addition to the improved accuracy with multiple quantifiable points, site-specific quantitative information of the modified sites could be obtained with fragment modified b, y ion pairs. Especially when multiple modifications occur on a single peptide, quantification on fragment ion level makes it possible to measure fold-changes at each individual modification site between samples. Herein, for the first time, we extended the concept of isobaric peptide termini labeling to large-scale quantitative ubiquitination analysis. A novel MS2 fragment ion based approach was developed that allowed quantifying ubiquitination at site level with high accuracy via isobaric K-ε-GG peptide labeling which can be achieved by SILAC and site-selective N-terminus dimethylation using different combinations of isotopic reagents. This proposed strategy was applied to analyze various proportions of mixed samples of two labeled MCF-7 cell lines, which showed that K-ε-GG peptides with one and two ubiquitinated lysine residues were both successfully quantified. And we demonstrated that quantification on MS2 fragment ion level is quite suitable for the quantitative analysis of K-ε-GG peptides containing multiple ubiquitinated lysine residues32,33 by the use of ubiquitinated b, y ion pairs.

Page 2 of 16

Tris-HCl (pH 8.0), and digested overnight at 37 oC with trypsin (sigma) at an enzyme to substrate ratio of 1:50 (wt/wt). Protease digestion was quenched by adding trifluoroacetic acid (TFA) to a final concentration of 0.5% (vol/vol). Equal amounts of K-ε-GG peptides (YDFFILNK(GG)LAK, NSSYVLLK(GG)TGK) were added as internal standard peptides (IS peptides). The digests were subsequently desalted using a C18 Sep-Pak SPE cartridge (Waters). Desalted peptides were lyophilized in vacuum. K-ε-GG Peptides Enrichment Enrichment. chment. Lyophilized peptides were reconstituted in 1.5 mL of ice-cold IAP buffer (10 mM Na2HPO4, 50 mM NaCl in 50 mM MOPS pH 7.2) and incubated with K-ε-GG ubiquitin remnant motif antibody bead conjugates (Cell Signaling Technologies) for 2 h at 4 oC on a rotation wheel. Antibody beads were washed three times with ice-cold IAP buffer followed by three times washes with ice-cold water. Immunoprecipitated peptides were eluted by adding 50 µL of 0.15% TFA two times. The eluted fraction and supernatant were both collected. N-terminus Dimethyl Labeling. Labeling. The eluted peptides were reacted with 4% formaldehyde (CH2O) or formaldehyde-d2 (CD2O), followed by the addition of 0.6 M sodium cyanoborohydride (NaCNBH3) in 1% acetic acid. The mixtures were incubated for 5 min at room temperature in a fume hood. Subsequently, 1% ammonium hydroxide was added and incubated to quench the reaction. Finally, the two differently labeled peptides were mixed, purified with µ-C18 ZipTips (Millipore) and analyzed by HPLC-MS/MS. Peptides in the supernatant were performed the same steps as the eluted peptides. Three experiments were performed with three pairs of MCF-7 cells respectively. NanoNano-HPLCHPLC-MS/MS Analysis. Analysis. The eluted peptide samples and the supernatant peptide samples were resuspended with solvent A respectively (water with 0.1% formic acid), separated by nano LC and analyzed by on-line electrospray tandem mass spectrometry. The experiments were performed on an EASY-nLC 1000 system (Thermo Fisher Scientific, Waltham, MA) connected to an Orbitrap Fusion mass spectrometer (Thermo Fisher Scientific, San Jose, CA) equipped with an online nano-electrospray ion source. Peptides were loaded onto the analytical column (Acclaim PepMap C18, 75 µm x 50 cm) and subsequently separated with a linear gradient, solvent B (ACN with 0.1% formic acid) from 2% to 30% in 105 min. The column was re-equilibrated at initial conditions for 15 min. The column flow rate was maintained at 200 nL/min. The electrospray voltage of 2.0 kV versus the inlet of the mass spectrometer was used. The Orbitrap Fusion mass spectrometer was operated in the data-dependent mode to switch automatically between MS and MS/MS acquisition. Survey full-scan MS spectra (m/z 350-1500) were acquired in Orbitrap with a mass resolution of 120 000 at m/z 200. The AGC target was set to 300 000, and the maximum injection time was 50 ms. MS/MS acquisition was performed in Orbitrap with 3 s cycle time, the resolution was 15 000 at m/z 200. The intensity threshold was 50 000, and the maximum injection time was 200 ms. The AGC target was set to 200 000, and the isolation window was 2 m/z. Ions were fragmented by higher energy collisional dissociation (HCD) with a normalized collision energy (NCE) of 35%, fixed first mass was set at 110. In all cases, one microscan was recorded using dynamic exclusion of 30 seconds.

