Comparative Analysis of Human Src-Family Kinase ... - ACS Publications

Sep 23, 2010 - Reverse Proteomics Research Institute, 1-9-11 Kaji, Chiyoda-ku, Tokyo ... Keywords: Src family kinases • protein tyrosine phosphoryla...
4 downloads 0 Views 4MB Size
Download PDF
Comparative Analysis of Human Src-Family Kinase Substrate Specificity in Vitro Hiroyuki Takeda,†,‡ Yoshifumi Kawamura,† Aya Miura,† Masatoshi Mori,† Ai Wakamatsu,| Jun-ichi Yamamoto,# Takao Isogai,|,# Masaki Matsumoto,⊥ Keiichi I. Nakayama,⊥ Tohru Natsume,§ Nobuo Nomura,*,§ and Naoki Goshima*,§ Japan Biological Informatics Consortium, TIME24 Building 10F 2-4-32 Aomi, Koto-ku, Tokyo 135-8073, Japan, Graduate School of Pharmaceutical Sciences, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan, Reverse Proteomics Research Institute, 1-9-11 Kaji, Chiyoda-ku, Tokyo 101-0044, Japan, Medical Institute of Bioregulation, Kyushu University, Maidashi, Higashi-ku, Fukuoka 812-8582, Japan, and National Institute of Advanced Industrial Science and Technology, 2-4-7 Aomi, Koto-ku, Tokyo 135-0064, Japan Received July 26, 2010

Src family kinases (SFKs) are the earliest known family of tyrosine kinases and are widely thought to play essential roles in cellular signal transduction. Although numerous functional analyses have been performed, no study has analyzed the specificity of all SFKs on an equal platform. To gain a better understanding of SFK phosphorylation, we designed a high-throughput in vitro kinase assay on the subproteome scale using surface plasmon resonance. We reacted each of the 11 human SFKs with 519 substrate proteins, and significant phosphorylation was detected in 33.6% (1921) of the total 5709 kinase-substrate combinations. A large number of novel phosphorylations were included among them. Many substrates were shown to be phosphorylated by multiple SFKs, which might reflect functional complementarity of SFKs. Clustering analysis of phosphorylation results grouped substrates into 10 categories, while the similarity of SFK catalytic specificity exhibited no significant correlation with that of amino acid sequences. In silico predictions of SRC-specific phosphorylation sites were not consistent with experimental results, implying some unknown SRC recognition modes. In an attempt to find biologically meaningful novel substrates, phosphorylation data were integrated with annotation data. The extensive in vitro data obtained in this study would provide valuable clues for further understanding SFK-mediated signal transduction. Keywords: Src family kinases • protein tyrosine phosphorylation • surface plasmon resonance • wheat germ expression system • substrate specificity • clustering

Introduction Protein tyrosine phosphorylation plays a key role in cell signaling, and substantial efforts to obtain phosphorylation data have been conducted to explore kinase substrates and understand phosphorylation-mediated signal transductions. Various assay approaches have been developed, including in vivo/in vitro methods, on chip/in solution, and detection using radioisotope/antibody/mass spectrum analysis.1-3 However, some approaches are not suitable for high throughput (HTP) assays because of low speed, protein denaturation or the narrow dynamic range of detection.4,5 * To whom correspondence should be addressed. Address: National Institute of Advanced Industrial Science and Technology, 2-4-7 Aomi, Kotoku, Tokyo 135-0064, Japan. Tel: +81-3-3599-8137; fax: +81-3-3599-8141; e-mail: [email protected] (N.G.); [email protected] (N.N.). † Japan Biological Informatics Consortium. | Graduate School of Pharmaceutical Sciences, The University of Tokyo. # Reverse Proteomics Research Institute. ⊥ Medical Institute of Bioregulation, Kyushu University. § National Institute of Advanced Industrial Science and Technology. ‡ Current address: Graduate School of Agriculture, Kyoto University, Gokasho, Uji, Kyoto 611-0011, Japan.

5982 Journal of Proteome Research 2010, 9, 5982–5993 Published on Web 09/23/2010

In vitro assays are advantageous in terms of ease of control of assay conditions, including kinase-substrate combinations and the applicability for multisample formats. Recent improvements in the construction of proteome-scale cDNA libraries6,7 and progression in protein expression8 have made it possible to supply numerous kinases and substrates for large-scale in vitro kinase assays.1,4,5,9-11 In addition, the in-solution kinase assay method could theoretically maintain protein tertiary structures and the authentic enzymatic actions of kinases to substrates, although the throughput is relatively lower than onchip methods. In order to balance extensiveness and accuracy of data in a large-scale assay, we have designed an in-solutionbased in vitro kinase assay method using HTP-surface plasmon resonance (SPR) as a phosphorylation detector, combining protection against protein denaturation and highly quantitative and wide-dynamic-range detection.12,13 Its real-time monitoring of quantifiable “on” and “off” rates during substrate capturing and antiphosphotyrosine (pTyr) antibody binding is of significant advantage over most competing high-throughput methodologies. 10.1021/pr100773t

 2010 American Chemical Society

Comparative Analysis of Human SFK Specificity in Vitro Src family kinases (SFKs) are known to regulate diverse cellular processes including proliferation, survival, differentiation, adhesion, and motility.14-17 The orthodox Src family consists of eight highly conserved kinases (FGR, FYN, SRC, YES, BLK, HCK, LCK, and LYN) that contain catalytic, SH2 and SH3 domains and an N-terminus consensus sequence for myristoylation and membrane association.16-18 Three other Src family like kinases (FRK, BRK, and SRM), which have a unique exon structure and lack a myristoylation site, are separated from orthodox SFKs in the phylogenic tree19,20 (Supplementary Figure 3, Supporting Information). We categorize these 11 kinases into one family according to Manning et al.,18 although some reports regard these kinases as two distinct families.16 Numerous studies have demonstrated that SFKs are highly conserved kinases but with broad substrate specificity.16,17,21,22 However, no study until now has examined the specificity of all SFKs on an equal platform, or explained their catalytic differences. In this study, we examined SFK phosphorylation extensively using an HTP kinase assay and elucidated the differences in catalytic specificities.

Materials and Methods Reagents. All chemicals were obtained from Wako Pure Chemical Industries (Osaka, Japan) except when noted. BLK kinase was purchased from Millipore (Billerica, MA, USA). FRK, LCK, SRC, SRM, and YES kinases were obtained from Carna Biosciences (Kobe, Japan). Anti-FLAG M2 antibody, Tween 20 and BSA were obtained from Sigma-Aldrich (St. Louis, MO, USA). Anti-pTyr antibody (P-Tyr-100) was obtained from Cell Signaling Technology (Danvers, MA, USA). Enzymes for the Gateway system, LDS sample buffer for SDS-PAGE, SDS-PAGE running buffer, NuPAGE Bis-Tris gels (4-12%), and Probond Ni-agarose resin were purchased from Invitrogen (Carlsbad, CA, USA). KOD Dash DNA polymerase was purchased from Toyobo (Osaka, Japan). RNase inhibitor and SP6 RNA polymerase were purchased from Promega (Madison, WI, USA). The wheat germ expression kit was obtained from CellFree Sciences (Yokohama, Japan). Glutathione Sepharose 4B, HBS-EP+ buffer, an aminecoupling kit, HRP-conjugated antimouse IgG and ECL Plus were purchased from GE Healthcare (Buckinghamshire, UK). Lysyl endopeptidase from Acromobactor was purchased from Wako. Trypsin was obtained from Thermo Fisher Scientific (Waltham, MA, USA). C18 Empore disk was procured from 3M (St. Paul, MN, USA). Preparation of Protein Samples. Figure 1A shows the outline for sample preparation. A total of 519 Gateway entry clones were selected from the human Gateway entry clone set harboring the ORF of cDNAs9 to express substrate proteins. Accession numbers of the clones are shown in Supplementary Table 1, Supporting Information (for details of each Gateway entry clone, see the Human Gene and Protein Database at http:// www.hgpd.jp).23 The destination vector for substrate protein expression contained a FLAG-GST tandem tag at the 5′-end of the attR1 site. After the LR reaction, DNA fragments harboring the SP6 promoter, tags, ORF, and 3′-untranslated region were amplified by PCR using KOD Dash DNA polymerase, and mRNA was generated from the PCR product using SP6 RNA polymerase. In vitro translation was performed by mixing the mRNA with wheat germ extract.8,9 Expressed substrates were purified by GST-affinity chromatography. Twenty microliters of 50% (v/v) glutathione Sepharose 4B slurry was added to 150 µL of crude expressed protein solution in a 96-well plate and shaken for 1 h by a temperature-controlled incubator shaker