MATERIALS AND EXPERIMENTAL PROCEDURE PROCEDURE Cell Culture. Culture. MCF-7 breast cancer cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, penicillin, and streptomycin in a 37 oC, 5% CO2 incubator. For SILAC labeling, cells were grown in media containing either normal arginine (Arg0) and normal lysine (Lys0) or arginine-15N4 (Arg4) and lysine-D4 (Lys4) (Cambridge Isotope Laboratories) for ~ 6 doublings. Prior to harvest, cells were treated with 5 µM MG-132 (sigma) for 4 h. Cell Lysis and Trypsin Digestion. Digestion. Collected cells were washed twice with phosphate-buffered saline (PBS) to remove residual cell culture medium and then lysed in a buffer containing 7 M urea, 2 M thiourea, 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 50 µM PR-619 (sigma), 1 mM chloroacetamide and protease inhibitors (Complete tablets; Roche Diagnostics). Lysates were sonicated on ice for 5 min and centrifuged at 20000 rcf for 20 min at 4 oC to collect the supernatant. Protein concentration was measured using the Bradford assay, and 15 mg proteins were reduced with 5 mM dithiothreitol (DTT) for 1 h at 37 oC and subsequently alkylated by 10 mM iodoacetamide for 45 min at 25 oC in the dark. Lysates were then diluted to 2 M urea with 50 mM 2


Page 3 of 16

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


MS Data Analysis. Analysis. All MS/MS spectra were searched twice with SEQUEST [v.28 (reversion 12)] against the human swissprot database (Homo sapiens, 20131 entries). Two sets of search parameters were used. For the first set of parameters, N-terminus (+28.0313 Da) and arginine (+3.9881 Da) were set up as a fixed modification and lysine (+4.0251 Da) were set up as a variable modification for CH2O-derived peptides. Whereas for the second set of parameters, N-terminus (+32.0564 Da) were set up as a fixed modification for CD2O-derived peptides. The same parameters of the two searches were set up as follows: fixed modification on the cysteine (+57.0215 Da), variable modification on the methionine (+15.9949 Da) and lysine (+146.0993 Da), partial tryptic cleavage with two missed cleavage sites, a mass tolerance of 10 ppm for the precursor ions and 0.05 Da for the fragment ions. To statistically validate the accuracy of peptide assignments to tandem mass spectra from SEQUEST, Trans Proteomic Pipeline software (revision 4.5, Institute of Systems Biology, Seattle, WA) was utilized to score the search results. Peptides with a Peptide Prophet probability over 0.90 and proteins with a Protein Prophet probability over 0.95 were retained. The false discovery rate was limited to less than 1%. Results of the two sets of database searches were combined to identify the peptides and proteins. Quantification was performed using an in-house software (ITMSQ),34 which is a convenient software tool to solve the issue of the co-fragmentation of isobaric peptides by spectrum splitting and provide the accurate quantification results. For quantifying ubiquitination, only those peptides containing modification on the lysine (+146.0993) were used for its quantification and two modes of the quantification operation were performed with ubiquitinated b, y ions and non-ubiquitinated b, y ions. The outlier data points of the ratios of b, y fragment ion pairs were removed using box plot. The ratio of a peptide was the median of ratios of assigned b, y fragment ion pairs in the MS2 spectra.

Figure 1. Schematic diagram of site-specific quantification of ubiquitination via isobaric K-ε-GG labeling. Cell states A and B were, respectively, cultured in media with lysine (K0) and arginine (R0) or lysine-D4 (K4) and arginine-15N4 (R4), digested by trypsin, and internal standard peptides (IS peptides) were added. Then, peptides from two samples were subjected to K-ε-GG immunoaffinity enrichment, separately. The N-termini of K-ε-GG peptides in different samples were dimethylated using CH2O and CD2O respectively. The mixed labeled peptides resulted in isobaric masses and showed single peak in MS1 scan, but exhibited multiple b, y fragment ion pairs with a mass difference of 4 Da (single charged fragment ion) or 2 Da (doubly charged fragment ion) for quantification. Red dots represent ubiquitin modified lysine (K0), blue dots represent ubiquitin modified D4-lysine (K4).

RESULTS AND DISCUSSION Strategy for SiteSite-Specific Quantification of Ubiquitination by isobaric isobaric KK-ε-GG peptide labeling. labeling. The basic idea of the proposed strategy is complementary isotopic labeling of ubiquitinated peptides to obtain the isobaric-labeled K-ε-GG peptides which could be used to produce b, y fragment ion pairs in the MS2 spectra for site-specific ubiquitination quantification. These requirements can be achieved via SILAC and site-selective N-terminus dimethylation using different combinations of isotopic reagents. Owing to the low abundance of protein ubiquitination, di-glycine ubiquitin remnant motif (K-ε-GG) immunoaffinity enrichment was applied. Dimethylation labeling was performed following the enrichment to prevent its influence on

immunoprecipitation with K-ε-GG antibodies, because the anti-di-glycine antibodies could not recognize dimethylated K-ε-GG peptides.14,22,23 An outline of the strategy is shown in Figure 1. First, cell states A and B were, respectively, cultured in media with lysine (K0) and arginine (R0) or lysine-D4 (K4) and arginine-15N4 (R4) for a sufficient period of time to obtain nearly complete protein labeling. Next, the cell lysates were digested with trypsin to generate peptides with lysine or arginine at the C-termini end. In order to demonstrate the technical variability of the workflow, especially the variability of di-glycine enrichment, equal amounts of K-ε-GG peptides were added into tryptic digests from each cell state as internal standard. Then, these two parts of peptides were subjected to K-ε-GG immunoaffinity enrichment, separately. The eluted fraction and supernatant

3



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 16

Figure 2. Identification and quantification of ubiquitinated proteome. Venn diagram showing the overlap of identified K-ε-GG peptides (A, left) and ubiquitinated proteins (A, right), quantified K-ε-GG peptides (B, left) and ubiquitinated proteins (B, right). Figure 3. Scatterplots for the quantified K-ε-GG peptide ratios by ubiquitinated and non-ubiquitinated b, y fragment ion pairs. “L” represents peptides in state A, “H” represents peptides in state B.

were both collected. The α-amine groups at the N-termini of eluted peptides in the two samples were subsequently dimethylated using CD2O (States A) and CH2O (States B) in acetic acid solution combining with reductive amination using sodium cyanoborohydride, respectively. Next, the differentially labeled peptides were mixed and subjected to LC-MS/MS analysis. It’s worth noting that the quantitative information of supernatant proteome could also be measured simultaneously in this research. The supernatant was labeled as the eluted fraction did. (Figure S-1). Peptides in isobaric form from two different samples showed single peak in MS1 scan, but exhibited multiple b, y fragment ion pairs with a mass difference of 4 Da (single charged fragment ion) or 2 Da (doubly charged fragment ion) in the MS2 spectra and their intensity ratios could be used for relative quantification. Here, the lower mass ion of every b fragment ion pair was derived from State B labeling with CH2O and the higher mass ion was derived from State A labeling with CD2O, and vice versa for every y fragment ion pair. It is noteworthy that peptides with missed cleavage site except K-ε-GG site are not isobaric.

compared with previously identified in other studies. In total, about 60% of the ubiquitinated sites identified in this study have been reported by other MS-based studies.15,16,20 The enrichment selectivity of K-ε-GG peptides was 45%, as determined by the number of modified peptides divided by the total number of all identified peptides in the eluted fraction. And ubiquitinated proteins accounted for 56% of all identified proteins, which is comparable with other studies.16,20 In the enriched dataset, about 2% of ubiquitinated peptides and over half of ubiquitinated proteins contained >1 ubiquitination sites (Figure S-2). In addition to the analysis of K-ε-GG proteome, peptides corresponding to the supernatant were also analyzed. In total, 3492 proteins and 18286 unique peptides were identified in the supernatant (Figure S-3A). In the supernatant, only 0.16% of identified peptides were ubiquitinated, among which 4 peptides were also identified in the eluted fraction (Figure S-4A). An overlap of 1005 proteins were identified in K-ε-GG modified proteome with K-ε-GG peptides and supernatant proteome with non-ubiquitinated peptides (Figure S-4B). 739 proteins were only identified in the K-ε-GG modified proteome, which demonstrated that K-ε-GG remnant enrichment could improve the identification of ubiquitination proteins and indicated the necessity of K-ε-GG remnant enrichment in the MS-based analysis of the ubiquitinated proteome in complex biological samples.