research articles (MBR-024, TAITEC, Saitama, Japan) at 900 rpm and 4 °C. The mixture was then transferred to a 96-well filter plate and centrifuged at 50g for 1 min. The resin was washed three times with 100 µL of wash buffer (50 mM Tris-HCl (pH 7.4) and 1 mM DTT). Purified protein was then extracted with 30 µL of elution buffer (50 mM Tris-HCl (pH 8.0), 2 mM reduced glutathione, and 1 mM DTT). The purified protein was mixed with the same volume of storage solution (50 mM Tris-HCl (pH 8.0), 1 mM DTT, 300 mM NaCl, 0.1% (w/v) Brij35, 20% (v/v) glycerol, and 2 mg/mL BSA), separated into tubes according to the amount needed in each experiment, frozen in liquid nitrogen, and then stored at -80 °C. Five SFKs (BRK, FGR, FYN, HCK, and LYN) among 11 kinases used in this study were expressed using a destination vector with a GST tag at the 5′end (Figure 1A) and purified as described above. Expressed enzymes were mixed with the same amount of storage solution and stored as described above. Six commercial His-tagged recombinant SFKs were respectively subdivided, frozen in liquid nitrogen, and stored at -80 °C. Protein quantification of the enzymes was performed by SDS-PAGE and CBB staining using BSA as a standard. In Vitro Tyrosine Phosphorylation and pTyr Detection by HTP-SPR. In vitro kinase assays with HTP-SPR were carried out as described previously.12,13 A kinase reaction mixture (30 µL in total) contained 6 µL of each substrate, 4 ng of tyrosine kinase, 10 µM ATP, 20 mM Hepes-NaOH (pH 7.2), 10 mM MgCl2, and 3 mM MnCl2. Reaction mixtures without tyrosine kinase were also prepared (mock-treated sample). The reaction was performed at 30 °C for 5 h in 384-well plates (Figure 1B), then terminated by adding 80 µL of EDTA solution (50 mM EDTA, 10 mM Hepes-NaOH (pH 7.4), 150 mM NaCl, and 0.05% (v/v) Tween 20). pTyr detection was performed using the Biacore A100 optical biosensor (GE Healthcare). The anti-FLAG M2 antibody for capturing FLAG-tagged proteins was diluted in 10 mM sodium acetate (pH 4.5) to 20 µg/mL and covalently immobilized on a Series S Sensor Chip CM5 (certified) (GE Healthcare) using standard amine coupling chemistry with 10 min of contact. The amount of immobilized anti-FLAG antibody was around 10 000 resonance units (RU). The 384-well plates containing the reactions were placed in the plate stacker of the Biacore A100 and kept at 4 °C while performing the detection. The running buffer (HBS-EP+ buffer) contained 10 mM Hepes-NaOH (pH 7.4), 150 mM NaCl, 0.05% (v/v) Tween 20, and 3 mM EDTA. The temperature in the flow cells was maintained at 25 °C. Reaction samples were injected separately into spot 1 or 5 of each flow cell at 30 µL/min for 150 s, and the FLAG-tagged substrate was selectively captured with immobilized anti-FLAG antibody (Figure 1C). Spots 2 and 4 were used as references for spots 1 and 5, respectively. Then, antipTyr antibody (1.5 µg/mL in HBS-EP+ buffer) was injected at 30 µL/min for 120 s. Finally, 10 mM NaOH was injected for 30 s to regenerate the capture antibody on the sensor chip. Data collection and analysis were performed using Biacore A100 evaluation software. Data Processing. A typical SPR sensorgram is shown in Figure 1D. Responses at the report points of “Capture level” and “Anti-pTyr ab binding” were monitored. “Anti-pTyr ab binding” was normalized by “Capture level” of the substrate as follows: “Normalized anti-pTyr ab binding” ) “Anti-pTyr ab binding” × “Capture level”-1 × N Journal of Proteome Research • Vol. 9, No. 11, 2010 5983

research articles

Takeda et al.

Figure 1. Experimental procedure. (A) Protein sample preparation. A total of 519 substrates and five SFKs were expressed from a Gateway entry clone set harboring cDNA ORFs using a wheat-germ cell-free expression system, and purified by GST-affinity beads. Six insect cell-expressed and his-tagged SFK enzymes were purchased. (B) In vitro kinase reaction. Tyrosine kinase (4 ng/reaction in all cases except LCK second experiment at 2 ng/reaction), substrate (6 µL/reaction) and ATP (100 µM) were mixed in 384-well plates and incubated for 5 h at 30 °C. (C) Phosphotyrosine detection by SPR. Biacore A100 has four flow cells in total with 20 detection spots on a sensor chip. Anti-FLAG antibody was preimmobilized on each detection spot except spot 3 by amine coupling. Spots 1 and 5 were used for capturing a sample and detection, and spots 2-4 were used as references. Detection step 1: reaction mix solutions in four wells were respectively injected into four flow cells and guided to spot 1. Substrate proteins with a FLAG tag were selectively captured with anti-FLAG antibody. Amount of captured substrates was monitored by SPR. Step 2: other four reaction solutions were injected and guided to spot 5 of each flow cell, and substrate proteins were also captured as in Step 1. Step 3: pTyr residue was detected by monitoring anti-pTyr antibody binding. Step 4: sensor chip surface was rinsed with alkaline to remove the substrate and anti-pTyr antibody. A regenerated sensor chip was reused for the next cycle. (D) Sensorgrams in one run. A total of 768 samples were processed in one run, and the SPR signal on each spot was monitored in real time. All sensorgrams are zeroed on the Y-axis to the mean baseline of anti-pTyr antibody injection. Steps described in panel C are indicated. Responses in “Capture level” and “Anti-pTyr ab binding” recorded in each sensorgram were used for subsequent data processing. (E) Criterion of reproducibility in duplicate experiments. “Normalized tyrosine phosphorylation” values in the duplicate experiments were compared considering their noise level. Error bar indicates the error range of the signal. (F) Criterion of signal significance. When [“Normalized anti-pTyr ab binding (kinase treated)” Error(kinase treated)] was greater than [“Normalized anti-pTyr ab binding (mock)” + Error (mock)], the phosphorylation signal was recognized as significant. Reference was a mock-treated sample assayed in the same run (Mock 1 for BRK, FGR, FRK, FYN, HCK, and LYN; Mock 3 for BLK, LCK, SRM, SRC, and YES. See Supplementary Table 1, Supporting Information). Error bar indicates error range of the signal.

where N (500) is the coefficient for normalization of the “Capture level”. Then, the baseline was subtracted as follows: 5984

Journal of Proteome Research • Vol. 9, No. 11, 2010

“Tyrosine phosphorylation” ) “Normalized anti-pTyr ab binding (kinase treated)” “Normalized anti-pTyr ab binding(mock)”

Comparative Analysis of Human SFK Specificity in Vitro In addition, standard samples (TOM1L1 (FLJ54992); treated with FYN and mock) were prepared and subjected to SPR at each run together with other samples. The “Tyrosine phosphorylation” value of the standard FLJ54992 protein was used to correct for variations of each experiment as follows: “Normalized tyrosine phosphorylation” ) “Tyrosine phosphorylation (sample)” “Tyrosine phosphorylation(standard TOM1L1)”-1 × M where M (200) is the coefficient. We used “Normalized tyrosine phosphorylation” value (Supplementary Table 1, e.g., column AA) to indicate the phosphorylation level of a substrate by a certain kinase. Several calculation values, such as “Normalized anti-pTyr ab binding” and “Tyrosine phosphorylation” are not shown, but are available upon request. The error range of SPR detection was determined to evaluate the reproducibility and significance of data. The definition of error is described in the “Error range” section of the Supplementary Results and Discussion, Supporting Information. The error of “Normalized anti-pTyr ab binding” was calculated as follows: error ) (1 + 0.1 × “Anti-pTyr ab binding”) × “Capture level”-1 × N where N (500) is the coefficient. Reproducibility of data was evaluated considering the error range. When the difference between two “Normalized anti-pTyr ab binding” values of a certain sample in duplicate experiments fell within the error range, the sample result was recognized as reproducible (Figure 1E). The significance of the phosphorylation signal mentioned below was not considered for evaluation of reproducibility. The significance of each phosphorylation was evaluated as described below. The error of “Normalized anti-pTyr ab binding” was calculated as described above. When the value obtained from “Normalized anti-pTyr ab binding (kinase treated)” minus Error (kinase treated) was greater than that obtained from “Normalized anti-pTyr ab binding (mock)” plus Error (mock), the phosphorylation signal was recognized as significant (Figure 1F). Relative activity was calculated and subjected to clustering analysis. The phosphorylation signals certified as significant phosphorylation events were used for calculation. Relative activity of a certain kinase was calculated as follows: Relative activity ) “Normalized tyrosine phosphorylation (Substrate)” × “Normalized tyrosine phosphorylation(CRK)”-1. Clustering was performed using the gCluto program (http:// www-users.cs.umn.edu/∼mrasmus/gcluto/index.shtml).24 Kmeans clustering was conducted for clustering of substrates using the repeated bisection method. The number of clusters was 10. Cosine correlation was used for similarity calculation for both substrates and kinases. Phosphorylation-site predictions were performed using NetPhos 2.0 (http://www.cbs.dtu.dk/services/NetPhos/).25 Predictions of kinase-specific phosphorylation sites were conducted with NetPhosK (http://www.cbs.dtu.dk/services/NetPhosK/).26