Identification of Endogenous Protein Ubiquitination. Dimethylation of K-ε-GG peptides caused a mass shift of 146.0993 Da for each modified lysine residue in these peptides, mass of dimethyl labeling reagent added to di-glycine ubiquitin remnant (28 Da + K4-ε-GG, 32 Da + K0-ε-GG). This increased mass was used as a variable modification of lysine in database search in order to determine protein ubiquitination sites. This search algorithm can avoid the misassignment of other modifications isobaric to the 114.043 Da di-glycine modification, such as artifactual alkylation with iodoacetamide on free amines in the side chain of lysines,35 the presence of Asn in the proximity of lysine residues and so on,17 which increased the confidence in the detection of protein ubiquitination. In order to identify more ubiquitinated peptides, cells were exposed to proteasome inhibitor MG-132 for 4 h. In total, we identified 3856 ubiquitinated sites on 4189 unique K-ε-GG peptides from 1744 proteins in the eluted fraction after K-ε-GG remnant enrichment across three experiments (Figure 2A). The identified ubiquitinated sites were

Quantification of Ubiquitination on MS2 Fragment Ion Level with b, y Fragment Ion pairs. To verify the feasibility of the application of isobaric peptide termini labeling to large-scale quantitative ubiquitination analysis, the same amount of proteins was extracted from MCF-7 cell lines grown in media supplemented with either light lysine/arginine (K0/R0) or heavy lysine/arginine (K4/R4), digested, enriched, separately labeled and analyzed by LC-MS/MS as described in Figure 1. Complete labeling of all peptides is essential for accurate quantification. To ensure the metabolic labeling efficiency, MCF-7 cells grown for 6 passages in K 4 R 4

4


Page 5 of 16

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


Figure 4. MS/MS spectrum of the peptide TITLEVEPSDTIENVK(GG)AK with the ratio of 1:1.

showed no chromatographic shift between the 1H and 2H-labeled fragment ions (Figure S-7A vs 7B), as well as the 14N and 15N-labeled y8 fragment ions (Figure S-7C vs 7D). This may be due to the fact that the deuterium atoms grouping around charged amino residues might reduce the isotope effect of dimethyl labeling.36,37 Thus, co-elution of the isobaric-labeled peptides in the proposed strategy ensured accurate and reliable quantification results. After N-terminus dimethylation of K-ε-GG peptides, the quantitative information could be located on the ubiquitinated lysine residues in the form of dimethylated-GlyGly. As a result, quantitative information can be obtained from two kinds of b, y ion pairs for the ubiquitinated proteome by an in-house edited script. One type is ubiquitinated b, y ions containing the ubiquitinated lysine residue, the other type is that the corresponding non-ubiquitinated b, y ions containing no ubiquitinated lysine residue of the same peptide, both of which could be used for ubiquitination quantification. Take TITLEVEPSDTIENVK(GG)AK for example, ubiquitinated b, y ions include b16, b17 and y3 to y17, non-ubiquitinated b, y ions include b1 to b15 and y1, y2. To evaluate the feasibility and reliability of these two quantification manners, a correlation analysis was conducted between the quantification results calculated by ubiquitinated and non-ubiquitinated b, y ion pairs with a theoretical ratio of 1:1 (Figure 3). In fact, the quantitative ratio of the same peptide from non-ubiquitinated and mono-ubiquitinated ions would be nearly identical. The consistency of the results of these two quantification manners indicated the reliability of the qualitative and quantitative results, and provided the possibility and guarantee for the quantification of different ubiquitinated sites in multi-ubiquitinated peptides. The information provided by these two types of quantification manners can play a complementary role in the proposed strategy. Ubiquitinated b, y ion pairs provided the quantification information of the exact modified sites. And non-ubiquitinated b, y ion pairs provided complementary quantification of the ubiquitinated peptides. In addition, the identification result of K-ε-GG peptides was further verified by these ubiquitinated b, y ion pairs because only K-ε-GG peptides

medium were extracted, digested and identified by LC-MS/MS. Over 99% of peptides were identified with heavily labeled lysine and arginine (Table S-1). Compared with the labeled peptides, the intensity of unlabeled peptides can be disregarded (Figure S-5), which indicated a complete incorporation of arginine-15N4 and lysine-D4. Site-selective peptide N-terminus dimethylation could be achieved by adjusting the pH of labeling solution.25 In the case of ubiquitinated proteome, a new type of isopeptide which leaves two glycine residues on the modified lysine residue contains an additional N-terminus compared with linear peptide. Thus, methyl groups will add to both the native N-terminus and the distal end of the di-glycine ubiquitin remnant after N-terminus dimethyl labeling. In order to evaluate the ubiquitinated peptide labeling efficiency of N-terminus dimethylation, standard K-ε-GG peptide “YDFFILNK(GG)LAK” (Figure S-6A) was derivatized with CH2O and CD2O, separately. After dimethylation, the peak of the unmodified peptide at m/z 1485 Da disappeared completely in both spectra. Peptides were incorporated into two dimethyl groups exhibiting a 56 Da (Figure S-6B) or 64 Da (Figure S-6C) mass increase, respectively, which indicated that two α-amino groups of K-ε-GG peptide N-terminus were dimethylated and derivatization reactions had run to completion. Additionally, we performed database search with dimethylation set up as a variable modification on peptide N-terminus to evaluate the dimethyl labeling efficiency in the supernatant and elute fractions. The efficiency of the N-terminus dimethylation were about 99% (Table S-2), which indicated highly labeling efficiency of peptides in a complex peptide mixture. To ensure the accuracy of the quantitative results, we examined the isotope effect of the isobaric labeled peptides. Lysine-ending peptides which the same numbers of deuterium atoms were incorporated into had no isotope effect. For those arginine-ending peptides, selective ion chromatograms of the lightly and heavily labeled b, y fragment ions from the MS2 spectra were extracted. Take the (+28)DLMVGDEASELR(+4) and labeled peptides of (+32)DLMVGDEASELR for example: extracted ion chromatograms of the lightly labeled b2-ion ((+28)DL, m/z 257.14) and the heavily labeled b2-ion ((+32)DL, m/z 261.17)

5



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

Figure 6. Relative abundance of seven polyubiquitin linkages.

71% for the identified K-ε-GG peptides could be accurately quantified. The median values of the quantified ratios for ubiquitinated peptides in three replicates were 1.05, 1.06 and 1.03. Furthermore, over 95% of quantified ratios ranged from 0.5 to 2 (on log2 scale it is -1 to 1) and the RSDs of the quantified K-ε-GG peptides from three replicates were 32%, 33% and 33%, which indicated good quantitative precision and considerable reproducibility of the proposed strategy. Histograms of the ratio distribution and scatter diagrams between these replicates were shown in Figure S-8. Samples with ratios of 1:5, 1:10 were analyzed to evaluate the dynamic range of the proposed quantitative strategy (Figure S-9). A dynamic range across nearly 2 orders of magnitude were obtained. Although some deviations were observed in quantification with a higher fold. The quantitative analysis of differentially regulated ubiquitination sites relies on the determination of not only the ubiquitinated peptides, but also other unmodified peptides representing the corresponding protein level. In some cases, differential ubiquitination expression was observed with unchanged protein abundance. In this proposed strategy, linear peptides without internal lysines and arginines from supernatant were also isobaric labeled and quantitative information of the supernatant proteome representing the relative protein abundances between two samples could be obtained by fragment ion pairs. Altogether 13696 non-redundant peptides from 3014 proteins were quantified in the supernatant proteome (Figure S-3B). To further assess variability of the workflow, especially the independent immunoaffinity of K-ε-GG peptides from two cell states, synthesized K-ε-GG peptides were added equally into two tryptic digests as internal standard. The quantitative results calculated by comparing the relative peak areas of these standard peptides were nearly identical to 1.00 (Figure S-10), which indicated that these peptides were almost enriched and labeled equally in parallel in different cell lysates, demonstrating the variability was relatively low by carefully controlled sample handling. Then, we investigated the influence of the number of ion pairs per peptide on the quantitative accuracy and compared the obtained ratios which was expected to 1:1 quantified with different numbers of ion pairs, including more than 4 ion pairs, 2-3 ion pairs and only 1 ion pair,

Figure 5. Influence of the number of fragment ion pairs per K-ε-GG peptide on the quantitative result. (A) Boxplots show the quantification result of all peptides by different numbers of ion pairs. (B) The number of K-ε-GG peptides and proteins quantified by different numbers of ion pairs.