research articles Annotation data of substrate proteins such as molecular class, phosphorylation, protein-protein interaction, and cellular localization were obtained from HPRD (http://www.hprd. org/).27 Kinase information was derived from Kinbase (http:// kinase.com/kinbase/).18 References of phosphorylation were obtained from HPRD, Phospho.ELM (http://phospho.elm.eu. org/),28 and PhosphoSitePlus (http://www.phosphosite.org/). The BodyMap-Xs database (http://bodymap.jp/) was used to collect expression information.29 Additionally, information on cellular localization was derived from databases such as Gene Ontology (http://www.geneontology.org/), SwissProt (http://au. expasy.org/sprot/), and LIFEdb (http://www.dkfz.de/LIFEdb/).30 Western Blotting. Kinase reaction was performed as described above. The reaction was terminated by adding NuPAGE LDS sample buffer with 2-mercaptoethanol following denaturation at 70 °C for 10 min. A portion of the mixture was subjected to electrophoresis using NuPAGE 4-12% Bis-Tris gels, then the gel was blotted using a PVDF membrane (BioRad, Hercules, CA, USA). The membrane was blocked with 5% (w/v) BSA in TBST (20 mM Tris-HCl (pH8.0), 134 mM NaCl and 0.1% (v/v) Tween20). The primary antibody (pTyr-100) and secondary antibody (HRP-conjugated antimouse IgG) were diluted at 1:15000 and 1:2000, respectively. Detection was performed using ECL-Plus and Fluor-S Multi Imager Max (BioRad). Protein quantification of a substrate was performed by SDS-PAGE and CBB staining using BSA as a standard. Mass Spectrum Analysis. Proteins phosphorylated by tyrosine kinases were precipitated with 10% (v/v) TCA, washed twice with ice cold acetone, and dissolved in 20 µL of 7 M guanidine hydroxide in 150 mM Tris-HCl (pH 8.8). The solution was boiled for 15 min, diluted with 3 vol of 150 mM Tris-HCl (pH 8.8), and digested with 100 ng of Lys-C for 16 h at 37 °C. After digestion, proteins were reduced with 2.5 mM Tris (2carboxyethyl) phosphine for 30 min at 85 °C, alkylated with 12.5 mM iodoacetoamide for 30 min at room temperature, diluted with equal volumes, and further digested with 100 ng of trypsin for 6 h. Resulting peptides were desalted by SepPAK C18 (50 mg) and subjected to phosphopeptide enrichment by Fe-charged Probond resin (Invitrogen) in a pipet tip fritted by a C18 membrane, as described previously.31 Loading/washing buffers were 0.1% (v/v) TFA and 60% (v/v) ACN. After washing, bound peptides were eluted with 1% (v/v) phosphate into C18, washed with 0.1% (v/v) TFA/2% (v/v) ACN, and eluted with 0.1% (v/v) TFA/60% (v/v) ACN. Eluted peptides were analyzed by a nanoLC-MS/MS system composed of a quadrupole-TOF hybrid mass spectrometer [QSTAR elite (Applied Biosystems, Foster City, CA, USA) and MDS-Sciex (Concord, ON, Canada)] and a nanoLC (Paradigm MS2, Michrom BioResources, Auburn, CA, USA). Peptide separation was carried out by in-housepulled fused silica capillary (0.1 mm inner diameter, 20 cm length, 0.05 mm tip inner diameter), packed with 3 µm C18 L-column (CERI Japan, Tokyo, Japan). Data acquisition was performed on IDA mode. All samples were dissolved in 0.1% (v/v) TFA and 2% (v/v) ACN, and injected into precolumns (Lcolumn micro: 0.3 mm inner diameter, 5 mm length, CERI Japan), washed with the same buffer, and eluted with a linear gradient 5-95%B for 30 min (A: 0.1% (v/v) formic acid and 2% (v/v) methanol, B: 0.1% (v/v) formic acid, 98% (v/v) methanol) at a flow rate of 200 nL/min. Survey MS spectra were acquired for 0.5 s, and the three most intense ions were isolated and fragmented with automatically optimized collision energy for variable MS/MS acquisition times. The peak list was obtained by the script in Analyst QS 2.0. MASCOT search (ver. 2.2.04) Journal of Proteome Research • Vol. 9, No. 11, 2010 5985

research articles

Takeda et al.

was performed against IPI human ver.3.1.6 (62322 entries) with the following parameter settings: trypsin was selected as the enzyme used, the allowed number of missed cleavages was set at 2, and carbamidomethylation on Cys was selected as the fixed modification. Oxidized methionine and phosphorylation on serine, threonine, and tyrosine were searched for as variable modifications. Precursor mass tolerances were 150 ppm and the tolerance of MS/MS ions was 0.2 Da. Peptide sequences, which had MASCOT score > 20 and were assigned to rank 1, were extracted. Using delta mass of highscored hits, recalibration was performed at post-MASCOT search using in-house script, and hits with more than 50 ppm error were removed. To remove false positive hits, we tested hits with low MASCOT scores ( HCK > LCK > others. Substrates of both clusters 6 and 7 were mainly phosphorylated by LCK and SRC, and the contrasts between LCK/SRC and other kinases were more remarkable in cluster 6 than in cluster 7. Proteins in cluster 8 were phosphorylated in a distinctive manner compared with others. Further cluster-

Comparative Analysis of Human SFK Specificity in Vitro

research articles them, the specificities of FGR and FRK differed from that of SRM. FGR and FRK phosphorylated LYN, p85, RasGAP, and Grb10, whereas SRM phosphorylated ARG, TOM1like1 and CRK associated substrate. FGR and FRK, similar to group 1, most strongly phosphorylated Docking protein 1 (Figure 4). However, SRM, which differed from all other SFKs, most intensively phosphorylated TOM1L1 (FLJ95985), implying its unique substrate specificity.

Figure 3. Overview of human SFK phosphorylation by clustering analysis. “Normalized tyrosine phosphorylation” value was calculated as described in Materials and Methods, and relative activity to CRK was subjected to clustering analysis using the gCluto program. The longitudinal axis corresponds to 519 substrates, and the horizontal axis to 11 tyrosine kinases. Each bar represents one phosphorylation event, and its color indicates the relative activity of each kinase to the substrate (CRK ) 1). White bar shows nonsignificant signal. The 519 substrates were grouped into 10 clusters. The tree represents similarity of substrate specificity of SFKs.

ing analysis divided cluster 8 into several subgroups, for example, (FYN, LCK) . others; (FYN, HCK, LCK) . others; (BRK, LCK) > (FYN, HCK, LYN, SRC) > (BLK, YES) (data not shown). Cluster 9 consisted of proteins hardly phosphorylated by SFKs. Clustering analysis categorized SFKs into three groups: group 1 (BRK, FYN, HCK, and LYN), group 2 (LCK, SRC, YES, and BLK), and group 3 (FGR, FRK, and SRM) according to the similarity of substrate specificities (Figure 3). Group 1 phosphorylated Docking protein 1 (FLJ31382) most intensively among the substrates, whereas group 2 phosphorylated CRK associated substrate (FLJ95582) 1.5 to 3 times more than Docking protein 1 (Figure 4). Kinases in group 3 significantly phosphorylated less than 5% of the 519 substrates. Among