would have characteristic ubiquitinated ions. The final quantitative results were the combination of these two quantification manners, that is, the quantitative ratio is a median of all the quantified peptide-spectrum matches (PSMs) from these two manners for a given peptide. For the isobaric labeled peptides, b, y fragment ion pairs with a mass difference of 4 Da (single charged fragment ion) or 2 Da (doubly charged fragment ion) observed in the MS2 spectra were used to calculate the quantitative ratios by comparing their intensities. For example, the precursor ion with m/z of 722.4095 (z=3) was fragmented and generated pairs of b and y ions in MS2 scan (Figure 4). These pairs of b and y ions were assigned to the K27-linked polyubiquitinated peptide of (+32)TITLEVEPSDTIENVK0(GG+32)AK0 and (+28)TITLEVEPSDTIENVK4(GG+28)AK4 after a SEQUEST search. The median of the intensity ratios of b, y ion pairs is 0.99, exhibiting good consistency with the expected value for 1:1 ratio, which validated the accuracy of the strategy. This can be ascribed to the use of multiple quantitative data points for the corresponding peptide and the reduced impact of potentially interfering fragment ions. In the eluted fraction, 2970 unique K-ε-GG peptides of 1383 proteins containing 2874 ubiquitinated sites were quantified in three experiments (Figure 2B). 60% of K-ε-GG peptides and 70% of ubiquitinated proteins were quantified in at least two replicates. In addition, a percentage of

6


Page 6 of 16

Page 7 of 16

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


Figure 7. MS/MS spectrum of the peptide TLTGK(GG)TITLEVEPSDTIENVK(GG)AK.

respectively. As shown in Figure 5A, the quantification accuracy and precise were improved with more quantification points per peptide. The median value was ranging from 1.38 to 1.14 and then to 1.05 with increased numbers of quantitative ion pairs. And we found that peptide quantified with no less than 2 ion pairs would provide a relatively accurate and precise result. Furthermore, we found that over 96% of peptides were quantified with more than 2 ion pairs (Figure 5B), indicating the reliability of the quantification results.

multiple ubiquitinated peptides may contribute to our understanding of regulation of heterogeneous forked ubiquitin chains and various aspects of multi-ubiquitination biology.

CONCLUSIONS In the present study, we have developed a novel approach for site-specific ubiquitination quantification with b, y fragment ion pairs based on K-ε-GG isobaric peptide labeling and performed it to quantitative analysis of ubiquitinated proteome in MCF-7 cell samples. In summary, 2970 unique K-ε-GG peptides of 1383 proteins containing 2874 ubiquitinated sites were successfully quantified. In addition to the known strengths of isobaric peptide labeling strategies such as multiple quantifiable points in MS2 scan, peptides in isobaric form from two different samples appear as one peak only in the MS1 spectra and the signal intensities of peaks are summed. This is very useful for the identification of K-ε-GG peptides in a complex sample. Moreover, site-selective N-terminus dimethylation of K-ε-GG peptides increases the confidence of determination of ubiquitinated peptides by alleviating the false positive discovery rate. The most important feature of the proposed strategy is quantifying each individual ubiquitinated lysine residue in K-ε-GG peptides with more than one ubiquitination sites by the use of specific ubiquitinated b, y ion pairs. At the same time, this strategy described in our study also provides complementary quantitative information of the supernatant proteome on the same sample. To our knowledge, this strategy represents the first site-specific quantitative analysis of protein ubiquitination with fragment b, y ion pairs. The approach described here allows us to quantify changes in ubiquitination on the exact modified lysine at site level, which is expected to be a promising quantitative strategy for the analysis of ubiquitination proteome and especially for the multiple ubiquitinated peptides and forked ubiquitin chain analysis. Moreover, quantification on MS2 fragment ion level by b, y ion pairs could be adopted for other types of posttranslational modifications, such as acetylation.

Quantitative Multi--ubiquitinated Quantitative Analysis of Multi Peptides.. An ubiquitin molecule may be ubiquitinated at any Peptides of the seven lysine residues (K6, K11, K27, K29, K33, K48 or K63). In our analysis, ubiquitinated at all these seven lysine residues were identified. Relative abundance of seven polyubiquitin linkages were quantified by comparing their corresponding peptide spectrum matches. The results were shown in Figure 6, which indicated that K48-linked polyubiquitin chains were the predominantly observed, followed by K33, K29, K63, K6, K11, and K27. Moreover, we identified several forked ubiquitin chains (K11 + K27, K27 + K29 and K29 + K33), where two ubiquitin molecules were linked to adjacent lysines on the proximal ubiquitin molecule at K11 and K27, K27 and K29, or K29 and K33.32,33 In the case of forked ubiquitin chains or other multi-ubiquitinated peptides, two di-glycine modified lysine residues were occurred on the same peptide. Quantifying ubiquitination changes at each modified site in those types of peptides is an analytical challenge. Whereas, our approach could describe quantitative changes of the exact modified site with their individual ubiquitinated b, y fragment ion pairs. For example, MS/MS spectrum of K11 + K27 forked ubiquitin chain TLTGK11(GG)TITLEVEPSDTIENVK27(GG)AK was shown in Figure 7. Fragment ion pairs from b7 to b12 could be assigned to the first ubiquitinated lysine and used for the quantification of ubiquitinated K11. While y3 to y12 ion pairs could be used to quantify ubiquitinated K27. Besides these forked ubiquitin chains, other multiple ubiquitinated peptides were also observed in the eluted dataset. In total, we quantified 39 peptides with two ubiquitination sites. By the employment of site-specific ubiquitinated fragment ion pairs, each ubiquitinated lysine residue in these ubiquitination peptides could be precisely quantified. Site-specific quantitative analysis of these