Although the primary structure of a kinase is widely regarded as an important factor in determining its substrate specificity, there was discordance between amino acid sequence similarity and substrate specificity (Figure 3 and Supplementary Figure 3, Supporting Information). For example, FYN and HCK showed the closest substrate specificity among the 11 SFKs in the clustering result (Figure 3), but nonetheless belong to different subfamilies, SrcA and SrcB, respectively (Supplementary Figure 3, Supporting Information).18 Phosphorylation sites of four common substrates of FYN and HCK were investigated by mass spectrometry to compare their specificity in detail. Seven phosphopeptides were identified from FYN-treated substrates, whereas four of them were also detected from the ones treated by HCK (Supplementary Table 3, Supporting Information). This suggests that there is not much difference in phosphorylation sites between these kinases, supporting the clustering results. Besides FYN (SrcA) and HCK (SrcB), a similar discrepancy between sequence similarity and clustering results was observed between SRC (SrcA) and LCK (SrcB). Moreover, although BRK is an “SFK-like” kinase distinct from typical SFKs in the phylogenic tree (Supplementary Figure 3, Supporting Information), clustering analysis classified BRK into group 1 with orthodox SFKs. SFK-Favorite Substrates. Several proteins were strongly phosphorylated by multiple kinases, and we named them “SFKfavorite substrates”. Figure 4 shows the representative SFKfavorite substrates, which were phosphorylated by at least nine SFKs in this study. For example, both CRK (FLJ81679) and Docking protein 1 (FLJ31382) are known to be phosphorylated by various tyrosine kinases besides SFKs and play important roles in cellular signal transduction.33,34 Several substrates in Figure 4 were previously known to be phosphorylated by a single SFK, although they were phosphorylated by multiple SFKs in this study. To the best of our knowledge, phosphorylation of four SFK-favorite substrates, namely, Cyclin G associated kinase (FLJ45425), ERK3 (FLJ80141), CDK6 (FLJ94044), and HEF-like protein (FLJ53493), has not been reported to date. Cyclin G associated kinase, ERK3 and CDK6 are S/T kinases involved in cell cycle progression and proliferation,35-37 while HEF-like protein is an adapter protein with an SH3 domain and histidine kinase domain. The substrates prepared in the present study include 125 SFK substrates reported previously (references are linked by hyperlink in Supplementary Table 1, Supporting Information). The percentage of SFK-favorite substrates differed between these previously reported substrates and others. Around half (48.8%) of the reported SFK substrates were phosphorylated by more than half of the SFKs, and 12.0% were phosphorylated by 9-11 SFKs (Figure 2B, middle). On the other hand, among 394 unreported substrates, the percentage of substrates phosphorylated by more than half of the SFKs decreased to 25.4% and the substrates phosphorylated by 9-11 SFKs accounted for only 1.3% (Figure 2B, bottom). In addition, more than half Journal of Proteome Research • Vol. 9, No. 11, 2010 5987

5988

Journal of Proteome Research • Vol. 9, No. 11, 2010 SRC, LCK, FYN

FYN

HCK, LYN, FYN YES

tyrosine kinase (TK/Tec) tyrosine kinase (TK/Src) tyrosine kinase (TK/Src)

adapter molecule

tyrosine kinase (TK/PDGFR) S/T kinase (AGC/PKA) S/T kinase (AGC/PKC/Alpha) S/T kinase (AGC/PKC/Eta) tyrosine kinase (TK/Fak)

tyrosine kinase (TK/Ret) RTK (TK/Ret) cytoplasmic domain tyrosine phosphatase

S/T kinase (CAMK/MLCK) transcription factor

tyrosine kinase (TK/Syk)

cytoskeletal associated protein

tyrosine kinase (TK/Src)

interferon, alpha receptor interleukin 2 receptor, beta

ITK LCK LYN

p85, phosphoinositide-3-kinase, regulatory subunit PDGFRb PKACa PKCa PKCe PYK2

RET RET-CDf SHP2

smMLCK STAT1

SYK

Wiskott-aldrich syndrome protein YES

FLJ80182

FLJ81091

FLJ53659 FLJ39367 FLJ95929 FLJ90539

FLJ80417 FLJ80417 FLJ75588

FLJ80123 FLJ80166 FLJ08071 FLJ94469 FLJ46514 FLJ80173

FLJ20114

FLJ93288 FLJ80078 FLJ75641

FLJ93015 FLJ93303

BC048960

BC012738

AK300610 AK096686 AK314722 AK075020

BC004257 BC004257 AK289854

BC032224 BC039846 NM_002737 AK313818 AK128371 BC042599

AK000121

AK312808 BC013200 AK290494

AK312597 AK312846

AK294750 AK094999 AK125143 AK091010

AK314619 AK055395 AK055944

FLJ95582 FLJ30833 FLJ31382

FLJ55514 FLJ37680 FLJ43153 FLJ33691

BC001718 AK124815

AK057105 AK314376 BC030594 BC053585

BC032229 BC032229

accession

FLJ81679 FLJ42825

FLJ32543 FLJ95155 FLJ80518 FLJ82825

FLJ80124 FLJ80124

FLJ ID

FYN SRC

SRC LCK LCK FYN LCK LCK SRC SRC SRC FYN LCK SRC SRC FYN FYN FGR FYN SRC

LCK FYN LYN FYN FYN FYN LCK YES HCK

FYN LCK LYN LCK SRC FGR HCK SRC YES BRK LCK SRC SRC

LCK LCK SRC HCK HCK SRC LCK

name

SFK

82.4 42.3

11.9 5.6 20.2 8.5 96.7 58.8 32.9 6.5 15.2 5.7 30.4 8.1 31.6 6.7 6.5 3.9 89.2 78.7

30.6 8.2 6.6 34.2 123.1 65.9 214.5 27.0 40.7

57.5 854.2 140.8 1008.5 19.5 35.1 329.0 250.8 170.2 18.0 324.7 7.3 40.4

18.4 92.7 34.3 24.9 29.3 16.8 14.9

“normalized tyrosine phosphorylation”d

1.43 0.43

0.12 0.03 0.11 0.15 0.52 0.32 0.34 0.07 0.16 0.10 0.16 0.08 0.32 0.12 0.11 0.34 1.55 0.81

0.17 0.14 0.11 0.59 2.14 1.15 1.16 0.72 0.66

1.00 4.61 2.34 5.44 0.20 3.07 5.37 2.58 4.53 0.41 1.75 0.08 0.42

0.10 0.50 0.35 0.41 0.48 0.17 0.08

relative activity (CRK ) 1)

phosphorylation (present study)

vitro vivo, vivo, vitro vivo, vivo, vivo, vivo vivo vitro vitro vivo vivo vivo vivo vivo, vivo, vitro

vivo, vivo vivo, vivo, vivo vivo, vivo vivo vitro

in vitro in vitro

in vitro in vitro in vitro

in vitro in vitro

in vitro

in vitro in vitro

in vitro

vivo vitro vivo, in vitro vitro vitro vitro vitro vitro vitro vivo vivo, in vitro vitro vivo, in vitro

vivo vivo vivo vitro vitro vitro vitro

in vivo in vivo

in in in in in in in in in in in in in in in in in in

in in in in in in in in in

in in in in in in in in in in in in in

in in in in in in in

experiment type

type

complex complex

direct direct direct direct direct direct direct direct direct direct direct direct direct direct direct direct direct direct

direct direct direct direct complex complex complex complex direct

direct direct direct direct direct direct direct direct direct direct direct direct direct

direct direct direct direct direct direct direct

interactione

a Detailed information on 49 kinase-substrate pairs (Figure 2A, shaded area) is shown. b Annotations are from HPRD and Kinbase database. c Reported phosphorylation information was collected using HPRD, Phospho.ELM, PhosphoSitePlus, and other literature search services. d Median value of duplicate experiments. e Information on protein-protein interaction was obtained from HPRD. f Substrate proteins that consist of the cytoplasmic domain of the receptor tyrosine kinase are shown as (kinase name)-CD.

LYN, LCK, SRC

SRC

LCK LCK LYN

LCK

SRC SRC, FYN FGR

RTK (TK/EGFR) cytoplasmic domain tyrosine kinase (TK/Fak) tyrosine kinase (TK/Src) adapter molecule, receptor signaling complex scaffold activity cytokine receptor cytokine receptor

EGFR-CDf FAK FGR Grb2

LCK, LYN, FYN

adapter molecule adapter molecule

tyrosine kinase (TK/Csk) adapter molecule, receptor signaling complex scaffold activity

SRC, FYN, HCK

cell surface receptor cytokine receptor

CD46 colony-stimulating factor 3 receptor CRK CRK associated substrate

CTK docking protein 1

LCK, YES HCK

tyrosine kinase (TK/Tec)

BTK

LYN

tyrosine kinase (TK/Axl) RTK (TK/Axl) cytoplasmic domain

categoryb

known upstream SFK(s)c

AXL AXL-CDf

protein name

substrate

Table 1. Novel Kinase-Substrate Pairs Whose Protein-Protein Interactions Were Previously Reporteda

research articles Takeda et al.

Comparative Analysis of Human SFK Specificity in Vitro

research articles

Figure 4. Representative SFK-favorite substrates. Substrates significantly phosphorylated by at least nine kinases are shown. Numbers indicate the “Normalized tyrosine phosphorylation” values. Colors indicate the relative activity of each kinase to the substrate (CRK ) 1). NS indicates a nonsignificant signal.

(54.6%) of unreported substrates were phosphorylated by less than half of the SFKs. Comparison of Experimental Results with in Silico Predictions. Another interesting question concerns the extent to which phosphorylation-site prediction is consistent with present experimental data. Tyrosine phosphorylation by SRC in this study showed no correlation with the total number of tyrosine residues of each substrate (r2 ) 0.07, p < 0.05) (Figure 5A), implying that SRC selectively phosphorylated certain tyrosine residues in the substrates. We conducted phosphoprediction using NetPhos 2.025 for general tyrosine phosphorylation sites, and NetPhosK26 for SRC-specific phosphorylation sites. The number of predicted pTyr residues was loosely correlated with our experimental result. NetPhos 2.0 prediction correlated most with experimental results when the threshold was 0.9 (r2 ) 0.31, p < 0.05) (Figure 5B). Prediction with a high threshold showed a higher correlation coefficient than that with a low threshold, suggesting that the incidence of false positive prediction increases with lower thresholds. The correlation between experimental results and Src-specific phosphorylation predicted by NetPhosK (threshold ) 0.5) (r2 ) 0.37, p < 0.05) was slightly higher than that predicted by NetPhos 2.0 (Figure 5C). It is noteworthy that some proteins were intensively phosphorylated by SRC, whereas NetPhosK failed to predict their SRC phosphorylation sites. We selected five proteins from such nonpredicted SRC substrates and confirmed tyrosine phosphorylation by Western blotting (Figure 5D) and MS analysis (Figure 5E). Our data demonstrated the usefulness of in vitro screening assays for the identification of novel kinaserecognition sites. Selection of Biologically Meaningful Novel Substrates Using Annotation Data. Among 1921 kinase-substrate pairs on which significant phosphorylation was observed, there were only 165 pairs whose phosphorylation was previously reported (Figure 2A). It is possible that the remaining significantly

phosphorylated proteins (1756) contained several bona fide substrates. However, the possibility cannot be ruled out that they contain biologically meaningless false positives. Other information, such as expression, localization, and proteinprotein interaction (PPI), may increase the certainty of the phosphorylation event. As an example of how to select promising substrate candidates by integrating phosphorylation and annotation data, we searched several candidates from kinasesubstrate pairs whose information was available in the public database. First, we attempted to use PPI information. Among the 5709 kinase-substrate combinations in the present study, 312 interactions were previously reported (Figure 2A, yellow circle). We detected 198 significant phosphorylation reactions among them, and 149 of these phosphorylated events had been reported. We thought that the remaining 49 reactions would occur in vivo (for details, see Table 1). As an example, we chose LCK-FAK and LCK-SHP2 combinations and analyzed their phosphorylation sites by mass spectrometry. FAK was phosphorylated on Tyr598, Tyr883, and Tyr947, which were reported as SRC recognition sites (Supplementary Figure 5, Supporting Information). The phosphorylation site of SHP2 was Tyr580, which is known as the phosphorylation site of PDGFR, Abl, and ALK.38-40 We failed to find any report of SFK phosphorylation on 12 substrates in Table 1, which implies their novelty. No substrates of BLK, FRK and SRM were found using this approach because of limited or absent PPI information in the database. Next, we examined novel substrate candidates of breast tumor kinase (BRK), which has recently attracted attention as a target for cancer therapy,41 by utilizing information about localization and expression. BRK is expressed in breast cancer cells and localized in the cytoplasm and nucleus, although typical SFKs are localized at the plasma membrane.19,42 We speculated that BRK-phosphorylated proteins that coexist with Journal of Proteome Research • Vol. 9, No. 11, 2010 5989

research articles

Takeda et al.

Figure 5. Analysis of correlation between in silico prediction and experimental data. Total numbers of tyrosine residues (A), pTyr residues predicted by NetPhos 2.0 (B), or Src phosphorylation sites predicted by NetPhosK (C) of each of the 519 substrates were normalized by the molecular weight, and plotted against SRC phosphorylation results (“Normalized tyrosine phosphorylation” by SRC). The thresholds used in NetPhos 2.0 and NetPhosK were 0.9 and 0.5, respectively. Fitted curve, correlation coefficient (r), and p-value are shown. (D) Confirmation of phosphorylation by Western blotting. Five substrate proteins, which were intensively phosphorylated by SRC but in which Src phosphorylation sites could not be predicted, were reacted with SRC and then subjected to Western blotting. pTyr residues were detected with anti-pTyr antibody. The position of each substrate (open arrowheads) was confirmed by another Western blotting with anti-GST antibody. Filled arrowheads indicate SRC protein used as a kinase. Amount of substrate protein was measured by SDS-PAGE and CBB staining using BSA as a standard. (E) Phosphopeptides identified by mass spectrum analysis. Phosphorylation sites of SRC treated substrates. The five substrates in panel D were treated by SRC, respectively, then phosphopeptides were identified by MS analysis.

BRK in vivo are possible downstream factors in the BRK signaling pathway. We examined the localization and expression pattern of the proteins newly phosphorylated by BRK in this study. Several proteins meeting all these requirements were found after intensive survey. One of them was FLJ31942 protein, which is C-terminally truncated BRD, a putative nuclear transcriptional regulator with S/T kinase activity that is implicated in epilepsy and lymphoma.43 Although tyrosine phosphorylation on BRK-treated BRD was confirmed by Western blotting, phosphorylation sites of BRK-treated BRD could not be identified by mass spectrometry (data not shown). Another protein was CK2a (FLJ80188), a catalytic subunit of casein kinase II (CKII) which is involved in cell cycle regulation and cancer development. It is known that FGR and LYN phosphorylate CK2a and activate CKII.44 It is also reported that different CKII activation patterns are observed in the cytoplasm 5990

Journal of Proteome Research • Vol. 9, No. 11, 2010

and nucleus, suggesting that CK2a might be phosphorylated by different upstream kinase(s). Therefore, it is possible that CKII activated by BRK contributes to CKII activation in the nucleus, and by typical SFK in the cytoplasm, such as FGR and LYN. Mass spectrometry analysis detected three phosphopeptides from BRK-treated CK2a (Supplementary Figure 5, Supporting Information). In addition to Tyr255, a major phosphorylation site of FGR and LYN, Tyr261, Tyr323, and Tyr325 were identified. The other novel substrate of BRK is GRB2 (FLJ33691), an adapter molecule with two SH3 domains and an SH2 domain, which is involved in intranuclear signaling. Two phosphorylation sites were identified by MS analysis. One was Tyr37, located in the N-terminus SH3 domain, and the other was Tyr134 in the SH2 domain; these phosphorylations might affect the function of GRB2 adapter protein by altering the binding mode of the domains.