ASSOCIATED CONTENT Supporting Information 7



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 16

Thiede, B. J. Proteome Res. 2009, 8, 4333-4341. (25) Qin, H.; Wang, F.; Zhang, Y.; Hu, Z.; Song, C.; Wu, R.; Ye, M.; Zou, H. Chem. Commun. 2012, 48, 6265-6267. (26) Yan, W.; Luo, J.; Robinson, M.; Eng, J.; Aebersold, R.; Ranish, J. Mol. Cell. Proteomics 2011, 10, 1-15. (27) Bamberger, C.; Pankow, S.; Park, S. K. R.; Yates, J. R., 3rd J. Proteome Res. 2014, 13, 1494-1501. (28) Xie, L.; Nie, A.; Yang, S.; Zhao, C.; Zhang, L.; Yang, P.; Lu, H. Analyst 2014, 139, 4497-4504. (29) Yang, S.; Nie, A.; Zhang, L.; Yan, G.; Yao, J.; Xie, L.; Lu, H.; Yang, P. J. Proteomics 2012, 75, 5797-5806. (30) Zhou, Y.; Shan, Y. C.; Wu, Q.; Zhang, S.; Zhang, L.; Zhang, Y. Anal. Chem. 2013, 85, 10658-10663. (31) Nie, A.; Zhang, L.; Yan, G.; Yao, J.; Zhang, Y.; Lu, H.; Yang, P.; He, F. Anal. Chem. 2011, 83, 6026-6033. (32) Tagwerker, C.; Flick, K.; Cui, M.; Guerrero, C.; Dou, Y.; Auer, B.; Baldi, P.; Huang, L.; Kaiser, P. Mol. Cell. Proteomics 2006, 5, 737-748. (33) Kim, H. T.; Kim, K. P.; Lledias, F.; Kisselev, A. F.; Scaglione, K. M.; Skowyra, D.; Gygi, S. P.; Goldberg, A. L. J. Biol. Chem. 2007, 282, 17375-17386. (34) Xie, L.; Zhang, L.; Nie, A.; Yan, G.; Yao, J.; Zhang, Y.; Yang, P.; Lu, H. Proteomics 2015, 15, 3755-3764. (35) Nielsen, M. L.; Vermeulen, M.; Bonaldi, T.; Cox, J.; Moroder, L.; Mann, M. Nat. Methods 2008, 5, 459-460. (36) Zhang, R.; Sioma, C. S.; Thompson, R. A.; Xiong, L.; Regnier, F. E. Anal. Chem. 2002, 74, 3662-3669. (37) Hsu, J.; Huang, S.; Chow, N.; Chen, S. Anal. Chem. 2003, 75, 6843-6852.

The Supporting Information is available free of charge on the ACS Publications website. Supplemental data as mentioned in the text (PDF).

AUTHOR INFORMATION Corresponding Author *Phone: (+) 86-21-54237618. Fax: (+) 86-21-54237961. E-mail: [email protected]. Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT The work was supported by the National Key Research and Development Program (2016YFA0501303 and 2015CB910401), NSF (Grants 21335002, 31670835 and 21675031), the Ph.D. Programs Foundation of Ministry of Education of China (20130071110034), Shanghai Projects (Eastern Scholar, Rising star 15QA1400600, 15JC1400700 and B109) and Shanghai Key Laboratory of Clinical Geriatric Medicine (13dz2260700).