research articles

Comparative Analysis of Human SFK Specificity in Vitro

Discussion In the present study, we conducted kinase reactions using 11 human SFKs and 519 substrates in a round-robin design, and found that 1921 (33.6%) of the 5709 kinase-substrate pairs exhibited significant phosphorylation signals. Among them, 1756 (91.4%) were novel phosphorylation events (Figure 2). It is thought that the substrate specificity of SFK is determined by several factors, such as catalytic specificity of the kinase domain and binding specificity of the SH2 and SH3 domains.20,45 Primary structure is widely regarded as the predominant factor that determines the higher structure and function of proteins. SFKs are classified into four subfamilies according to their amino acid similarity: SrcA (FGR, FYN, SRC, and YES), SrcB (BLK, HCK, LCK, and LYN), Frk (FRK) and Srm (BRK and SRM)18 (Supplementary Figure 3, Supporting Information). In the present study, however, FYN showed closest substrate similarity to HCK, so did SRC to LCK (Figure 3), even though they belong to different subfamilies within the phylogenic tree (Supplementary Figure 3, Supporting Information). In another instance, BRK, which is grouped in the SRM subfamily, showed a similar specificity to orthodox SFKs such as FYN, HCK, and LYN. This discrepancy implies that similarity of amino acid sequences is not the only critical factor that determines SFK substrate specificity. The wheat germ expression system, with which all substrates and several SFKs were expressed in the present study, is suitable for in vitro screening assays because of its efficiency of protein expression and absence of endogenous tyrosine kinase activity. We have shown that wheat germ expressed proteins (cytokines,9 phosphatases,9 and others (Goshima et al., unpublished work) have similar biological activities to commercial ones or published data. The activities of wheat germ expressed and commercial FYN were turned out to be comparable (Supplementary Table 4: see also Supplementary Results and Discussion, Supporting Informaton). All substrates and several SFKs were expressed by the wheat germ expression system as recombinant proteins harboring FLAG-GST and GST tags, respectively (Figure 1A). As the GST protein is reported to form homodimer, we examined homodimer formation of GST tags directly and observed no significant binding under the assay condition employed in this study (Goshima et al., unpublished work). In addition, we found that many substrate proteins that were treated by GST-tagged SFKs (e.g., FYN or HCK) were not bound with anti-pTyr antibody, even though some of them were captured on the sensor chip in large amount (over 1000 RU). This suggests that occurrence of homodimerization of GST tags can be excluded in this assay. Denaturation of substrates is another factor that should be taken into consideration46,47 (Supplementary Results and Discussion, Supporting Information). In the present study, substrate proteins were freshly expressed using wheat-germ expression, and the occurrence of denaturation was minimized by temperature control during purification and storage. In addition, the HTP-SPR method is an in-solution kinase assay that offers advantages in preventing the denaturation of protein samples during assay. In fact, (i) there was no correlation between the number of tyrosine residues on each substrate and the phosphorylation result in this study (Figure 5A); (ii) the majority of known kinase-substrate pairs were phosphorylated in the present study (Figure 2A); and (iii) high reproducibility was observed in duplicate experiments (Supplementary Table 2, Supporting Information). These observations suggest that the kinase assay results obtained by the HTP-SPR method would

reflect, at least to some extent, the bona fide substrate specificity of kinases. In this study, five wheat germ expressed kinases and six commercial available insect-cell expressed ones were used (Supplementary Table 2, Supporting Information). Since the activities of the wheat germ expressed FYN and the commercial FYN harbor comparable activities (see Supplementary Results and Discussion, Supporting Information), and since SFKs are a highly conserved kinase family, we speculate that other wheat germ expressed SFKs also exert similar activities of commercial ones. The sources of kinases, however, may be more or less influential in some cases. The possibility cannot be ruled out that false-positive signals may occur when an autophosphorylated SFK strongly interacts with a substrate protein. Protein-protein interactions between kinases and substrates can be verified by SPR. Western blotting using anti-pTyr antibody also should be effective to confirm phosphorylation of substrates. The assays in the present study were performed in vitro, and thus reaction conditions may be different from intracellular ones. One fourth of previously reported phosphorylation events were not confirmed in the present study (Figure 2A). However, some of them might be phosphorylated in vitro with some modifications in assay conditions such as pH or ionic strength, which vary significantly with respect to subcellular microenvironment and organelle types both in space and time. Although we found several promising substrate candidates by integrating phosphorylation data with annotation data from public databases such as protein-protein interaction, localization, and expression, the majority of phosphorylation events remain to be confirmed by further analysis. A number of reported SFK substrates were SFK-favorite substrates (Figure 2B, middle), suggesting that SFK-favorite substrates might have been identified preferentially in the past. In the unreported substrates, on the contrary, the percentage of substrates phosphorylated by many SFKs decreased considerably (Figure 2B, middle). This implies that many “rare” substrates, specifically phosphorylated by a small number of SFKs, remain to be identified. Large-scale in vitro screening would be a suitable tool for identifying such “rare” substrates and unknown kinase recognition sites, because every combination of kinase-substrate pairs could theoretically be prepared in vitro even if their amounts are limited in vivo. Of course, the possibility should also be considered that several kinase-substrate pairs would not meet in vivo due to temporal and/or spacial differences in expression and localization.

Conclusion This study conducted a large-scale in vitro kinase assay using the HTP-SPR method and wheat germ-expressed substrates to compare substrate specificity of all SFKs on an equal platform. In short, SFKs phosphorylated a variety of substrates, emphasizing the complexity of SFK-mediated signaling pathways. Clustering analysis visualized differences in SFK substrate specificity and revealed the discrepancy between amino acid sequence similarity and substrate specificity of SFKs. It was also shown that possible substrate candidates and novel kinase recognition sites could be discovered using large-scale in vitro screening assay data.

Acknowledgment. We acknowledge New Energy and Industrial Technology Development Organization, Japan, for support. We thank the members of our laboratory for Journal of Proteome Research • Vol. 9, No. 11, 2010 5991

research articles

Takeda et al.

providing Gateway entry clones, and Research Association for Biotechnology, Japan and Helix Research Institute, Japan for providing FLJ cDNA clones. We thank K. Takeda for pertinent comments in preparing the manuscript.

Supporting Information Available: Supplementary Table 1, detailed data obtained from HTP-SPR in MS excel format; Supplementary Table 2, summary table of SFK phosphorylation; Supplementary Table 3, phosphorylation sites of FYN and HCK treated substrates; Supplementary Table 4, phosphorylation sites of commercially available or wheat germexpressed FYN treated substrates. Supplementary Figure 1, criteria of signal reproducibility and signal significance; Supplementary Figure 2, effect of heat denaturing of substrate protein on phosphorylation; Supplementary Figure 3, phylogenic tree of SFKs; Supplementary Figure 4, SDS-PAGE of SFKs; Supplementary Figure 5, MS/MS spectra of phosphopeptide identified in the present study. Error range, quality check of wheat germ expressed kinases, and denaturation of substrates are described in Supplementary Results and Discussion. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Johnson, S. A.; Hunter, T. Kinomics: methods for deciphering the kinome. Nat. Methods 2005, 2 (1), 17–25. (2) Olsen, J. V.; Blagoev, B.; Gnad, F.; Macek, B.; Kumar, C.; Mortensen, P.; Mann, M. Global in vivo, and site-specific phosphorylation dynamics in signaling networks. Cell 2006, 127 (3), 635–648. (3) Ptacek, J.; Snyder, M. Charging it up: global analysis of protein phosphorylation. Trends Genet 2006, 22 (10), 545–554. (4) Mitchell, P. A perspective on protein microarrays. Nat. Biotechnol. 2002, 20 (3), 225–229. (5) Kambhampati, D., Ed. Protein Microarray Technology: Wiley-VCH: Weinheim, 2004. (6) Ota, T.; Suzuki, Y.; Nishikawa, T.; Otsuki, T.; Sugiyama, T.; Irie, R.; Wakamatsu, A.; Hayashi, K.; Sato, H.; Nagai, K.; Kimura, K.; Makita, H.; Sekine, M.; Obayashi, M.; Nishi, T.; Shibahara, T.; Tanaka, T.; Ishii, S.; Yamamoto, J.; Saito, K.; Kawai, Y.; Isono, Y.; Nakamura, Y.; Nagahari, K.; Murakami, K.; Yasuda, T.; Iwayanagi, T.; Wagatsuma, M.; Shiratori, A.; Sudo, H.; Hosoiri, T.; Kaku, Y.; Kodaira, H.; Kondo, H.; Sugawara, M.; Takahashi, M.; Kanda, K.; Yokoi, T.; Furuya, T.; Kikkawa, E.; Omura, Y.; Abe, K.; Kamihara, K.; Katsuta, N.; Sato, K.; Tanikawa, M.; Yamazaki, M.; Ninomiya, K.; Ishibashi, T.; Yamashita, H.; Murakawa, K.; Fujimori, K.; Tanai, H.; Kimata, M.; Watanabe, M.; Hiraoka, S.; Chiba, Y.; Ishida, S.; Ono, Y.; Takiguchi, S.; Watanabe, S.; Yosida, M.; Hotuta, T.; Kusano, J.; Kanehori, K.; Takahashi-Fujii, A.; Hara, H.; Tanase, T. O.; Nomura, Y.; Togiya, S.; Komai, F.; Hara, R.; Takeuchi, K.; Arita, M.; Imose, N.; Musashino, K.; Yuuki, H.; Oshima, A.; Sasaki, N.; Aotsuka, S.; Yoshikawa, Y.; Matsunawa, H.; Ichihara, T.; Shiohata, N.; Sano, S.; Moriya, S.; Momiyama, H.; Satoh, N.; Takami, S.; Terashima, Y.; Suzuki, O.; Nakagawa, S.; Senoh, A.; Mizoguchi, H.; Goto, Y.; Shimizu, F.; Wakebe, H.; Hishigaki, H.; Watanabe, T.; Sugiyama, A.; Takemoto, M.; Kawakami, B.; Watanabe, K.; Kumagai, A.; Itakura, S.; Fukuzumi, Y.; Fujimori, Y.; Komiyama, M.; Tashiro, H.; Tanigami, A.; Fujiwara, T.; Ono, T.; Yamada, K.; Fujii, Y.; Ozaki, K.; Hirao, M.; Ohmori, Y.; Kawabata, A.; Hikiji, T.; Kobatake, N.; Inagaki, H.; Ikema, Y.; Okamoto, S.; Okitani, R.; Kawakami, T.; Noguchi, S.; Itoh, T.; Shigeta, K.; Senba, T.; Matsumura, K.; Nakajima, Y.; Mizuno, T.; Morinaga, M.; Sasaki, M.; Togashi, T.; Oyama, M.; Hata, H.; Komatsu, T.; Mizushima-Sugano, J.; Satoh, T.; Shirai, Y.; Takahashi, Y.; Nakagawa, K.; Okumura, K.; Nagase, T.; Nomura, N.; Kikuchi, H.; Masuho, Y.; Yamashita, R.; Nakai, K.; Yada, T.; Ohara, O.; Isogai, T.; Sugano, S. Complete sequencing and characterization of 21,243 full-length human cDNAs. Nat. Genet. 2004, 36 (1), 40–45. (7) Strausberg, R. L.; Feingold, E. A.; Grouse, L. H.; Derge, J. G.; Klausner, R. D.; Collins, F. S.; Wagner, L.; Shenmen, C. M.; Schuler, G. D.; Altschul, S. F.; Zeeberg, B.; Buetow, K. H.; Schaefer, C. F.; Bhat, N. K.; Hopkins, R. F.; Jordan, H.; Moore, T.; Max, S. I.; Wang, J.; Hsieh, F.; Diatchenko, L.; Marusina, K.; Farmer, A. A.; Rubin, G. M.; Hong, L.; Stapleton, M.; Soares, M. B.; Bonaldo, M. F.; Casavant, T. L.; Scheetz, T. E.; Brownstein, M. J.; Usdin, T. B.; Toshiyuki, S.; Carninci, P.; Prange, C.; Raha, S. S.; Loquellano, N. A.; Peters, G. J.; Abramson, R. D.; Mullahy, S. J.; Bosak, S. A.; McEwan,

5992

Journal of Proteome Research • Vol. 9, No. 11, 2010

(8) (9)

(10) (11)

(12)

(13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23)

(24) (25) (26)

P. J.; McKernan, K. J.; Malek, J. A.; Gunaratne, P. H.; Richards, S.; Worley, K. C.; Hale, S.; Garcia, A. M.; Gay, L. J.; Hulyk, S. W.; Villalon, D. K.; Muzny, D. M.; Sodergren, E. J.; Lu, X.; Gibbs, R. A.; Fahey, J.; Helton, E.; Ketteman, M.; Madan, A.; Rodrigues, S.; Sanchez, A.; Whiting, M.; Young, A. C.; Shevchenko, Y.; Bouffard, G. G.; Blakesley, R. W.; Touchman, J. W.; Green, E. D.; Dickson, M. C.; Rodriguez, A. C.; Grimwood, J.; Schmutz, J.; Myers, R. M.; Butterfield, Y. S.; Krzywinski, M. I.; Skalska, U.; Smailus, D. E.; Schnerch, A.; Schein, J. E.; Jones, S. J.; Marra, M. A. Generation and initial analysis of more than 15,000 full-length human and mouse cDNA sequences. Proc. Natl. Acad. Sci. U. S. A. 2002, 99 (26), 16899–16903. Sawasaki, T.; Ogasawara, T.; Morishita, R.; Endo, Y. A cell-free protein synthesis system for high-throughput proteomics. Proc. Natl. Acad. Sci. U. S. A. 2002, 99 (23), 14652–14657. Goshima, N.; Kawamura, Y.; Fukumoto, A.; Miura, A.; Honma, R.; Satoh, R.; Wakamatsu, A.; Yamamoto, J.; Kimura, K.; Nishikawa, T.; Andoh, T.; Iida, Y.; Ishikawa, K.; Ito, E.; Kagawa, N.; Kaminaga, C.; Kanehori, K.; Kawakami, B.; Kenmochi, K.; Kimura, R.; Kobayashi, M.; Kuroita, T.; Kuwayama, H.; Maruyama, Y.; Matsuo, K.; Minami, K.; Mitsubori, M.; Mori, M.; Morishita, R.; Murase, A.; Nishikawa, A.; Nishikawa, S.; Okamoto, T.; Sakagami, N.; Sakamoto, Y.; Sasaki, Y.; Seki, T.; Sono, S.; Sugiyama, A.; Sumiya, T.; Takayama, T.; Takayama, Y.; Takeda, H.; Togashi, T.; Yahata, K.; Yamada, H.; Yanagisawa, Y.; Endo, Y.; Imamoto, F.; Kisu, Y.; Tanaka, S.; Isogai, T.; Imai, J.; Watanabe, S.; Nomura, N. Human protein factory for converting the transcriptome into an in vitroexpressed proteome. Nat. Methods 2008, 5 (12), 1011–1017. Mann, M.; Jensen, O. N. Proteomic analysis of post-translational modifications. Nat. Biotechnol. 2003, 21 (3), 255–261. Ptacek, J.; Devgan, G.; Michaud, G.; Zhu, H.; Zhu, X.; Fasolo, J.; Guo, H.; Jona, G.; Breitkreutz, A.; Sopko, R.; McCartney, R. R.; Schmidt, M. C.; Rachidi, N.; Lee, S. J.; Mah, A. S.; Meng, L.; Stark, M. J.; Stern, D. F.; De Virgilio, C.; Tyers, M.; Andrews, B.; Gerstein, M.; Schweitzer, B.; Predki, P. F.; Snyder, M. Global analysis of protein phosphorylation in yeast. Nature 2005, 438 (7068), 679– 684. Takeda, H.; Fukumoto, A.; Miura, A.; Goshima, N.; Nomura, N. High-throughput kinase assay based on surface plasmon resonance suitable for native protein substrates. Anal. Biochem. 2006, 357 (2), 262–271. Takeda, H.; Goshima, N.; Nomura, N. High-throughput kinase assay based on surface plasmon resonance. Methods Mol. Biol. 2010, 627, 131–145. Machida, K.; Mayer, B. J.; Nollau, P. Profiling the global tyrosine phosphorylation state. Mol. Cell. Proteomics 2003, 2 (4), 215–233. Hunter, T. Tyrosine phosphorylation in cell signaling and disease. Keio J. Med. 2002, 51 (2), 61–71. Robinson, D. R.; Wu, Y. M.; Lin, S. F. The protein tyrosine kinase family of the human genome. Oncogene 2000, 19 (49), 5548–5557. Tatosyan, A. G.; Mizenina, O. A. Kinases of the Src family: structure and functions. Biochemistry (Mosc) 2000, 65 (1), 49–58. Manning, G.; Whyte, D. B.; Martinez, R.; Hunter, T.; Sudarsanam, S. The protein kinase complement of the human genome. Science 2002, 298 (5600), 1912–1934. Mitchell, P. J.; Barker, K. T.; Shipley, J.; Crompton, M. R. Characterisation and chromosome mapping of the human non receptor tyrosine kinase gene, brk. Oncogene 1997, 15 (12), 1497–1502. Serfas, M. S.; Tyner, A. L. Brk, Srm, Frk, and Src42A form a distinct family of intracellular Src-like tyrosine kinases. Oncol. Res. 2003, 13 (6-10), 409–419. Bose, R.; Holbert, M. A.; Pickin, K. A.; Cole, P. A. Protein tyrosine kinase-substrate interactions. Curr. Opin. Struct. Biol. 2006, 16 (6), 668–675. Songyang, Z.; Cantley, L. C. Recognition and specificity in protein tyrosine kinase-mediated signalling. Trends Biochem. Sci. 1995, 20 (11), 470–475. Maruyama, Y.; Wakamatsu, A.; Kawamura, Y.; Kimura, K.; Yamamoto, J.; Nishikawa, T.; Kisu, Y.; Sugano, S.; Goshima, N.; Isogai, T.; Nomura, N. Human Gene and Protein Database (HGPD): A novel database presenting a large quantity of experiment-based results in human proteomics. Nucleic Acids Res. 2009, 37 (Database issue), D762–D766. Rasmussen, M., Karypis, G. gCLUTO: an interactive clustering, visualization, and analysis system. UMN-CS TR-04-021. 2004. Blom, N.; Gammeltoft, S.; Brunak, S. Sequence and structure-based prediction of eukaryotic protein phosphorylation sites. J. Mol. Biol. 1999, 294 (5), 1351–1362. Blom, N.; Sicheritz-Ponten, T.; Gupta, R.; Gammeltoft, S.; Brunak, S. Prediction of post-translational glycosylation and phosphory-