REFERENCES (1) Swatek, K. N.; Komander, D. Cell Res. 2016, 26, 399-422. (2) Varshavsky, A. Annu. Rev. Biochem. 2012, 81, 167-176. (3) Xu, P.; Duong, D. M.; Seyfried, N. T.; Cheng, D.; Xie, Y.; Robert, J.; Rush, J.; Hochstrasser, M.; Finley, D.; Peng, J. Cell 2009, 137, 133-145. (4) Yau, R.; Rape, M. Nat. Cell Biol. 2016, 18, 579-586. (5) Husnjak, K.; Dikic, I. Annu. Rev. Biochem. 2012, 81, 291-322. (6) Low, T. Y.; Magliozzi, R.; Guardavaccaro, D.; Heck, A. J. Proteomics 2013, 13, 526-537. (7) Reinstein, E.; Ciechanover, A. Ann. Intern. Med. 2006, 145, 676-684. (8) Cohen, P.; Tcherpakov, M. Cell 2010, 143, 686-693. (9) Peng, J.; Schwartz, D.; Elias, J. E.; Thoreen, C. C.; Cheng, D.; Marsischky, G.; Roelofs, J.; Finley, D.; Gygi, S. P. Nat. Biotechnol. 2003, 21, 921-926. (10) Xu, G.; Paige, J. S.; Jaffrey, S. R. Nat. Biotechnol. 2010, 28, 868-873. (11) Bustos, D.; Bakalarski, C. E.; Yang, Y.; Peng, J.; Kirkpatrick, D. S. Mol. Cell. Proteomics 2012, 11, 1529-1540. (12) Kirkpatrick, D. S.; Denison, C.; Gygi, S. P. Nat. Cell Biol. 2005, 7, 750-757. (13) Wagner, S. A.; Beli, P.; Weinert, B. T.; Schölz, C.; Kelstrup, C. D.; Young, C.; Nielsen, M. L.; Olsen, J. V.; Brakebusch, C.; Choudhary, C. Mol. Cell. Proteomics 2012, 11, 1578-1585. (14) Udeshi, N. D.; Mertins, P.; Svinkina, T.; Carr, S. A. Nat. Protoc. 2013, 8, 1950-1960. (15) Udeshi, N. D.; Svinkina, T.; Mertins, P.; Kuhn, E.; Mani, D. R.; Qiao, J. W.; Carr, S. A. Mol. Cell. Proteomics 2013, 12, 825-831. (16) Wagner, S. A.; Beli, P.; Weinert, B. T.; Nielsen, M. L.; Cox, J.; Mann, M.; Choudhary, C. Mol. Cell. Proteomics 2011, 10, M111.013284. (17) Xu, G.; Jaffrey, S. R. Biotechnol. Genet. Eng. 2013, 29, 73-109. (18) Carrano, A. C.; Bennett, E. J. Mol. Cell. Proteomics 2013, 12, 3521-3531. (19) Yu, K.; Phu, L.; Varfolomeev, E.; Bustos, D.; Vucic, D.; Kirkpatrick, D. S. J. Mol. Biol. 2015, 427, 2121-2134. (20) Udeshi, N. D.; Mani, D. R.; Eisenhaure, T.; Mertins, P.; Jaffe, J. D.; Clauser, K. R.; Hacohen, N.; Carr, S. A. Mol. Cell. Proteomics 2012, 11, 148-159. (21) Kim, W.; Bennett, E. J.; Huttlin, E. L.; Guo, A.; Li, J.; Possemato, A.; Sowa, M. E.; Rad, R.; Rush, J.; Comb, M. J.; Harper, J. W.; Gygi, S. P. Mol. Cell 2011, 44, 325-340. (22) Rose, C. M.; Isasa, M.; Ordureau, A.; Prado, M. A.; Beausoleil, S. A.; Jedrychowski, M. P.; Finley, D. J.; Harper, J. W.; Gygi, S. P. Cell Syst. 2016, 3, 395-403. (23) Thomas, S. N.; Zhang, H.; Cotter, R. J. Clin. Proteom. 2015, 12, 14-28. (24) Koehler, C. J.; Strozynski, M.; Kozielski, F.; Treumann, A.;

8


Page 9 of 16

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


For TOC only

9



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

190x269mm (300 x 300 DPI)


Page 10 of 16

Page 11 of 16

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


420x267mm (300 x 300 DPI)



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

212x181mm (300 x 300 DPI)


Page 12 of 16

Page 13 of 16

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


360x150mm (300 x 300 DPI)



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

129x167mm (300 x 300 DPI)


Page 14 of 16

Page 15 of 16

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


236x175mm (300 x 300 DPI)



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

360x140mm (300 x 300 DPI)


Page 16 of 16

Site-Specific Quantification of Protein Ubiquitination on MS2 Fragment

Recommend Documents