research articles

Comparative Analysis of Human SFK Specificity in Vitro

(27)

(28)

(29)

(30)

(31)

(32)

(33) (34)

lation of proteins from the amino acid sequence. Proteomics 2004, 4 (6), 1633–1649. Peri, S.; Navarro, J. D.; Amanchy, R.; Kristiansen, T. Z.; Jonnalagadda, C. K.; Surendranath, V.; Niranjan, V.; Muthusamy, B.; Gandhi, T. K.; Gronborg, M.; Ibarrola, N.; Deshpande, N.; Shanker, K.; Shivashankar, H. N.; Rashmi, B. P.; Ramya, M. A.; Zhao, Z.; Chandrika, K. N.; Padma, N.; Harsha, H. C.; Yatish, A. J.; Kavitha, M. P.; Menezes, M.; Choudhury, D. R.; Suresh, S.; Ghosh, N.; Saravana, R.; Chandran, S.; Krishna, S.; Joy, M.; Anand, S. K.; Madavan, V.; Joseph, A.; Wong, G. W.; Schiemann, W. P.; Constantinescu, S. N.; Huang, L.; Khosravi-Far, R.; Steen, H.; Tewari, M.; Ghaffari, S.; Blobe, G. C.; Dang, C. V.; Garcia, J. G.; Pevsner, J.; Jensen, O. N.; Roepstorff, P.; Deshpande, K. S.; Chinnaiyan, A. M.; Hamosh, A.; Chakravarti, A.; Pandey, A. Development of human protein reference database as an initial platform for approaching systems biology in humans. Genome Res. 2003, 13 (10), 2363–2371. Diella, F.; Cameron, S.; Gemund, C.; Linding, R.; Via, A.; Kuster, B.; Sicheritz-Ponten, T.; Blom, N.; Gibson, T. J. Phospho.ELM: a database of experimentally verified phosphorylation sites in eukaryotic proteins. BMC Bioinformatics 2004, 5, 79. Ogasawara, O.; Otsuji, M.; Watanabe, K.; Iizuka, T.; Tamura, T.; Hishiki, T.; Kawamoto, S.; Okubo, K. BodyMap-Xs: anatomical breakdown of 17 million animal ESTs for cross-species comparison of gene expression. Nucleic Acids Res. 2006, 34 (Database issue), D628–D631. Mehrle, A.; Rosenfelder, H.; Schupp, I.; del Val, C.; Arlt, D.; Hahne, F.; Bechtel, S.; Simpson, J.; Hofmann, O.; Hide, W.; Glatting, K. H.; Huber, W.; Pepperkok, R.; Poustka, A.; Wiemann, S. The LIFEdb database in 2006. Nucleic Acids Res. 2006, 34 (Database issue), D415–D418. Matsumoto, M.; Oyamada, K.; Takahashi, H.; Sato, T.; Hatakeyama, S.; Nakayama, K. I. Large-scale proteomic analysis of tyrosinephosphorylation induced by T-cell receptor or B-cell receptor activation reveals new signaling pathways. Proteomics 2009, 9 (13), 3549–3563. Nagai, T.; Ibata, K.; Park, E. S.; Kubota, M.; Mikoshiba, K.; Miyawaki, A. A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat. Biotechnol. 2002, 20 (1), 87–90. Feller, S. M.; Ren, R.; Hanafusa, H.; Baltimore, D. SH2 and SH3 domains as molecular adhesives: the interactions of Crk and Abl. Trends Biochem. Sci. 1994, 19 (11), 453–458. Carpino, N.; Wisniewski, D.; Strife, A.; Marshak, D.; Kobayashi, R.; Stillman, B.; Clarkson, B. p62(dok): a constitutively tyrosine-

(35) (36) (37) (38)

(39) (40)

(41) (42) (43) (44)

(45) (46) (47)

phosphorylated, GAP-associated protein in chronic myelogenous leukemia progenitor cells. Cell 1997, 88 (2), 197–204. Eisenberg, E.; Greene, L. E. Multiple roles of auxilin and hsc70 in clathrin-mediated endocytosis. Traffic 2007, 8 (6), 640–646. Coulombe, P.; Meloche, S. Atypical mitogen-activated protein kinases: structure, regulation and functions. Biochim. Biophys. Acta 2007, 1773 (8), 1376–1387. Grossel, M. J.; Hinds, P. W. From cell cycle to differentiation: an expanding role for cdk6. Cell Cycle 2006, 5 (3), 266–270. Keegan, K.; Cooper, J. A. Use of the two hybrid system to detect the association of the protein-tyrosine-phosphatase, SHPTP2, with another SH2-containing protein, Grb7. Oncogene 1996, 12 (7), 1537–1544. Mitra, S.; Beach, C.; Feng, G. S.; Plattner, R. SHP-2 is a novel target of Abl kinases during cell proliferation. J. Cell Sci. 2008, 121 (Pt 20), 3335–3346. Voena, C.; Conte, C.; Ambrogio, C.; Boeri Erba, E.; Boccalatte, F.; Mohammed, S.; Jensen, O. N.; Palestro, G.; Inghirami, G.; Chiarle, R. The tyrosine phosphatase Shp2 interacts with NPM-ALK and regulates anaplastic lymphoma cell growth and migration. Cancer Res. 2007, 67 (9), 4278–4286. Harvey, A. J.; Crompton, M. R. The Brk protein tyrosine kinase as a therapeutic target in cancer: opportunities and challenges. Anticancer Drugs 2004, 15 (2), 107–111. Derry, J. J.; Prins, G. S.; Ray, V.; Tyner, A. L. Altered localization and activity of the intracellular tyrosine kinase BRK/Sik in prostate tumor cells. Oncogene 2003, 22 (27), 4212–4220. Denis, G. V.; Vaziri, C.; Guo, N.; Faller, D. V. RING3 kinase transactivates promoters of cell cycle regulatory genes through E2F. Cell Growth Differ. 2000, 11 (8), 417–424. Donella-Deana, A.; Cesaro, L.; Sarno, S.; Ruzzene, M.; Brunati, A. M.; Marin, O.; Vilk, G.; Doherty-Kirby, A.; Lajoie, G.; Litchfield, D. W.; Pinna, L. A. Tyrosine phosphorylation of protein kinase CK2 by Src-related tyrosine kinases correlates with increased catalytic activity. Biochem. J. 2003, 372 (Pt 3), 841–849. Ubersax, J. A.; Ferrell, J. E., Jr. Mechanisms of specificity in protein phosphorylation. Nat. Rev. Mol. Cell. Biol. 2007, 8 (7), 530–541. Davis, R. J. The mitogen-activated protein kinase signal transduction pathway. J. Biol. Chem. 1993, 268 (20), 14553–14556. Sondhi, D.; Xu, W.; Songyang, Z.; Eck, M. J.; Cole, P. A. Peptide and protein phosphorylation by protein tyrosine kinase Csk: insights into specificity and mechanism. Biochemistry 1998, 37 (1), 165–172.

PR100773T

Journal of Proteome Research • Vol. 9, No. 11, 2010 5993